Page 1

Table of Contents Agenda Speaker Bios Congressional Budget Office Updated Budget Projections: Fiscal Years 2013 to 2023

Medicare Select Provisions of Various Medicare Reform Proposals Academy Publications A Guide to Analyzing the Issues: Medicare Premium Support (October 2012) A Guide to Analyzing the Issues: What Voters Should Know About Medicare (April 2012) An Actuarial Perspective on Proposals to Improve Medicare’s Financial Condition (May 2011) Letter on Medicare to Joint Select Committee on Deficit Reduction (August 2011) Medicare’s Financial Condition: Beyond Actuarial Balance (June 2013) Revising Medicare’s Fee-For-Service Benefit Structure (March 2012)


National Flood Insurance Program Select Provisions of Various National Flood Insurance Reform Proposals Academy Publications Letter on Proposal to Reform the National Flood Insurance Program (June 2012) The National Flood Insurance Program: Past, Present…and Future? (July 2011)

Social Security Select Provisions of Various Social Security Reform Proposals Academy Publications A Guide to Analyzing Social Security Reform (2012) Significance of the Social Security Trust Funds (May 2012) Testimony on Proposed Adjustments to Social Security Benefits (May 2013) Understanding the Assumptions Used to Evaluate Social Security’s Financial Condition (May 2012)

Additional Topics of Interest Academy Publications Actuarial Soundness Special Report (May 2012) Letter to Federal Insurance Office on Current State of the Market for Natural Catastrophe Insurance (June 2013) Letter to President’s Working Group on Financial Markets in Response to Request for Analysis of Terrorism Risk Insurance (August 2010)


Agenda 8:00–9:00 Oriental Ballroom B&C

Registration and Breakfast

9:00–9:05 Oriental Ballroom B&C

Welcome: Remarks–President-Elect Tom Terry Introduction of Keynote Speaker

9:05–9:50 Oriental Ballroom B&C

9:50 –10:30 Oriental Ballroom B&C

10:30–10:45 10:45–12:00 Oriental B&C, Hirshorn, Portrait, Sackler 12:00–12:30 Oriental Ballroom B&C 12:30–1:00 Oriental Ballroom B&C

1:00–1:45 Oriental Ballroom B&C

Keynote Address: Hon. Kent Conrad / Q&A Former U.S. Senator, ND–Chair of Senate Budget Committee The Public Interest: Bringing the Profession’s Focus to Bear on the Federal Budget Process as it affects key Direct Spending Programs Where Are We Now: Review of Academy’s Work/Approach to Social Security Reform, Medicare Reform, Medicaid, National Flood Insurance Break Cross-Practice Breakout Sessions (4 Breakout Groups) Breakout Groups Report Back to Full Group Buffet Lunch Introduction of Joyce Manchester, Congressional Budget Office, Chief of the Long-Term Analysis Unit–Health, Retirement, and Long-Term Analysis Division Remarks: Joyce Manchester / Q&A

1:45–2:45 Oriental Ballroom B&C 2:45–3:30 Oriental Ballroom B&C

Plenary Discussion Facilitated by Tom Terry Introduction of U.S. Rep. Jim Cooper (D-TN) Remarks: Rep. Jim Cooper

3:30–4:00 Oriental Ballroom B&C 1

Closing Discussion; Adjourn Summer Summit

2013 Summer Summit Agenda | American Academy of Actuaries


SPEAKERS Senator Kent Conrad Independent Director of Genworth Financial, Inc. Former Chairman of the Senate Budget Committee Former U.S. Senator (D-ND) As former chairman of the Senate Budget Committee and new independent director of Genworth Financial Inc., Senator Kent Conrad (D-ND) has earned bipartisan respect as an expert on the nation’s fiscal issues. Selected by TIME as one of “America’s 10 Best Senators” and rated by The American as one of the ten most economically knowledgeable members of Congress, Conrad has played a vital role in shaping our nation’s fiscal policies during the last three decades. During his retirement announcement, Conrad stated, “It was more important that ‘I spend my time and energy trying to focus on solving the nation’s budget woes than be distracted by another campaign.’ ” Exclusively represented by Leading Authorities speakers bureau, Senator Kent Conrad addresses fiscal policy, the fiscal cliff, the economy, energy, and healthcare. In addition to being the chairman of the Budget Committee, Conrad was a member of the National Commission on Fiscal Responsibility and Reform and a strong proponent of the Simpson-Bowles plan. According to the Washington Post, “Conrad almost singlehandedly forced President Obama to create the commission known as Bowles-Simpson, which produced a debt-reduction plan now hailed as a model of bipartisan compromise on taxes and government spending.” Since its release in December 2010, that plan has become the standard against which all other debt-reduction proposals are measured. Self-described as a “deficit hawk” because of his calls for a balanced federal budget, Conrad also served on the six-member bipartisan group, the so-called “Gang of Six.”

1

2013 Summer Summit Speakers | American Academy of Actuaries


Congressman Jim Cooper Congressman Jim Cooper was born and raised in Tennessee. He and Martha, his wife of twenty eight years, live in Nashville and have three children. A New York Times columnist recently called him "the House's conscience, a lonely voice for civility in this ugly era" and a "tarttongued moderate" who "seeks bipartisanship on fiscal matters and other issues in a polarized political climate." USA Today named him one of the "Brave 38" of a "tiny band of heroes" in Congress for his work on a bipartisan budget plan. In Congress, he's known for his work on the federal budget, health care and government reform. He's also a businessman, attorney and part-time Vanderbilt professor when Congress is not in session. Joyce Manchester Chief of the Long-Term Analysis Unit, Congressional Budget Office Joyce Manchester is Chief of the Long-Term Analysis Unit at the Congressional Budget Office. Her group is responsible for longterm projections of Social Security, Medicare, Medicaid, and the federal budget as well as long-term analysis of reforms in those areas. Her research has examined the work behavior of Disability Insurance applicants, responses in claiming behavior following changes in Social Security policy, and long-term earnings losses among job losers. She previously held positions in the Office of Policy at the Social Security Administration, the Social Security Advisory Board, the Macroeconomics Analysis Division at CBO, and the Economics Department at Dartmouth College. She has a B.A. in economics and mathematics from Wesleyan University and a Ph.D. in economics from Harvard.

2

2013 Summer Summit Speakers | American Academy of Actuaries


CONGRESS OF THE UNITED STATES CONGRESSIONAL BUDGET OFFICE

CBO Updated Budget Projections: Fiscal Years 2013 to 2023 Percentage of GDP 28

Actual

26

Average Outlays, 1973 to 2012 (21.0%)

Outlays

24

Projected

Total Revenues and Outlays

22 20 18

Revenues

16

Average Revenues, 1973 to 2012 (17.9%)

14 0 1973

1978

1983

1988

1993

1998

2003

2008

2013

2018

2023

Percentage of GDP 120

Actual

Projected

100 80

Federal Debt Held by the Public

60 40 20 0 1940

1950

1960

1970

MAY 2013

1980

1990

2000

2010

2020


Notes Unless otherwise indicated, years referred to in this report are federal fiscal years (which run from October 1 to September 30). Numbers in the text and tables may not add up to totals because of rounding. Supplemental data for this analysis are available on CBO’s website (www.cbo.gov).

CBO

Pub. No. 4722


Contents Overview

1

CBO’s Baseline Budget Projections

1

Revenues

2

Outlays

2

Changes in CBO’s Baseline Projections Since February 2013

5

Revenues

5

Outlays

6

About This Document

Tables 1. CBO’s Baseline Budget Projections

18

8

2. Mandatory Outlays Projected in CBO’s Baseline

10

3. Discretionary Budget Authority for Fiscal Year 2013

13

4. Discretionary Spending Projected in CBO’s Baseline

14

5. Federal Debt Projected in CBO’s Baseline

16

6. Changes in CBO’s Baseline Projections of the Deficit Since February 2013

17

Figures 1. Total Revenues and Outlays 2. Federal Debt Held by the Public 3. Projected Spending for Major Budget Categories

9 9 12

CBO


Updated Budget Projections: Fiscal Years 2013 to 2023

Overview If the current laws that govern federal taxes and spending do not change, the budget deficit will shrink this year to $642 billion, the Congressional Budget Office (CBO) estimates, the smallest shortfall since 2008. Relative to the size of the economy, the deficit this year—at 4.0 percent of gross domestic product (GDP)—will be less than half as large as the shortfall in 2009, which was 10.1 percent of GDP. Because revenues, under current law, are projected to rise more rapidly than spending in the next two years, deficits in CBO’s baseline projections continue to shrink, falling to 2.1 percent of GDP by 2015 (see Table 1 on page 8). However, budget shortfalls are projected to increase later in the coming decade, reaching 3.5 percent of GDP in 2023, because of the pressures of an aging population, rising health care costs, an expansion of federal subsidies for health insurance, and growing interest payments on federal debt. By comparison, the deficit averaged 3.1 percent of GDP over the past 40 years and 2.4 percent in the 40 years before fiscal year 2008, when the most recent recession began. During the next 10 years, both revenues and outlays are projected to be above their 40-year averages as a percentage of GDP (see Figure 1 on page 9). For the 2014–2023 period, deficits in CBO’s baseline projections total $6.3 trillion. With such deficits, federal debt held by the public is projected to remain above 70 percent of GDP—far higher than the 39 percent average seen over the past four decades. (As recently as the end of 2007, federal debt equaled 36 percent of GDP.) Under current law, the debt is projected to decline from about 76 percent of GDP in 2014 to slightly below 71 percent in 2018 but then to start rising again; by 2023, if current laws remain in place, debt will equal 74 percent of GDP and continue to be on an upward path (see Figure 2 on page 9).

Such high and rising debt later in the coming decade would have serious negative consequences: When interest rates return to higher (more typical) levels, federal spending on interest payments would increase substantially. Moreover, because federal borrowing reduces national saving, over time the capital stock would be smaller and total wages would be lower than they would be if the debt was reduced. In addition, lawmakers would have less flexibility than they would have if debt levels were lower to use tax and spending policy to respond to unexpected challenges. Finally, a large debt increases the risk of a fiscal crisis, during which investors would lose so much confidence in the government’s ability to manage its budget that the government would be unable to borrow at affordable rates. CBO’s estimate of the deficit for this year is about $200 billion below the estimate that it produced in February 2013, mostly as a result of higher-than-expected revenues and an increase in payments to the Treasury by Fannie Mae and Freddie Mac.1 For the 2014–2023 period, CBO now projects a cumulative deficit that is $618 billion less than it projected in February. That reduction results mostly from lower projections of spending for Social Security, Medicare, Medicaid, and interest on the public debt.

CBO’s Baseline Budget Projections CBO’s baseline projections are not meant to be a forecast of future outcomes. CBO constructs its baseline in accordance with provisions set forth in the Balanced Budget and Emergency Deficit Control Act of 1985 and the Congressional Budget and Impoundment Control Act of 1. CBO’s previous projections were published in Congressional Budget Office, The Budget and Economic Outlook: Fiscal Years 2013 to 2023 (February 2013), www.cbo.gov/publication/43907.

CBO


2

UPDATED BUDGET PROJECTIONS: FISCAL YEARS 2013 TO 2023

1974. As specified in law, CBO’s baseline projections are constructed under the assumption that current laws generally remain unchanged; in that way, they can serve as a benchmark against which potential changes in law can be measured. However, even if federal laws remained unchanged for the next decade, actual budgetary outcomes would differ from CBO’s baseline projections, perhaps significantly, because of unanticipated changes in economic conditions and other factors that would affect federal revenues and spending.

Revenues Under current law, revenues are expected to increase by 15 percent in 2013, about 4 percentage points more than CBO projected in February and substantially more than the expected growth of about 3 percent in nominal GDP this year. As a result, revenues in CBO’s baseline will climb from 15.8 percent of GDP in 2012 to 17.5 percent in 2013, which is slightly below their average of 17.9 percent of GDP over the past 40 years. Revenues have increased robustly so far this year in part because of the expiration of the 2 percentage-point payroll tax cut in January 2013. In addition, receipts of individual income taxes have been boosted by three factors:  Beginning in January, tax rates on personal income above certain thresholds went up;  Some high-income taxpayers realized more income late in calendar year 2012 in anticipation of changes in tax law and therefore paid taxes on that income in fiscal year 2013; and  Personal income rose for reasons not related to the changes in tax provisions.

CBO also attributes some of the growth in revenues this year to increases in the average tax rate on domestic economic profits, which boosted receipts from corporate income taxes.2 CBO projects that revenues will increase further under current law, to 18.3 percent of GDP in 2014 and 19.3 percent in 2015, and then remain near 19 percent of 2. The average tax rate is the ratio of corporate income taxes paid to domestic economic profits. An increase in that measure typically occurs because taxable corporate profits increase faster than domestic economic profits. Domestic economic profits do not take into account certain factors that affect corporate income taxes, such as deductions of bad debts, income from firms’ capital gains realizations, and deductions for accelerated depreciation.

CBO

MAY 2013

GDP from 2016 through 2023. About half of the expected increase in revenues over the next two years stems from changes in tax rules, such as the scheduled expiration at the end of December 2013 of enhanced depreciation deductions allowed for certain business investments. Accounting for the other half of the expected increase in revenues are factors related mainly to the strengthening economy, including increases relative to GDP in some components of taxable income (such as wages and salaries, capital gains realizations, proprietors’ income, and domestic economic profits) and a continued rise in the average tax rate on domestic economic profits to more normal levels. CBO projects that revenues will grow at close to the same rate as GDP after 2015. Individual income tax receipts are projected to rise relative to GDP as increases in taxpayers’ real (inflation-adjusted) income push more income into higher tax brackets; in contrast, corporate income tax receipts and remittances to the U.S. Treasury from the Federal Reserve are projected to fall relative to GDP.

Outlays The Deficit Control Act requires that CBO make its projections for most mandatory programs under the assumption that current laws continue unchanged. For that reason, CBO’s baseline projections of mandatory spending reflect the automatic spending reductions required by the Budget Control Act of 2011 (Public Law 112-25) and expected changes in the economy, demographics, and other factors. For discretionary spending, CBO’s baseline incorporates the caps put in place by the Budget Control Act and accounts for further reductions in those caps under the law’s automatic enforcement procedures. On that basis, total outlays are projected to decline slightly relative to GDP through 2015 and then to rise in most years through 2023, reaching 22.6 percent of GDP—above the 21.0 percent of GDP that has been the average for the past 40 years. Mandatory Spending. CBO projects that mandatory spending (net of offsetting receipts, which are certain collections that are treated as reductions in outlays) will increase from $2.2 trillion in 2014 to $3.6 trillion by 2023 (see Table 2 on page 10). Between 2014 and 2021, mandatory spending is projected to remain about the same as a share of the economy—between 13.1 percent and 13.5 percent. But under current law, mandatory spending will accelerate in the final two years of the projection period, reaching 14.0 percent of GDP in


MAY 2013

2022, CBO estimates.3 During the past 40 years, mandatory spending has averaged 10.2 percent of GDP. Most mandatory spending consists of outlays for Social Security and the federal government’s major health care programs.4 By 2023, net outlays for those components of mandatory spending will total $3.0 trillion, CBO projects, and will account for more than half of all federal spending, growing from 10.1 percent of GDP in 2014 to 11.6 percent in 2023 (see Figure 3 on page 12). In CBO’s baseline projections:  Spending for Social Security and Medicare is projected to remain relatively stable as a share of GDP over most of the projection period. That spending begins to rise relative to GDP in the final few years of the period, however, with Social Security outlays reaching 5.5 percent and net Medicare outlays totaling 3.5 percent by 2023, compared with 4.9 percent and 3.0 percent, respectively, in 2012.  Federal outlays for Medicaid rise steadily as a share of GDP over the next 10 years, from 1.8 percent in 2014 to 2.1 percent in 2023, by CBO’s estimate. That rise is attributable in part to expected increases in the cost of Medicaid benefits per beneficiary and in part to the fact that many states are expected to expand Medicaid coverage significantly under provisions of the Affordable Care Act.5  Spending on subsidies that will help people purchase health insurance through exchanges (which will 3. Under CBO’s baseline, mandatory outlays are projected to remain steady at 14.0 percent of GDP in 2022 and 2023. That result stems largely from a shift in the timing of certain payments. Because both October 1, 2022, and October 1, 2023, fall on weekends, certain payments that are due on those days will instead be made at the end of September, thus shifting them into the previous fiscal year. Without that timing shift, mandatory outlays would increase from 13.7 percent of GDP in 2022 to 13.9 percent in 2023 under CBO’s baseline. 4. Major health care programs include Medicare, Medicaid, the Children’s Health Insurance Program, and subsidies for the purchase of health insurance through newly established exchanges (and related spending). 5. The Affordable Care Act comprises the Patient Protection and Affordable Care Act (P.L. 111-148) and the health care provisions of the Health Care and Education Reconciliation Act of 2010 (P.L. 111-152) and, in the case of this document, the effects of subsequent related judicial decisions, statutory changes, and administrative actions.

UPDATED BUDGET PROJECTIONS: FISCAL YEARS 2013 TO 2023

3

become available starting in 2014 for individuals and families who meet income and other eligibility criteria), along with related spending, is projected to increase from 0.1 percent of GDP in 2014 to 0.5 percent 10 years from now.6 Discretionary Spending. For 2013, CBO estimates that $1.15 trillion in budget authority has been provided for discretionary programs (see Table 3 on page 13). That total comprises the $1.21 trillion provided in appropriation acts reduced by $64 billion to reflect the sequestration (cancellation of budgetary resources) that was ordered by the Office of Management and Budget under the provisions of the Budget Control Act.7 Following the rules governing the baseline, CBO projects that discretionary budget authority will total $1.11 trillion in 2014, 3 percent less than the funding estimated for 2013 (see Table 4 on page 14). That drop occurs mainly because the automatic spending reductions in 2014—which will lower the caps on discretionary spending for that year—will be larger than those in 2013. In CBO’s baseline, appropriations for programs not constrained by those caps—overseas contingency operations, activities receiving emergency designations, disaster relief, and spending designed to enhance program integrity by reducing overpayments in certain benefit programs— are assumed to grow with inflation from the amounts provided in 2013. With that combination of compliance with the funding caps for “regular” appropriations and inflation adjustments for programs that are unconstrained by the caps, discretionary budget authority would rise by about 2 percent a year, on average, from 2014 to 2023 under CBO’s baseline assumptions. Discretionary outlays in CBO’s baseline are projected to fall from $1.29 trillion in 2012 to $1.21 trillion in 2013, a drop of almost 6 percent. In the baseline, those outlays 6. The subsidies to be provided through new health insurance exchanges will also reduce revenues. Taking the spending and revenue portions together, the total cost of the exchange subsidies and related spending is projected to be 0.6 percent of GDP in 2023. 7. On March 1, 2013, the Administration ordered a sequestration of both discretionary and mandatory budgetary resources for fiscal year 2013; see Office of Management and Budget, OMB Report to the Congress on the Joint Committee Sequestration for Fiscal Year 2013 (March 1, 2013, http://go.usa.gov/TMKz (pdf, 1.3 MB). After that report was released, final appropriations for this fiscal year were enacted, and the amount of the sequestration was reduced for certain accounts by a total of nearly $5 billion.

CBO


4

UPDATED BUDGET PROJECTIONS: FISCAL YEARS 2013 TO 2023

fall further in 2014—to $1.17 trillion (a reduction of another 4 percent). By 2023, such outlays would total $1.41 trillion under CBO’s baseline assumptions. Measured as a share of GDP, discretionary outlays are projected to drop from 8.3 percent in 2012 to 5.5 percent in 2023; over the past 40 years, such outlays have averaged 8.6 percent. Interest and Debt. The increase in debt, along with an anticipated substantial rise in interest rates as the economy strengthens, is expected to sharply boost interest payments. CBO projects that, under current law, the government’s net interest spending will more than double as a share of GDP in the coming decade—from 1.4 percent in 2014 to 3.2 percent in 2023, a percentage that has been exceeded only once in the past 50 years. Debt held by the public consists mostly of securities that the Treasury issues to raise cash to fund the federal government’s activities and to pay off its maturing liabilities. The amount the Treasury borrows by selling securities (net of the amount of maturing securities that it redeems) is influenced primarily by the annual budget deficit. In addition, the Treasury needs to borrow to provide financing for student loans and other credit programs; CBO projects that such additional borrowing, often called other means of financing, will average $69 billion per year during the 2014–2023 period (see Table 5 on page 16). After accounting for all of the government’s borrowing needs, CBO projects that, under current law, debt held by the public will increase from 73 percent of GDP at the end of fiscal year 2012 to 75 percent at the end of this year and 76 percent at the end of 2014. Under the assumptions that govern CBO’s baseline, debt will fall to a low of 71 percent of GDP in 2018 and then rise for the remainder of the projection period, measuring 74 percent of GDP at the end of 2023. Lawmakers have suspended the previous limit on the amount of debt that the Treasury can issue to the public and government accounts through May 18, 2013. On May 19, the debt limit will be raised to its previous value—$16.394 trillion—plus the amount of borrowing that occurred while the limit was suspended (that is, from early February to May 18). If no further action is taken before May 19, the Treasury will resort to what are known as extraordinary measures for managing cash and borrowing to allow the government to continue operating

CBO

MAY 2013

normally. To avoid defaulting on the federal government’s obligations, the debt ceiling will need to be raised before those extraordinary measures are exhausted, most likely in October or November. Alternative Assumptions About Fiscal Policy. Alternative fiscal policies would lead to different budget outcomes than those indicated by the baseline. For example, the baseline projections of spending for overseas contingency operations are derived from an extrapolation of the amount of funding provided for such operations this year. In coming years, the funding required for overseas contingency operations—in Afghanistan or other countries—may be smaller than the amounts in the baseline if the number of deployed troops and the pace of operations diminish over time.8 Such lower funding would reduce outlays relative to CBO’s baseline projections. Likewise, CBO’s baseline includes an extrapolation of $39 billion in funding (including the effects of sequestration) declared an emergency requirement in response to Hurricane Sandy; removing that extrapolation would reduce discretionary outlays by $291 billion over the 2014–2023 period relative to CBO’s baseline projections. In the other direction, if the automatic spending reductions in place through 2021 were reversed, in whole or in part, projected outlays would be higher than the amounts shown in the baseline. Similarly, if the Congress and the President decided to extend tax provisions that are scheduled to expire over the next decade without making offsetting changes in other tax policies, revenues would be lower than those in the baseline. The expiring tax provisions include one that allows an immediate deduction for businesses of 50 percent of the cost of new investments in equipment and also include certain aspects of refundable tax credits. In recent years, CBO has presented an alternative fiscal scenario that illustrated the impact on projected deficits and debt of maintaining policies that were then in place but that were scheduled to change under then-current law. Many components of the alternative fiscal scenario (including many with the largest budgetary effects) have now been made permanent. The remaining components consist of holding constant Medicare’s payment rates for 8. In CBO’s baseline, budget authority for overseas contingency operations in 2013 totals $93 billion (including the effects of sequestration). Outlays projected in the baseline for such operations total $999 billion over the 2014–2023 period.


MAY 2013

physicians’ services (now scheduled to be reduced in January 2014), undoing the automatic spending reductions scheduled for 2014 through 2021, and extending certain expiring tax provisions. If lawmakers were to make those changes to current law, and if other changes in policies with offsetting effects on budget deficits were not enacted, deficits and debt would be higher than the amounts shown in CBO’s current baseline. Relative to the baseline projections for 2014 to 2023, deficits would rise by a total of $2.4 trillion (including debt-service costs) to yield cumulative deficits of $8.8 trillion. Debt held by the public would reach 83 percent of GDP by the end of 2023, the largest share since 1948.

Changes in CBO’s Baseline Projections Since February 2013 The deficit that CBO now estimates for 2013, in the absence of further changes to tax and spending laws, is $203 billion smaller than the $845 billion figure projected in February. CBO’s new baseline projections for the 2014–2023 period show cumulative deficits that are $618 billion less than the 10-year shortfall of $7.0 trillion the agency projected in February. Those revisions are all categorized as technical changes—revisions that are made for reasons other than updated economic information or the enactment of new laws. For example, changes in estimates of the rate at which a program will obligate and spend funds or adjustments to projected participation rates for benefit programs are shown as technical changes. CBO’s updated budget projections use the same economic projections that the agency had developed for the February baseline, and legislation enacted since February has not had a significant effect on the budget totals.9 The decline in the projected deficit for 2013 stems largely from a boost in estimated revenues as well as from expected payments to the Treasury by Fannie Mae and Freddie Mac. The changes in CBO’s baseline for the 2014–2023 period result mainly from lower projections of outlays (primarily for Social Security, Medicare, 9. The new baseline encompasses the effects of the Consolidated and Further Continuing Appropriations Act, 2013 (P.L. 113-6), which was enacted after the release of CBO’s February baseline. However, the February baseline had already incorporated assumptions about aggregate amounts of discretionary spending that were very similar to the amounts resulting from the law, and the small remaining differences were difficult to distinguish from the effects of sequestration.

UPDATED BUDGET PROJECTIONS: FISCAL YEARS 2013 TO 2023

5

and Medicaid). As a result of those changes, CBO now projects that, under current law, debt held by the public will total 74 percent of GDP in 2023, down from the 77 percent the agency projected in February.

Revenues CBO has raised its revenue projections by $105 billion (or about 4 percent) for 2013 and by $95 billion (or 0.2 percent) for the following 10 years (see Table 6 on page 17). Those increases stem primarily from higherthan-expected collections of individual and corporate income taxes this year. CBO believes that the causes of the higher collections are largely temporary, so it has increased projected revenues in subsequent years by much smaller amounts. Receipts from individual income taxes this year are now estimated to be $69 billion (or 5 percent) higher than CBO anticipated in February. The higher-than-expected collections mainly represent amounts that were paid with income tax filings in April 2013. Although the specific reasons for the added collections will not be known until detailed information from tax returns becomes available over the next year, the unexpected payments probably reflect higher taxable income in 2012, in part because higher-income taxpayers realized more income than CBO expected in anticipation of increased tax rates for tax year 2013. CBO expects that taxable personal income will revert to its historical level relative to GDP in coming years. CBO has raised its estimate of corporate tax receipts in 2013 by $40 billion (or 16 percent). Those collections reflect tax liabilities from 2012 and 2013. As is the case with the collections of individual income taxes, the sources of the recent payments of corporate income taxes will not be known until information from tax returns becomes available in the future. CBO projected in February that the average tax rate on domestic economic profits would increase over the next several years to a level more consistent with the historical average. The recent strength in corporate tax receipts probably represents a faster-than-expected reversion, rather than a long-term change in the average tax rate. Hence, CBO has boosted its estimate of corporate tax receipts in 2014 and 2015 by smaller amounts and left its projections for subsequent years unchanged. Changes in estimated revenues associated with the health insurance coverage provisions of the Affordable Care Act

CBO


6

UPDATED BUDGET PROJECTIONS: FISCAL YEARS 2013 TO 2023

increased projected revenues by about $24 billion between 2014 and 2023. That change is the net result of two partially offsetting revisions—an increase of $138 billion in estimated revenues because smaller subsidies in the form of tax reductions are now expected to be provided for the purchase of insurance through the new exchanges and a decrease of $114 billion in estimated revenues resulting from other technical changes.10

Outlays Updates since February have reduced estimated outlays in the current year and the following 10-year period by $98 billion and $522 billion, respectively. The reduction over the 10-year period accounts for about 1 percent of total outlays. Most of the change in the current year results from a $95 billion increase in estimated payments from Fannie Mae and Freddie Mac (which would be recorded in the budget as offsetting receipts). The companies are required to make quarterly payments to the Treasury in amounts related to their net worth; CBO increased its estimates of those payments for 2013 after certain accounting changes significantly raised the entities’ estimated net worth for this year. Changes over the 2014–2023 period are almost entirely composed of reductions in projected outlays for three mandatory programs—Social Security, Medicare, and Medicaid—and for net interest costs. Social Security. CBO has decreased its projection of outlays for Social Security over the 2014–2023 period by $86 billion (or 1 percent); three-fourths of the reduction occurred in the Disability Insurance program. On the basis of recent months’ application and award rates and 10. The reduction in exchange subsidies that take the form of tax reductions arises from two main factors: One is that more people are projected to obtain coverage through Medicaid and fewer through the exchanges than previously estimated (see the Medicaid discussion below), which lowers the projected total amount of exchange subsidies; the other is that a larger share of the exchange subsidies is projected to take the form of outlays, which further lowers projected exchange subsidies that take the form of tax reductions. The other technical changes are a $58 billion reduction in estimated revenues owing to the excise tax on high-premium insurance plans (from incorporating more recent data on the premiums firms pay) and a $56 billion reduction in projected revenues stemming mostly from a shift in the expected mix of taxable and nontaxable compensation.

CBO

MAY 2013

average payments, CBO expects that fewer individuals will be newly awarded Disability Insurance benefits through 2015 and that average award amounts and retroactive benefits will be lower than previously projected for those years. However, for the final five years of the baseline period, CBO anticipates that the program will make more awards to new beneficiaries and that the reduction in average retroactive benefits will begin to taper off. In addition, because of more recent information on benefit payments for Old-Age and Survivors Insurance, CBO expects that the number of people receiving benefits under that program will be slightly smaller than previously estimated, lowering benefit payments in all years of the projection period; that reduction was partially offset by a slight increase in the amount of the average expected benefit. Medicare. CBO’s current projection of net mandatory spending for Medicare is $85 billion (or 1.2 percent) lower over the 2014–2023 period than it projected in February 2013. The major components of that change are a reduction of $143 billion in projected gross spending for benefits, partially offset by reductions of $48 billion in collections of offsetting receipts and $10 billion in Medicare savings as a result of sequestration. (Those changes in projected Medicare savings from sequestration are largely offset by compensating changes in estimated spending for other programs.) In particular, additional data on spending in 2013 caused CBO to reduce projected spending by about 1 percent for Medicare benefits this year and over the 2014–2023 period. Compared with amounts in the February 2013 baseline, estimated spending for Medicare benefits over the 2014– 2023 period is lower for all major components of the program—Part A (hospital insurance), Part B (medical insurance), and Part D (outpatient prescription drug benefits). Medicaid. CBO reduced projected spending for Medicaid between 2014 and 2023 by $77 billion (or 2 percent) relative to its February 2013 estimates. That drop represents the net effect of a number of adjustments. The largest change is a reduction in spending for long-term care services. CBO lowered estimated spending for such services over the 2014–2023 period by $119 billion on the basis of an analysis of recent growth in such spending, which slowed from an average annual growth rate of 6 percent between 1999 and 2009 to a rate of 2 percent over the past three years.


MAY 2013

Partially offsetting the reduction in projected spending for long-term care services were a number of changes, mostly to projected Medicaid enrollment, that together increased estimated spending by $42 billion over the 2014–2023 period. In 2023, for example, CBO now projects average monthly Medicaid enrollment that is roughly 6 percent greater than it estimated in February. Those changes stem primarily from two factors:  Updated data on current enrollment (which is higher than previously projected), and  An increase in the proportion of potentially newly eligible Medicaid beneficiaries who will reside in states that are expected to choose to fully extend Medicaid coverage under the Affordable Care Act (reflecting recent developments with state governments).11

The reductions in CBO’s projections of spending for Medicare and Medicaid continue a recent trend. During the past several years, health care spending has grown much more slowly both nationally and for federal programs than it did historically and more slowly than CBO had projected. As a result, in 2012, federal spending for Medicare and Medicaid was about 5 percent below the amount that CBO had estimated in March 2010. In response to the observed slowdown, CBO has made a series of downward adjustments to its projections of spending for Medicare and Medicaid. From the March 2010 baseline to the current baseline, CBO has lowered its estimates of federal spending for the two programs in 2020 for technical reasons by about $225 billion—in

UPDATED BUDGET PROJECTIONS: FISCAL YEARS 2013 TO 2023

7

particular, by $138 billion for Medicare and by $89 billion for Medicaid—or by roughly 15 percent for each program. Those reductions mostly reflect the slower growth in the programs’ spending in recent years. Net Interest. CBO has reduced its projection of net interest costs over the 2014–2023 period by $193 billion. Most of that decrease ($173 billion) is attributable to lower projected debt-service costs as a result of smaller deficits. The remaining $20 billion stems mainly from modifications to CBO’s assumptions about the mix of securities that the Treasury is expected to issue to finance future deficits.

11. CBO currently expects that, by 2023, 70 percent of such newly eligible Medicaid beneficiaries will reside in states that choose to fully extend Medicaid coverage (that is, making eligible most nonelderly people with income below 138 percent of the federal poverty level), 10 percent will reside in states that partially extend Medicaid, and 20 percent will reside in states that do not expand Medicaid at all. The increase in CBO’s estimate of the proportion of potentially newly eligible Medicaid beneficiaries who will reside in states that are expected to fully extend coverage under the Affordable Care Act, along with other small modeling changes, led to a net decrease in the number of people expected to take up health insurance through exchanges and thus in the projected total amount of exchange subsidies. At the same time, as discussed above, the share of those subsidies that is projected to take the form of outlays was increased. In all, those factors turned out to be almost exactly offsetting over the 2014–2023 period, so projected outlays for exchange subsidies are nearly unchanged relative to the February 2013 estimate.

CBO


8

UPDATED BUDGET PROJECTIONS: FISCAL YEARS 2013 TO 2023

MAY 2013

Table 1.

CBO’s Baseline Budget Projections Actual, 2012

2013

2014

2015

2016

2017

2018

2019

2020

2021

2022

Total 2014- 20142023 2018 2023

In Billions of Dollars Revenues

Individual income taxes Social insurance taxes Corporate income taxes Other Total

On-budget Off-budgeta

1,132 1,333 1,380 1,558 1,691 1,826 1,942 2,051 2,168 2,291 2,422 2,560 8,398 19,890 845 952 1,020 1,066 1,126 1,192 1,253 1,309 1,366 1,428 1,492 1,559 5,656 12,811 242 291 380 455 489 511 512 498 492 493 499 506 2,348 4,836 230 _____ 237 _____ 262 _____ 319 _____ 300 _____ 249 _____ 237 _____ 245 _____ 253 _____ 282 _____ 319 _____ 333 ______ 1,367 ______ 2,800 _____ 2,450

2,813

3,042

3,399

3,606

3,779

3,943

4,103

4,280

4,494

4,732

1,881 570

2,144 670

2,311 731

2,634 765

2,796 811

2,919 860

3,038 905

3,155 948

3,289 990

3,459 1,034

3,653 1,079

4,959 17,769 40,336

3,834 13,698 1,125 4,071

31,089 9,247

Outlays

Mandatory Discretionary Net interest Total

On-budget Off-budgeta Deficit (-) or Surplus

On-budget Off-budgeta Debt Held by the Public

2,031 2,020 2,196 2,326 2,519 2,633 2,737 2,893 3,053 3,225 3,470 3,617 12,412 28,670 1,285 1,213 1,168 1,187 1,206 1,229 1,250 1,286 1,316 1,347 1,386 1,415 6,041 12,790 220 _____ 223 _____ 237 _____ 264 _____ 313 _____ 398 _____ 497 _____ 573 _____ 644 _____ 703 _____ 764 _____ 823 ______ 1,710 ______ 5,216 _____ 3,537

3,455

3,602

3,777

4,038

4,261

4,485

4,752

5,012

5,275

5,620

3,030 508

2,816 640

2,890 712

3,022 755

3,235 803

3,408 853

3,581 904

3,793 959

3,993 1,020

4,191 1,084

4,468 1,153

5,855 20,163 46,677

4,628 16,135 1,226 4,027

-1,087

-642

-560

-378

-432

-482

-542

-648

-733

-782

-889

-895 -2,394 -6,340

-1,149 62

-672 30

-579 19

-388 10

-440 8

-489 7

-542 1

-637 -11

-704 -29

-732 -50

-815 -74

37,207 9,469

-794 -102

-2,437 44

-6,118 -222

11,281 12,036 12,685 13,156 13,666 14,223 14,827 15,537 16,330 17,168 18,118 19,070

n.a.

n.a.

Memorandum:

Gross Domestic Product

15,549 16,034 16,646 17,632 18,792 19,959 20,943 21,890 22,854 23,842 24,858 25,910 93,972 213,326 As a Percentage of Gross Domestic Product

Revenues

Individual income taxes Social insurance taxes Corporate income taxes Other

7.3 5.4 1.6 1.5 ____

8.3 5.9 1.8 1.5 ____

8.3 6.1 2.3 1.6 ____

8.8 6.0 2.6 1.8 ____

9.0 6.0 2.6 1.6 ____

9.2 6.0 2.6 1.2 ____

9.3 6.0 2.4 1.1 ____

9.4 6.0 2.3 1.1 ____

9.5 6.0 2.2 1.1 ____

9.6 6.0 2.1 1.2 ____

9.7 6.0 2.0 1.3 ____

9.9 6.0 2.0 1.3 ____

8.9 6.0 2.5 1.5 ____

9.3 6.0 2.3 1.3 ____

15.8

17.5

18.3

19.3

19.2

18.9

18.8

18.7

18.7

18.8

19.0

19.1

18.9

18.9

12.1 3.7

13.4 4.2

13.9 4.4

14.9 4.3

14.9 4.3

14.6 4.3

14.5 4.3

14.4 4.3

14.4 4.3

14.5 4.3

14.7 4.3

14.8 4.3

14.6 4.3

14.6 4.3

Mandatory Discretionary Net interest

13.1 8.3 1.4 ____

12.6 7.6 1.4 ____

13.2 7.0 1.4 ____

13.2 6.7 1.5 ____

13.4 6.4 1.7 ____

13.2 6.2 2.0 ____

13.1 6.0 2.4 ____

13.2 5.9 2.6 ____

13.4 5.8 2.8 ____

13.5 5.6 2.9 ____

14.0 5.6 3.1 ____

14.0 5.5 3.2 ____

13.2 6.4 1.8 ____

13.4 6.0 2.4 ____

Total

22.7

21.5

21.6

21.4

21.5

21.3

21.4

21.7

21.9

22.1

22.6

22.6

21.5

21.9

19.5 3.3

17.6 4.0

17.4 4.3

17.1 4.3

17.2 4.3

17.1 4.3

17.1 4.3

17.3 4.4

17.5 4.5

17.6 4.5

18.0 4.6

17.9 4.7

17.2 4.3

17.4 4.4

-7.0

-4.0

-3.4

-2.1

-2.3

-2.4

-2.6

-3.0

-3.2

-3.3

-3.6

-3.5

-2.5

-3.0

-7.4 0.4

-4.2 0.2

-3.5 0.1

-2.2 0.1

-2.3 *

-2.4 *

-2.6 *

-2.9 -0.1

-3.1 -0.1

-3.1 -0.2

-3.3 -0.3

-3.1 -0.4

-2.6 *

-2.9 -0.1

72.6

75.1

76.2

74.6

72.7

71.3

70.8

71.0

71.5

72.0

72.9

73.6

n.a.

n.a.

Total

On-budget Off-budgeta Outlays

On-budget Off-budgeta Deficit (-) or Surplus

On-budget Off-budgeta Debt Held by the Public

Source: Congressional Budget Office. Note: n.a. = not applicable; * = between zero and 0.05 percent. a. The revenues and outlays of the Social Security trust funds and the net cash flow of the Postal Service are classified as off-budget.

CBO


MAY 2013

UPDATED BUDGET PROJECTIONS: FISCAL YEARS 2013 TO 2023

9

Figure 1.

Total Revenues and Outlays (Percentage of gross domestic product) 26

Actual Average Outlays, 1973 to 2012 (21.0%)

24

Projected

Outlays

22

20

18

16

Average Revenues, 1973 to 2012 (17.9%)

14

0 1973

1978

1983

1988

1993

1998

Revenues

2003

2008

2013

2018

2023

Source: Congressional Budget Office.

Figure 2.

Federal Debt Held by the Public (Percentage of gross domestic product) 120

Actual

Projected

100

80

60

40

20

0 1940

1945

1950

1955

1960

1965

1970

1975

1980

1985

1990

1995

2000

2005

2010

2015

2020

Source: Congressional Budget Office.

CBO


10

UPDATED BUDGET PROJECTIONS: FISCAL YEARS 2013 TO 2023

MAY 2013

Table 2.

Mandatory Outlays Projected in CBO’s Baseline (Billions of dollars) Total Actual, 2014- 20142012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2018 2023

Social Security Old-Age and Survivors Insurance Disability Insurance Subtotal Health Care Programs Medicarea Medicaid Health insurance subsidies and related spending MERHCF Children's Health Insurance Program Other Subtotala Income Security Supplemental Nutrition Assistance Program Supplemental Security Income Unemployment compensation Earned income and child tax credits Family supportb Child nutrition Foster care Miscellaneous tax creditsc Subtotal Federal Civilian and Military Retirement Civiliand Military Other Subtotal

632 136 ___ 768

668 141 ___ 809

704 145 ___ 848

745 149 ___ 894

790 839 892 949 1,011 1,073 1,138 1,207 154 160 ____ 165 ____ 171 ____ 178 ____ 187 ____ 197 ____ 207 ___ ____ 944 999 1,057 1,121 1,189 1,260 1,335 1,414

3,970 9,349 773 _____ 1,714 ____ 4,743 11,062

551 251

586 265

597 298

615 328

671 369

3,301 1,808

* 9 9 ___7

1 9 9 ___8

827

878

23 44 76 95 104 108 115 122 128 135 9 10 10 11 12 12 13 14 15 16 14 15 8 6 6 6 6 6 6 6 25 ____ 22 ____ 28 ____ 27 ____ 29 ____ 30 ____ 31 ____ 32 ____ 33 ___7 ____ 947 1,037 1,157 1,231 1,288 1,391 1,479 1,576 1,719 1,808

341 949 51 122 49 77 109 _____ 263 ____ 5,659 13,632

80 47 92 77 24 19 7 ___7

83 53 70 77 24 20 7 ___6

80 54 52 80 25 21 7 ___6

79 55 45 82 25 22 7 ___6

79 61 43 82 25 23 7 ___6

78 59 42 83 25 24 7 ___6

77 55 43 84 25 25 7 ___7

76 62 46 73 25 26 8 ___0

75 64 49 74 25 27 8 ___0

74 66 53 75 25 28 8 ___0

73 74 56 77 25 29 8 ___0

73 70 58 78 25 30 8 ___0

394 284 225 410 123 116 35 31 ____

764 622 488 787 248 255 75 31 ____

354

341

325

321

328

323

322

315

321

329

342

343

1,619

3,270

87 49 ___7

91 54 ___7

93 56 ___7

95 57 ___7

98 63 ___7

101 61 ___7

105 58 ___8

108 64 ___9

111 66 10 ___

115 68 10 ___

119 75 10 ___

122 72 10 ___

492 294 36 ___

1,067 639 86 ____

144

152

155

159

168

169

171

181

187

193

204

205

822

1,792

56 12 __ 68

65 14 __ 79

70 13 __ 83

72 13 __ 85

81 14 __ 95

78 14 __ 92

74 14 __ 88

82 16 __ 97

83 16 ___ 100

84 17 ___ 101

92 19 ___ 111

86 19 ___ 105

375 69 ___ 443

801 156 ___ 957

12 25 -19 7 58 __

25 -9 -36 6 -18 __

14 -1 -20 -10 59 __

16 2 -20 -10 56 __

16 1 -12 -11 58 __

16 1 -3 -12 57 __

15 * 3 -13 53 __

16 * 4 -18 52 __

16 * 3 -18 51 __

16 * 3 -12 50 __

16 0 2 -14 58 __

16 0 2 -14 57 __

77 3 -52 -55 283 ___

157 3 -37 -132 551 ___

82

-32

41

43

53

58

60

54

52

57

62

61

256

542

695 396

722 418

794 441

849 466

911 1,018 1,064 493 521 554

7,938 4,283

e

Veterans Income security Other Subtotal Other Programs Agriculture Troubled Asset Relief Program Higher education Deposit insurance Other Subtotal

Continued

CBO


MAY 2013

UPDATED BUDGET PROJECTIONS: FISCAL YEARS 2013 TO 2023

Table 2.

11

Continued

Mandatory Outlays Projected in CBO’s Baseline (Billions of dollars) Total Actual, 2014- 20142012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2018 2023

Offsetting Receipts Medicaref Federal share of federal employees' retirement Social Security Military retirement Civil service retirement and other Subtotal Receipts related to natural resources MERHCF Other Subtotal Total

-85

-90

-92

-94

-100

-108

-117

-125

-133

-143

-156

-170

-510 -1,237

-16 -22 -30 __

-16 -21 -29 __

-16 -21 -30 __

-17 -21 -31 __

-18 -22 -32 __

-18 -23 -33 __

-19 -24 -34 __

-20 -25 -35 __

-21 -26 -37 __

-21 -27 -38 __

-22 -28 -40 __

-23 -29 -41 __

-88 -111 -159 ___

-195 -243 -351 ___

-67

-66

-67

-69

-71

-74

-77

-80

-83

-86

-89

-93

-358

-789

-13 -11 -33 ___

-15 -9 -27 ___

-14 -7 -22 ___

-14 -8 -29 ___

-15 -9 -32 ___

-14 -9 -33 ___

-15 -10 -30 ___

-19 -10 -31 ___

-17 -11 -31 ___

-18 -12 -32 ___

-18 -12 -27 ___

-18 -13 -27 ___

-73 -43 -146 ____

-163 -101 -293 ____

-210

-207

-203

-214

-226

-239

-248

-265

-275

-290

-303

-320 -1,130 -2,584

2,031 2,020 2,196 2,326 2,519 2,633 2,737 2,893 3,053 3,225 3,470 3,617 12,412 28,670

Memorandum:

Mandatory Spending Excluding Offsetting Receipts

2,241 2,227 2,399 2,540 2,745 2,872 2,986 3,158 3,328 3,516 3,773 3,937 13,542 31,254

Medicare Spending Net of Offsetting Receipts

466

496

505

521

Spending for Major Health Care Programs Net of Offsetting Receiptsg

726

771

840

909 1,024 1,083 1,133 1,224 1,303 1,388 1,516 1,590

572

587

605

669

717

768

861

894

2,791

6,700

4,989 12,010

Source: Congressional Budget Office. Notes: Data on spending for benefit programs in this table generally exclude administrative costs, which are discretionary. * = between zero and $500 million; MERHCF = Department of Defense Medicare-Eligible Retiree Health Care Fund (including TRICARE for Life). a. Excludes offsetting receipts from premium payments and amounts paid by states from savings on Medicaid’s prescription drug costs. b. Includes Temporary Assistance for Needy Families and various programs that involve payments to states for child support enforcement and family support, child care entitlements, and research to benefit children. c. Includes outlays for the American Opportunity Tax Credit and other tax credits. d. Includes Civil Service, Foreign Service, Coast Guard, and other, smaller retirement programs as well as annuitants’ health care benefits. e. Income security includes veterans’ compensation, pensions, and life insurance programs. Other benefits are primarily education subsidies. f.

Includes Medicare premiums and amounts paid by states from savings on Medicaid’s prescription drug costs.

g. Includes Medicare (net of receipts from premiums), Medicaid, the Children’s Health Insurance Program, and subsidies offered through new health insurance exchanges and related spending.

CBO


12

UPDATED BUDGET PROJECTIONS: FISCAL YEARS 2013 TO 2023

MAY 2013

Figure 3.

Projected Spending for Major Budget Categories (Percentage of gross domestic product) 7

Major Health Care Programsa

6

5

Social Security Defense Discretionary Spending

4

Nondefense Discretionary Spending 3

Other Mandatory Spendingb

2

Net Interest 1

0 2012

2013

2014

2015

2016

2017

2018

2019

2020

2021

2022

2023

Source: Congressional Budget Office. a. Includes Medicare (net of receipts from premiums), Medicaid, the Children’s Health Insurance Program, and subsidies offered through new health insurance exchanges and related spending. b. Other than mandatory spending for major health care programs and Social Security.

CBO


MAY 2013

UPDATED BUDGET PROJECTIONS: FISCAL YEARS 2013 TO 2023

13

Table 3.

Discretionary Budget Authority for Fiscal Year 2013 (Billions of dollars) Before Sequestration

Amount a Sequestered

After Sequestration

Discretionary Budget Authority Subject to the Caps Established by the Budget Control Act b

1,043

n.a.

n.a.

Adjustments to Budget Authority as Estimated by CBO When Legislation Was Enactedc

14 _____

n.a.

n.a.

Budget Authority Subject to the Caps With Modifications to Original Scoring

1,057

-55

1,002

Adjustments to the Caps Overseas contingency operationsd Emergency requirementse Disaster relieff Program integrityg

99 41 12 * ____

-5 -2 -1 * __

93 39 11 * ____

152

-8

144

1,209

-64

1,145

Total Total Discretionary Budget Authority in CBO's Baseline

Source: Congressional Budget Office. Note: n.a. = not applicable; * = between -$500 million and $500 million. a. The amount sequestered includes a reduction in unobligated balances for defense programs. After final appropriations for 2013 were enacted, the Administration reduced its original calculation of the sequestration of discretionary budget authority by a total of nearly $5 billion. Subtracting the amount sequestered in 2013 from the caps specified in the Deficit Control Act of 1985, as amended by the Budget Control Act of 2011, would result in a total of $988 billion in budget authority for 2013. b. The amount shown here is budget authority as estimated by CBO when appropriations were enacted. The caps on discretionary budget authority for 2013 apply to security spending ($684 billion) and nonsecurity spending ($359 billion). The security category comprises discretionary appropriations for the Departments of Defense, Homeland Security, and Veterans’ Affairs; the National Nuclear Security Administration; the intelligence community management account (Treasury account 95-0401-0-1-054); and discretionary accounts related to international affairs (budget function 150). The nonsecurity category comprises all other discretionary appropriations. c. The amount of budget authority in CBO’s baseline is $14 billion more than the amount that CBO estimated when appropriations were enacted, for two main reasons. First, about $19 billion in savings from changes to mandatory programs included in the final appropriation act were credited against discretionary spending when the legislation was enacted; in CBO’s baseline, those savings appear in their normal mandatory accounts. Second, current estimates of the receipts of the Federal Housing Administration are about $5 billion higher than the amounts initially credited to the legislation. The figures in this table are based on provisions of the Consolidated and Further Continuing Appropriations Act, 2013 (Public Law 113-6), and the Disaster Relief Appropriations Act, 2013 (P.L. 113-2). d. This category consists of funding for war-related activities in Afghanistan or for similar activities. e. This category consists mostly of funding for relief and recovery from Hurricane Sandy that was designated as an emergency requirement by the Congress. About $5 billion in funding related to Hurricane Sandy was declared disaster relief, and about $3 billion was not declared either as an emergency requirement or as disaster relief. This category also contains $0.5 billion in rescissions of previous emergency funding, resulting from the Dodd-Frank Wall Street Reform and Consumer Protection Act (P.L. 111–203). f.

For the purposes of adjustments to the caps, “disaster relief” refers to activities carried out pursuant to section 102(2) of the Robert T. Stafford Disaster Relief and Emergency Assistance Act (42 U.S.C. 5122(2)); such activities may result from a natural disaster that causes damage of sufficient severity to warrant federal assistance.

g. Program integrity initiatives identify and reduce overpayments in benefit programs, such as Disability Insurance, Supplemental Security Income, Medicare, Medicaid, and the Children’s Health Insurance Program. For 2013, funding for program integrity initiatives thus far has been provided only for Disability Insurance and Supplemental Security Income.

CBO


14

UPDATED BUDGET PROJECTIONS: FISCAL YEARS 2013 TO 2023

MAY 2013

Table 4.

Discretionary Spending Projected in CBO’s Baseline (Billions of dollars)

2014

2015

2016

2017

2018

2019

2020

2021

2022

2023

Total, 20142023

Budget Authority

Defense Caps established by the Budget Control Act, including automatic spending reductions a Adjustments to the capsb Overseas contingency operationsc Emergency designationd Subtotal, Adjustments Total, Defense

Nondefense Caps established by the Budget Control Act, including automatic spending reductions a Adjustments to the capsb Overseas contingency operationsc Disaster relief e Emergency designationd Program integrity f Subtotal, Adjustments Total, Nondefense

All Defense and Nondefense Budget Authority Caps established by the Budget Control Act, including automatic spending reductions a Adjustments to the capsb Total Discretionary Budget Authority

498

512

523

536

549

562

576

590

605

621

5,573

84 * __ 84

86 * __ 86

88 * __ 88

90 * __ 90

92 * __ 92

94 * ___ 94

96 * ___ 96

98 * ___ 99

101 * ___ 101

103 * ___ 103

933 ___1 934

582

598

611

626

641

656

673

689

706

724

6,507

469

483

493

504

517

531

544

557

571

586

5,255

11 11 40 * __

11 12 40 * __

12 10 41 __1

12 8 42 __1

12 9 43 __1

12 8 44 __1

13 9 45 __1

13 10 46 __1

13 10 47 __1

14 10 48 __1

123 97 438 ___6

63

64

64

63

65

65

67

69

71 a

72 a

663

531

547

557

567

581

596

611

626

642

658

5,917

967 147

995 150

1,016 152

1,040 153

1,066 157

1,093 160

1,120 164

1,147 168

1,177 172

1,207 175

10,828 1,597

1,114 1,145 1,168 1,193 1,223 1,252 1,284 1,315 1,348 1,383

12,424

Continued Source: Congressional Budget Office. Note: * = between zero and $500 million. a. The Deficit Control Act of 1985, as amended by the Budget Control Act of 2011, specifies caps on discretionary appropriations through 2021. CBO has extrapolated the totals for 2022 and 2023 on the basis of its projected rates of inflation for wages and prices. Automatic spending reductions, which are already incorporated into CBO's projections, are slated to reduce the caps for 2014 through 2021. b. The statutory caps do not constrain funding for overseas contingency operations, emergencies, disaster relief, and certain program integrity initiatives (which identify and reduce overpayments in some benefit programs); the caps are therefore adjusted to accommodate funding for those purposes.

CBO


MAY 2013

UPDATED BUDGET PROJECTIONS: FISCAL YEARS 2013 TO 2023

Table 4.

15

Continued

Discretionary Spending Projected in CBO’s Baseline (Billions of dollars)

2014

2015

2016

2017

2018

2019

2020

2021

2022

2023

Total, 20142023

Outlaysg

Defense Outlays under the caps with automatic spending reductionsa Adjustments to the capsb Overseas contingency operationsc Emergency designationd Subtotal, Adjustments Total, Defense

Nondefense Outlays under the caps with automatic spending reductionsa Adjustments to the capsb Overseas contingency operationsc Disaster relief e Emergency designationd Program integrity f Subtotal, Adjustments Total, Nondefense

All Defense and Nondefense Budget Authority Outlays under the caps with automatic spending reductionsa Adjustments to the capsb Total Discretionary Outlays

526

516

523

527

534

551

564

577

597

607

5,523

70 * __

80 __*

85 * __

88 * __

89 * __

92 * __

94 * __

96 * __

99 * __

101 * ___

893 ___1

70

80

85

88

89

92

94

96

99

101

894

596

596

608

615

623

643

658

674

696

708

6,417

552

559

555

563

572

584

597

610

624

639

5,855

6 3 10 * __

8 6 18 __*

10 8 24 __1

11 10 31 __1

11 9 34 __1

11 9 38 __1

12 9 40 __1

12 9 42 __1

12 9 44 __1

13 9 45 __1

106 81 326 ___6

20

32

43

51

55

59

61

64

66 a

67 a

518

572

591

597

614

627

643

658

673

690

706

6,373

1,079 89

1,075 112

1,078 128

1,091 139

1,106 145

1,135 151

1,160 156

1,187 160

1,221 165

1,247 168

11,378 1,412

1,168 1,187 1,206 1,229 1,250 1,286 1,316 1,347 1,386 1,415

12,790

c. This category consists of funding for war-related activities in Afghanistan or for similar activities. d. This category consists mostly of funding for relief and recovery from Hurricane Sandy that was designated as an emergency requirement by the Congress. About $5 billion in funding related to Hurricane Sandy was declared disaster relief, and about $3 billion was not declared either as an emergency requirement or as disaster relief. This category also contains $0.5 billion in rescissions of previous emergency funding, resulting from the Dodd-Frank Wall Street Reform and Consumer Protection Act (Public Law 111–203). e. For the purposes of adjustments to the caps, “disaster relief” refers to activities carried out pursuant to section 102(2) of the Robert T. Stafford Disaster Relief and Emergency Assistance Act (42 U.S.C. 5122(2)); such activities may result from a natural disaster that causes damage of sufficient severity to warrant federal assistance. Under the Deficit Control Act, as amended, the limits on discretionary budget authority can be raised to reflect funding for disaster relief. However, the total increase in the cap in any year for that reason can be no more than the average funding for disaster relief over the previous 10 years (excluding the highest and lowest amounts) plus any amount by which the prior year’s appropriation was below the maximum allowable cap adjustment for that year. In CBO’s baseline, such funding exceeds the average, beginning in 2016; the required reduction in the cap adjustment is included in the totals shown for disaster relief. f.

Program integrity initiatives identify and reduce overpayments in benefit programs, such as Disability Insurance, Supplemental Security Income, Medicare, Medicaid, and the Children’s Health Insurance Program. For 2013, funding for program integrity initiatives thus far has been provided only for Disability Insurance and Supplemental Security Income.

g. Outlays for the 2014–2023 period include the effects of sequestration in 2013.

CBO


16

UPDATED BUDGET PROJECTIONS: FISCAL YEARS 2013 TO 2023

MAY 2013

Table 5.

Federal Debt Projected in CBO’s Baseline (Billions of dollars) Actual, 2012 2013

2014

2015

2016

2017

2018

2019

2020

2021

2022

2023

Debt Held by the Public at the Beginning of the Year

10,128 11,281 12,036 12,685 13,156 13,666 14,223 14,827 15,537 16,330 17,168 18,118

Changes in Debt Held by the Public Deficit Other means of financing

642 113 ___

560

378

432

482

542

648

733

782

889

895

66 _____

89 ___

93 ___

78 ___

74 ___

63 ___

61 ___

60 ___

57 ___

62 ___

56 ___

1,153

755

649

471

510

556

605

710

793

838

950

952

Total Debt Held by the Public at the End of the Year

1,087

11,281 12,036 12,685 13,156 13,666 14,223 14,827 15,537 16,330 17,168 18,118 19,070

Memorandum:

Debt Held by the Public at the End of the Year (As a percentage of GDP)

72.6

75.1

76.2

74.6

72.7

71.3

70.8

71.0

71.5

72.0

72.9

73.6

Debt Held by the Public Excluding Financial Assetsa In billions of dollars As a percentage of GDP

10,391 11,031 11,565 11,924 12,334 12,793 13,311 13,935 14,643 15,400 16,262 17,130 66.8 68.8 69.5 67.6 65.6 64.1 63.6 63.7 64.1 64.6 65.4 66.1

Gross Federal Debt b

16,051 16,887 17,616 18,185 18,805 19,529 20,321 21,200 22,148 23,125 24,173 25,228

Debt Subject to Limit

c

16,027 16,863 17,591 18,160 18,779 19,503 20,295 21,173 22,120 23,096 24,144 25,198

Source: Congressional Budget Office. Note: GDP = gross domestic product. a. Subtracts from debt held by the public the value of outstanding student loans and other credit transactions, financial assets (such as preferred stock) purchased from institutions participating in the Troubled Asset Relief Program, cash balances, and other financial instruments. b. Comprises federal debt held by the public and Treasury securities held by federal trust funds and other government accounts. c. The amount of federal debt that is subject to the overall limit set in law. Debt subject to limit differs from gross federal debt because most debt issued by agencies other than the Treasury and the Federal Financing Bank is excluded from the debt limit. The debt limit was most recently set at $16.4 trillion but has been suspended through May 18, 2013. On May 19, the debt limit will be raised to its previous level plus the amount of borrowing that occurred while the limit was suspended.

CBO


MAY 2013

UPDATED BUDGET PROJECTIONS: FISCAL YEARS 2013 TO 2023

17

Table 6.

Changes in CBO’s Baseline Projections of the Deficit Since February 2013 (Billions of dollars) Total 2014- 20142013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2018 2023

Deficit in CBO's February 2013 Baseline

-845

-616

-430

-476

-535

-605

-710

-798

-854

-957

Changes in Revenues Individual income taxes Corporate income taxes Social insurance taxes Other

69 40 -1 -3 ____

25 24 -2 -8 ___

19 7 -3 3 ___

17 0 -3 1 ___

16 0 -3

13 0 -4 -2 ___

11 0 -5 -4 ___

10 0 -6 -5 ___

9 0 -6 -6 ___

10 0 -6 -7 ___

12 0 -5 -9 ___

90 31 -14 -7 ____

143 31 -42 -37 ___

105

39

26

15

6

2

*

-2

-2

-2

100

95

-2 -6

-8 -8 -3 *

-9 -7 -3 2

-9 -15 -5

-95 ___7

-6 -4 1 -2 2 __

2 ___

-10 -12 -10 -2 -1 ___

-9 -11 -13 -2 -2 ___

-9 -1 -15 -3 -3 ___

-8 -9 -18 -3 -3 ___

-40 -41 -13 1 1 ___

-86 -85 -77 -11 -9 ____

-97

-9

*

-2

All Changes in Revenues

Changes in Outlays Mandatory Social Security Net Medicare Medicaid Fannie Mae and Freddie Maca Other Subtotal Discretionary

*

* ___ 14

-978 -2,661 -6,958

* ___

-9 -8 -4 1 -1 ___

-2 ___

-10 -11 -8 -2 -1 ___

-16

-16

-21

-30

-32

-34

-38

-31

-41

-92

-267

-2

-3

-4

-7

-7

-8

-9

-10

-9

-18

-62

*

Net interest Debt service Other

*

-2

-3

-5

-10

-15

-20

-23

-28

-32

-36

-35

-173

-1 ___

-4 ___

-5 ___

-5 ___

-4 ___

-4 ___

-1 ___

* ___

* ___

1 ___

2 ___

-22 ___

-20 ____

Subtotal

-1

-6

-8

-10

-14

-20

-20

-23

-28

-31

-34

-57

-193

-98

-16

-26

-29

-39

-57

-60

-65

-75

-71

-84

-168

-522

Total Effect on the Deficit

203

56

51

44

53

63

62

66

72

68

82

267

618

Deficit in CBO's May 2013 Baseline

-642

-560

-378

-432

-482

-542

-648

-733

-782

-889

All Changes in Outlays b

-895 -2,394 -6,340

Source: Congressional Budget Office. Note: * = between -$500 million and $500 million. a. To provide CBO’s best estimate of what the Treasury will ultimately report as the federal deficit for the current year, CBO’s baseline includes an estimate of net cash transactions between the Treasury and Fannie Mae and Freddie Mac for fiscal year 2013. For 2014 through 2023, CBO’s baseline shows the projected subsidy costs of credit assistance offered by Fannie Mae and Freddie Mac. b. Positive numbers indicate a decrease in the deficit.

CBO


18

UPDATED BUDGET PROJECTIONS: FISCAL YEARS 2013 TO 2023

About This Document

T

his document is one of a series of reports on the state of the budget that the Congressional Budget Office (CBO) issues each year. It satisfies the requirement of section 202(e) of the Congressional Budget and Impoundment Control Act of 1974 that CBO submit to the Committees on the Budget periodic reports about fiscal policy and its baseline projections of the federal budget. Amber Marcellino of CBO’s Budget Analysis Division prepared the report, with assistance from Mark Booth, and with guidance from Jeffrey Holland, Theresa Gullo, Holly Harvey, Peter Fontaine, and Frank Sammartino. The estimates described here were the work of more than 100 people at CBO and many people on the staff of the Joint Committee on Taxation. In keeping with CBO’s mandate to provide objective, impartial analysis, this report makes no recommendations. It and other CBO publications are available on the agency’s website (www.cbo.gov).

Douglas W. Elmendorf Director May 2013

CBO

MAY 2013


Selected Provisions of Various Medicare Reform Proposals --Bipartisan Partnership and Coalition Proposals-Bipartisan Policy Center A Bipartisan Rx for Patient-Centered Care and System-Wide Cost Containment (April 2013) http://bipartisanpolicy.org/library/report/health-care-cost-containment 

Preserve and improve Medicare care delivery and payment systems o Create “Medicare Networks” option under traditional Medicare through improved ACOs that would also share savings with beneficiaries. o Pay MA plans according to competitive-priced bidding. o Expand voluntary bundled payment demonstration to standard Medicare payment method. o Implement (no earlier than 2020) a fallback spending limit to restrain per-beneficiary spending growth to GDP per capita + 0.5. Strengthen and modernize the Medicare benefit o Add a catastrophic cap, cerate combined deductible, replace coinsurance with copayments, while maintaining same aggregate cost-sharing as today o New Medicare supplement (including employer sponsored retiree health) requirements: minimum $250 deductible, minimum $2500 OOP max, can’t cover more than half of copayments and coinsurance o Increase cost-sharing assistance for low-income beneficiaries o Increase share of beneficiaries who pay income-related premiums Other Medicare and health system reforms that improve care and lower cost growth o Implement DME competitive bidding o Equalize E&M and certain other outpatient payments across settings o End the CMS MA star rating demonstration; discontinue star bonus payments after conversion to competitive bids o Make certain cost sharing and other changes to encourage high-quality, low-cost drug utilization

Partnership for Sustainable Health Care http://rwjf.org/content/dam/farm/reports/reports/2013/rwjf405432 (April 2013) (AHIP, Ascension Health, Families USA, National Coalition on Health Care, Pacific Business Group on Health) 1

2013 Summer Summit Materials | American Academy of Actuaries


  

Transform the current payment paradigm—Move away from fee-for-service payment systems and toward value-based payment approaches (e.g., payments for patient safety, bundled payments, ACOs, global payments). Pay for care that is proven to work—Reduce payments for services that are proven to be less effective. Incentivize consumer engagement in care—Use tiered cost-sharing to encourage the use of high-value services and providers. Improve the infrastructure needed for an effective health care market—Such efforts should include accelerating research on treatment effectiveness, developing uniform measures, ensuring adequate work force, streamlining administrative processes, reducing and resolving medical malpractice disputes, increasing transparency, encouraging competitive markets. Incentivize states to partner with public and private stakeholders to transform the health care system.

Simpson-Bowles, A Bipartisan Path Forward to Securing America’s Future http://library.constantcontact.com/download/get/file/11020350683061504/Updated+Full+Plan.pdf (April 2013) 

       

Delivery system and payment reforms o Reform SGR and encourage physician participation in coordinated care o Expand penalties for avoidable complications and readmissions o Expand payment bundling o Expand competitive bidding for DME and other devices and services o Create alternative benefit packages and increase transparency to encourage care coordination o Give IPAB more flexibility Reform FFS cost-sharing rules o Unified deductible, uniform coinsurance, and out-of-pocket max; vary deductible and out-of-pocket max with income o Restrict Medigap plans from covering near first dollar costs o Impose surcharge on retiree health plans while offering seniors the option to “cash out” and use value to subsidize their Medicare premium Enact Medicare malpractice reform Encourage state innovation to reduce costs Expand income-relating of Medicare premiums Increase Medicare eligibility age with income-related buy-in at age 65 Reduce and reform post-acute care payments Reduce various payments to hospitals Reduce costs of prescription drugs in Medicare: extend Medicaid prescription drug rebates to dual eligibles (but not other LIS Part D enrollees); prohibit “pay for delay” Reduce fraud, abuse, and excessive payments

2

2013 Summer Summit Materials | American Academy of Actuaries


--Congressional Proposals-Sens. Burr and Coburn—Seniors’ Choice Act (February 2012) http://www.coburn.senate.gov/public/index.cfm?a=Files.Serve&File_id=dd0753e9-e62b-46409659-75099f9bd1a9 • • • • • • • • •

Unify the Part A and B deductibles, establish uniform coinsurance rates, and institute an annual income-related out-of-pocket cap. Prohibit first-collar coverage in Medigap and limit coverage to only one half of cost sharing. Repeal the IPAB. Increase Part B premiums (with a hold-harmless provision for low-income seniors). Raise the Medicare eligibility age incrementally to 67 by 2027. Freeze physician payment rates for the near future until new premium support model is implemented. Increase the share of beneficiaries subject to income-related Part B and D premiums. Offer a transitional voluntary care coordination benefit, directed to high-risk beneficiaries. Implement premium support program in 2016. o Traditional Medicare and private plans would participate in competitive bidding o Fixed government contribution (adjusted for income) would be tied to the weighted average bid

Sens. Corker and Alexander--Fiscal Sustainability Act of 2013 (S. 11) http://thomas.loc.gov/cgi-bin/bdquery/D?d113:1:./temp/~bdY16x::|/home/LegislativeData.php • • • •

Unify the Part A and B deductibles, establish uniform coinsurance rates, and institute an annual out-of-pocket cap. Prohibit first-dollar coverage in Medigap; no new Medigap enrollment beginning in 2018. Replace MA with a system based on competitive bids. Premiums subsidized at 85% of second lowest quintile. Raise the Medicare eligibility age incrementally to 67 by 2027.

Sen. Hatch http://www.hatch.senate.gov/public/index.cfm/releases?ID=7fa4c651-1d83-48ef-b3c45b23215be2f5 (1/24/2013)    

Increase the Medicare eligibility age gradually from 65 to 67. Modernize Medigap by limiting first-dollar coverage. Unify the Part A and B deductibles, establish uniform coinsurance rates, and institute an annual out-of-pocket cap. Medicare competitive bidding (premium support-type approach). The government contribution would be based on competitive bids between traditional Medicare and private plans.

3

2013 Summer Summit Materials | American Academy of Actuaries


Sen. Wyden http://www.wyden.senate.gov/news/press-releases/wyden-outlines-new-medicare-reforms (6/13/2013)  

 

Modify the ACO “attribution rule” which bars providers targeting chronically ill seniors who would benefit most from coordinated chronic care. Ensure high-quality, coordinated care is available to chronically ill seniors throughout the country by reconfiguring Medicare reimbursement to target areas with the highest incidence of chronic illness, and reward practitioners, in those areas, who improve care and hold down costs. Make individual care plans the rule, rather than the exception, for seniors with more than one chronic condition. Provide incentives to help keep seniors as healthy as possible.

Reps. Blackburn and Ellmers http://energycommerce.house.gov/press-release/responding-seniors%E2%80%99-needs-andimproving-medicare-choices (4/11/2013)      

Repeal the SGR and replace it with a reformed physician payment system Modernize the FFS plan design—unify the Parts A & B deductibles, add a catastrophic outof-pocket cap, and modernize the Medigap program. Reduce subsidies for high-income beneficiaries. Protect Medicare Advantage against payment cuts and Medicare Part D against price controls or government price negotiations. Reform the medical liability system. Eliminate waste, fraud, and abuse.

--Pres. Obama FY2014 Budget Proposal-http://www.whitehouse.gov/omb/budget/Overview Includes cuts of $306 billion in projected Medicare payments to health care providers and $57 billion of higher payments by Medicare beneficiaries.  Require drug manufactures to provide rebates to Medicare for drugs purchased by lowincome Part D beneficiaries, similar to those provided in Medicaid.  Accelerate manufacturer discounts and federal subsidies to more quickly close the Part D coverage gap.  Replace the SGR with a reformed physician payment system.  Reduce payment rates to post-acute care providers.  Strengthen IPAB by reducing target from GDP+1 to GDP+.5.  Reduce Medicare coverage of bad debts.  Reduce Medicare Advantage payments to reflect documentation and coding overpayments.  Cut waste, fraud, and abuse.  Increase income-related premiums for Part B and Part D.  Premium surcharge on certain Medigap plans.  Institute a copayment for home health care.

4

2013 Summer Summit Materials | American Academy of Actuaries


--Other Policy Option Compendiums-Kaiser Family Foundation Policy Options to Sustain Medicare for the Future (January 2013) http://kaiserfamilyfoundation.files.wordpress.com/2013/02/8402.pdf This is a review of options and their savings/cost estimates rather than a set of recommendations. The volume includes sections on changes to:     

Medicare eligibility, beneficiary costs, and program financing Medicare payments to plans and providers (including medical malpractice) Delivery system reform and care for high-need beneficiaries Medicare program structure (e.g., benefit redesign, premium support) Medicare program administration (e.g., spending caps, coverage policy, governance and management, program integrity)

Congressional Budget Office Reducing the Deficit: Spending and Revenue Options (March 2011) http://www.cbo.gov/sites/default/files/cbofiles/ftpdocs/120xx/doc12085/03-10reducingthedeficit.pdf This is a compendium of options, as opposed to recommendations, aiming to reduce the deficit. Medicare-related options include: • • • • • • •

Raise the Medicare eligibility age to 67 Impose cost sharing for skilled nursing facilities and home health care Change the cost-sharing structures of traditional Medicare FFS and Medigap Increase Part B premiums to 35 percent of program costs Reduce Medicare’s payment rates across the board in high-spending areas Eliminate the Critical Access Hospital, Medicare-Dependent Hospital, and Sole Community Hospital programs Require manufacturers to pay a minimum rebate on drugs covered under Medicare Part D for low-income beneficiaries

5

2013 Summer Summit Materials | American Academy of Actuaries


ISSUE GUIDE

A Guide to Analyzing Medicare Premium Support


A GUIDE TO ANALYZING THE ISSUES: MEDICARE PREMIUM SUPPORT

P

remium support is a reform option that has been proposed as a way to improve Medicare’s financial condition. Medicare, the federal program providing health insurance to virtually all Americans 65 and older as well as many younger individuals with long-term disabilities, is currently inadequately financed to sustain the program for the long term. In addition, over time it will impose larger financial demands on both beneficiaries and the federal budget. This guide is intended to help voters understand what premium support is and the potential implications of shifting Medicare to a premium support program.

The American Academy of Actuaries is a 17,000-member professional association whose mission is to serve the public and the U.S. actuarial profession. The Academy assists public policymakers on all levels by providing leadership, objective expertise, and actuarial advice on risk and financial security issues. The Academy also sets qualification, practice, and professionalism standards for actuaries in the United States. Copyright Š2013 American Academy of Actuaries. All rights reserved.


What is premium support? Under a typical premium support approach, Medicare beneficiaries would receive a government contribution (sometimes referred to as a voucher) to apply toward the premium of a health plan of their choice, perhaps with the traditional Medicare program being one of the choices. Beneficiaries choosing a plan with a premium greater than the government contribution would be responsible for paying the difference. The federal government contribution could change over time, for example in accordance with inflation or average premium growth.

How would premium support change the structure of the current Medicare program? Medicare beneficiaries today can choose to enroll either in the traditional fee-for-service (FFS) Medicare program or in a private Medicare Advantage (MA) plan. In Medicare Advantage, plans submit bids based on the same benefits that are available in the FFS program. Bids reflect each plan’s expected cost of providing these benefits. Government payments to plans are tied to benchmarks that reflect costs under the FFS program. Plan bids are compared to the benchmarks. If an MA plan’s bid exceeds the benchmark, beneficiaries choosing that plan must pay an additional premium. If an MA plan’s bid falls below the benchmark, a portion of the difference is provided to the plan to fund benefits in addition to those provided by traditional Medicare. The Medicare program today essentially

follows a defined benefit approach. In other words, the government pays whatever is needed to cover a defined benefit package and bears the risk of health spending growth. Premium support proposals would change the nature of the Medicare program from a defined benefit approach to what is considered a defined contribution approach. Under a defined contribution approach, depending on how the federal contribution is defined, government spending may be capped and beneficiaries could bear the risk of health spending growing faster than the cap. Advocates of premium support reforms argue that capping the government contribution could encourage insurers to develop and beneficiaries to choose more cost-effective health plans. Opponents of premium support have argued that rather than reducing overall Medicare spending, premium support may shift costs to beneficiaries and make coverage less affordable.

Are there any premium supporttype approaches currently used for health insurance? The current Medicare Part D prescription drug program contains elements of a premium support approach. In particular, it uses a competitive bidding approach to determine how much the government will contribute toward the plan premiums. Private prescription drug plans submit bids that reflect the expected premiums they require to provide prescription drug benefits to Medicare beneficiaries. The government contribution toward these plans is approximately 75 percent of the average premium bid for basic coverage.1 Beneficiaries who choose

Plan bids and enrollee premiums are based on a standardized population. Government payments to plans, however, are risk-adjusted to reflect enrollee characteristics and health conditions that can affect their prescription drug spending. 1

▲ ISSUE GUIDE: MEDICARE PREMIUM SUPPORT

WWW.ACTUARY.ORG

1


plans with higher premiums or benefits beyond basic coverage pay higher premiums. Beneficiaries who choose plans with below-average bids pay lower premiums. Under the Affordable Care Act, insurance coverage in state health insurance exchanges also will contain elements of a premium support approach. In particular, premium subsidies will be available for low- and moderateincome individuals and families, and these subsidies will be based on the second-lowestcost silver tier plan available in the geographic rating area. Participants choosing plans with higher costs would have to pay the difference. Some employer-based health insurance coverage also can have premium support elements. There are employers who offer multiple plan options to their employees, but set a fixed employer premium contribution cap regardless of the plan chosen. Employees choosing higher-cost plans would have to pay higher premiums. The health plan for federal government workers is one example.

What details matter most when designing a Medicare premium support program? There are different ways to design a premium support program. How the details are developed will affect how beneficiaries fare and whether Medicare costs are contained. HOW THE GOVERNMENT CONTRIBUTION IS SET AND HOW IT GROWS OVER TIME

Under a premium support program, not only would an initial government contribution need to be determined, but also how that contribution grows over time. One option would be to set the initial contribution at the estimated average per-beneficiary government cost under â–˛ ISSUE GUIDE: MEDICARE PREMIUM SUPPORT

the current Medicare program. Another option would be to use competitive bidding to determine the government contribution (e.g., set the government contribution at some percentage of the average premium bid). Under either option, depending on the premiums for plans offered in the premium support program, beneficiary premiums could be greater or less than those they would have paid under the current Medicare program. Perhaps even more important than how the initial government contribution would be set is how it would increase over time. Over the past several decades, spending on health care services has increased faster than general inflation and the economy as a whole. Indexing the government contributions to general inflation, the economy, or some other index that doesn’t keep pace with health spending growth could put pressure on insurance plans to contain costs. But if the government contribution does not increase at least as much as the health spending underlying the plan premiums, then a greater share of Medicare costs would be shifted to beneficiaries over time, either in the form of higher premiums or in the form of higher cost sharing if they choose less generous plans. Increased cost sharing likely would result in reduced health care utilization, but also could result in beneficiaries foregoing needed care, particularly lower-income beneficiaries. Tying the government contributions to the increases in the average premium bids would help prevent costs from being shifted to beneficiaries because bids would track better to changes in health spending, yet still would provide an incentive for beneficiaries to move to lower-cost plans. WWW.ACTUARY.ORG

2


WHETHER THE TRADITIONAL MEDICARE FFS PROGRAM IS RETAINED AS A PLAN OPTION

Under the current Medicare program, beneficiaries have the option of choosing either the traditional FFS plan or one of the available private Medicare Advantage plans. The premium support program could be structured such that the FFS plan remains available to all Medicare beneficiaries, is available only to beneficiaries already enrolled in Medicare at the time premium support is implemented, or is not available to any beneficiaries, including those already enrolled. Retaining the FFS plan option for all current and future Medicare beneficiaries would provide greater continuity with the current program. Rules may be needed, however, to ensure fair competition between FFS and the private plan options. Allowing only current Medicare enrollees to continue having the FFS option would mean that over time the FFS program would consist of older beneficiaries who would likely have more costly health care needs. That could have negative consequences for the financing of the program unless funds are shifted from the other plans to the FFS program to reflect its higher-cost population. Eliminating the FFS program altogether could have implications for the costs of the private plans. The Congressional Budget Office has estimated that rates paid to health care providers are higher for private health insurance plans than for Medicare. With no FFS plans these higher costs would not be fully offset by savings from greater utilization management in private plans.2 Depending on local-area market dynamics, the presence of the FFS plan could provide leverage to private plans

in their rate negotiations with providers, thus reducing the cost of claims, and therefore premiums, below what they otherwise would be in absence of the FFS option. HOW THE BENEFIT PACKAGE IS DEFINED

Medicare Advantage plans must cover at least the same benefits offered in the traditional Medicare FFS option. Premium support plans could be subject to these same types of requirements or new standardized or minimum benefit packages could be required. As an alternative, plans could be provided more leeway in designing their benefit packages. Allowing plan flexibility in benefit designs could allow more timely adoption of innovative benefits and designs. But allowing more flexibility could be confusing for beneficiaries and could also lead to the unintended consequence of plans with benefit packages intentionally designed to avoid appealing to beneficiaries with relatively high-cost health care needs. To mitigate these potential consequences, it would be necessary to implement a risk-adjustment mechanism to ensure that plans are appropriately paid for the risks they bear. Additional requirements also could be considered, such as prohibiting discriminatory plan designs or marketing practices, ensuring an adequate provider network, and developing insurance exchanges to better facilitate the beneficiary decision-making process. WHETHER THERE IS ADDITIONAL FINANCIAL PROTECTION FOR LOW-INCOME BENEFICIARIES

Low-income individuals especially can be at

Congressional Budget Office, “Long-Term Analysis of a Budget Proposal by Chairman Ryan,” April 5, 2011 (revised April 8, 2011). Available at: http://cbo.gov/sites/default/files/cbofiles/ftpdocs/121xx/doc12128/04-05-ryan_letter.pdf. 2

▲ ISSUE GUIDE: MEDICARE PREMIUM SUPPORT

WWW.ACTUARY.ORG

3


risk for avoiding or delaying health care due to costs. Under the current Medicare program, certain low-income beneficiaries receive premium subsidies and some receive cost-sharing subsidies as well. A premium support program could be structured to include premium and/or cost-sharing subsidies for low-income beneficiaries. These could come in the form of direct payments to the health plans or as deposits to health savings accounts that are held by the beneficiaries. The degree to which such subsidies would ensure access to affordable care for low-income beneficiaries would depend on their form and amount. WHEN THE TRANSITION TO PREMIUM SUPPORT TAKES PLACE

Some proposals would implement the transition to a premium support model fairly quickly, within the next few years, and others would delay the implementation for a longer period, for example 10 years. The timing of the transition would affect the plan options available to current and future Medicare beneficiaries as well as which generations share the burden of any lower Medicare spending. A longer transition prior to implementation would allow beneficiaries who eventually would be affected by the change more time to understand and adapt to the new program. Delaying the implementation would shield current beneficiaries and those near retirement from any changes, especially if they can continue in their current plans after the transition. Delaying changes, however, would mean that future Medicare enrollees would be the ones to face any Medicare changes, either positive or negative. And any spending reductions necessary to ensure ▲ ISSUE GUIDE: MEDICARE PREMIUM SUPPORT

long-term Medicare solvency and sustainability would need to be greater if borne only by future Medicare enrollees. OTHER DESIGN DECISIONS

Similar to the current Medicare program, most, if not all, Medicare premium support proposals would prohibit plans from denying coverage or charging higher premiums based on age or health status. To ensure that plans are adequately compensated to reflect the health costs of their enrollee populations, it would be necessary for the government contribution (as opposed to the beneficiary premium) to be risk adjusted so that it varies across plans based on age, health conditions, and other factors that are correlated with health spending. In addition, decisions would need to be made regarding whether the government contribution would differ by region to reflect geographical variations in health spending. If the government contribution doesn’t vary, then beneficiaries’ premiums would vary not only depending on what plan they choose, but also based on where they live. Another consideration would be whether the government contribution would differ depending on a how a plan rates on quality-related measures.

Would a premium support approach reduce Medicare spending? The specifics of the premium support approach would affect whether and to what degree it could reduce Medicare spending. Many of these factors are discussed above. Depending on how the government contribution is set, federal Medicare spending could be lower than currently projected. Whether those savWWW.ACTUARY.ORG

4


ings result in lower overall Medicare savings or instead a shift in costs from the government to Medicare beneficiaries also depends on how utilization management, administrative costs, and provider payment rates under private plans would compare to those under traditional Medicare over time. Ensuring overall Medicare savings rather than just savings to the government may require that plans are structured to facilitate higher quality care and more costeffective health care payment and delivery systems. In addition, effective incentives for beneficiaries to become more cost-conscious health care consumers may be required.

More information on Medicare The more you know about how Medicare works, its financial condition, and the options available for reform, the better equipped you will be to evaluate what candidates have to say about the program. You may want to further your understanding with the following Academy publications: n A Guide to Analyzing the Issues: What Voters Should Know About Medicare n Medicare’s

Financial Condition: Beyond Actuarial Balance

n An

Actuarial Perspective on Proposals to Improve Medicare’s Financial Condition

n Revising

Medicare’s Fee-For-Service Benefit Structure

▲ ISSUE GUIDE: MEDICARE PREMIUM SUPPORT

WWW.ACTUARY.ORG

5


Mary Downs, Executive Director Craig Hanna, Director of Public Policy Cori Uccello, Senior Health Fellow Heather Jerbi, Assistant Director of Public Policy Charity Sack, Director of Communications Members of the Health Practice Council Communications Task Force: Thomas F. Wildsmith, MAAA, FSA, chairperson; Mark J. Jamilkowski, MAAA, FSA; Darrell D. Knapp, MAAA, FSA; Donna C. Novak, MAAA, ASA, FCA; David A. Shea, Jr., MAAA, FSA; Cori E. Uccello, MAAA, FSA, FCA, MPP; Shari A. Westerfield, MAAA, FSA.

For further information, contact us at:

American Academy of Actuaries 1850 M Street NW, Suite 300 Washington, DC 20036-5805 Telephone 202 223 8196 Facsimile 202 872 1948 WWW.ACTUARY.ORG

â–˛ ISSUE GUIDE: MEDICARE PREMIUM SUPPORT

WWW.ACTUARY.ORG

6


Campaign 2 012 VOTER GUIDE

A Guide to Analyzing the Issues: What Voters Should Know About Medicare


WHAT VOTERS SHOULD KNOW ABOUT MEDICARE

M

edicare plays a critically important role in ensuring that older and certain younger disabled Americans have access to health care. But the program faces serious, long-term financing problems. As a result, the American Academy of Actuaries believes that policymakers need to take action to restore the long-term solvency and financial sustainability of the program. The sooner such corrective measures are enacted into law, the more flexible and gradual the approach can be. Due to the importance of the Medicare program and the magnitude of the financial challenges, Medicare-related issues should figure prominently in the 2012 elections. This guide is intended to help voters understand how Medicare is financed, the financial challenges facing the program, and some of the options available to improve Medicare’s financial condition. Voters can use this information to encourage candidates to advance concrete proposals to improve the program’s fiscal sustainability for current and future generations of Americans.

The American Academy of Actuaries is a 17,000-member professional association whose mission is to serve the public and the U.S. actuarial profession. The Academy assists public policymakers on all levels by providing leadership, objective expertise, and actuarial advice on risk and financial security issues. The Academy also sets qualification, practice, and professionalism standards for actuaries in the United States. Copyright ©2012 American Academy of Actuaries. All rights reserved.


UNDERSTANDING MEDICARE What Is Medicare? Medicare is the federal program providing health insurance to virtually all Americans over the age of 65 and many younger long-term disabled individuals. Medicare beneficiaries can access benefits through either the traditional Medicare program or Medicare Advantage (MA) plans offered by private health insurers. Beneficiaries may purchase supplemental coverage (e.g., Medigap) to help fill in the gaps in Medicare coverage. n The traditional Medicare program provides coverage for inpatient hospital services (Part A), and for physicians and outpatient care services (Part B). n Medicare

Advantage plans are offered by private insurers. The plans must cover at least all of the services that the traditional program covers; however, they may offer extra benefits either at no additional cost or for an additional premium.

n Medicare

prescription drug benefits (Part D) are available through private insurers as a stand-alone plan to supplement traditional Medicare or as part of a Medicare Advantage plan.

How Is Medicare Funded? Medicare is funded through two separate trust funds. The two trust funds support different parts of the Medicare program and are financed in different ways. n The Hospital Insurance (HI) trust fund covers Part A and is financed largely through earmarked payroll taxes. n The

Supplementary Medical Insurance

â–˛ CAMPAIGN 2012: MEDICARE

(SMI) trust fund covers Parts B and D. The SMI trust fund is financed through beneficiary premiums, which cover roughly one-fourth of the cost, and federal general tax revenues, which cover the remaining three fourths of the cost. Both beneficiary premiums and the amount of general revenues allocated to the trust fund are reset annually based on the projected cost for the coming year. n Medicare

Advantage plans are funded through both the HI and SMI trust funds.

How Is Medicare Doing Financially? Medicare’s current financing is inadequate to sustain the program for the long term, and over time it will place increasing financial demands on both beneficiaries and the federal budget. Medicare has three fundamental long-term financial problems: 1. The payroll tax revenues supporting the HI trust fund are inadequate to fund the HI portion of Medicare benefits. Spending is projected to exceed revenues in all future years, which means the trust fund will have to draw down assets each year to pay benefits. The HI trust fund is projected to run out of assets in 2024. At that point, payroll tax revenues will cover only 87 percent of program costs and even less thereafter. Ensuring that payroll taxes would be sufficient to pay benefits over the next 75 years would require an immediate 47 percent increase in payroll taxes, an immediate 26 percent decrease in benefits, or some combination of the two. 2. Increases in SMI costs will place increasing financial pressures on both beneficiary WWW.ACTUARY.ORG

1


household budgets and the federal budget. The SMI trust fund will remain solvent because its financing is reset each year to meet future costs. Projected increases in SMI expenditures, however, will require increases in beneficiary premiums and general revenue contributions. SMI costs are projected to grow from 2.0 percent of Gross Domestic Product (GDP) in 2011 to 4.0 percent of GDP in 2085.

3. Increases in total Medicare spending threaten the program’s sustainability. Overall Medicare spending—including both HI and SMI—is projected to consume an ever-growing share of the nation’s economy, threatening the program’s longterm sustainability. Total Medicare costs are projected to grow from 3.7 percent of GDP in 2011 to 6.7 percent of GDP in 2085.

Will the Recent Health Care Reform Law Fix the Problem? Not fully. The Affordable Care Act (ACA) included a number of provisions designed to reduce Medicare spending, increase Medicare revenues, and develop new health care delivery systems and payment models to improve health care quality and cost efficiency. These have improved projections for Medicare’s financial condition. But, while this was an important first step, it did not go far enough to put Medicare back on a sound financial footing. The financial challenges described above already reflect the anticipated improvements from the ACA.

The sooner corrective measures are enacted, the more flexible and gradual the approach can be.

What’s Causing the Problem? Both the number of Americans enrolled in Medicare and the cost per enrollee are increasing. Medicare is challenged by the same rising health spending that is affecting the overall health care system. In the case of Medicare, the problem of rising health care costs is compounded by the aging of the population and the retirement of the baby boom generation.

▲ CAMPAIGN 2012: MEDICARE

WWW.ACTUARY.ORG

2


OPTIONS FOR REFORMING MEDICARE There is no one solution to Medicare’s financial problems; any solution will require policymakers to make difficult choices and will involve a combination of options. Any solution likely will require taxpayers, Medicare beneficiaries, health care providers, and private insurers to share the burden. In simple terms, improving Medicare’s financial condition will require: n Increasing revenues, n Reducing n Some

spending, or

combination of both.

What Options Are Available for Reforming Medicare? A number of specific options for improving Medicare’s financial position have been included in recent debt and deficit reduction proposals but have not been enacted. Some of these are: n Set spending targets to limit the growth in health spending. n Expand

the authority of the Independent Payment Advisory Board (IPAB).

n Transition

to a premium support or voucher

program. n Reform

the physician payment system.

n Reduce

spending for prescription drugs.

n Revise

the traditional Medicare benefit design and cost-sharing requirements.

n Raise

the Medicare eligibility age.

n Increase

Medicare Part B premiums for some or all beneficiaries.

More information on each of these options is provided below. ▲ CAMPAIGN 2012: MEDICARE

How Do You Decide Which Options Are Best? Improving the sustainability of the health system requires slowing the growth in overall health spending, rather than just shifting the costs to from one payer to another. This means that unless system-wide spending is addressed, implementing options to control Medicare spending will have limited long-term effectiveness. And while controlling costs is vital to the sustainability of the program, it is not the only consideration. Slowing the growth in health spending, while maintaining or improving the quality of care, will require provider payment and health care delivery systems that encourage integrated and coordinated care. To evaluate the various options for reforming Medicare, you should consider: n How does the reform option affect the cost of the program? n How does it affect beneficiaries’ access to care? n How

does it affect the quality of care?

n Does

it slow the growth in health spending, rather than just shifting costs from one payer to another?

n Does

it give providers the incentives to provide, and their patients the incentives to obtain, the kind of integrated and coordinated care that could help control costs and improve quality?

Options to Reform Medicare n Set Spending Targets to Limit the Growth

in Medicare Spending. Specific spending targets could be established either for WWW.ACTUARY.ORG

3


Medicare in particular or for all federal health spending. If spending exceeded the targets, it could trigger specific automatic actions such as benefit reductions or revenue increases. As an alternative, the trigger could be structured to require the president or a commission to submit proposals that would have to be considered by Congress on an expedited basis. n Expand the Authority of the Independent

Payment Advisory Board. The ACA created the IPAB to make recommendations to reduce the growth in per capita Medicare spending if that spending exceeds a targeted growth rate. IPAB recommendations would be implemented automatically unless Congress passes legislation that produces comparable savings. The type of recommendations IPAB can make, however, are limited. The IPAB could be given authority to consider a wider range of recommendations, and the expansion of scope could be tied to more ambitious targets for reducing spending growth. n Transition to a Premium Support or

Voucher Program. These proposals would change Medicare from a defined benefit plan to a defined contribution plan, meaning government would limit the amount it contributes to Medicare coverage (or private plans). Beneficiaries would pay the difference between plan premiums and the government contribution. n Reform the Physician Payment System.

Physician payment rates are governed by the Sustainable Growth Rate (SGR) system, which aims to limit the growth in Medicare spending for physician services. ▲ CAMPAIGN 2012: MEDICARE

Congress, however, typically overrides the physician fee cuts that the SGR formula would require (this override is known as the “doc fix”). Had Congress not continued to override the fee cuts, physicians would have seen a reduction in payments of almost 30 percent in 2012. Such a large decrease could have threatened beneficiary access to care. One approach to reforming Medicare physician payments would eliminate the SGR, temporarily freeze physician fees at their current level, and replace the SGR with a new physician payment system. Such an option would increase Medicare spending, however, unless it is offset by Medicare spending reductions. n Reduce Spending for Prescription

Drugs. Proposals to reduce spending for prescription drugs would require Medicare to negotiate drug prices under Part D, extend drug rebates to individuals who are eligible for both Medicare and Medicaid, or establish a government-run Part D option. Reducing prescription drug prices would lower Part D spending and beneficiary premiums. n Revise the Design of Traditional Medicare.

The benefit design for beneficiaries enrolling in traditional Medicare (as opposed to private Medicare Advantage plans) has several shortcomings. The lack of an outof-pocket maximum leaves beneficiaries unprotected against catastrophic costs; most beneficiaries have supplemental cov­ erage (e.g., Medigap) with low cost-sharing requirements that reduce incentives to seek cost-effective care; and the cost-sharing structure is not ideal for influencing consumer behavior. Updating the traditional WWW.ACTUARY.ORG

4


cost-sharing features could help better align beneficiary incentives to seek cost-effective care. Meeting this goal, however, may require changes to supplemental coverage as well. Proposals to update the traditional benefit design would change or combine the Part A and B cost-sharing requirements, add a maximum out-of-pocket limit, and/or eliminate first-dollar coverage in supplemental plans (or apply an additional charge to those plans). n Raise the Medicare Eligibility Age.

The current eligibility age for Medicare is 65; the normal retirement age for Social Security has been increased to age 67. There have been proposals to increase the Medicare eligibility age and perhaps even index it to increases in longevity. This would reduce Medicare costs, but the savings would be offset partially by increased federal spending in other areas such as Medicaid and the premium subsidies available through the new health insurance exchanges created by the ACA.

n Increase Part B Premiums for Some or All

Beneficiaries. Most Medicare beneficiaries pay the standard Part B premium, currently set to cover25 percent of the average cost of Part B benefits. Higher-income beneficiaries, however, pay between 35 and 80 percent of the average cost. Some proposals would increase Part B premiums for those not already subject to higher premiums or raise them higher for those who are already paying relatively higher premiums. This would increase Medicare revenues by shifting costs to beneficiaries, but would not affect Medicare spending.

▲ CAMPAIGN 2012: MEDICARE

What You Can Do Understand that There Is No Silver Bullet

There is no one, simple solution for shoring up Medicare. Ensuring that Medicare benefits are payable in the future almost certainly will require shared responsibility from Medicare beneficiaries, taxpayers, health care providers, and private insurers. When it comes to reform, sooner is better than later. Improving Medicare’s long-term solvency and sustainability ultimately will require slowing the growth in health spending rather than just shifting costs from one payer to another. Slowing the growth in health spending, while maintaining or improving quality, will require provider payment and health care delivery systems that encourage integrated and coordinated care. Learn as Much as You Can The more you know about how Medicare works, its financial condition, and the options available for reform, the better equipped you will be to evaluate what candidates have to say about the program. You may want to start with the following Academy publications: n Medicare’s Financial Condition: Beyond Actuarial Balance (May 2012) n An

Actuarial Perspective on Proposals to Improve Medicare’s Financial Condition (May 2011)

n Revising

Medicare’s Fee-For-Service Benefit Structure (March 2012)

Speak Out Encourage candidates for federal office to detail their approaches for putting Medicare on a sound financial footing. Ask Congress and the president to ensure the long-term future of the program.

WWW.ACTUARY.ORG

5


Mary Downs, Executive Director Craig Hanna, Director of Public Policy Cori Uccello, Senior Health Fellow Heather Jerbi, Senior Health Policy Analyst Mark Cohen, Director of Communications Members of the Health Practice Council Communications Task Force: Thomas F. Wildsmith, MAAA, FSA, chairperson; Mark J. Jamilkowski, MAAA, FSA; Darrell D. Knapp, MAAA, FSA; Donna C. Novak, MAAA, ASA, FCA; David A. Shea, Jr., MAAA, FSA; Cori E. Uccello, MAAA, FSA, FCA, MPP; Shari A. Westerfield, MAAA, FSA.

For further information, contact us at:

American Academy of Actuaries 1850 M Street NW, Suite 300 Washington, DC 20036-5805 Telephone 202 223 8196 Facsimile 202 872 1948 WWW.ACTUARY.ORG

â–˛ CAMPAIGN 2012: MEDICARE

WWW.ACTUARY.ORG

6


American Academy of Actuaries

MARCH 2009

May 2011

Key Points n When evaluating proposals to improve

Medicare’s financial condition, it’s important to recognize that improving the sustainability of the health system as a whole requires slowing the growth in overall health spending rather than shifting costs from one payer to another. n New payment and delivery system models

have the potential to control costs and improve quality by better aligning incentives to encourage integrated and coordinated care.

Additional Resources Medicare Advantage Payment Reform: http:// www.actuary.org/pdf/health/ma_oct09.pdf Health Insurance Coverage and Reimbursement Decisions: Implications for Increased Comparative Effectiveness Research: http:// www.actuary.org/pdf/health/comparative. pdf Value-Based Insurance Design: http://www. actuary.org/pdf/health/vbid_june09.pdf

An Actuarial Perspective on Proposals to Improve Medicare’s Financial Condition

M

edicare plays a critically important role in ensuring access to health care among Americans age 65 and older and certain younger adults with permanent disabilities. Yet, rising health care spending threatens the sustainability of the Medicare program and the overall health system. Moreover, rising health spending threatens the nation’s fiscal health. Provisions in the Affordable Care Act (ACA) have improved Medicare’s financial condition. Nevertheless, putting the country on a more sustainable fiscal path requires additional efforts to slow health spending growth. To this end, debt and deficit reduction proposals put forward by various groups, such as the National Commission on Fiscal Responsibility and Reform, include provisions to control health spending.1 Related legislation has also been introduced. The American Academy of Actuaries’ Medicare Steering Committee supports continuing efforts by the president and Congress to address these challenges and urges further action to restore the long-term solvency and sustainability of Medicare. To assist in those efforts, this paper outlines many of the Medicare-related provisions in the various debt and deficit reduction proposals. For each proposal, a summary of the key cost, access, and quality issues from an actuarial perspective is provided. In future work, the 1 See The Henry J. Kaiser Family Foundation, “Comparison of Medicare Provisions in Deficit-Reduction Proposals,” (last modified April 4, 2011) for a side-by-side comparison of key Medicare changes recommended by various debt and deficit reduction proposals.

The American Academy of Actuaries is a 17,000-member professional association whose mission is to serve the public and the U.S. actuarial profession. The Academy assists public policymakers on all levels by providing leadership, objective expertise, and actuarial advice on risk and financial security issues. The Academy also sets qualification, practice, and professionalism standards for actuaries in the United States. ©2011 The American Academy of Actuaries. All Rights Reserved.

1850 M Street NW, Suite 300, Washington, DC 20036 Tel 202 223 8196, Fax 202 872 1948 www.actuary.org Mary Downs, Executive Director Mark Cohen, Director of Communications Craig Hanna, Director of Public Policy Cori Uccello, Senior Health Fellow Heather Jerbi, Senior Health Policy Analyst


committee plans to examine many of these options in more detail. When evaluating proposals to improve Medicare’s financial condition, it’s important to recognize that improving the sustainability of the health system as a whole requires slowing the growth in overall health spending rather than shifting costs from one payer to another. So unless system-wide spending is addressed, implementing options to control Medicare spending will have limited long-term effectiveness.

Limit the Growth in Medicare Spending Some current proposals would set spending targets, either for Medicare in particular, or for federal health spending in total. Exceeding those targets could trigger specific actions, such as automatically reducing benefits or increasing revenues. The trigger, alternatively, could be structured to require the president or a commission to submit proposals that would be considered by Congress on an expedited basis. One approach, for instance, would set target spending for all federal health expenditures at the growth in gross domestic product (GDP) plus 1 percent. If the target is exceeded, the president would be required to submit proposals to reduce spending. Another approach automatically would reduce fee-forservice provider payments by 1 percent if general revenue contributions to Medicare exceed 45 percent of Medicare funding. (As discussed

below, the ACA created the Independent Payment Advisory Board, or IPAB, which focuses on reducing Medicare spending if it exceeds a targeted growth rate. As currently structured, the IPAB is somewhat restricted on what options it can recommend.) Cost: Medicare savings would depend on how aggressively the spending targets are set. Savings to the health system overall, however, would be offset to the extent that costs are instead shifted to Medicare beneficiaries or other payers. Access/Quality: The impact on the access to and quality of care would depend on the specific recommendations made. Depending on how the reductions are structured, reducing provider payment rates could reduce beneficiary access to care and/or the quality of care. Other specific options for reducing benefit costs or increasing revenues are examined in other sections of this paper.

Transition to a Premium Support or Voucher Program Some proposals would transition Medicare to a premium support or voucher program, while others offer such an approach as an option if certain measures to reduce Medicare spending growth are not deemed adequate. These approaches would change the Medicare program from a defined benefit plan to a defined contribution plan. Under a premium support approach, the

Members of the Medicare Steering Committee include: Edwin C. Hustead, MAAA, FSA, EA, Chairperson; Dennis J. Hulet, FSA, MAAA, FCA, Vice Chairperson; Jill H. Brostowitz, FSA, MAAA; Janet M. Carstens, FSA, MAAA, FCA; Michael V. Carstens, FSA, MAAA; April S. Choi, FSA, MAAA; Randall S. Edwards, FSA, MAAA; P. Anthony Hammond, ASA, MAAA; Troy M. Filipek, FSA, MAAA, FCA; Harry Hotchkiss, FSA, MAAA; Joel C. Kabala, FSA, MAAA; Margot D. Kaplan, ASA, MAAA, FCA; Laura Beth Lieberman, FSA, MAAA; Walter T. Liptak, ASA, MAAA, EA; Mark E. Litow, FSA, MAAA; Donna C. Novak, ASA, MAAA, FCA; Susan E. Pierce, MAAA, FSA; Anna M. Rappaport, FSA, MAAA, EA; Steven Rubenstein, MAAA, ASA; Jeremiah D. Reuter, ASA, MAAA; Jay C. Ripps, FSA, MAAA; John Sardelis, MAAA; Paul A. Schultz, FSA, MAAA; Gordon R. Trapnell, FSA, MAAA; Cori E. Uccello, FSA, MAAA, FCA, MPP; John A. Wandishin, FSA, MAAA; Thomas F. Wildsmith, FSA, MAAA; Carl Wright, MAAA, FSA. 2

Issue Brief May 2011


government would limit the amount it contributes toward Medicare coverage, with beneficiaries paying additional premiums to cover any difference between plan premiums and the government contribution. The growth in government contributions would be indexed by inflation or some other factor. Under a voucher-type approach, individuals would receive a voucher to purchase private health insurance. The voucher could be adjusted by various beneficiary characteristics—such as age, health status, geographic location, and/or income—and would be indexed by inflation or some other factor. Cost: Moving to a defined contribution approach would shift the risk of health spending growth away from the government and toward beneficiaries. Depending on how the government contribution is set, federal Medicare spending could be lower than currently projected. To the extent that health spending growth exceeds the increase in the government contribution, costs would be shifted to beneficiaries through higher premiums and/ or higher cost sharing. As discussed below, increased cost-sharing requirements could lower spending growth due to reduced utilization. The impact of such an approach on overall health spending would also depend on how utilization management, administrative costs, and provider payment rates under private plans would compare to those under traditional Medicare. Access/Quality: Access to Medicare or private insurance would depend on the difference between the government contribution and the premium. The greater the share

of costs that are shifted from the government to beneficiary premiums, the more likely that beneficiaries will opt for less generous plans. Although this could encourage beneficiaries to seek more cost-effective care, some may forgo needed care. In addition, to bring costs down, care quality might be compromised. Such a system, for instance, might lead to a less-expensive second tier delivery system, which may be much more limited in the types of providers available.

Expand the Authority of the Independent Payment Advisory Board (IPAB) The ACA created the IPAB, which is similar to the Medicare Payment Advisory Commission (MedPAC).2 The IPAB is charged with preparing recommendations to reduce the growth in Medicare per capita expenditures if spending exceeds a targeted growth rate. The targets are based on inflation until 2019, and on GDP plus 1 percent thereafter. Unlike MedPAC recommendations, IPAB recommendations would be implemented automatically unless the Congress passes legislation producing comparable reductions. The board is somewhat restricted in its recommendations—it cannot propose to ration health care, raise revenues, increase beneficiary premiums or cost sharing, or otherwise restrict benefits or modify eligibility criteria.3 In addition, until 2020 most hospital services are excluded from the scope of payment changes that can be recommended. Provisions included in various fiscal proposals would expand the scope of the IPAB, by eliminating the temporary carve-outs for hospital services, allowing options related to

2 MedPAC would continue its role as advisor to Congress on issues affecting the Medicare program and would review any IPAB proposals. 3 Section 3403 of the Affordable Care Act: http://docs.house.gov/energycommerce/ppacacon.pdf.

Issue Brief May 2011

3


cost sharing and benefit design, and giving it authority over all federal health spending. The expansion of scope could be tied to directing IPAB to meet more ambitious spending growth targets. Cost: To the extent that the spending growth targets are tightened, additional Medicare cost savings could be achieved, compared to current law. However, total savings would be offset to the extent that costs are shifted to beneficiaries. Access/Quality: The impact on the access to and quality of care would depend on the specific recommendations made. Options to revise Medicare’s plan design are examined in more detail below.

Reform the Sustainable Growth Rate System The sustainable growth rate (SGR) system was enacted as part of the Balanced Budget Act of 1997 to limit the growth in spending for Medicare physician services. The system compares actual cumulative spending for Medicare physician services to a specified spending target. If actual spending exceeds the target, then physician payment updates are adjusted downward. With the exception of 2002, the first year that physician fee cuts were called for under the SGR formula, the fee cuts have been temporarily overridden each year by Congress (i.e., the “doc fix”). As a result of the cumulative shortfall, physician payment rates will be reduced by nearly 30 percent in 2012, barring another override from Congress. By putting pressure on physician payment

updates, the SGR system might have resulted in slower growth in physician payment updates than would have occurred otherwise. There are calls, nevertheless, to reform or eliminate the SGR system due to concerns regarding beneficiary access to care under large fee cuts, provider frustration regarding the short-term nature of payment fixes, the growing budgetary costs of further overrides, and the way the system’s across-the-board fee cuts poorly target those providers with the highest volume increases.4,5 One approach would eliminate the SGR, temporarily freeze physician payments, and develop a new physician payment system. The proposal would pay for the elimination of the SGR by other reductions in Medicare and Medicaid spending. Cost: Officially eliminating the SGR would increase Medicare spending over baseline projections including the SGR, unless offset by other spending reductions. Access/Quality: Eliminating the SGR could help maintain beneficiaries’ access to care. Depending on how a new physician payment system would be developed, it could better align payments with the provision of highvalue care.

Reduce Spending for Prescription Drugs Provisions included in various proposals would reduce payments for prescription drugs. One option would be to increase drug rebates by requiring Medicare to use its bargaining power to negotiate drug prices under the Part D program. Another option would extend drug rebates to those eligible

4 Medicare Payment Advisory Commission (MedPAC), Report to the Congress: Medicare Payment Policy (Chapter 4), March 2011. 5 The Congressional Budget Office (CBO) estimates that replacing the SGR with a 10-year physician payment freeze would cost about $250 billion; if payments were increased over time, the cost would be even greater. (The Budget and Economic Outlook: Fiscal Years 2011 to 2021, January 2011.)

4

Issue Brief May 2011


for both Medicare and Medicaid. Another approach would establish a government-run Part D option that would be offered alongside Part D private plans. The Centers for Medicare and Medicaid (CMS) would negotiate prices with prescription drug companies. However, as with Medicare Parts A and B, this ultimately could lead to CMS setting prescription drug prices. Cost: By reducing the prices paid for prescription drugs, these options would lower Part D spending and reduce its growth rate. To the extent that prescription drug companies can respond by increasing their prices in the private sector, costs would be shifted from Medicare to the private sector. Lowering Part D spending would also reduce beneficiary premiums for Part D plans. In some cases the copayments for some prescription drugs could also be reduced. Access/Quality: Reducing the prices paid for prescription drugs potentially could reduce research and development in the pharmaceutical industry. Introducing a government-run Part D option could lead to some current Part D providers leaving the market, especially if the government-run plan sets drug prices— thereby reducing the choices available to enrollees.

Revise Medicare’s Fee-For-Service (FFS) Benefit Design and Cost-Sharing Requirements Medicare, like most other health insurance plans, uses patient cost-sharing requirements (e.g., deductibles, copayments, coinsurance) to help balance plan affordability with the com-

prehensiveness of coverage. Patient cost sharing directly lowers Medicare spending by shifting a share of medical costs to the beneficiary. In addition, cost sharing can lower spending overall by reducing utilization. Patient costsharing requirements ideally align beneficiary incentives with program goals to provide quality and cost-effective care. However, Medicare’s fee-for-service (FFS) cost-sharing requirements are not currently structured to meet these goals. In particular: n The FFS cost-sharing requirements are skewed more toward less discretionary services, with high deductibles for Part A inpatient services and lower deductibles for Part B physician and outpatient services; n

Most beneficiaries have supplemental policies to fill in most or all FFS cost-sharing requirements, thereby reducing the incentives for beneficiaries to seek cost-effective care;6

n

The lack of an out-of-pocket maximum under FFS leaves beneficiaries unprotected against catastrophic health costs.

Provisions in various proposals would increase and/or restructure Medicare’s costsharing requirements. A number of proposals would combine or restructure the Part A and Part B cost-sharing requirements and add a new maximum out-of-pocket limit. (Medicare Advantage plans have some flexibility on how to structure cost-sharing requirements, and as of 2011, are required to cap out-of-pocket spending.) Some of these proposals would also eliminate first-dollar coverage in Medigap plans and/or prohibit supplemental insurance from covering any new or increased cost-shar-

6 MedPAC reports that 89 percent of FFS beneficiaries in 2005 had supplemental coverage: 33 percent had individually purchased Medigap coverage, 37 percent had employer-sponsored coverage, 17 percent had Medicaid, and 2 percent had other public coverage. See Report to the Congress: Improving Incentives in the Medicare Program (Chapter 6), June 2009.

Issue Brief May 2011

5


ing amounts. Taken together, these changes could help encourage Medicare beneficiaries to seek cost-effective care. A value-based insurance design (VBID) also could encourage the use of cost-effective care. A VBID approach would lower the cost sharing for high-value services and increase the cost sharing for lowvalue services. The ACA moved Medicare in this direction by covering certain preventive services with no cost sharing. Comparative effectiveness research can facilitate the identification of low- and high-value services. Cost: Increasing Medicare’s cost-sharing requirements would reduce Medicare spending by shifting more of the costs to beneficiaries. Savings could also result by lowering utilization, especially if supplemental plans are prohibited from covering the difference. Adding an out-of-pocket cap would offset cost savings. Adjusting cost sharing to align incentives with effective use of services has shown promise in reducing spending in the non-Medicare market—most often for prescription drugs.7 Access/Quality: A restructuring of Medicare’s cost-sharing requirements could better align beneficiary incentives for high-quality and cost-effective care. In addition, incorporating a maximum out-of-pocket limit would provide the catastrophic protection that the FFS program currently lacks. Such a restructuring would increase out-of-pocket spending for many beneficiaries, but decrease it for those with the greatest health care needs. Broad increases in cost sharing, rather than targeted increases, have been shown to reduce not only unnecessary care, but also necessary care, especially among the low income and chronically ill. For this reason, some propos-

als would exempt lower-income beneficiaries from cost sharing increases. In addition, a VBID approach could incorporate lower costsharing requirements for chronic treatments.

Raise the Medicare Eligibility Age Since the program began in 1965, beneficiaries have been eligible for full Medicare benefits at age 65, consistent with Social Security’s normal retirement age at that time. Since that time, the normal retirement age for Social Security has been increased to age 67 and there are currently proposals to increase it beyond 67. Similarly, there are proposals to gradually increase the Medicare eligibility age (e.g., to age 67 or 69), and some also would index the eligibility age for increased longevity. Cost: Raising the Medicare eligibility age would reduce the cost of the Medicare program and could increase payroll tax revenues by encouraging individuals to work beyond age 65. However, the increased revenues would be offset by increased federal spending to the extent that individuals between age 65 and the new eligibility age receive premium subsidies through the health insurance exchanges or coverage through Medicaid. In addition, some costs would be shifted to employers, states, and individuals. Access/Quality: People between age 65 and the new eligibility age would have to find a new source of health insurance—through employer coverage, the individual market or health insurance exchanges, or other public coverage such as Medicaid—or go uninsured. Provisions in the ACA increase the availability of other coverage sources. In particular,

7 See for instance, “Evidence That Value-Based Insurance Can Be Effective,” Michael E. Chernew, et al. Health Affairs 29(3): 530-536, March 2010.

6

Issue Brief May 2011


beginning in 2014, the ACA requires that private health insurance coverage be offered on a guaranteed-issue basis, prohibits preexisting condition exclusions, and limits premium variations by age. Low- and moderate-income individuals may be eligible for premium and cost-sharing subsidies or Medicaid coverage. Shifting individuals between age 65 and the new eligibility age into private plans would increase average premiums for private plans. This could potentially reduce insurance coverage among younger individuals if their premiums increase as a result.

Increase Medicare Part B Premiums Medicare Part B premiums, initially set at 50 percent of Part B costs, currently are set at 25 percent of costs. Beginning in 2007, premiums for higher-income beneficiaries were raised to between 35 and 80 percent of costs, depending on income. The ACA temporarily freezes the index on income thresholds used to determine the premiums, which means more beneficiaries will be subject to higher premiums over time. Some proposals would increase the Part B premiums for those not already subject to higher premiums or raise them higher for those already subject to higher premiums. Cost: Increasing Medicare premiums would increase program revenues by shifting costs to beneficiaries. But it would not reduce Medicare spending (unless some beneficiaries decide to opt out of Medicare Part B due to the higher premiums). Access/Quality: Beneficiaries who are unwilling or unable to pay higher Part B premiums may face reduced access to care.

Next Steps This paper provides a brief overview of the various Medicare-related provisions put forward as part of proposals aimed at improving the nation’s fiscal condition. In future work, the American Academy of Actuaries’ Medicare Steering Committee plans to explore in more detail many of these and other options. The focus will be not only on whether an option helps improve Medicare’s financial condition, but also on whether it improves the sustainability of the health system as a whole by slowing the growth in overall health spending. In addition, the committee intends to examine new programs in the ACA that were included to jumpstart reforms to the health care delivery system. The Medicare Shared Savings Program, for instance, will facilitate the creation of Accountable Care Organizations (ACOs). The newly created Center for Medicare and Medicaid Innovation (CMI) will identify and test new models of health care delivery and payment and speed the expansion of successful models. By better aligning incentives to encourage integrated and coordinated care, ACOs and other new payment and delivery system models have the potential to control costs and improve quality.

Issue Brief May 2011

7


1850 M Street NW Suite 300 Washington, DC 20036 Tel 202 223 8196 Fax 202 872 1948 www.actuary.org


August 31, 2011 Dear Senator/Representative: As you begin the challenging work of the Joint Select Committee on Deficit Reduction, I urge you to use this opportunity to develop sound public policy proposals to improve the long-term solvency and sustainability of the Medicare program. With Medicare’s critical role of ensuring access to health care for Americans age 65 and older and certain younger adults with permanent disabilities, we as a nation cannot afford to have this important program continue on its current financial path. Rising health care spending threatens not only the sustainability of the Medicare program and the overall health system but also the nation’s fiscal health. As you consider potential Medicare reforms in the context of deficit reduction, it is important to evaluate the impact those reforms could have on the viability of the Medicare program including cost, access, and quality of care. In addition, improving Medicare’s long-term sustainability requires slowing the growth in overall health spending—not simply shifting costs from one payer to another. Medicare’s sustainability is a core issue for the American Academy of Actuaries1 and, as one of our highest public policy priorities, we would welcome the opportunity to be of assistance to you as you consider reforms that would affect Medicare and/or the health system as a whole. We are committed to providing objective actuarial information and analysis related to Medicare and other health-related issues. To orient you to some of our ongoing work, our Medicare Steering Committee has issued several recent publications that we commend to your attention, including: 

An Actuarial Perspective on Proposals to Improve Medicare’s Financial Condition (May 2011) summarizes the key cost, access, and quality issues associated with some of the Medicare-related provisions in the various debt- and deficit-reduction proposals. http://www.actuary.org/pdf/Medicare_Financial_IB_Final_051211.pdf Medicare’s Financial Condition: Beyond Actuarial Balance (May 2011) highlights the key findings in the 2011 Medicare Trustees’ Report. http://www.actuary.org/pdf/health/Medicare%20Financial%20IB%20Final%20052511.pdf Presentation from a recent Capitol Hill briefing the Academy hosted that break down some of the trustees’ key findings as well as outlining several options for reforming the program. http://www.actuary.org/pdf/health/MedicareTrusteesBriefing_Presentation_110527.pdf

In addition to these documents, we have a number of other publications on our website that address payment and delivery system reforms, which have the potential to control costs and improve quality

1

The American Academy of Actuaries is a 17,000-member professional association whose mission is to serve the public and the U.S. actuarial profession. The Academy assists public policymakers on all levels by providing leadership, objective expertise, and actuarial advice on risk and financial security issues. The Academy also sets qualification, practice, and professionalism standards for actuaries in the United States.


through better alignment of incentives to encourage integrated and coordinated care. All of these publications can be found at http://www.actuary.org/issues/health_reform_implementation.asp. If you have any questions or if you would like to discuss in more detail any implications various reforms may have on the Medicare program, please contact Heather Jerbi, the Academy’s senior health policy analyst (202.785.7869; Jerbi@actuary.org). Sincerely, Edwin C. Hustead, MAAA, FSA, EA Chairperson, Medicare Steering Committee American Academy of Actuaries Cc: Mark Prater, Staff Director, Joint Select Committee on Deficit Reduction


American Academy of Actuaries

MARCH 2009

JUNE 2013

Key Points n

n

n

n

The HI trust fund will be depleted in 2026, two years later than projected last year. Total Medicare expenditures will make up an increasing share of federal outlays and the gross domestic product (GDP), threatening the program’s long-term sustainability. An alternative scenario illustrates the potential understatement of current-law projections if cur­rently scheduled provider payment reductions are not realized. Changes are needed to improve Medicare’s longterm solvency and sustainability. The sooner such corrective measures are enacted, the more flexible the approach and the more gradual the implementation can be.

Additional Resources Revising Medicare’s Fee-For-Service Benefit Structure: http://www.actuary.org/pdf/health/ Medicare_Fee_Struc_Issue_Brief_022712.pdf A Guide to Analyzing Medicare Premium Support: http://www.actuary.org/files/Issue_Guide_Medicare_Premium_021113.pdf An Actuarial Perspective on Proposals to Improve Medicare’s Financial Condition: http://www. actuary.org/pdf/Medicare_Financial_IB_Fi­ nal_051211.pdf

Medicare’s Financial Condition: Beyond Actuarial Balance

E

ach year, the Boards of Trustees of the Federal Hospital Insurance (HI) and Supplementary Medical Insurance (SMI) Trust Funds report to Congress on the Medicare program’s financial condition. Medicare plays a critically important role in ensuring access to health care among Americans age 65 and older and certain younger adults with permanent disabilities. The program is operated through two trust funds. The HI trust fund (Medicare Part A) pays primarily for hospital services. The SMI trust fund includes accounts for the Medicare Part B program, which covers physician and outpatient hospital services, and the Medicare Part D program, which covers the prescription drug program. The trustees’ report is the primary source of information on the financial status of the Medicare program, and the American Academy of Actuaries proudly recognizes the important contribution that members of the actuarial profession have made in preparing the report and educating the public about the important issues surrounding the program’s solvency and sustainability. The projected financial condition of Medicare as identified in the 2013 Medicare trustees’ report has improved compared with the projections from the 2012 report. The year in which the HI trust fund is projected to be depleted is now 2026, two years later than projected last year. The 75-year HI deficit decreased from 1.35 percent of taxable payroll in last year’s report to 1.11 percent of taxable payroll in the 2013 report. This improvement is due to

The American Academy of Actuaries is a 17,000-member professional association whose mission is to serve the public and the U.S. actuarial profession. The Academy assists public policymakers on all levels by providing leadership, objective expertise, and actuarial advice on risk and financial security issues. The Academy also sets qualification, practice, and professionalism standards for actuaries in the United States. ©2013 The American Academy of Actuaries. All Rights Reserved.

1850 M Street NW, Suite 300, Washington, DC 20036 Tel 202 223 8196, Fax 202 872 1948 www.actuary.org Mary Downs, Executive Director Charity Sack, Director of Communications Craig Hanna, Director of Public Policy Cori Uccello, Senior Health Fellow Heather Jerbi, Assistant Director of Public Policy


more recent data and technical changes in projection methods. HI expenditures are projected to exceed HI revenues in most years of the 75-year projection period. Total Medicare expenditures will make up an increasing share of federal outlays and the gross domestic product (GDP). As required by statute, the trustees’ projections of Medicare’s financial outlook are based on benefits and revenues scheduled under current law. The trustees acknowledge, however, that these estimates likely understate the seriousness of Medicare’s financial condition. In the Statement of Actuarial Opinion that accompanies the trustees’ report, Paul Spitalnic, the acting chief actuary of the Centers for Medicare & Medicaid Services (CMS), specifically notes that actual Medicare expenses are likely to exceed the current-law projections. He states, “the financial projections shown in [the] report for Medicare do not represent a reasonable expectation for actual program operations in either the short range…or the long range…” In particular, the trustees and the acting chief actuary point to scheduled reductions in provider payments that are unlikely to occur. Currently scheduled physician payment reductions in accordance with

the sustainable growth rate (SGR) mechanism are considered likely to be overridden by Congress (i.e., the Medicare “doc fix”). In addition, current law requires downward adjustments in payment updates for most non-physician providers to reflect productivity improvements; these adjustments might not be sustainable in the long term. At the request of the trustees, the CMS Office of the Actuary developed an alternative analysis that provides an illustration of the potential understatement of current-law Medicare cost projections if the physician payment reductions are overridden, the productivity adjustments are phased down, and there are no savings from the Independent Payment Advisory Board (IPAB). Although the illustrative alternative projections are not intended to be interpreted as the official best estimates of future Medicare costs, they do, as noted in the alternative analysis, “help illustrate and quantify the potential magnitude of the cost understatement under current law.” This issue brief presents projections based on both current law and the illustrative alternative projections.1 The trustees conclude: “The projections in this year’s report continue to demonstrate the need for timely and effective

Both the 2013 Medicare Trustees Report and the CMS Office of the Actuary’s illustrative alternative scenario analysis are available at: http://www.cms.gov/ReportsTrustFunds/. 1

Members of the Medicare Steering Committee include: Thomas F. Wildsmith, FSA, MAAA, Chairperson; Dennis J. Hulet, FSA, MAAA, FCA, Vice Chairperson; Joe Bawazer, ASA, MAAA; Jill H. Brostowitz, FSA, MAAA; Michael V. Carstens, FSA, MAAA; April S. Choi, FSA, MAAA; Randall S. Edwards, FSA, MAAA; Troy M. Filipek, FSA, MAAA, FCA; Joel C. Kabala, ASA, MAAA; Margot D. Kaplan, ASA, MAAA, FCA; Mark E. Litow, FSA, MAAA; Steve Niu, FSA, MAAA, EA; Susan E. Pierce, MAAA, FSA; Robert J. Pipich, FSA, MAAA; Anna M. Rappaport, FSA, MAAA, EA; Jeremiah D. Reuter, ASA, MAAA; Gordon R. Trapnell, FSA, MAAA; Cori E. Uccello, FSA, MAAA, FCA, MPP; John A. Wandishin, FSA, MAAA; Carl Wright, MAAA, FSA. 2

AMERICAN ACADEMY OF ACTUARIES ISSUE BRIEF JUNE 2013


action to address Medicare’s remaining financial challenges—including the projected depletion of the HI trust fund, this fund’s long-range financial imbalance, and the issue of rapid growth in Medicare expenditures. Furthermore, if the lower prices payable for health services under Medicare cannot be sustained, then these further policy reforms will have to address much larger financial challenges than implied by the current-law projections.” This issue brief more closely examines the findings of the trustees’ report with respect to program solvency and sustainability. The American Academy of Actuaries’ Medicare Steering Committee concurs that the Medicare program faces serious financing problems. As highlighted in the 2013 Medicare trustees’ report and its accompanying illustrative alternative analysis: n The HI trust fund is projected to be depleted in 2026, two years later than projected in last year’s report. n The HI trust fund faces serious longterm funding challenges. HI expenditures are expected to exceed HI revenues in most future years. In the year that the trust fund is projected to be depleted—2026—tax revenues would cover only 87 percent of program costs. n The projected HI deficit over the next 75 years is 1.11 percent of taxable payroll.2 Eliminating this deficit would require an immediate 38 percent increase in standard payroll taxes or an immediate 22 percent reduction in benefits—or some combination of the

two. Delaying action would require more drastic tax increases or benefit reductions in the future. n Under the illustrative alternative scenario, the HI trust fund would be depleted a few months earlier in 2026 and the 75-year HI deficit would be 2.17 percent of taxable payroll. n The SMI trust fund is expected to remain solvent because its financing is reset each year to meet projected future costs. Projected increases in SMI expenditures will require significant increases over time in beneficiary premiums and general revenue contributions. Under current-law projections, SMI spending is expected to grow from 2.0 percent of GDP in 2012 to 4.0 percent of GDP in 2085. Under the illustrative alternative scenario, SMI spending is expected to reach 5.6 percent of GDP in 2085. n Total Medicare expenditures also are projected to increase as a share of GDP, thereby threatening Medicare’s longterm sustainability. Under current-law projections, total Medicare spending as a share of GDP is expected to grow from 3.6 percent in 2012 to 6.5 percent in 2085. Under the illustrative alternative scenario, total Medicare spending is projected to reach 9.6 percent of GDP in 2085. Because Medicare plays a critically important role in ensuring that older and certain disabled Americans have access to health care, the American Academy of Actuaries’ Medicare Steering Committee

The current HI payroll tax rate is 1.45 percent of taxable earnings, payable by both employees and their employers for a total of 2.90 percent. Self-employed individuals pay both shares. Beginning in 2013, earnings exceeding $200,000 for individuals and $250,000 for married couples filing jointly are subject to an additional HI tax of 0.9 percent. 2

AMERICAN ACADEMY OF ACTUARIES ISSUE BRIEF JUNE 2013

3


urges action to restore the long-term solvency and financial sustainability of the program. The sooner such corrective measures are enacted, the more flexible the approach and the more gradual the implementation can be. Failure to act now will necessitate far more drastic actions later.

improvement results from more recent data and technical changes in projection methods. The projected trust fund exhaustion date is two years later than in last year’s report, and the 75-year HI deficit decreased from 1.35 percent of taxable payroll to 1.11 percent. n

HI expenditures currently exceed HI revenues. The gap is projected to narrow over the next few years, becoming a surplus for a few years before HI expenditures are expected to exceed revenues, including interest income, for the remainder of the 75-year projection period. The HI trust fund assets, therefore, will need to be redeemed. If the federal government is experiencing unified budget deficits, funding the redemptions will require that additional money be borrowed from the public, thereby increasing the federal deficit and debt.

n

The HI trust fund is projected to be depleted in 2026. At that time, tax revenues are projected to cover only 87 percent of program costs, with the share declining to 71 percent in 2050. In 2085, payroll tax revenues are projected to cover 73 percent of program costs. There is no current provision for general fund transfers to cover HI expenditures in excess of dedicated revenues.

n

The projected HI deficit over the next 75 years is 1.11 percent of taxable payroll. Eliminating this deficit would require an immediate 38 percent increase in standard payroll taxes or a 22 percent reduction in benefits—or some combination of the two. Delaying action would require more drastic changes in the future.

MEDICARE FINANCING PROBLEMS The Medicare program has three fundamental long-range financing challenges: 1. Income to the HI trust fund is not adequate to fund the HI portion of Medicare benefits; 2. Increases in SMI costs increase pressure on beneficiary household budgets and the federal budget; 3. Increases in total Medicare spending threaten the program’s sustainability. Each of these problems is discussed in more detail below.

Medicare HI Trust Fund Income Falls Short of the Amount Needed To Fund HI Benefits Like Social Security, Medicare relies on trust funds to account for all income and expenditures. The HI and SMI programs operate separate trust funds with different financing mechanisms. General revenues, payroll taxes, premiums, and other income are credited to the trust funds, which are used to pay benefits and administrative costs. Any unused income is required by law to be invested in U.S. government securities for use in future years. In effect, the trust fund assets represent loans to the U.S. Treasury’s general fund. The HI trust fund, which pays for hospital services, is funded primarily through earmarked payroll taxes. The projections of Medicare’s financial outlook in the trustees’ report must be based on current law. Under these current-law projections, the financial condition of the HI trust fund has improved since the 2012 trustees’ report. This 4

AMERICAN ACADEMY OF ACTUARIES ISSUE BRIEF JUNE 2013

Current-law projections, however, likely understate the fiscal challenges to the Medicare HI trust fund. In particular, the scheduled reductions in provider payment rate updates to reflect productivity adjustments may not be sustainable in the long term. At the request of the trustees,


the CMS Office of the Actuary provided an illustrative alternative analysis that phases down the productivity adjustments gradually over 15 years, beginning in 2020, from about 1.1 percent to 0.4 percent and assumes no savings from IPAB. Under the illustrative alternative scenario, the HI trust fund also would be depleted in 2026, but the projected deficit over the next 75 years would be 2.17 percent of taxable payroll—compared to 1.11 percent under current-law projections. Eliminating this deficit would require an immediate 75 percent increase in standard payroll taxes or a 36 percent reduction in benefits— or some combination of the two.

Increases in SMI Costs Increase Pressure on Beneficiary Household Budgets and the Federal Budget The SMI trust fund includes accounts for the Medicare Part B program, which covers physician and outpatient hospital services, and the Medicare Part D program, which covers the prescription drug program. Approximately one-quarter of SMI spending is financed through beneficiary premiums, with federal general tax revenues covering the remaining three-quarters.3 The SMI trust fund is expected to remain solvent because its financing is reset each year to meet projected future costs. As a result, increases in SMI costs will require increases in beneficiary premiums and general revenue contributions. Increases in general revenue contributions will put more pressure on the federal budget.

Premium increases similarly will place pressure on beneficiaries, especially when considered in conjunction with increasing beneficiary cost-sharing expenses. The average beneficiary expenses (premiums and cost sharing) for Parts B and D combined currently are 23 percent of the average Social Security benefit. These expenses will increase to 40 percent of the average Social Security benefit by 2085. These expenses do not include cost sharing under Part A. The 2013 trustees’ report projects that under current law, SMI spending will continue to grow faster than GDP, increasing from 2.0 percent of GDP in 2012 to 3.1 percent of GDP in 2030, and to 4.0 percent of GDP in 2085. As acknowledged by the Trustees, the current-law projections likely understate the increases in Part B spending. Given that SGRrelated physician payment reductions have been overridden every year since 2003, it is considered unlikely that future scheduled reductions will take effect in full.4 In addition, the scheduled reductions in non-physician provider payment rate updates to reflect productivity adjustments might not be sustainable in the long term. The CMS Office of the Actuary’s illustrative alternative analysis sets physician payment updates to 0.7 percent per year throughout the short range projection period (the average update for the last 10 years), thereafter phasing down to the growth in per capita national health expenditures. In addition, the alternative analysis phases down the productivity adjustments gradually over 15 years, beginning in 2020, from about 1.1

Part B beneficiaries pay monthly premiums covering approximately 25 percent of program costs; general revenues cover the remaining 75 percent of costs. Part D premiums are set at approximately 25 percent of Part D costs. Because of low-income premium subsidies, however, beneficiary premiums will cover only approximately 14 percent of total Part D costs in 2013. State payments on behalf of certain beneficiaries will cover approximately 12 percent of costs and general revenues will cover the remaining 74 percent of costs. 4 The sustainable growth rate (SGR) system was enacted as part of the Balanced Budget Act of 1997 to limit the growth in spending for physician services. The system compares actual cumulative spending to a specified spending target. If actual spending exceeds the target, then physician payment updates are adjusted downward. A cumulative reduction of 25 percent is estimated for next year 3

AMERICAN ACADEMY OF ACTUARIES ISSUE BRIEF JUNE 2013

5


percent to 0.4 percent, and assumes no savings from IPAB. The alternative scenario projections assume no changes to the current-law Part D projections. Under the illustrative alternative scenario projections, SMI spending would increase from 2.0 percent of GDP in 2012 to 3.3 percent of GDP in 2030, and to 5.6 percent of GDP in 2085.

Table 1: SMI Expenditures as a Percent of GDP Calendar Year

2013 Report (current law)

2013 Alternative Projection

2012

2.0

2.0

2020

2.2

2.3

2030

3.1

3.3

2040

3.4

3.9

2050

3.5

4.2

2060

3.7

4.7

2070

3.9

5.1

2080

3.9

5.5

2085

4.0

5.6

Sources: 2013 Medicare Trustees’ Report, CMS Office of the Actuary

Increases in Total Medicare Spending Threaten the Program’s Sustainability A broader issue related to Medicare’s financial condition is whether the economy can sustain Medicare spending in the long run. To help gauge the future sustainability of the Medicare program, we examine the share of GDP that will be consumed by Medicare. Because Medicare spending is expected to continue growing faster than GDP, greater shares of the economy will be devoted to Medicare over time, meaning smaller shares of the economy will be available for other priorities. According to the current-law projections, Medicare expenditures as a percentage of GDP will grow from 3.6 percent of GDP in 6

AMERICAN ACADEMY OF ACTUARIES ISSUE BRIEF JUNE 2013

2012 to 6.5 percent of GDP in 2085. Under the CMS Office of the Actuary alternative scenario, however, total Medicare expenditures would increase to 9.6 percent of GDP in 2085.

Table 2: Total Medicare Expenditures as a Percent of GDP Calendar Year

2013 Report (current law)

2013 Alternative Projection

2012

3.6

3.6

2020

3.9

4.0

2030

5.1

5.4

2040

5.8

6.5

2050

6.0

7.2

2060

6.1

7.9

2070

6.4

8.7

2080

6.5

9.3

2085

6.5

9.6

Sources: 2013 Medicare Trustees’ Report, CMS Office of the Actuary

CONCLUSION The Affordable Care Act (ACA), enacted in 2010, contains numerous provisions designed to reduce Medicare costs, increase Medicare revenues, and develop new health care delivery systems and payment models that improve health care quality and cost efficiency. Additional steps need to be taken, however, to solve the long-term financial challenges to Medicare. The HI trust fund is projected to be depleted in 2026, and Medicare spending will continue to grow faster than the economy— increasing the pressure on beneficiary household budgets and the federal budget and threatening the program’s sustainability. In addition, Medicare’s financial challenges are likely to be much more severe than projected in the trustees’ report. The report’s Medicare spending projections are considered


understated to the extent that currently scheduled reductions in physician payments are expected to be overridden by Congress, as they have been every year since 2003, and the ACA’s provisions for downward adjustments in provider payment updates to reflect productivity improvements are unsustainable in the long term. If Medicare projections are calculated using assumptions that the physician payment reductions are overridden and the productivity adjustments are phased down, Medicare’s financial condition is shown to be even worse than under current-law projections. The American Academy of Actuaries’ Medicare Steering Committee continues to have significant concerns about Medicare’s financing problems, even under the currentlaw projections, and strongly recommends that policymakers implement changes to improve Medicare’s financial outlook.

We concur with the 2013 trustees when they say: The Board of Trustees believes that solutions can and must be found to ensure the financial integrity of HI in the short and long term and to reduce the rate of growth in Medicare costs through viable means. Consideration of such reforms should occur in the near future. The sooner the solutions are enacted, the more flexible and gradual they can be. Moreover, the early introduction of reforms increases the time available for affected individuals and organizations—including health care providers, beneficiaries, and taxpayers—to adjust their expectations. Congress and the executive branch must work closely together with a sense of urgency to address these challenges. And we wish to underscore this call for action.

Medicare Provisions in the Affordable Care Act The Affordable Care Act (ACA), enacted into law in 2010, includes many provisions designed to reduce Medicare costs, increase Medicare revenues, and develop new health care delivery systems and payment models that improve health care quality and cost efficiency. Major provisions include: n Reductions to provider payment updates.

The annual updates for fee-for-service provider payment rates are adjusted downward to reflect productivity improvements. n Basing Medicare Advantage plan pay-

ments on fee-for-service rates. Medicare Advantage plan payments are being reduced gradually relative to fee-for service costs. n Health care payment and delivery system

improvements. Pilot programs, demonstration projects, and other reforms are being implemented to increase the focus on

delivering high quality and cost-effective care. These include initiatives on bundled payments and accountable-care organizations. n Increases in Medicare revenues. Provisions

to increase Medicare revenues include: increasing the HI payroll tax for earnings above an unindexed threshold, temporarily freezing the income thresholds for Part B incomerelated premiums, and increasing Part D premiums for higher-income beneficiaries. n Creation of the Independent Payment Ad-

visory Board (IPAB). Beginning in 2014, the board will submit recommendations to make changes to provider payments if Medicare spending exceeds a target per capita growth rate. Unless legislative action overrides the recommendations, they will be implemented automatically.

AMERICAN ACADEMY OF ACTUARIES ISSUE BRIEF JUNE 2013

7


American Academy of Actuaries

MARCH 2009

MARCH 2012

Revising Medicare’s Fee-ForService Benefit Structure

Key Points n Medicare’s current FFS cost-sharing requirements have several shortcomings: — The lack of a cost-sharing limit leaves beneficiaries unprotected against catastrophic costs; —Most beneficiaries have supplemental coverage with low cost-sharing requirements that reduce incentives to seek cost-effective care; —The cost-sharing structure is not ideal for influencing consumer behavior. n Updating the FFS cost-sharing features could help better align beneficiary incentives to seek cost-effective care. Meeting this goal, however, may require changes to supplemental coverage as well.

Additional Resources An Actuarial Perspective on Proposals to Improve Medicare’s Financial Condition: http://www.actuary.org/pdf/Medicare_Financial_IB_Final_051211.pdf Health Insurance Coverage and Reimbursement Decisions: Implications for Increased Comparative Effectiveness Research: http:// www.actuary.org/pdf/health/comparative.pdf Value-Based Insurance Design: http://www. actuary.org/pdf/health/vbid_june09.pdf

I

mproving the quality and cost-effectiveness of care under the Medicare program is a key health policy challenge. Many Medicare reform proposals in recent years have focused on realigning financial incentives in Medicare’s provider payment and delivery system. However, a comprehensive package of reforms to improve Medicare sustainability also should consider better aligning incentives on the beneficiary side. To accomplish this, there have been calls to update the program’s traditional fee-for-service (FFS) benefit design (i.e., its cost-sharing features) and to address other issues related to beneficiary incentives. Such changes could deal with some of the shortcomings of the current benefit structure, including its lack of a cost-sharing maximum, and could help encourage Medicare beneficiaries to seek more cost-effective care. This brief expands upon the analysis of potential changes to the Medicare FFS benefit design included in the American Academy of Actuaries’ Medicare Steering Committee issue brief, An Actuarial Perspective on Proposals to Improve Medicare’s Financial Condition, including a brief examination of value-based insurance design (VBID).

Current Medicare Fee-For-Service Benefit Design Like most other health insurance plans, Medicare uses patient costsharing requirements, such as deductibles, copayments, and coinsurance, to help balance the cost of the program with the comprehensiveness of the benefits provided (see Text Box 1). Patient cost sharing directly lowers Medicare spending by shifting a share of medical costs

The American Academy of Actuaries is a 17,000-member professional association whose mission is to serve the public and the U.S. actuarial profession. The Academy assists public policymakers on all levels by providing leadership, objective expertise, and actuarial advice on risk and financial security issues. The Academy also sets qualification, practice, and professionalism standards for actuaries in the United States. ©2012 The American Academy of Actuaries. All Rights Reserved.

1850 M Street NW, Suite 300, Washington, DC 20036 Tel 202 223 8196, Fax 202 872 1948 www.actuary.org Mary Downs, Executive Director Mark Cohen, Director of Communications Craig Hanna, Director of Public Policy Cori Uccello, Senior Health Fellow Heather Jerbi, Senior Health Policy Analyst


to the beneficiary. In addition, cost sharing can lower spending overall by reducing health care utilization. While Medicare’s patient cost-sharing requirements perform the same basic functions as similar requirements in other health insurance programs, their structures vary greatly. Medicare’s hybrid nature—which combines a mandatory hospital insurance program with voluntary coverage for physician and outpatient services as well as voluntary prescription drug coverage—is directly reflected in the structure of the Medicare fee-for-service benefits. Whereas private health insurance programs typically have integrated benefit structures that are designed to manage hospital and non-hospital expenses in a coordinated fashion, the Medicare Part A (hospital) and Part B (physician and outpatient) benefits are structured very differently from each other— and the patient cost-sharing provisions are not coordinated between the two. This lack of

coordination in the design of Medicare’s FFS benefits has important consequences for both beneficiaries and taxpayers. In an ideal situation, patient cost-sharing requirements align beneficiary incentives with program goals to provide high-quality and cost-effective care. Medicare’s current FFS cost-sharing requirements, however, are not well structured to meet these goals and have other drawbacks. In particular: n MEDICARE

DOES NOT PLACE AN ANNUAL LIMIT ON BENEFICIARY COST-SHARING LIABILITY. The lack of an annual limit on cost

sharing under the FFS option leaves beneficiaries unprotected against catastrophic health costs. n MOST

MEDICARE BENEFICIARIES HAVE SUPPLEMENTAL POLICIES. Because there is

no cost-sharing limit, supplemental coverage is a necessity for beneficiaries who desire protection against the costs associated with catastrophic illness. Most Medicare benefi-

Selected Part A and Part B Cost-Sharing Requirements PART A

Hospital stay:

$1,156 deductible for days 1–60 per benefit period $289/day copayment for days 61–90 $578/day copayment for days 91–150

Skilled nursing facility stay:

$0 for the first 20 days each benefit period $144.50 per day for days 21–100 each benefit period All costs for each day after 100 each benefit period

PART B

Annual deductible: $140 Physician services: 20 percent coinsurance Outpatient hospital 20 percent coinsurance (up to hospital deductible services: of $1,156) Note: See www.medicare.gov for cost-sharing requirements for additional Medicare covered services.

Members of the Medicare Steering Committee include: Edwin C. Hustead, MAAA, FSA, EA, Chairperson; Dennis J. Hulet, FSA, MAAA, FCA, Vice Chairperson; Joe Bawazer, ASA, MAAA; Jill H. Brostowitz, FSA, MAAA; Michael V. Carstens, FSA, MAAA; April S. Choi, FSA, MAAA; Randall S. Edwards, FSA, MAAA; Troy M. Filipek, FSA, MAAA, FCA; Harry Hotchkiss, FSA, MAAA; Joel C. Kabala, FSA, MAAA; Margot D. Kaplan, ASA, MAAA, FCA; Laura Beth Lieberman, FSA, MAAA; Walter T. Liptak, ASA, MAAA, EA; Mark E. Litow, FSA, MAAA; Steve Niu, FSA, MAAA, EA; Susan E. Pierce, MAAA, FSA; Robert J. Pipich, FSA, MAAA; Anna M. Rappaport, FSA, MAAA, EA; Jeremiah D. Reuter, ASA, MAAA; Steven Rubenstein, MAAA, ASA; Paul A. Schultz, FSA, MAAA; Gordon R. Trapnell, FSA, MAAA; Cori E. Uccello, FSA, MAAA, FCA, MPP; John A. Wandishin, FSA, MAAA; Thomas F. Wildsmith, FSA, MAAA; Carl Wright, MAAA, FSA. 2

ISSUE BRIEF MARCH 2012


ciaries have supplemental coverage that also fills in the FFS cost-sharing requirements for non-catastrophic illnesses, which reduces the incentives for beneficiaries to seek cost-effective care. n THE

FFS DEDUCTIBLES ARE HIGHER FOR INPATIENT CARE. Cost-sharing requirements

aim, in part, to influence consumer behavior. Medicare’s cost-sharing provisions, however, are not structured in an ideal way to do this. Part A inpatient stays, which are less likely to be influenced by cost-sharing requirements, require fairly high deductibles—$1,156 in 2012 and additional copayments for hospital stays lasting beyond 60 days. In contrast, Part B physician and outpatient services, which are more likely to be influenced by cost-sharing requirements, require a fairly low annual deductible of $140 in 2012. Thereafter, a 20 percent coinsurance is required on most Part B services. In contrast to traditional Medicare FFS plans, Medicare Advantage plans have some flexibility on how to structure cost-sharing requirements—and very few use the FFS costsharing structure. In addition, all Medicare Advantage plans now are required to provide an annual cost-sharing limit, which in 2012 can be no more than $6,700.

Restructuring the Fee-ForService Benefit Design To address the problems with the current FFS benefit design, proposals have been developed that would combine a new cost-sharing limit1

with a unified Part A and Part B deductible. The copayment and coinsurance requirements also could be restructured. These changes would result in more coordinated Part A and B costsharing requirements and would bring the FFS benefit design more in line with the structure of private health insurance programs. Unifying the Part A and B deductibles has the potential to better align beneficiary incentives designed to reduce unnecessary care and promote more cost-effective care. But, as discussed in more detail below, the majority of Medicare beneficiaries have supplemental coverage that can limit the effectiveness of the incentives in Medicare’s cost-sharing requirements. In addition, beneficiaries need more access to price and quality information to better facilitate more cost-effective beneficiary behavior. And perhaps most important, provider incentives need to be consistent with beneficiary incentives and more information regarding treatment effectiveness is needed. Adding an annual cost-sharing limit could be a significant benefit enhancement that would, absent other changes, increase the cost of the program. When combined with the introduction of a unified Part A and B deductible, however, such a restructuring could be achieved in a budget neutral way. In other words, the out-of-pocket limit and combined deductible could be chosen so that costs to the Medicare program would be the same under the new structure as they are projected currently. As an alternative, this restructuring can be done in a way that reduces (or increases)

Beneficiary Cost-Sharing Liability vs. Out-of-Pocket Costs Medicare beneficiaries who receive medical services are responsible for meeting any applicable cost-sharing requirements. These beneficiary cost-sharing liabilities, however, may not reflect what a beneficiary actually pays out of pocket to meet those requirements. For instance, beneficiaries with supplemental coverage (e.g., Medigap, employer-sponsored retiree health coverage) have all or a portion of their cost-sharing liabilities covered. A full accounting of how a change in the Medicare

FFS plan design would affect beneficiary out-of-pocket costs (including premiums for supplemental coverage) therefore would need to incorporate not only the specific changes to the benefit design, but also whether and how changes in Medicare supplemental coverage are required and whether and how beneficiaries change their supplemental coverage purchases and health care utilization in response to the changes.

1 Even with a cost-sharing limit, beneficiaries would remain responsible for all costs associated with benefits that are not covered by Medicare.

ISSUE BRIEF MARCH 2012

3


Medicare costs. An annual out-of-pocket limit would reduce cost-sharing for those beneficiaries with the highest health care spending. In any year, however, even if the plan design changes are made to be budget neutral, the majority of beneficiaries who have lower health care spending would face higher costsharing amounts.2 A recent report from the Medicare Payment Advisory Commission (MedPAC) provides insights on the effects of adding a catastrophic limit on cost sharing and combining the Part A and B deductibles, assuming other cost-sharing requirements remain unchanged (Table 1). Under current law, which does not include a cap on cost sharing, a combined deductible of $595 would have been required in 2011 to remain budget neutral compared with the separate plan deductibles. Under this approach, 6 percent of Medicare beneficiaries would have experienced a reduction in out-of-pocket spending of $50 or more and 28 percent would have experienced an increase in spending of $50 or more. About two-thirds of beneficiaries would have experienced no change or a change of $50 or less. Implementing a cap on cost sharing would require higher combined deductibles to remain budget neutral. The lower the cost-sharing cap,

the higher the combined deductible and the more likely it is that beneficiaries would experience an increase in out-of-pocket costs. For instance, a $3,000 cap on cost-sharing would have required a $1,635 combined deductible and 36 percent of beneficiaries would have faced increased out-of-pocket costs of $50 or more. Nevertheless, the increased catastrophic protection would result in large savings for many of those exceeding the cap. With a combined deductible and a cap on cost sharing, beneficiaries who are more likely to face increased cost sharing include those with no hospitalizations and high Part B spending, but not enough to exceed the catastrophic cap, since the combined deductible exceeds the current Part B deductible. Beneficiaries who are more likely to face a reduction in cost sharing are those with hospitalizations and spending exceeding the cap. Note that this analysis reflects the change in cost-sharing liability over a one-year period only. Over a longer time period, it is likely that beneficiaries would have some years during which they are hospitalized and would incur a lower cost-sharing liability under a combined deductible and cost-sharing cap, and some years during which they would have a lower cost-sharing liability under the current FFS

Table 1. Level of Combined FFS Deductible Required to Hold Constant Medicare Program Spending in 2011

Catastrophic limit on cost sharing

Combined deductible required to break even

None— current law

$595

$7,000

How FFS beneficiaries’ out-of-pocket spending would differ from baseline

Nonspenders

Change of $50 or less

Spending increase of $50 or more

Spending decrease of $50 or more

5%

61%

28%

6%

960

5

56

33

6

$5,000

1,170

5

54

34

7

$4,000

1,328

5

53

35

6

$3,000

1,635

5

52

36

7

Notes: Out-of-pocket spending includes only cost-sharing amounts paid by the beneficiary. It excludes spending paid by supplemental coverage as well as premiums for Medicare and supplemental coverage. Changes in out-of-pocket spending incorporate changes in utilization due to the revised cost-sharing requirements, but not any changes in supplemental coverage. Categories may not sum to 100 percent due to rounding. Source: Actuarial Research Corporation (as published by MedPAC in Report to the Congress: Medicare and the Health care Delivery System, June 2011, Chapter 3). 2 Setting the unified deductible below the “budget neutral” level would reduce the number of beneficiaries who would face higher cost-sharing requirements, but would increase the cost of the Medicare program unless offset by other spending reductions or revenue increases.

4

ISSUE BRIEF MARCH 2012


plan design. In other words, using a one-year basis to estimate the change in cost sharing understates the value to beneficiaries of adding a cost-sharing cap on a budget neutral basis. When adding a cost-sharing limit along with a unified deductible, other cost-sharing requirements could remain unchanged. As an alternative, service-specific copayment and coinsurance requirements could be replaced with a uniform coinsurance rate for all services. Or, flat copayments, which are more typical among Medicare Advantage plans, could be used. Moving toward flat copayments in FFS Medicare could have the advantage of being more understandable and predictable to beneficiaries than coinsurance, in which cost sharing varies depending on the cost of the service. Depending on the copayment levels set, however, moving toward copayments rather than coinsurance could require higher unified deductibles to stay budget neutral. Although not the focus of this issue brief, the costs of adding a cost-sharing limit could be offset in ways other than increasing other cost-sharing requirements, such as through premium increases.3 An issue that would need to be addressed if Part A and B deductibles are combined is how to treat beneficiaries who sign up for Part A coverage but not Part B coverage (or vice versa). Applying the combined deductible would allow Part A-only beneficiaries to meet a lower deductible, but they would not be subject to potentially higher cost sharing under Part B. Moreover, allowing Part A-only beneficiaries to benefit from the addition of a cost-sharing limit could create equity concerns. An alternative would be to maintain the current higher Part A deductibles for beneficiaries choosing to enroll only in Part A and not allow them to benefit from the cost-sharing limit.

Impact on Medicare Trust Funds and Part B Premiums A redesign of the Medicare FFS benefit package that is budget neutral still could have important implications for the funding of the Medicare program. This would occur, for instance, if a combined deductible and cost-sharing cap

shifts costs between Parts A and B. In turn, this could affect not only the trust fund finances, but also Part B premiums. For instance, a combined deductible that is less than the Part A deductible and greater than the Part B deductible could mean that Medicare spending (net of cost sharing) would shift from Part B to Part A. How costs shift between the two parts is complicated by the costsharing cap, which could change the distribution of net costs between Part A and Part B. In addition, issues arise regarding the timing of claims during a year. If a beneficiary has physician care early in the year and inpatient care later in the year, the deductible first would apply to the physician care, with any remaining deductible applicable to the inpatient care. This would result in different net spending in the Part A and Part B programs than if inpatient care was received earlier in the year with the deductible first applying to that care. With hospital stays early in the year, which are usually accompanied by physician services, it may be difficult to determine how to split the deductible between Part A and Part B. Which services are received after the cost-sharing cap is reached, rather than before, similarly could affect the distribution between Part A and Part B spending. It may be appropriate for CMS to perform a retrospective adjustment at the end of the year to redistribute spending between Parts A and B to better reflect the true split between Part A and B spending, rather than the timing of claims. If the implementation of a combined deductible and a cost-sharing cap results in a net shift in Medicare spending from Part B to Part A, then Part B premiums, which are set at a percentage of Part B costs, would be lower than they are under current law. If a plan design change were to shift costs from Part A to Part B, however, Part B premiums would be higher. The Part A trust fund exhaustion date also could be affected. An increase in Part B premiums could result in a decrease in Part B enrollment. Part B is a voluntary program, and, although the vast majority of Medicare Part A enrollees also enroll in Part B, participation rates have been de-

In addition, cost-sharing requirements could be increased without moving to a unified deductible.

3

ISSUE BRIEF MARCH 2012

5


clining somewhat over the years. The Medicare trustees project that participation rates will continue to fall due to the higher premiums that apply to higher-income beneficiaries as well as the younger aged who are still working and have coverage from an employer.4 Nevertheless, Part B participation rates are projected to exceed 90 percent throughout the current 75-year projection period. If Part B participation rates decline more substantially, Part B premiums could increase even further, assuming that those enrolling would have higher health care needs than those who forgo coverage. At some point, it might be appropriate to consider additional measures to increase participation. Such measures could include increasing the penalty for those forgoing coverage, mandating Part B coverage, or allowing individuals to choose higher cost-sharing requirements in return for lower premiums. The latter approach, which also could allow individuals to choose lower cost-sharing requirements in return for higher premiums, in effect could combine FFS plan design changes with a premium support approach. Aside from any potential shifts in costs between the Medicare Part A and Part B programs associated with changing the FFS cost-sharing requirements, it is also important to consider any interactions between Medicare and Medicaid. Although changing the Medicare costsharing requirements likely would have little or no direct effects on beneficiaries dually eligible for Medicare and Medicaid, there is a potential shift in costs between the two programs.

Medicare Supplemental Insurance Because Medicare imposes significant costsharing requirements, most beneficiaries have some type of supplemental coverage to fill in the gaps. According to data compiled by MedPAC, 89 percent of FFS beneficiaries in 2007 had supplemental coverage: 43 percent had employer-sponsored coverage; 29 percent had individually purchased Medigap coverage; 16

percent had Medicaid, and 1 percent had other public coverage.5 Supplemental coverage can remove the financial incentives for beneficiaries to control their health spending, and some research suggests that filling in Medicare’s cost-sharing gaps results in higher Medicare spending than would have been incurred otherwise.6 As a result, there have been calls to limit the extent to which Medigap plans are allowed to cover Medicare’s cost-sharing requirements. For instance, the ACA directs the secretary of HHS to request the National Association of Insurance Commissioners to include nominal cost sharing for Medigap plans that provide first dollar coverage. Some proposals would place further limitations on Medigap plans. Others would levy an excise tax on Medigap plans or a Part B premium surcharge for beneficiaries with Medigap plans with low cost-sharing requirements. Such an excise tax or Part B premium surcharge would be a way for Medicare to recoup some of the costs of higher utilization among beneficiaries with Medigap plans and would encourage beneficiaries to choose plans that fill in less of Medicare’s cost-sharing requirements. If changes to the FFS plan design are implemented, insurance products that coordinate with Medicare may need to be modified so that they do not limit the desired impact of any FFS restructuring. For instance, Medigap plans could be prohibited from covering the higher deductibles. Or, the cost-sharing caps could be implemented on a true out-of-pocket basis, meaning that beneficiary cost sharing covered through supplemental coverage would not count toward the cost-sharing limit. Reducing the richness of Medigap plans available to beneficiaries, either directly through legislative/regulatory changes or indirectly through levying a Medigap excise tax or Part B premium surcharge, could result in an increased understanding among beneficiaries of their benefit choices, lower insurance premiums (due to reduced plan generosity

See Table III.A3 of the 2011 Medicare Trustees Report. Percentages calculated from Figure 3-1 in MedPAC, Report to the Congress: Medicare and the Health Care Delivery System, June 2011. 6 Although much research agrees that Medicare spending is higher among beneficiaries with supplemental coverage, there is less agreement regarding whether this difference is due to cost-sharing differences or other factors, such as the tendency of beneficiaries with higher health care needs to obtain supplemental coverage. For a review of the literature, see MedPAC Report to the Congress: Aligning Incentives in Medicare (Chapter 2, June 2010). 4 5

6

ISSUE BRIEF MARCH 2012


and increased administrative and marketing efficiencies), and a reduction of unnecessary utilization. Reducing the share of costs that Medigap plans can cover would shift costs at the point of service to beneficiaries, increasing the incentives to seek more cost-effective care and avoid unnecessary care. This has the potential to lower both Medicare and beneficiary costs, but the extent to which costs would decline is unclear. Changes in the rules governing Medigap plans should be structured carefully to avoid unintended consequences. Research suggests that broad increases in cost sharing, rather than targeted increases, reduce not only unnecessary care, but also necessary care, especially among the low income and chronically ill.7 Preventive care could be exempted from any new cost-sharing requirements, and additional protection for low-income and/or chronically ill beneficiaries who are not eligible for Medicaid should be considered. Other issues that should be addressed when considering changes to Medigap plan requirements include: n Policymakers would need to decide whether required changes in Medigap plans would apply to new coverage purchases only or to all existing policies as well. Medigap benefits are contractually guaranteed and cannot be cancelled for reasons other than premium non-payment. Besides the potential legal issues that may arise due to a violation of the contractual agreement, customer and insurer issues arise from changes to existing policies. A consumer’s premiums collected to date might have reflected prefunding for future services. Accordingly, insurers have accounted for this prefunding in the form of reserves. If changes are made to policies already in force, a clear transition plan to maintain fairness to insureds and reserve

adequacy for insurers would need to be developed. n

Many Medicare beneficiaries may be enrolling in Medigap plans to make their cost sharing more predictable and to avoid the inconveniences and complexities associated with paying providers directly. Any changes to Medigap plans, and to Medicare costsharing requirements more broadly, should incorporate ways to minimize beneficiary inconvenience or confusion as well as additional administrative burdens on providers for payment collections.

n

Medigap plans are only one source of private supplemental coverage. Even more beneficiaries are covered by employersponsored supplemental policies. While employer-sponsored plans typically do not provide first dollar coverage, it still may be appropriate to consider the role of employer-sponsored plans in supplementing Medicare and whether changes are needed.8 Note that employer-sponsored supplemental plans often include drug coverage and take the place of Part D as well as supplementing Parts A and B.

n

The addition of a cost-sharing limit for the traditional FFS program in itself could reduce the demand for supplemental coverage. Reducing the ability of supplemental plans to provide first dollar coverage further could reduce enrollment in these plans. Lower enrollment in supplementary coverage would mean that more beneficiaries would face the financial incentives inherent in the FFS benefit design, without those incentives being limited by supplemental coverage that fills in cost-sharing requirements.

7 The RAND Health Insurance Experiment found that although the reduction in services resulting from higher cost sharing did not lead to poorer health outcomes for the average person, low-income individuals in poor health were more likely to suffer poorer health outcomes. See Joseph P. Newhouse and the Health Insurance Experiment Group Free for all? Lessons from the RAND Health Insurance Experiment. Cambridge, Mass: Harvard University Press (1993). More recently, Amitabh Chandra et al found evidence that the savings associated with raising cost sharing for physician visits and prescription drugs is offset modestly by increased hospital utilization. The offsets are more substantial, however, for the chronically ill. See “Patient Cost-Sharing and Hospitalization Offsets in the Elderly,� American Economic Review 100(1): 193-213 (2010). 8 In the same manner, it may be appropriate to consider the role of Medicare when it is the secondary payer to other coverage, such as employer coverage for active workers aged 65 and older. In these instances, Medicare coverage in effect supplements other coverage. There are limits, however, as to how much Medicare will pay, and therefore, the extent to which Medicare fills in cost-sharing requirements. For instance, if the primary plan already pays more for a service than Medicare does, then Medicare would pay nothing more.

ISSUE BRIEF MARCH 2012

7


Value-Based Insurance Design Redesigning the FFS benefit structure could be a step in the direction of better aligning beneficiary incentives to seek cost-effective care. As discussed earlier, broad changes in cost sharing, however, will not necessarily target reductions in unnecessary or ineffective care. In the longer-term, moving to a value-based insurance design (VBID) could structure beneficiary financial incentives more effectively. A VBID approach would lower the cost sharing for highvalue services and increase the cost sharing for low-value services. The ACA moved Medicare in this direction by covering certain preventive services with no cost sharing. Adjusting cost sharing to align incentives with effective use of services has shown promise in reducing spending in the non-Medicare market—most often for prescription drugs.9 Comparative effectiveness research can provide more guidance to help distinguish lowvalue and high-value services. Although lowering cost sharing for high-value services could gain widespread acceptance, it likely would be more difficult to implement higher costsharing requirements for treatments deemed of lower value. One potential way to increase cost sharing for lower-value services is to use reference pricing. Under reference pricing, Medicare would pay the costs of the lowestprice option when multiple treatment options achieve similar results. Beneficiaries choosing a higher-cost option would pay the difference.

would provide protection against catastrophic health costs and has the potential to encourage beneficiaries to seek cost-effective care. Restructuring the FFS benefit design could be done in a budget neutral manner, or it could be done in a way that reduces Medicare spending overall. For Medicare to achieve savings beyond the amounts shifted to beneficiaries, the plan design changes would need to encourage beneficiaries to take a more active role in their health care, seek care when necessary, and learn more about the cost and expected outcomes of their care. Restructuring, however, will affect only the few beneficiaries who do not have supplemental coverage, unless insurance products that coordinate with Medicare are modified so that they do not limit the desired effects of any FFS restructuring. In addition, provider incentives need to be consistent with beneficiary incentives and more information regarding costs, quality, and treatment effectiveness is needed. Redesigning the FFS plan design is more of a short-term solution, with transitioning to a VBID a longer-term approach. Even under a VBID approach, however, a more comprehensive restructuring of not just the benefit design but also the payment and delivery systems is needed to move Medicare toward a more integrated, coordinated, and cost-effective system.

Conclusion The current Medicare FFS benefit design has several drawbacks. It lacks a cap on cost sharing, making supplemental coverage a necessity if beneficiaries are to be protected against the costs associated with catastrophic illnesses. Since most beneficiaries have supplemental policies to cover their FFS cost-sharing requirements, their incentives to seek cost-effective care are reduced. In addition, the Medicare FFS cost-sharing requirements are skewed toward less discretionary services. Restructuring the FFS benefit design by unifying the Part A and B deductibles and adding a cost-sharing limit

1850 M Street NW Suite 300 Washington, DC 20036 Tel 202 223 8196 Fax 202 872 1948 www.actuary.org

9 See for instance, “Evidence That Value-Based Insurance Can Be Effective,� Michael E. Chernew, et al. Health Affairs 29(3): 530-536, March 2010.

8

ISSUE BRIEF MARCH 2012


Summary of the National Flood Insurance Program The U.S. Congress passed the National Flood Insurance Act (NFIA) of 1968 to provide a means by which the risk of flood could be insured in the United States. Private-sector insurance companies had long viewed the risk of flood events as uninsurable. The NFIA of 1968 created the National Flood Insurance Program (NFIP), in which the federal government acts as the insurer. Today, flood insurance in the U.S. is primarily provided by the federal government via the NFIP, in partnership with private insurers and servicing contractors. The NFIP offers flood insurance to homeowners, renters, and business owners, if the communities in which their properties are located participate in the program. Participating communities agree to adopt and enforce ordinances that meet or exceed the requirements set forth by the Federal Emergency Management Agency (FEMA) to reduce the risk of flooding. The NFIP’s enabling legislation applied these minimum standards only to new construction; owners of existing properties did not have to rebuild, and many of them received subsidized premium rates that did not reflect their properties’ true flood risks. The NFIA of 1968 included a sunset provision; as such, the program lapses unless Congress acts to extend it. The 2004 and 2005 hurricanes that struck the East and Gulf coasts of the United States, and the substantial losses suffered in those storms, served as a wake-up call to Congress that comprehensive changes to the NFIP were needed. The NFIA of 1968 had established a cap on the amount of money the program could borrow from the U.S. Treasury Department; the cap was periodically increased over the years. Its 2004-05 losses were so great that Congress increased the NFIP’s borrowing authority several times in the aftermath of those storm seasons. Including the debt incurred as a result of Superstorm Sandy losses, the NFIP’s current debt to Treasury stands at $24 billion. In July 2012, the Biggert-Waters Flood Insurance Reform Act (P.L. 112-141) was passed, just a few months before Superstorm Sandy caused record-setting flooding in the Northeast. BiggertWaters extended the NFIP for another five years (until September 30, 2017) and established a process, including a gradual decrease in existing subsidizations, by which all NFIP insureds will eventually be charged full risk rates.

1

2013 Summer Summit Materials | American Academy of Actuaries


Selected Provisions of Various National Flood Insurance Reform Proposals H.R. 960, The Flood Victim Premium Relief Act of 2013 http://www.gpo.gov/fdsys/pkg/BILLS-113hr960ih/pdf/BILLS-113hr960ih.pdf This bill would amend the NFIA of 1968 to limit premium rate increases on specified residential properties in federally declared major disaster areas. On such properties, it would require Biggert-Waters-imposed premium rate increases or new premium rates to be phased in over an eight-year period, at a rate of five percent for each of the first four years and 20 percent for each of the last four years. It defines applicable properties as those located within a major disaster area: (1) for which revised or updated flood maps became effective during a 2-year period beginning when the disaster occurred; or (2) that, upon enactment of this bill, are eligible for preferred risk rate method premiums for flood insurance coverage, or that were eligible for them at any time during the 12-month period immediately preceding the relevant disaster. It also requires such a property to be: (1) owned by the same owner who owned the property at the time of the disaster; and (2) the owner’s primary residence since the disaster (except for any non-occupancy resulting from it). This bill was introduced on March 5, 2013 and was immediately referred to the House Financial Services Committee; it has seen no action since. H.R. 968, The Fire Damaged Home Rebuilding Act of 2013 http://www.gpo.gov/fdsys/pkg/BILLS-113hr968ih/pdf/BILLS-113hr968ih.pdf This bill would amend the NFIA of 1968 to allow local variances for certain residential structures located in special flood hazard areas (SFHAs) and substantially damaged by fire. Specifically, it would prohibit certain land use and control measures from being deemed inadequate or inconsistent with land management criteria for participation in the NFIP because such measures authorize granting a variance for such structures. It would also prohibit the administrator of FEMA from suspending a community from participation in the NFIP or placing said community on probation based on a perceived failure of a community structure, eligible for variance pursuant to this bill, to abide by the NFIP participation criteria. The bill defines acceptable variances as those granted by an appropriate community official permitting noncompliance with certain land use and control measures and permitting repair and restoration of a covered structure to its pre-damaged condition, without requiring the elevation of the structure. It also requires the official to have made specified determinations as a prerequisite to granting such a variance and requires chargeable flood insurance premium rates for a residential structure granted such a variance to equal, after repair and restoration, the rates that would have applied if the structure had not been substantially damaged, repaired, and restored in accordance with the variance. This bill was introduced on March 5, 2013 and was immediately referred to the House Financial Services Committee; it has seen no action since. 2

2013 Summer Summit Materials | American Academy of Actuaries


H.R. 1035, Untitled http://www.gpo.gov/fdsys/pkg/BILLS-113hr1035rfs/pdf/BILLS-113hr1035rfs.pdf This bill would direct the FEMA administrator to study and report to Congress about options, methods, and strategies to make voluntary, community-based flood insurance policies available through the NFIP. The directive would include a strategy to implement means of encouraging communities to engage in flood mitigation activities. The bill was introduced on March 7, 2013 and was immediately referred to the House Financial Services Committee. It was voted on and passed in the House on March 12, 2013. On March 13, it was received in the Senate and immediately referred to the Senate Banking Committee. H.R. 1194, The National Flood Insurance Program Termination Act http://www.gpo.gov/fdsys/pkg/BILLS-113hr1194ih/pdf/BILLS-113hr1194ih.pdf This bill would prohibit the FEMA administrator from providing any new flood insurance coverage after Dec. 31, 2013 or from renewing any coverage provided before that date. It would terminate the NFIP and provide repeal amendments to the NFIA of 1968, the Biggert-Waters Flood Insurance Reform Act of 2012 and several NFIP-related laws passed in the interim. It would also give congressional consent to any two or more states entering into compacts to make flood insurance available to interested parties for loss and damage arising from any flood occurring in the U.S. The bill was introduced on March 14, 2013 and was immediately referred to the House Financial Services Committee and the House Judiciary Committee. On April 15, it was referred to the House Judiciary Committee’s Subcommittee on Regulatory Reform, Commercial and Antitrust Law. H.R. 1267, The Flood Insurance Premium Relief Act of 2013 http://www.gpo.gov/fdsys/pkg/BILLS-113hr1267ih/pdf/BILLS-113hr1267ih.pdf This bill would delay implementation of the Biggert-Waters premium rate increases for properties purchased between July 6, 2012 (the date of Biggert-Waters enactment) and Jan. 1, 2015. (Biggert-Waters applies full-risk flood insurance premium rates immediately to all NFIP properties purchased after its enactment.) During the first year after the date of purchase, premium rates on such properties would remain the same as they were prior to the date of purchase. Beginning one year from the date of purchase, rate increases would be phased in at a rate of 10 percent per year over 10 years. The bill would also make conforming amendments to the NFIA of 1968. The bill was introduced on March 19, 2013 and was immediately referred to the House Financial Services Committee; it has seen no action since.

3

2013 Summer Summit Materials | American Academy of Actuaries


H.R. 1268, The Flood Mitigation Expense Relief Act of 2013 http://www.gpo.gov/fdsys/pkg/BILLS-113hr1268ih/pdf/BILLS-113hr1268ih.pdf This bill would amend the IRC to allow qualified taxpayers a tax credit of up to $5,000 per taxable year for flood mitigation expenses. Qualified taxpayers may be individuals or small businesses (those with under 50 employees) who: 1) hold flood insurance policies under the NFIA of 1968 2) own applicable property: a. for which the chargeable flood insurance premium rate was increased or will increase as a result of Biggert-Waters b. with an elevation lower than the base flood elevation as determined by the applicable flood insurance rate map or located in an area that, after the enactment of Biggert-Waters, has been designated as having a higher flood risk than the risk designated for said area prior to the enactment of Biggert-Waters. Such credit would terminate after 2022. This bill would also authorize appropriations to the FEMA administrator to carry out the predisaster hazard mitigation program authorized by the Robert T. Stafford Disaster Relief and Emergency Assistance Act and the flood mitigation assistance program authorized by the National Flood Insurance Act of 1969. The bill specifies that such funds may only be used for mitigation activities and acquisition by states or communities of properties located in areas that, after the enactment of Biggert-Waters, have been designated as having a higher flood risk than the risk designated for said areas prior to the enactment of Biggert-Waters. The bill was introduced on March 19, 2013 and was immediately referred to the House Ways and Means Committee, the House Committee on Transportation and Infrastructure, the House Financial Services Committee, and the House Energy and Commerce Committee. On March 20, it was referred to the House Committee on Transportation and Infrastructure’s Subcommittee on Economic Development, Public Buildings and Emergency Management. On March 22, it was referred to the House Energy and Commerce Committee’s Subcommittee on Energy and Power. H.R. 1485, Untitled http://www.gpo.gov/fdsys/pkg/BILLS-113hr1485ih/pdf/BILLS-113hr1485ih.pdf This bill would amend the NFIA of 1968 to authorize the FEMA administrator to provide flood insurance to some prospective insureds at rates lower than certain statutory estimates. It would increase by 12.5 percent per year the chargeable risk premium rate for flood insurance for business properties and non-primary residences until the average risk premium rate for such properties was the same as the average risk premium rate for specified other properties. It would also increase by 12.5 percent per year the chargeable risk premium rate for flood insurance for primary residences purchased after Biggert-Waters was enacted, as if such an increase began on the date Biggert-Waters was enacted, until the average risk premium rate for such properties was the same as the average risk premium rate for specified other properties. (Again, this would be a change from Biggert-Waters, which applies full-risk flood insurance premium rates immediately to all NFIP properties purchased after its enactment.)

4

2013 Summer Summit Materials | American Academy of Actuaries


The bill would also require the FEMA administrator to adjust the chargeable risk premium rates for flood insurance, for properties subject to the 12.5 percent increase, as if the changes enacted by this bill were enacted concurrently with Biggert-Waters. This bill was introduced on April 11, 2013 and was immediately referred to the House Financial Services Committee; it has seen no action since. H.R. 2199, The Flood Insurance Implementation Reform Act of 2013 http://www.gpo.gov/fdsys/pkg/BILLS-113hr2199ih/pdf/BILLS-113hr2199ih.pdf This bill would delay until three years after its enactment the Biggert-Waters requirement that any property located in an NFIP-participating area have its flood insurance risk premium rate adjusted to accurately reflect its current flood risk. The bill would also amend the NFIA of 1968 to delay, until five years after the enactment of Biggert-Waters, the prohibition against FEMA providing prospective insureds with subsidized rates (lower than the full actuarial estimates) for property purchased after the enactment of Biggert-Waters. This bill would also prohibit FEMA from taking into account federal funding or participation when determining whether a community has made sufficient progress in its development or improvement of flood protection systems. The bill would make flood insurance available at special rates to riverine and coastal levees located in communities that FEMA identifies as being in the process of restoring a flood protection system, previously accredited on a flood insurance rate map (FIRM) as providing 100year frequency flood protection, but that no longer does so. The bill requires such rates to apply without regard for federal funding or participation. This bill would amend Biggert-Waters to allow FEMA to use other funds in addition to those specified in the Act to carry out the affordability study the Act requires. Upon notice to the congressional committees of jurisdiction that FEMA cannot submit its study report by its deadline, the bill requires that FEMA identify an alternative method of information-gathering and to subsequently submit the information so obtained. The bill directs FEMA to identify, review, maintain, update, and publish FIRMs pertaining to areas protected by non-structural flood mitigation features and to work with states, local communities, and property owners to identify such areas and features. This bill was introduced on May 23, 2013 and was immediately referred to the House Financial Services Committee; it has seen no action since. H.R. 2217, The Department of Homeland Security Appropriations Act, 2014 http://www.gpo.gov/fdsys/pkg/BILLS-113hr2217rfs2/pdf/BILLS-113hr2217rfs2.pdf This bill makes appropriations for Fiscal Year (FY) 2014 for, among other agencies, FEMA, including for the flood hazard mapping and risk analysis program and the National Flood Insurance Fund. It prohibits the use of funds to administer or enforce provisions regarding adjusted premium rates under the NFIA of 1968. 5

2013 Summer Summit Materials | American Academy of Actuaries


The bill was introduced on May 29, 2013 and was immediately referred to the House Appropriations Committee. It was voted on and passed in the House on June 6, 2013. On June 7, it was received in the Senate and was ultimately referred to the Senate Appropriations Committee. S. 667, The Fire-Damaged Home Rebuilding Act of 2013 http://www.gpo.gov/fdsys/pkg/BILLS-113s667is/pdf/BILLS-113s667is.pdf This bill would amend the NFIA of 1968 to allow local variances for certain residential structures substantially damaged by fire or other disaster, excluding flood. Specifically, it would allow state or local authorities to grant variances from compliance with certain mandatory adequate land use and control measures to allow repair and restoration of eligible structures to their pre-damaged condition without requiring the structure’s elevation. The repaired and restored structure must be located on the same site as it was before it was damaged, and the number of floors of the restored structure cannot exceed the number of floors of the original. A maximum of 10 such variances would be authorized per calendar year. The bill would also prohibit the administrator of FEMA from finding that land use and control measures are inadequate or inconsistent with NFIP participation criteria and would prohibit the FEMA administrator from suspending a community from participation in the NFIP or placing said community on probation based on a perceived failure of an eligible community structure to abide by the NFIP participation criteria. The bill also prohibits the chargeable flood insurance premium rate for a residential structure granted variance in accordance with this bill from being lower than the rate that would have applied if the structure had not been substantially damaged by a fire or non-flood disaster, repaired, and restored pursuant to the variance. This bill was introduced on April 8, 2013 and was immediately referred to the Senate Banking Committee; it has seen no action since. S. 996, Untitled http://www.gpo.gov/fdsys/pkg/BILLS-113s996is/pdf/BILLS-113s996is.pdf This bill would amend the NFIA of 1968 to repeal the prohibition against FEMA providing prospective insureds with subsidized flood insurance premium rates (lower than the full actuarial estimates) for properties purchased after the enactment of Biggert-Waters. (This would be a departure from Biggert-Waters, which applies full-risk flood insurance premium rates immediately to all NFIP properties purchased after its enactment.) This bill would delay the effective date of any flood insurance rate changes until 180 days after FEMA submits a report on methods to establish an affordability framework for the NFIP or gives notice to the congressional committees of jurisdiction of an alternative method of informationgathering for such a report, if the affordability report cannot be submitted by its due date.

6

2013 Summer Summit Materials | American Academy of Actuaries


This bill would also direct the FEMA administrator to study and report to Congress on options, methods, and strategies to make voluntary, community-based flood insurance policies available through the NFIP. This bill would also prohibit FEMA from taking into account federal funding or participation when determining whether a community has made sufficient progress in its development or improvement of flood protection systems. This bill requires replacement structures or facilities to be relocated to alternative sites if FEMA identifies a practicable alternative outside a coastal high hazard area that provides better protection against the hazards associated with such areas. This bill was introduced on May 21, 2013 and was immediately referred to the Senate Banking Committee; it has seen no action since. S. 1075, The Saving Homeowners from Onerous Rate Escalations (SHORE) Act of 2013 http://www.gpo.gov/fdsys/pkg/BILLS-113s1075is/pdf/BILLS-113s1075is.pdf This bill would amend the NFIA of 1968 to direct the FEMA administrator to phase in, over eight years, any increase in flood insurance risk premium rates caused by the prohibition against extending subsidies to new or lapsed policies. This bill also extends from five years to 10 years the phase-in period for premium adjustment increases in flood insurance risk premium rates. It sets forth a phase-in rate of five percent for each of the first five years after the effective date of an update and 15 percent for each of the five subsequent years, and five percent for each of the first five years after the effective date of a new designation as a SFHA, and 15 percent for each of the five subsequent years. This bill was introduced on May 23, 2013 and was immediately referred to the Senate Banking Committee; it has seen no action since.

S. 1098, Responsible Implementation of Flood Insurance Reform Act of 2013 http://www.gpo.gov/fdsys/pkg/BILLS-113s1098is/pdf/BILLS-113s1098is.pdf As previously mentioned, Biggert-Waters applies full-risk flood insurance premium rates to newly-purchased properties. This bill would provide newly-purchased properties with a gradual increase in rates similar to the gradual increases set forth elsewhere in Biggert-Waters. Specifically, it would increase the rate of a newly-purchased property by 20 percent each year, beginning one year after purchase, until the average risk premium rate for such properties is equal to the average of the risk premium rate for specified other properties. This bill was introduced on June 6, 2013 and was immediately referred to the Senate Banking Committee; it has seen no action since.

7

2013 Summer Summit Materials | American Academy of Actuaries


June 28, 2012 The Honorable Harry Reid Majority Leader, U.S. Senate S-221 Capitol Building Washington, DC 20510-7020 The Honorable Mitch McConnell Minority Leader, U.S. Senate S-230 Capitol Building Washington, DC 20510-7010 The Honorable John Boehner Speaker, U.S. House of Representatives H-232 Capitol Building Washington, DC 20515-6501 The Honorable Nancy Pelosi Minority Leader, U.S. House of Representatives H-204 Capitol Building Washington, DC 20515-6537 Re: Reform of the National Flood Insurance Program Dear Majority Leader Reid, Minority Leader McConnell, Speaker Boehner, and Minority Leader Pelosi: As you prepare to consider comprehensive reform to the National Flood Insurance Program (NFIP), the American Academy of Actuaries’1 Flood Insurance Subcommittee appreciates this opportunity to provide an actuarial perspective on this legislation and its possible effect on the NFIP. First and foremost, we support reauthorization of the NFIP for at least 5 years. We believe that multiyear reauthorization will help to strengthen the market for flood insurance as the program regains stability. Additionally, we strongly support efforts to provide a better financial base for the NFIP. We are encouraged by the provisions of this bill that move toward a rate structure that will better match potential loss payments and premium income, including an increase in deductibles for subsidized rate properties, an increase in the maximum annual premium change allowed, and new rules to phase in “full actuarial rates.”

1

The American Academy of Actuaries is a 17,000-member professional association whose mission is to serve the public and the U.S. actuarial profession. The Academy assists public policymakers on all levels by providing leadership, objective expertise, and actuarial advice on risk and financial security issues. The Academy also sets qualification, practice, and professionalism standards for actuaries in the United States.


Terms such as “full actuarial rates” and “actuarially sound rates” are often cited to describe the premiums for programs like the NFIP. There are no standard definitions for these terms, but the Statement of Principles Regarding Property and Casualty Insurance Ratemaking, promulgated by the Casualty Actuarial Society, says that a rate should provide “for all costs associated with the transfer of risk.” The NFIP ratemaking process identifies a “full actuarial rate” as one that appropriately covers the expected average annual loss for a property, plus a small risk load. Also, the Academy’s Actuarial Soundness Task Force recently published a public policy report on actuarial soundness2, which deals, in part, with this issue as it applies to public entities. The NFIP’s existing $17.75 billion debt, its potential for future borrowing, and its rating structure bring into sharp focus a central issue to consider when debating the NFIP’s financial future. Congress’s stated purpose in authorizing the creation of the program was, in part: “…so that such flood insurance may be based on workable methods of pooling risks, minimizing costs, and distributing burdens equitably among those who will be protected by flood insurance and the general public.” The NFIP allows for funding using both insurance premiums and other public sources. Flood insurance premiums for the program have been developed using an approach that focuses on keeping premiums at a level below market rates to encourage participation. Also, many of the non-insurance activities of the NFIP that benefit the public at large are funded with premium revenues. This structure does not facilitate minimization of the need for borrowing. While there is always the potential need to borrow, if that need is to be minimized, the ratemaking approach and the regulations that circumscribe that approach may need to change. Observations by Section Section 100211 specifies the use of “generally accepted actuarial principles” in calculating premium rates. Generally accepted actuarial principles are designed to estimate the expected cost of future losses. However, requiring premiums to cover the average historical loss year, including catastrophic losses, does not align with generally accepted actuarial practice. Although such premium rate estimates would be informed by historical losses, requiring coverage of those losses could, after an unusually large event, such as Hurricane Katrina, result in rates much higher than the expected cost. Any effort to put the NFIP on sound financial footing would benefit from addressing the program’s existing $17.75 billion debt, which was incurred following the catastrophic losses suffered in the 2005 Gulf Coast hurricanes. Resolving the issue of the outstanding debt is critical to the program becoming more financially sound; however, the bill does not address this issue directly. Currently, a significant portion of the NFIP premium goes to paying down the debt and interest on the debt. Even so, it will be decades until this debt is paid off, even if no extreme catastrophic events occur in the meantime. Ultimately, for the NFIP’s financial stability to be improved, the debt would need to be waived, or a specific cost provision to be used for repayment could be added to the premiums. This debt situation demonstrates a difference between private insurance and the NFIP. Actuarial practice in private insurance would not support the reimbursement of past deficits by future policyholders, unless specific surcharge, outside the “actuarial rate,” was established and dedicated to debt repayment. 2

http://actuary.org/files/publications/Actuarial%20Soundness%20Special%20Report%20FINAL%205%2010%2012.pdf

2


Section 100212 requires the NFIP administrator to establish a reserve fund, and, when its balance is less than the specified reserve ratio, the reserve fund would be added to at a rate of 7.5 percent of that ratio, per year. We have three concerns about this provision. First, given the current scope of the program, it has been estimated that this would add an annual charge to policyholders of approximately $800 million, or about 25 percent of the NFIP’s total annual premium. Second, that $800 million annual charge to policyholders would be increased by an additional 30 to 40 percent because of loadings for various insurance expenses. Third, it is unclear how this provision would work in combination with the provisions in Section 100213 concerning debt repayment. If payments to the reserve are not made before the debt is paid down, there seems to be little chance that the reserve will actually be funded, because debt repayment is such a longterm process. Section 100225 requires a $90 million annual contribution from the Flood Fund for mitigation projects, but prohibits collecting revenue to offset that cost. Such an expense will reduce the amount available to pay for flood losses and will create a destabilizing element in the flood insurance program. If the program is to move toward financial soundness, there needs to be an offsetting source of revenue for mitigation projects. Section 100232 provides for participation from the private market by allowing the administrator to purchase reinsurance or reinsurance equivalents. The NFIP could potentially benefit from these purchases in certain situations. However, for the reasons described above, the NFIP’s current rating structure likely will not provide enough revenue to purchase a significant level of reinsurance. Reinsurance providers will likely require a return commensurate with the risk associated with the coverage provided. Current NFIP rates do not incorporate and reflect a cost of capital commensurate with this risk. However, enactment of this bill should provide some funds that could be used for reinsurance. The American Academy of Actuaries’ Flood Insurance Subcommittee hopes that you will find these comments helpful and would be pleased to assist you in your efforts to reform and reauthorize the NFIP. If you have any questions, please feel free to contact Lauren Pachman, the Academy’s casualty policy analyst, at pachman@actuary.org. Again, thank you for this opportunity to comment on the proposed legislation. Sincerely,

Stuart B. Mathewson, FCAS, MAAA Chair, Flood Insurance Subcommittee

CC:

Members, U.S. Senate Members, U.S. House of Representatives 3


A P u b l i c P o l i c y M O n o g r ap h

The National Flood Insurance Program: Past, Present...and Future? July 2011

American Academy of Actuaries Flood Insurance Subcommittee


A PUBLIC POLICY MONOGRAPH

The National Flood Insurance Program: Past, Present ‌ and Future? July 2011

Developed by the Flood Insurance Subcommittee of the American Academy of Actuaries

The American Academy of Actuaries is a 17,000-member professional association whose mission is to serve the public and the U.S. actuarial profession. The Academy assists public policymakers on all levels by providing leadership, objective expertise, and actuarial advice on risk and financial security issues. The Academy also sets qualification, practice, and professionalism standards for actuaries in the United States.


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

2011 Flood Insurance Subcommittee Stuart Mathewson, Chairperson

Patrick Causgrove Alex Krutov

Sara Frankowiak

1850 M Street N.W., Suite 300 Washington, D.C. 20036-5805

Š 2011 American Academy of Actuaries. All rights reserved.


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

TABLE OF CONTENTS

1. Executive Summary....................................................................................... 1 2. Purpose and Scope ....................................................................................... 3 3. Background—History and Intent of Program ................................................. 4 4. Structure of the National Flood Insurance Program....................................... 4 5. The Premium Rate Structure of the NFIP...................................................... 5 6. Actuarial Principles, Actuarial Soundness, and the NFIP .............................. 7 7. The Write-Your-Own (WYO) and Direct Programs ...................................... 12 8. Key Differences Between Private-Sector Insurance and the NFIP .............. 14 9. Legal Issues ................................................................................................ 17 10. Mandatory Purchase Requirement .............................................................. 19 11. Interactions Between Federal Disaster Aid and Flood Insurance ................ 21 12. Multiple-Loss Properties .............................................................................. 23 13. Riverine and Flash Flooding Issues Versus Coastal Flooding..................... 24 14. Map Modernization ...................................................................................... 25 15. FEMA Evaluations ....................................................................................... 26 16. Potential Congressional Reforms ................................................................ 28 17. Summary and Conclusions.......................................................................... 30 Appendix A ......................................................................................................... 31 Appendix B ......................................................................................................... 39 Appendix C ......................................................................................................... 40 Appendix D ......................................................................................................... 46 Appendix E ......................................................................................................... 48

American Academy of Actuaries

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

1. Executive Summary Flood insurance in the United States primarily is provided by the federal government via the National Flood Insurance Program (NFIP), in partnership with private insurers and servicing contractors. In the aftermath of the 2004 and 2005 hurricanes that struck the East and Gulf coasts of the United States, and in consideration of the substantial losses suffered in those storms, there have been calls for reform of the program. But since the NFIP is substantively different from typical insurance, few insurance professionals and public policymakers are sufficiently familiar with the NFIP to recognize the broad consequences of changing it. This monograph is focused on the background and the current structure of the program and the primary issues surrounding the program today. The U.S. Congress passed the National Flood Insurance Act of 1968 (NFIA) to provide a means by which the risk of flood could be insured in the United States. Private-sector insurance companies long had viewed the risk of flood events as uninsurable. This act created the NFIP, a mechanism by which the federal government could act as the insurer. The NFIP is a far-reaching program sponsored by the federal government and administered, in large part, by nearly 90 property and casualty insurers. The program connects and influences the behavior of many constituencies in the United States, including home and business owners; builders; local building permitting and flood plain management officials; lenders; insurers; state, local, and federal government officials; and regulators. Because the NFIP is a public program that encompasses public policy as well as insurance goals, developing actuarially sound premium rate structures has been one of the program’s many important objectives. The NFIP, as a public insurance program, poses some significant actuarial challenges, as the structural differences between it and private sector insurers result in differences in what constitutes an “actuarially sound” premium level. These and other key differences are outlined in this paper. From its inception in 1968, the NFIP has been guided by its three foundations:   

Flood risk identification. Mapping the flood risks of each community and publishing the Flood Insurance Rate Maps (FIRMs); Flood plain management. Promulgating minimum building and flood plain management standards and encouraging communities to exceed the minimum standards; Flood insurance. Providing a mechanism for individuals to prefund the risk of flood losses.

In addition, a long-term goal of the NFIP has been to reduce the demand for and reliance on disaster assistance after floods.

American Academy of Actuaries

1

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

Several issues are highlighted in this monograph: 

 

1

Pursuant to the NFIA, the enabling legislation of the NFIP, there are two basic types of premium rates: (1) “risk premium,” “full-risk,” or “actuarial” rates, and (2) “other than risk premium,” “discounted,” or “subsidized” rates. Each type presents a challenge to actuarial principles. Actuarial rates, for which expected future losses are used as a base, reflect the other costs of risk transfer differently than private sector insurance. Subsidized rates are lower than actuarial rates and thus will be inherently inadequate to fully fund future losses. Within the NFIA, flood risk premium rates are to be “based on consideration of the risk involved and accepted actuarial principles … to make such insurance available on an actuarial basis …” The NFIP policy form provides that all disputes arising from claims-handling are governed by federal regulations. Regarding agent activities at the point of sale, however, the NFIA states that an agent or broker is not to be held harmless for error or omission. In addition, extra-contractual causes of action have been used to file claims arising from flood policies in state court. A mandatory purchase requirement exists for certain property owners in special flood hazard areas (SFHAs). Since its inclusion in the Flood Disaster Protection Act of 1973, the mandatory purchase requirement has increased the market penetration rate. Despite this increase, challenges to its enforcement remain. The availability of assistance from disaster aid programs after a flood event does not lessen the importance of maintaining flood insurance. The viewpoint of some at-risk consumers, however, is that flood insurance is not necessary. That impression might have been exacerbated by the extent to which government aid flowed to victims in the wake of the hurricane events of 2005. Repetitive loss properties seem to be much more at risk than the average property insured by the NFIP. Whether the disproportionate cost to the NFIP from these properties should be addressed through pricing and/or process changes is controversial. Properties most at risk for flooding are those located near rivers and/or coasts. The requirements of these two groups of properties differ, and those differences can cause conflicts regarding political and funding concerns. The differences between riverine and coastal flooding also contribute to a perception that the premium rates in hurricane-prone states are subsidized by those that are not hurricane-prone. Flood hazard maps must be updated continually to better identify properties located within SFHAs. The Risk Mapping, Assessment and Planning (MAP) program provides digital access to and dissemination of new maps, which is expected to reduce the costs associated with such tasks. For the first time, the new flood hazard maps accurately reflect environmental changes and technological advancements.1 The Federal Emergency Management Agency’s (FEMA) view is that the current Risk MAP program, like the Flood Map Modernization Program before it, will enhance the quality, reliability, and availability of the maps.

http://www.dhs.gov/ynews/testimony/testimony_1299855117667.shtm (last visited on June 29, 2011).

American Academy of Actuaries

2

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

The NFIP is dependent upon congressional action to remain operational past its statutory sunset date. The current sunset date is Sept. 30, 2011, the result of the most recent of several temporary extension bills, some of which were retroactively passed by Congress and signed by President Obama following the program’s scheduled expiration. In early 2010, the NFIP was permitted to lapse three separate times before extensions were passed. In each instance, enacting legislation extending the NFIP was made retroactive to the lapse dates. To ensure the ongoing viability of the NFIP following unprecedented loss activity in 2004 and 2005, members of Congress advanced several proposals during subsequent legislative sessions. None of the proposed reforms passed in that session of Congress or since. At the time of this publication, in the 112th Congress, neither the House nor the Senate had brought legislation to their respective floors, although such action has been anticipated. This monograph provides actuarial insights on key issues addressed by proposed reforms, as well as examination of the perspectives of several major stakeholder groups.

2. Purpose and Scope This monograph is presented to inform the taxpaying public, federal and state policymakers and regulators, actuaries, agents, and other insurance professionals about the NFIP so that they may participate in and contribute to the public debate with a comprehensive, financial frame of reference. It contains discussion of the background and intent of the program, an outline of federal legislative and regulatory actions that have affected flood insurance in the United States, and an examination of how the program has evolved over time. And, perhaps most significantly, this monograph also identifies key differences between the NFIP and conventional, privately-offered insurance found in the marketplace. In addition, there is an examination of some of the important issues that underlie recent discussions, with explanation of the background and specific considerations of each issue and a description of how it would affect the financial condition of the NFIP. This monograph is not intended to be an in-depth examination of the actuarial soundness2 of the ratemaking and financial structure of the NFIP. Rather, this paper is intended primarily to provide an educational foundation upon which to discuss the key issues affecting the NFIP. While an in-depth analysis of the ratemaking and financial structure of the NFIP from an actuarial standpoint would be a valuable study, it is beyond the scope of this monograph.

2

As used herein, actuarial soundness is explained in the Casualty Actuarial Society (CAS) Statement of Principles Regarding Property and Casualty Insurance Ratemaking, located at http://www.casact.org/standards/princip/sppcrate.pdf (last visited on June 29, 2011).

American Academy of Actuaries

3

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

3. Background—History and Intent of Program3 There have been numerous changes to the program since its inception in 1968. Many of the changes were prompted by large flood-loss events, many of which triggered significant claims payments under NFIP policies. A large proportion of the payments have been made for losses caused by hurricanes, tropical storms, and major riverbank flooding. The most costly event to date was Hurricane Katrina, which made landfall on Aug. 29, 2005, and caused losses greater than the prior total amount paid out by the program.4 A comprehensive summary of the background, history, and intent of the program can be found in Appendix A of this monograph.

4. Structure of the National Flood Insurance Program The NFIP is administered, in large part, by nearly 90 property and casualty insurers.5 A separate direct flood vendor administers policies written directly with the government. The program affects many constituencies nationwide, including home and business owners; builders; realtors; local building permitting and flood plain management officials; lenders; insurers; insurance agents; state, local, and federal government officials; and regulators. In its role as the manager of the NFIP, FEMA identifies and maps areas of flood risk, promotes the appropriate management of the flood plain, and provides insurance for properties insured by the NFIP. These services are intended to reduce disaster aid payouts by requiring flood-exposed property owners to contribute to the cost of their potential losses through the purchase of insurance. The structure and administration of the NFIP is complicated. The NFIP is directed by the Federal Insurance and Mitigation Administration of FEMA, which is part of the U.S. Department of Homeland Security. The insurance operations of the NFIP are carried out mostly by the participating property and casualty insurers (the write-your-own [WYO] companies), which operate under a business arrangement with FEMA governed by statute and regulation.6

3

“A Chronology of Major Events Affecting The National Flood Insurance Program,” December 2005. Completed for the Federal Emergency Management Agency Under Contract Number 282-98-0029. The American Institutes for Research, the Pacific Institute for Research and Evaluation, Deloitte & Touche LLP. 4 http://www.gao.gov/new.items/d07169.pdf (last visited on June 29, 2011). 5

http://www.fema.gov/nfipInsurance/search.do;jsessionid=B77A973206922C733DE236E1EA5B5C6F.Wor kerPublic2 (last visited on June 29, 2011); “The Choice is Yours – WYO Companies Actively Writing Flood Insurance 2010-2011,” FEMA F-073 (3/11). 6

http://www.fema.gov/library/file;jsessionid=24FA340AF1B550553838D4AD8C1F8841.Worker2Library?t ype=originalAccessibleFormatFile&file=guidewyo.txt&fileid=6ebeff10-0941-11e0-a865-001cc4568fb6 (last visited on June 29, 2011).

American Academy of Actuaries

4

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

Two important programmatic features of the NFIP, both of which have operational impacts, are the existence of its sunset provision and its periodic need to borrow money from the U.S. Treasury to pay claims. Sunset Provision: The NFIA contains a sunset provision.7 The NFIP’s expiration date has been extended from time to time by Congress, generally for five-year periods, but sometimes on a more temporary, stopgap basis. The stopgap extensions generally have been adopted when Congress was in the midst of a more comprehensive NFIP review of reauthorization, with the intention of drafting reform legislation that would extend the program for longer periods of time. The sunset provision has the potential to cause concern to NFIP stakeholders if Congress delays in setting a new expiration date as the sunset date approaches, primarily because the sunset provision leaves open the possibility of a lapse in the NFIP. NFIP Borrowing Authority: The National Flood Insurance Act of 1968 contains a specific cap on the NFIP’s borrowing authority. The cap originally was $1 billion. In 1996, Congress raised it to $1.5 billion.8 After the catastrophic claims from the 2005 hurricanes, especially Dennis, Katrina, Rita, and Wilma, Congress raised the NFIP’s borrowing authority several times. The current NFIP borrowing cap is $20.725 billion, established in June 2010 by the National Flood Insurance Extension Act of 2010.9 The amount that NFIP borrows from the U.S. Treasury cannot exceed the existing cap. When NFIP borrowing approaches the existing cap, the NFIP warns the WYO companies and the NFIP servicing agent to be prepared to stop making claims and other payments related to their flood programs. The sunset provision and the borrowing authority cap can be perceived at times as critical weaknesses of the NFIP. Congress can and does delay extending the NFIP or delay increasing the borrowing authority cap. If Congress were to postpone such decisions during critical times, such as after major flooding events, unfortunate dislocations could occur, such as delays in payments to claimants and discontinuation of claims-handling activities. Additional details of the NFIP’s oversight structure are provided in Appendix B.

5. The Premium Rate Structure of the NFIP The NFIP is a public program that encompasses social goals through insurance goals. Developing actuarially sound premium rate structures has been an important consideration of the program.

7

National Flood Insurance Act of 1968, Section 1319. US Code, Title 42, Chapter 50, Subchapter I, Section 4016, Amendments, 1996 – Subsection (a)(2). 9 http://www.gpo.gov/fdsys/pkg/PLAW-111publ196/pdf/PLAW-111publ196.pdf (last visited on June 29, 2011). 8

American Academy of Actuaries

5

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

The original NFIA established the NFIP providing for classes of business to be priced using risk premium rates (usually referred to as full-risk or actuarial rates) as well as classes that have “other than risk premium rates” (usually referred to as subsidized or discounted rates). Sections 4014 and 4015, respectively, of Title 42 of the U.S. Code provide the legislative basis for those two general premium classifications.10 At times, Congress and FEMA have prioritized societal and marketing goals, such as increasing the policies in force and gaining acceptance of new FIRMs by affected communities, over developing and maintaining full-risk rates. FEMA’s actuarial staff annually publishes an NFIP Actuarial Rate Review memorandum.11 This memorandum describes the NFIP’s premium-rate determination methodology and provides explanations for rate changes, along with updated statistics. For example, the memo published in support of the Oct. 1, 2010, rate changes included Exhibits A and D, which are reproduced in Appendix C here. Those exhibits provide NFIP policy distribution data and other information about premiums. Approximately 80 percent of NFIP policyholders receive full-risk rates. NFIP premium rates are reset annually by class of business, with about half of the full-risk rates determined using a hydrologic/financial model originally developed by the U.S. Army Corps of Engineers. The other half of the full-risk rate properties primarily are located outside the SFHAs, where not enough detailed information exists to use the hydrologic/financial model for rate-setting. Actuarial and engineering judgments and underwriting experience are used to set rates for those areas. The NFIP actuarial staff periodically conducts analyses of claims, trends of in force growth, and expenses by class of business to update the model. To determine rate classifications, structures are categorized by flood zone according to their location on a FIRM, their elevation relative to the base flood elevation (BFE), and by occupancy type (e.g., by residential versus nonresidential), along with other specific determinants of risk. Subsidized premium rates are determined differently than full-risk rates. Details about the NFIP’s rate-making process can be found in Appendix C. The rationale for allowing subsidized classes of business was to permit the large inventory of structures (known as pre-FIRM structures) that were built in SFHAs prior to the general implementation (in approximately 1974) of FIRMs and flood-related building codes to be covered by flood insurance at reasonable rates. Also, a goal of the NFIP always has been one of encouraging participation, even if that meant that some property owners would pay actuarially inadequate premiums. Those subsidized property owners have been and are prefunding at least part of the cost of their flood losses. This provides additional NFIP premium reserves to fund losses as well as, ideally, lessening the public burden of providing future disaster assistance. In addition, widespread participation in the 10

http://codes.lp.findlaw.com/uscode/42/50/I/4014 (last visited on June 29, 2011).

11

http://www.fema.gov/library/file;jsessionid=A6C64BEBC128FC0F1F83DE142E0D724A.Worker2Library ?type=publishedFile&file=actuarial_rate_rev2009.pdf&fileid=55182cb0-94ea-11df-a6e7-001cc4568fb6 (last visited on June 29, 2011).

American Academy of Actuaries

6

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

NFIP engenders public awareness of flood dangers and encourages local officials to take the flood plain-management actions necessary to make their communities safer. The NFIP estimates that approximately one-fifth of current policyholders are paying subsidized rates.12

6. Actuarial Principles, Actuarial Soundness, and the NFIP The enabling legislation that established the NFIP specifically provided for two distinct classes of business, which were differentiated by two types of premium rates: 

Risk premium rates, more commonly known as full-risk or actuarial rates, which would be “based on consideration of the risk involved and accepted actuarial principles,”13 and

Other than risk premium rates, more commonly known as subsidized rates, “which would be reasonable, would encourage prospective insureds to purchase flood insurance and would be consistent with the purposes of” the legislation.14

The enabling statute and the actions of the NFIP determined that the program would not be actuarially sound in the aggregate, because the premiums for the policies that receive subsidized rates are not expected to match their full long-term costs. In fact, even some classes of policies subject to full-risk rates may not be considered actuarially sound because of statutory requirements to provide premium rates that ignore specific known risks (sometimes temporarily), for specific groups of policyholders.15 Those known inadequacies, however, can be compensated for in the aggregate by increasing the overall level of rates. Although full-risk rates generally are intended to be calculated according to accepted actuarial principles, agreement among the various stakeholders about what accepted actuarial principles should be for a program like the NFIP remains elusive. There are many differences between the NFIP and private-sector insurance programs that affect appropriate application of actuarial principles and determination of actuarial soundness. One key difference involves the cost of capital. Because private-sector insurance companies must prefund their losses, they place large amounts of capital at risk, and they generally must earn a profit. This creates a need for those insurers to include a risk load component that includes the cost of capital in their rates. The NFIP does not require a risk load because the federal government could, theoretically at least, provide unlimited liquidity and credit. 12

http://www.gao.gov/new.items/d11429t.pdf (last visited on June 29, 2011). Title 42 of the U.S. Code, Section 4014, paragraph (a)(1)(A). 14 Title 42 of the U.S. Code, Section 4014, paragraph (a)(2). 15 Title 42 of the U.S. Code, Section 4014, paragraphs (e) and (f), provide that, under certain specific conditions, if a community is making “adequate progress on the construction of a flood protection system,” or if the community is actively in the process of adequately restoring such a flood protection system (primarily referring to dams or levees), the buildings so protected are eligible for flood insurance rates as if the protection system was already accredited to meet FEMA’s standards of protection. 13

American Academy of Actuaries

7

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

Instead of prefunding their losses, the NFIP can handle deficits after major events by borrowing funds from the U.S. Treasury. For the NFIP, the differences that affect ratemaking include mandatory purchase requirements, specific statutory items like the 10 percent cap on premium increases, the legislative mandate of FEMA, and the fact that the NFIP is not expected to hold required capital or earn a profit. Actuarial Standards of Practice and Statements of Principles There are a large number of Actuarial Standards of Practice (ASOPs), promulgated by the Actuarial Standards Board housed within the American Academy of Actuaries; and Statements of Principles (SOPs), promulgated by the Casualty Actuarial Society (CAS);16 which constitute the body of currently accepted actuarial principles for property insurance ratemaking and risk classification. Much of that guidance is applied similarly by NFIP actuaries as by those in the private sector. The public nature of the program and FEMA’s public policy goals, however, sometimes conflict with the goal of achieving actuarial soundness. The following are the key standards and other resources that are particularly relevant to those sometimes conflicting goals: 

ASOP No. 30, Treatment of Profit and Contingency Provisions and the Cost of Capital in Property/Casualty Insurance Ratemaking

ASOP No. 12, Risk Classification (for all practice areas)

CAS, Statement of Principles Regarding Property and Casualty Insurance Ratemaking

CAS, Statement of Principles—Risk Classification

ASOP No. 41, Actuarial Communications

ASOP No. 38, Using Models Outside the Actuary’s Area of Expertise (Property and Casualty)

As discussed below, the NFIP’s actuarial methodology differs from private-sector actuarial practice primarily in two areas: cost of capital; and the classification of risks. Actuarial Principles Regarding NFIP Rates Because the NFIP’s overall premiums are inadequate by design, the program should be expected in some years to produce deficits that will not be made up over time. In the early 16

The CAS’s purposes “are to advance the body of knowledge of actuarial science applied to property, casualty and similar risk exposures, to establish and maintain standards of qualification for membership, to promote and maintain high standards of conduct and competence for the members, and to increase the awareness of actuarial science.” See www.casact.org (last visited on June 29, 2011).

American Academy of Actuaries

8

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

years of the NFIP, the Federal Insurance Administration reduced rates several times to encourage participation.17 Then, in 1981, the NFIP initiated a multiyear series of rate increases for all subsidized policies, which made the program more fiscally sound.18 From the mid-1980s until August 2005, the NFIP essentially was self-supporting; it was able to pay back to the U.S. Treasury all loans incurred over that period.19 Hurricane Katrina changed that. (During 2008 and 2009, Congress debated whether to forgive the NFIP’s $17.75 billion debt to the U.S. Treasury in the post-Katrina environment. At least in the initial period of the 112th Congress, legislative proposals did not include the disposition of the outstanding Treasury debt.) In accordance with relevant actuarial principles, the basis for the NFIP’s full-risk rates is the expected values of annual losses, including those due to catastrophic events, differentiated by rating class. Also, net premiums incorporate the expected values of all expenses of the NFIP, including the annual expenses of maintaining the FIRMs (though not the prior Map Modernization program20, which was funded by a congressional appropriation). Investment income is not considered in the rates; it is assumed to be immaterial. In a departure from actuarial principles as recommended by ASOP No. 30, there has not historically been calculated an explicit cost of capital in the NFIP ratemaking process. It is argued that the federal government provides the capital backing of the NFIP in the form of its guarantee that all legitimate claims will be paid. But the NFIP is not expected to earn any return on capital. The NFIP does have contingency loadings in its gross premium rates.21 Actuarial Principles Regarding NFIP Risk Classes Another deviation from private-sector actuarial practices is in the classification of risks. The largest variation in practice is evidenced in the subsidized rates discussed above. The NFIP has a number of subsidized risk classes, comprising more than 20 percent of the in force policy base.22 Subsidized risk classes aside, additional deviations from typical private-sector programs regarding risk classification include:

17

http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 19 http://www.fas.org/sgp/crs/misc/R40650.pdf (last visited on June 29, 2011); see also http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 20 Created and funded by Congress, the Map Modernization program was intended to create flood maps for use by the NFIP that more accurately reflect recent developmental and natural changes in the environment. The program utilized revised data and improved technologies to identify flood hazards and better reflect actual risk. See http://www.fema.gov/plan/prevent/fhm/mm_why.shtm (last visited on June 29, 2011). 21 The current loadings are 20 percent of net premiums for the most risky buildings, considered to be those located in the V-zones (buildings exposed to the water velocities due to wave motion), and 10 percent for all other risks. Those loadings are primarily designed as a cushion to mitigate the extreme volatility in losses from flood events, but they also serve to compensate for possible underestimations of catastrophic losses and other assumptions that may turn out to be non-conservative in the long run. 22 http://www.gao.gov/new.items/d11429t.pdf (last visited on June 29, 2011). 18

American Academy of Actuaries

9

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

Encouragement of sound flood plain management practices and the rapid adoption of FIRMs by local communities have led to the practice of permanently grandfathering, on a less than full-rate basis, buildings that were built in compliance with an existing FIRM at the time but are now no longer compliant, based on a subsequent FIRM. FEMA, however, compensates for the grandfathered buildings by raising rates in the B, C, and X zones, such that overall rates for those zones are actuarially adequate.

The NFIP is subject to a statutory cap on annual premium increases of 10 percent by risk class.23 That restriction may have, at times, led to inadequate premiums for certain risk classes.

Due to market forces and the need to mitigate against adverse selection, privatesector insurance programs tend to have a large number of relatively homogeneous risk classes. As a public program, the NFIP generally is not subject to the same market forces as the private sector. To facilitate the operations of the program, and because of its unique public policy goals, NFIP risk classes therefore are very broad.24 As a result, there are only five major risk classes nationally with separately differentiated rates: o AE zone, which describes “areas subject to inundation by the 1-percentannual-chance flood event determined by detailed methods.”25 Rates are differentiated by elevation relative to base flood elevation (BFE) o VE zone, which describes “areas subject to inundation by the 1-percentannual-chance flood event with additional hazards due to storm-induced velocity wave action.”26 Rates are differentiated by elevation relative to BFE. o X zone, which describes “moderate flood hazard areas … and are the areas between the limits of the base flood and the 0.2-percent-annual-chance (or 500-year) flood.27 Describes standard risks outside of special flood hazard areas (SFHA), that is, in B, C, and X-zones o Preferred risk policies (PRPs)—for preferred risks, located in B, C, and X zones o Subsidized—for pre-FIRM buildings located in SFHAs

23

http://www.cbo.gov/ftpdocs/86xx/doc8648/hr3121.pdf (last visited on June 29, 2011). When the NFIP was first created, rating distinctions were much finer. For example, the AE and VE zones each were divided into separate subzones based on topographies and were refined further based on community-specific rating factors. In the late 1970s, the rating scheme was simplified to its current state. 25 http://www.fema.gov/plan/prevent/floodplain/nfipkeywords/zone_ae.shtm (last visited on July 7, 2011). 26 http://www.fema.gov/plan/prevent/floodplain/nfipkeywords/zone_ve.shtm (last visited on July 7, 2011). 27 http://www.fema.gov/plan/prevent/floodplain/nfipkeywords/flood_zones.shtm (last visited on July 7, 2011). 24

American Academy of Actuaries

10

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

For example, an AE-zone building located in a West Virginia river valley at a specific elevation would be charged the same premium as a similar AE-zone building with the same coverage details and elevation rating that was located in a flat South Carolina flood plain— regardless of whether the two buildings had significantly different flood-loss histories and assuming both had the same community rating system status. Actuarial Principles Regarding Actuarial Soundness The Casualty Actuarial Society’s Statement of Principles Regarding Property and Casualty Insurance Ratemaking sets forth the following four principles for a set of premium rates to be considered actuarially sound: 

Principle 1: A rate is an estimate of the expected value of future costs.

Principle 2: A rate provides for all costs associated with the transfer of risk.

Principle 3: A rate provides for the costs associated with an individual risk transfer.

Principle 4: A rate is reasonable and not excessive, inadequate, or unfairly discriminatory if it is an actuarially sound estimate of the expected value of all future costs associated with an individual risk transfer.

It is also stated in actuarial guidance that, if a law or regulation conflicts with the provisions of an actuarial standard, the actuary should develop rates in accordance with the law or regulation.28 For policies that are subject to full-risk rates, the NFIP’s rate making process can be said to follow Principle 1 above. The lack of a cost of capital provision in the NFIP rates could be viewed as falling short of Principle 2. On the other hand, the NFIP’s unique position as an insurance program backed by the federal government could preclude the need for a cost of capital element in the premium rates. The ubiquity of grandfathering and the NFIP’s wide rate classes could be interpreted as counter to Principle 3 because there are cross-subsidies within rating classes. On the other hand, in administering any insurance system, managers should balance the cost of estimating an individual risk transfer and the expense of maintaining a system of extensive rate classifications. Private sector insurance systems tend to have substantially more detailed data and therefore can develop more refined rate structures. Even then, most probably could be found to contain some cross-subsidies within their rate classifications. To make the NFIP rating scheme more specific, it would have to collect more refined data. The extent to which Principle 4 may be violated depends on conclusions reached about whether the NFIP’s structure violates Principles 1-3. 28

Introduction to the Actuarial Standards of Practice, Section 4.6.2, http://www.actuarialstandardsboard.org/pdf/asops/Introduction_113.pdf (last visited on June 29, 2011).

American Academy of Actuaries

11

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

The administration of the NFIP involves many unique considerations that differentiate it from any private sector insurance program.

7. The Write-Your-Own (WYO) and Direct Programs The NFIP’s interface with insurance producers and the general public is through the WYO program and the direct program. The WYO program, a cooperative undertaking of FEMA and the private sector insurance industry, began in 1983.29 Since that time, the WYO program gradually has become the dominant distributive and administrative arm of the NFIP. As a result, the WYO program accounts for a large majority of the approximately 5 million NFIP policies currently in force.30 The federal government backs the insurance contracts of the NFIP entirely. Although the participating insurers have farreaching operational involvement in the WYO program, they bear none of the underwriting risk. In June 2010, one of the WYO companies, which handled approximately 15 percent of NFIP policies, announced its withdrawal from the WYO program, effective as of Sept. 2010. The company’s stated reason for withdrawing was based in part on the numerous program interruptions caused by delays in reauthorization and reform of the NFIP. Since 2002, there have been 11 last-minute reauthorizations of the NFIP, and, on several occasions, the program was allowed to lapse. These lapses and resumptions of coverage require a large company to dedicate significant resources to coordinating communications with its customers, employees, and agents. The withdrawal could strain the resources of the NFIP Direct program, which must service the affected policyholders. This withdrawal is an example of the potential market disruptions that can result from failure to reauthorize the program for an extended period of time in a timely fashion. The stated goals of the WYO program31 are to:   

Increase the NFIP policy base and the geographic distribution of policies; Improve service to NFIP policyholders through the infusion of insurance industry knowledge; and Provide the insurance industry direct operating experience with flood insurance.

The WYO program operates within the NFIP and is subject to its rules and regulations. The WYO program allows participating property and casualty insurance companies to write and service federal flood insurance in their own names. The companies receive an expense allowance for marketing, policies written, policy administration, and claims processed, while the federal government retains responsibility for underwriting and claims policy, product design, and pricing and underwriting losses. Individual WYO companies may, to the extent possible, and consistent with WYO program rules and regulations, conform their flood business to their normal business practices for other lines of insurance. But given the differences between the rules and regulations to which the 29

http://www.fema.gov/business/nfip/wyowhat.shtm (last visited on June 29, 2011). http://www.gao.gov/new.items/d11297.pdf (last visited on June 29, 2011). 31 http://www.fema.gov/business/nfip/wyowhat.shtm (last visited on June 29, 2011). 30

American Academy of Actuaries

12

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

WYO companies are subject and those of the rest of the NFIP, and the substantive congressional changes to the NFIP, conformity is not always possible.32 FEMA sets the rates, coverage limitations, and eligibility requirements, while the WYO companies perform all of the policy administration for their customers. Flood insurance coverage is issued by the WYO companies as a separate policy from all other coverages provided by the WYO companies. The WYO companies, essentially, are fiscal agents of the federal government, while the federal government, essentially, is the guarantor of all flood insurance coverage. The companies are directly responsible for their obligations to insureds. The federal government is precluded from being made a party to any lawsuit arising out of distributional and/or coverage disputes within the WYO program. The WYO company is responsible for administering and defending claims. But with respect to the direct program, in which flood insurance policies are placed directly with the NFIP, those flood policies are contracts directly between FEMA and the insured. FEMA would be the defendant in the case of any such direct program-related lawsuit. For further detail on how disputes are handled, please refer to Legal Issues, addressed in Section 9 below. Each of the WYO companies is responsible for following the laws and rules set forth for the distribution and underwriting of the flood policies that it issues and for settling the claims of those policies. For these services, they are reimbursed under the terms of the Financial Assistance/Subsidy Arrangement (the standard insurance contract used by WYO companies, which is described in greater detail in Appendix A), for various expense allowances, fees, and production bonuses. The WYO companies collect premiums from policyholders. Under the Financial Control Plan, described in Appendix D, the WYO companies must keep NFIP funds separate from the rest of their accounts. In accordance with the Financial Assistance/Subsidy Arrangement, WYO companies deduct their expense allowances from the premiums, and the servicing agent deducts its agent commissions from the premiums. Other NFIP-related payables also are deducted from the premiums. When congressional authorization or appropriation of funds for the NFIP is withdrawn, a WYO company could be required to discontinue issuing new policies immediately. There are those that have argued that the levels of expense reimbursements to WYO companies are too generous, while others have argued that the reimbursement levels are insufficient to cover all expenses associated with servicing flood policies under the rules and regulations required by Congress and FEMA. Within the past few years, proposed congressional reforms have included a requirement that the expenses of WYO companies be studied in detail.33 From the perspective of stakeholders’ competing interests, there is simultaneous need for WYO companies to be provided with incentives to participate in the NFIP and for the premiums to have a reasonable expense provision. The direct program allows individual insurance producers to submit flood insurance business directly to the NFIP rather than through a WYO company. This program is 32

NFIP Flood Insurance Manual, http://www.fema.gov/pdf/nfip/manual200705/02ref.pdf (last visited on June 29, 2011). 33 http://www.policyarchive.org/handle/10207/bitstreams/19249.pdf (last visited on June 29, 2011).

American Academy of Actuaries

13

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

administered by a federal contractor known as the NFIP Servicing Agent. Like the WYO companies, the NFIP Servicing Agent collects premiums from policyholders. The NFIP servicing agent is a contractor chosen through a periodic bidding process. Further detail on the operations of the WYO companies is provided in Appendix D.

8. Key Differences Between Private-Sector Insurance and the NFIP The NFIP is a public insurance program. There are significant differences between the NFIP and private sector property and casualty insurance. Key differences are outlined below. 

The goals of the NFIP are very different from the goals of private sector insurance companies. As stated previously, the purposes of the NFIP are: 1) identifying flood risk, 2) regulating flood plain management, and 3) providing flood insurance. A fourth longer-term goal of the NFIP has been to reduce federal expenditures on disaster assistance after floods.34 The NFIP also has the power to require coverage in some cases and to require certain flood plain management practices for communities to participate in the NFIP.35 By contrast, a primary motivation for private sector insurance companies is to earn a profit by providing for the needs of their customers through appropriate insurance coverages. Unlike the NFIP, private sector insurance companies have no power to require that their customers buy coverage from a particular company or to take specific actions to manage their risks.



NFIP flood policy contract language is provided by federal statute and/or regulation.36 The insured may not be able to assert that he/she did not know or understand the policy in coverage disputes. The NFIP requires that coverage disputes arising under the program be litigated in federal courts.37 In the private sector insurance industry, litigation often arises over ambiguities in policy language, and, because of the general principle in contract law that a contract must be construed against the drafter, the courts interpret disputed policy

34

Aug. 1, 2002 NFIP Program Description, FEMA, page 2, http://www.fema.gov/library/file;jsessionid=CCE6DB0C006D7874AE4633336B5A985A.Worker2Library ?type=publishedFile&file=nfipdescrip_1_.pdf&fileid=e6fdab40-80bd-11db-9aa6-000bdba87d5b (last visited on June 29, 2011). 35 Aug. 1, 2002 NFIP Program Description, FEMA, page 29, http://www.fema.gov/library/file;jsessionid=CCE6DB0C006D7874AE4633336B5A985A.Worker2Library ?type=publishedFile&file=nfipdescrip_1_.pdf&fileid=e6fdab40-80bd-11db-9aa6-000bdba87d5b (last visited on June 29, 2011). 36 http://www.fema.gov/good_guidance/download/10040 (last visited on June 29, 2011). 37

https://www.floridabar.org/divcom/jn/jnjournal01.nsf/c0d731e03de9828d852574580042ae7a/c5c4744b3ca 071088525741a004ac73d!OpenDocument&Highlight=0,* (last visited on June 29, 2011).

American Academy of Actuaries

14

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

language in favor of the insured.38 Most litigated coverage disputes arising out of private sector insurance policies, unlike those pertaining to the NFIP, are heard in state courts. 

The types of coverages and insurance limits provided by the NFIP are set by statute and regulation, and they differ from coverage provided under a typical personal lines property policy in the private sector. o As of the time of publication, the NFIP had a maximum coverage limit ($250,000 building/$100,000 personal property on dwelling policies, $500,000 building/$500,000 personal property on non-residential buildings).39 On the other hand, the limits available in the private sector insurance market are as high as any company is willing to sell. o The personal property coverage in the NFIP is actual cash-value coverage.40 In the private market, on the other hand, replacement-costvalue coverage typically is available, at least as an option. o Additional living expenses are not covered by the NFIP, and business interruption coverage is not presently covered by NFIP commercial policies.41 On the other hand, most private policies offer some coverage for such expenses at an additional cost.

NFIP policy rates are developed differently than those in the private sector. o The NFIP’s flood policy rates do not include a profit provision that includes the cost of capital.42 Private-sector insurance policies include a profit provision sufficient to cover all costs of risk transfer. A privatesector insurance company must maintain and build capital to preserve its solvency. The NFIP can use a lower standard partly because it has the ability to avoid running a deficit by borrowing from the U.S. Treasury, when necessary. o NFIP’s flood program rate changes do not need approval by state regulatory authorities. Conversely, in the private sector, rates are closely monitored by state regulators and are subject to filing and approval requirements that can vary by state. o Flood insurance rates for pre-FIRM properties are promulgated by regulation to be subsidized at a level that is below that of actuarially sound

38

http://courts.phila.gov/pdf/cpcvcomprg/sylva-dh.pdf (last visited on June 29, 2011).

39

http://www.fema.gov/library/file;jsessionid=CF770CB79157EBBD1E5C20AE72DD4F71.WorkerLibrary? type=publishedFile&file=f679_sumcov0709.pdf&fileid=e0da41c0-10fb-11df-921d-001cc456982e (last visited on June 29, 2011). 40 Standard Flood Insurance Policy as of May 1, 2005. 41 Standard Flood Insurance Policy as of May 1, 2005. 42 Aug. 1, 2002 NFIP Program Description, FEMA, page 28, http://www.fema.gov/library/file;jsessionid=CCE6DB0C006D7874AE4633336B5A985A.Worker2Library ?type=publishedFile&file=nfipdescrip_1_.pdf&fileid=e6fdab40-80bd-11db-9aa6-000bdba87d5b (last visited on June 29, 2011).

American Academy of Actuaries

15

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

rates.43 In the private sector, insurers typically charge actuarially sound rates. o There are no regulatory capital requirements in the NFIP. Flood insurance is backed by the full faith and credit of the United States.44 Private-sector insurance companies, on the other hand, are monitored by state regulators for solvency and meet various capital requirements designed to maintain their standing with rating agencies. 

As noted above, significant changes to NFIP’s coverage, policy administration, and operations are accomplished largely by federal statute and/or regulation. As has been demonstrated in the past several years, the implementation of changes to the NFIP often takes a significant amount of time. In private-sector insurance, however, individual companies regularly adjust in response to the market.

Congress provides oversight of the NFIP. The congressional committees with NFIP oversight authority are the House Committee on Financial Services and the Senate Committee on Banking, Housing and Urban Affairs. The NFIP is also overseen by the executive branch via FEMA and the Department of Homeland Security. Unlike the NFIP, private-sector companies are overseen by their boards of directors and owners as well as to state regulators.

The NFIP is not allowed to refuse to cover an “eligible” property, regardless of the property’s loss history. Ineligible structures are few and are proscribed by the federal program. The private-sector insurance industry, on the other hand, is able to accept or reject applications for policies based on the underwriting guidelines of each individual company (subject to the constraints of applicable state statutes or regulations).

The NFIP is not authorized to operate indefinitely. The continuation of the NFIP depends upon congressional action prior to each established sunset date. Should Congress fail to reauthorize the NFIP, it is possible that all existing flood insurance policies would cease to be enforceable, expiring policies would not be renewed, and new policies would not go into effect until the NFIP was reauthorized. Unlike the NFIP, private companies may operate indefinitely, so long as they are financially solvent.

43

Aug. 1, 2002 NFIP Program Description, FEMA, page 26, http://www.fema.gov/library/file;jsessionid=CCE6DB0C006D7874AE4633336B5A985A.Worker2Library ?type=publishedFile&file=nfipdescrip_1_.pdf&fileid=e6fdab40-80bd-11db-9aa6-000bdba87d5b (last visited on June 29, 2011). 44 Aug. 1, 2002 NFIP Program Description, FEMA, page 28, http://www.fema.gov/library/file;jsessionid=CCE6DB0C006D7874AE4633336B5A985A.Worker2Library ?type=publishedFile&file=nfipdescrip_1_.pdf&fileid=e6fdab40-80bd-11db-9aa6-000bdba87d5b (last visited on June 29, 2011).

American Academy of Actuaries

16

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

9. Legal Issues As previously discussed, the NFIP is unique in that it functions as a federal program, not as a private-sector insurance program. Congress established the NFIP to share the risk of flood losses by underwriting flood insurance coverage in communities that adopt and enforce local flood plain regulations that meet or exceed NFIP criteria. The administrator of the NFIP is FEMA, which has established a comprehensive regulatory structure setting forth the rights and responsibilities of insureds and insurers under the NFIP. The subsequent creation of the WYO program allowed private-sector insurance companies to issue standard government policies and collect policy premiums. Under the WYO program, private-sector insurance companies essentially become fiscal agents of the United States. The federal government is not a party to any lawsuit arising out of WYO program-related distributional and/or coverage disputes. FEMA regulations require a WYO company to defend claims against it, but FEMA reimburses the WYO company for its defense costs. In the direct program, by contrast, the policy is a contract directly between FEMA and the insured, and FEMA defends in any subsequent lawsuit. Most courts that have considered the issue have concluded that the NFIA’s language confers federal district court jurisdiction on suits that are based on the handling and disposition of NFIP claims. These decisions recognize the intent of Congress to create a national program for flood insurance, noting the federal government’s extensive participation in the NFIP, its administrative and financial responsibilities pursuant to the NFIA, and the absence of statutory language allowing claims under the NFIA to be brought in state court.45 The NFIP policy form provides that all disputes arising from claims handling are governed by federal regulations and the National Flood Insurance Act of 1968, as amended. Disputes alleging improper administration or adjustment of NFIP claims are governed exclusively by federal jurisdiction. Such claims are essentially breach-ofcontract claims, and claimant remedies are limited therefore to those provided pursuant to the policy itself. The flood insurance policy form specifically states the following conditions for filing a lawsuit (note that the terms “us” and “we” refer to the WYO company): You may not sue us to recover money under this policy unless you have complied with all the requirements of the policy. If you do sue, you must start the suit within 1 year after the date of the written denial of all or part of the claim, and you must file the suit in the United States District Court of the district in which the insured property was located at the time of loss. This requirement applies to

45

Craig, Lee, and Wegryzn, Lisa E., “Federal Preemption of Extracontractual Claims Under Flood Insurance Policies,” Mealey’s Litigation Report: Bad Faith, Vol. 12, #16, Dec. 15, 1998; http://www.butlerpappas.com/showarticle.aspx?Show=534#N_33_ (last visited on July 7, 2011).

American Academy of Actuaries

17

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

any claim that you may have under this policy and to any dispute that you may have arising out of the handling of any claim under the policy.46 Although FEMA covers the expenses of WYO insurers in paying out claims and in litigating challenges in federal court, the NFIP’s enabling legislation states a clear exception to this rule: “[t]he Director of the Federal Emergency Management Agency may not hold harmless or indemnify an agent or broker for his or her error or omission.”47 Federal funds are not at stake in cases against WYO insurers in which the cause of action relates to the procurement of flood insurance. Thus, such claims have been considered state-law tort claims rather than federal-law contract claims. Insured claimants have used various extra-contractual causes of action to file claims arising from a flood policy in state court.48 Such causes of action have included bad faith, fraudulent misrepresentation, unfair trade practices, and, in some of those contexts, requests for noneconomic damages.49 Pursuant to the WYO program, FEMA has elected to have state-licensed insurance companies, agents, and brokers sell flood insurance to consumers. Private-sector insurance companies participating in the WYO program must be licensed and regulated to engage in the business of property insurance in states in which they wish to sell flood insurance.50 State regulations require that insurance company agents and brokers provide NFIP customers with the same service that the states require of them in selling other lines of insurance.51 In its role as coordinator of the NFIP, FEMA must ensure, through monitoring and oversight, that its programs are implemented across the nation in accordance with statutory and regulatory requirements. The Flood Insurance Reform Act of 2004 mandated the implementation of several NFIP management reforms intended to improve transparency for policyholders. Reforms included:

46

NFIP Standard Flood Insurance Policy, as of May 1, 2011; http://www.fema.gov/library/file;jsessionid=AD2236DC678374257A2A6A26200BF7A5.WorkerLibrary?t ype=publishedFile&file=f122dwellingform0809.pdf&fileid=7c7d32a0-11a5-11df-921d-001cc456982e (last visited on June 29, 2011). 47 42 U.S.C. 4081(c), http://codes.lp.findlaw.com/uscode/42/50/II/C/4081 (last visited on June 29, 2011). 48 Craig, Lee, and Wegryzn, Lisa E., “Federal Preemption of Extracontractual Claims Under Flood Insurance Policies,” Mealey’s Litigation Report: Bad Faith, Vol. 12, #16, Dec. 15, 1998; http://www.butlerpappas.com/showarticle.aspx?Show=534#N_33_ (last visited on July 7, 2011). 49 Craig, Lee, and Wegryzn, Lisa E., “Federal Preemption of Extracontractual Claims Under Flood Insurance Policies,” Mealey’s Litigation Report: Bad Faith, Vol. 12, #16, Dec. 15, 1998; http://www.butlerpappas.com/showarticle.aspx?Show=534#N_33_ (last visited on July 7, 2011). 50 Frequently-Asked Questions About the NFIP: http://www.floodsmart.gov/floodsmart/pages/faqs.jsp (last visited on June 29, 2011). 50 108 U.S.C.S. 2238-16. 51 Id.

American Academy of Actuaries

18

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

    

Addition of supplemental forms explaining the specific coverage being purchased; Addition of a flood insurance claims handbook describing the process for filing and appealing claims; Establishment of a regulatory appeals process; Requiring NFIP education and training for insurance agents; Implementation of a claims-sampling strategy that provides FEMA management with information used to assess the overall performance of the WYO companies, including the overall accuracy of the underwriting of NFIP policies and the adjustment of claims.52

10. Mandatory Purchase Requirement53 The Flood Disaster Protection Act of 1973 (P.L. 93-234) made the purchase of flood insurance mandatory for certain property owners and the National Flood Insurance Reform Act of 1994 (P.L. 103-325) made adjustments to that requirement.54 Property owners in SFHAs who obtain loans from federally-regulated lending institutions are required to purchase and retain flood insurance for the life of their mortgage loans.55 Mortgages from nonregulated lenders, typically private mortgage companies, are not subject to the requirement, unless such mortgages are issued by subsidiaries of regulated lenders or subsequently are sold to the Federal National Mortgage Association (Fannie Mae) or the Federal Home Loan Mortgage Corporation (Freddie Mac). The NFIP is a voluntary program based on an agreement between the federal government and agents of the participating community; the community’s governing body must pass a protective land-use ordinance establishing protective floodplain development standards.56 Before doing so, communities assess their flood hazard and determine whether flood insurance and flood plain management would benefit their residents and economy. If a community does not participate in the NFIP, a lender can offer only a conventional loan and is required to inspect any flood maps, determine flood hazard risk, and provide notice of such risk. However, as noted above, the purchase of flood insurance is a mandatory prerequisite to obtaining mortgage loans from federally-regulated lending institutions on buildings located in a SFHA. GAO studies of the mandatory purchase requirement found mixed levels of compliance, although compliance appears to have increased considerably since 52

42 U.S.C. 4011. “The National Flood Insurance Program’s Mandatory Purchase Requirement: Policies, Processes, and Stakeholders,” Mar. 2005. The American Institutes for Research.

53

54

http://www.fema.gov/library/file;jsessionid=EAB659D3BED51A78ADCF525E30295264.Worker2Library ?type=publishedFile&file=frm_rf94.pdf&fileid=ceaa94b0-9821-11db-b057-000bdba87d5b (last visited on June 29, 2011). 55

http://www.wvdhsem.gov/WV_Disaster_Library/Library/FLOODS/NFIP%20Info%20for%20Prospective %20Buyers.htm (last visited on June 29, 2011). 56 http://www.commerce.state.ak.us/dca/logon/plan/planning-flood.htm (last visited on July 7, 2011).

American Academy of Actuaries

19

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

the passage of the 1994 act.57 To address situations in which coverage is required and not purchased, lenders and servicers have several options. First, regulated lenders are obligated to refuse to extend a loan until the building that secures the loan is covered by flood insurance. Second, if a loan complies with the purchase requirement at origination but later is found to be noncompliant, lenders can purchase a standard flood insurance policy (SFIP) for the property if the property owner has not done so. Third, lenders and servicers who discover that an SFHA property lacks flood insurance can obtain coverage after loan origination through the Mortgage Portfolio Protection Program (MPPP).58 FEMA created the MPPP in 1991 to facilitate lenders’ and servicers’ efforts to ensure that affected property owners have coverage when underwriting information is limited or unavailable. Policies purchased through the MPPP almost always are more expensive than standard flood insurance policies. The rates for MPPP policies are high due to a lack of information and the resulting unknown risk level; high rates also discourage forced placement.59 Because “force-placing” a policy pursuant to the MPPP requires lenders and servicers to wait for the expiration of a mandated gap in coverage, many banks prefer lender-placed, private market flood insurance over MPPP coverage.60 Private-sector insurance can be used in lieu of NFIP coverage at loan origination to provide additional coverage when the value of a property exceeds the amount of coverage available through the NFIP or when a lender or servicer concludes that coverage through the NFIP is not commensurate with the value of the property or the law’s requirements for coverage. Private-sector insurance also can be used in situations in which NFIP coverage is not available, in communities that are suspended from participation in the NFIP due to their failure to adopt or enforce flood plain management regulations, and in all units of the Coastal Barrier Resource System.61 The mandatory purchase requirement is restricted to properties in SFHAs, thus emphasizing the importance of accurate FIRMs. If SFHAs accurately identify areas of high risk, then a majority of claims should originate from within these areas. The claims history in several Eastern states displays this pattern. Across the United States and within many states, however, a different pattern emerges. Nearly 69 percent of all NFIP-insured properties are in SFHAs, but only 64 percent of claims are from these areas. A more dramatic pattern emerges in 13 states, where more than half of all claims have been from policyholders outside of SFHAs.62 These data call into question the exemption of property owners outside SFHAs from the mandatory purchase requirement, even though many of them appear to face at least as much risk as owners within SFHAs. The data also underscore the importance of FEMA’s 57

http://www.fema.gov/doc/nfip/mandpurch_0305.doc (last visited on July 6, 2011). http://tinyurl.com/4yfb9jm (last visited on July 6, 2011). 59 Call for Issues Status Report. Washington, DC: FEMA. Available at http://tinyurl.com/3z2ehut (last visited on July 6, 2011). 60 http://tinyurl.com/3aopare (last visited on July 6, 2011). 61 More information on the Coastal Barrier Resource System is available in Appendix E. See http://www.fema.gov/plan/prevent/floodplain/nfipkeywords/cbrs.shtm (last visited on June 29, 2011). 62 The National Flood Insurance Program’s Mandatory Purchase Requirement: Policies, Processes, and Stakeholders, Mar. 2005. The American Institutes for Research, p. 58. 58

American Academy of Actuaries

20

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

Risk MAP program, which could improve the delineation and understanding of areas at high risk of flooding.

11. Interactions Between Federal Disaster Aid and Flood Insurance At times, government emergency aid is available to disaster victims after a flood event. As a result, a misconception arises in the public’s perception that the purchase of flood insurance is not necessary because money from disaster aid programs will be available to bail them out. That viewpoint may have been burnished further, in part, by the widely publicized large amounts of government aid made available to victims in the wake of hurricane events that have occurred since Hurricane Katrina made landfall in 2005.

Source: Hartwig, Robert P. Written testimony delivered to the United States Senate Committee on Banking, Housing and Urban Affairs, Oct. 18, 2005.

But the availability of assistance from disaster aid programs does not lessen the importance of maintaining flood insurance. Pursuant to the Robert T. Stafford Disaster Relief and Emergency Assistance Act (42 U.S.C. 5121-5207), before federal disaster assistance can be offered, the president must declare the area a major disaster.63 The Stafford Act authorized FEMA to create a Federal Response Plan (FRP) to address the provision of federal aid after disasters. The FRP subsequently was superseded, first by the National Response Plan and, later, by the National Response Framework. 63

http://training.fema.gov/EMIWeb/downloads/is7unit_3.pdf (last visited on June 29, 2011).

American Academy of Actuaries

21

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

A disaster declaration only can come in response to a request by the governor of the state affected by the disaster. A gubernatorial request typically is based on a damage assessment to determine loss and recovery needs. The gubernatorial request is evaluated by a group of regional FEMA representatives, which makes a recommendation to the President. Localized flooding usually does not qualify as a major disaster. Flood insurance policyholders are eligible to receive government aid, but that aid typically is reduced by the amount of any insurance proceeds a claimant receives. Assistance from FEMA may be provided to cover losses that are uninsured and are otherwise eligible for aid. But insurance deductibles, paid for by insureds, are not covered under FEMA’s programs. If a claimant does not have NFIP flood insurance, and he/she receives government aid for flood damage, he/she is required to obtain flood insurance coverage for the future. Excluding the aid available through the NFIP’s enabling statute and the statutes that have modified it since, other government assistance, when and if available, comes primarily in the form of low-interest loans that must be repaid.64 The U.S. Small Business Administration may make federally subsidized loans to repair or replace homes, personal property, or businesses that sustained damages not covered by insurance. Disaster grants and housing from FEMA might be available to meet serious disaster-related needs that are not met in other ways, including but not limited to:     

Temporary housing, Repairs to or replacement of damaged property, Medical costs, Clothes and household items, Limited disaster grants.

Only a relatively small fraction of overall disaster assistance overlaps with the insurance coverage available from the NFIP. Outright grants, provided by FEMA’s Individuals and Households Program, usually are small and intended to meet only essential needs that are not otherwise covered.65 Financial assistance to repair property damage is limited to an amount that will make a home inhabitable as quickly as possible. Replacement of damaged personal property is limited to “items or services that help prevent or overcome a disaster-related hardship.”66 There is no focused, institutionalized oversight or accountability for the money made available by the federal government for post-disaster assistance. The Government Accountability Office (GAO) recommends that Congress identify and track the types and amount of federal assistance provided for addressing catastrophic events and develop metrics to inform its oversight.67

64

http://www.iii.org/media/updates/archive/press.739566 (last visited on June 29, 2011). http://www.fema.gov/pdf/assistance/process/help_after_disaster_english.pdf (last visited on June 29, 2011). 66 http://www.fema.gov/assistance/dafaq.shtm#qd_15 (last visited on June 29, 2011). 67 http://www.gao.gov/new.items/d07235r.pdf (last visited on June 29, 2011). 65

American Academy of Actuaries

22

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

12. Multiple-Loss Properties One area of controversy involves NFIP coverage of properties that have suffered multiple flood losses. These properties have shown that they are at greater risk than the average risk insured by the NFIP. These properties, known as severe repetitive-loss (SRL) properties, have been defined in various ways. SRL properties currently are defined as properties that have experienced:  Four or more paid NFIP losses, with the amount of each exceeding $5,000 and the total exceeding $20,000 or  At least two such claims, with the cumulative amount exceeding the value of the property68 In its 2004 report on the subject, the GAO looked at all locations that had two or more claims over the past 10 years.69 These properties had accounted for some 38 percent of all claim dollars since 1978. But about half of them were still insured, amounting to only about 1 percent of the then-insured properties. From 1978 to 2003, the total cost of multiple-loss properties’ claims was approximately $4.6 billion. There are two major concerns with this situation. One, obviously, is the cost to the program. The other is the question of whether the NFIP should continue to insure properties that are likely to have further losses and whether these properties are being subsidized by the rest of the NFIP insureds. The NFIP currently cannot refuse to cover an eligible property, and ineligible properties are few. The 1994 NFIP Reform Act authorized mitigation grants and post-flood insurance as part of the standard flood insurance policy. This means that if a home or business is damaged by a flood, the property owner may be required to meet certain building requirements in the community to reduce future flood damage before repairing or rebuilding. To help cover the costs of meeting those requirements, the NFIP includes Increased Cost of Compliance (ICC) coverage for all new and renewed Standard Flood Insurance Policies. It was expected that this increased cost of compliance built into the standard flood insurance policy would encourage post-flood mitigation. Some argue, however, that ICC methods have not been utilized as effectively as anticipated to increase post-flood mitigation. ICC claimants must have received substantial damage declarations from their communities, and the ICC benefit only can be used to bring structures into compliance with existing building code flood requirements. While there have been proposals in the past to impose a surcharge on multiple-loss properties, Congress has not chosen to approve any surcharges except those in the 2004 68

As set forth in the Flood Insurance Reform Act of 2004; applies to one in four family residential properties. 69 “National Flood Insurance Program, Actions to Address Repetitive Loss Properties,” testimony before the Subcommittee on Economic Policy, Committee on Banking, Housing and Urban Affairs, U. S. Senate, Mar. 25, 2004, Statement of William O. Jenkins Jr., director homeland security & justice issues, U. S. GAO.

American Academy of Actuaries

23

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

NFIP Reform Act, which dealt with the disproportionate cost of insuring SRL properties in a limited way. FEMA accordingly asserts that it is somewhat limited in its ability to help mitigation of SRL properties without congressionally mandated changes like the 2004 Act.70 FEMA does have several mitigation grant programs that provide funding to reduce the flood-risk profiles of eligible buildings. One such program, the Severe Repetitive Loss (SRL) Pilot Program, specifically targets SRL structures.71 The SRL pilot program assists communities in acquisition, elevation, relocation, demolition, and flood-proofing of these properties. Owners who refuse mitigation assistance could have their flood premiums increased 150 percent. In addition, each future claim in excess of $1,500 may trigger an additional rate increase of 150 percent, though this option is limited to properties that are subject to nonsubsidized full-risk rates. The results of this pilot program may help determine future actions on other repetitiveloss properties. Properties identified as severe repetitive loss properties are handled by the NFIP Special Direct Facility, not a WYO company. This allows NFIP to supervise these properties more closely and make appropriate mitigation decisions.

13. Riverine and Flash Flooding Issues Versus Coastal Flooding Properties at greatest risk of flooding are those located near rivers or coasts. Many other types of properties are at risk of flash flooding. Several issues arise because of the differences between properties near coasts and other properties. 

Riverine flooding occurs when rivers and streams overflow their banks, breach levees, or breach dams and flood adjacent land. Flash floods occur when very heavy rainfall overwhelms storm drainage systems, causing localized but very heavy storm flooding. Flash floods inundate buildings with water, typically causing damage to first floors and below-ground floors. Floods in narrow valleys occasionally can carry enough force to destroy entire buildings, but such areas are typically small, with relatively few buildings at risk. Coastal storm surge accompanies intense ocean storms and damages buildings within the run-up of the surge. Storm surge destroys buildings on coasts by the combined force of the inundation topped with damaging waves while undermining the ground beneath buildings. With the dense concentration of highvalued structures along most of our hurricane-exposed coasts, high claims payouts are a virtual certainty.

70

“The NFIP and Repetitive Loss Properties: Issues, Strategies and Proposed Actions,” testimony before the Subcommittee on Economic Policy, Committee on Banking, Housing and Urban Affairs, U.S. Senate, Mar. 25, 2004, presented by Chad Berginnis, chair, Association of State Floodplain Managers, Inc. 71 The SRL Pilot Program provides funds to NFIP-participating communities to reduce or eliminate longterm flood risks to SRL properties, in turn reducing expenditures by the NFIF. Additional information about these and other programs is available at http://www.fema.gov/government/grant/mit_grant_fact_sheet.shtm (last visited on June 29, 2011).

American Academy of Actuaries

24

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

There is a large difference in market penetration between areas with riverine flooding exposure and those on the coast. Recent NFIP statistics show that 69 percent of policies are in the 10 hurricane-exposed states between Virginia and Texas. (Florida alone has 38 percent of the policies.) Those policies account for 60 percent of the in force premium for the NFIP; those same states have accounted for 80 percent of the losses paid from Jan. 1, 1978 through April 30, 2011. Another 11 percent of the policies come from the nine Northeast coastal states between Maryland and Maine. The rest of the country, therefore, accounts for only 20 percent of the policies.72

14. Map Modernization73 Flood Map Modernization (Map Mod) was a multiyear initiative that concluded in 2010. Map Mod was designed to improve and update flood hazard identification maps. The successor program to Map Mod is Risk Mapping, Assessing and Planning (MAP). FEMA’s flood maps have been produced and used for 35 years in identifying flood hazard areas and setting flood insurance rates. The maps also have been more widely used for other purposes, including community planning and emergency preparedness. The flood hazard identification maps originally were created using traditional paper mapmaking methods. The modernization of the maps was first addressed in the National Flood Insurance Reform Act of 1994. Over the years, changes in mapmaking technology and in flood hazard conditions, along with increased knowledge of those conditions, resulted in the near-obsolescence of many of the NFIP maps. These maps are relied upon for proper risk classification and assessment of community risk levels. Inaccuracies in the maps could lead to premium inequities and flawed community decisions. Updated maps will account for revised data and use improved technologies to identify flood hazards. The quality, reliability, and availability of the maps are to be increased through improved methodologies and technology. FEMA also developed a five-year plan called the Multi-Year Flood Hazard Identification Plan (MHIP) for providing periodically updated digital flood hazard maps. The plan’s funding expired in 2010, though the opportunity to comment on its efficacy remains available on the FEMA website.74

72

See http://www.fema.gov/business/nfip/statistics/pcstat.shtm (last visited on June 28, 2011; data as of Apr. 30, 2011). 73 Information in this section was provided by www.fema.gov. 74 http://www.floodmaps.fema.gov/fhm/scripts/mh_surv.asp (last visited on June 29, 2011).

American Academy of Actuaries

25

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

15. FEMA Evaluations The NFIP Evaluation The Evaluation of the National Flood Insurance Program was a major project initiated by FEMA in 2000 to review the progress of the NFIP in working toward achieving its legislative mandate and to obtain recommendations for future actions and policies that would enhance the NFIP. It was completed in 2006, and all of the evaluation research papers are publicly available online.75 According to the Evaluation’s final report: One purpose of the National Flood Insurance Act of 1968 was to authorize “continuing studies of flood hazards…in order to provide for a constant reappraisal of the flood insurance program and its effect on land-use requirements.” This clear call for evaluation and the fact that the NFIP had never been the subject of a comprehensive evaluation led FEMA in 2000 to contract with the American Institutes for Research (AIR), an independent, not-for-profit corporation, to design, lead, and manage the Evaluation of the NFIP. The Evaluation consisted of more than a dozen individual research studies, using widely varying methods, that focused on a range of subjects determined to be critical to assessment of the NFIP’s progress …76 The Evaluation includes 15 reports, including 13 major research papers. The totality is a compilation of research and data that comprises more than 1,500 pages and more than 200 recommendations. One item of interest is contained in NFIP Evaluation Report 14, The National Flood Insurance Program’s Market Penetration Rate: Estimates and Policy Implications, by the RAND Corp., presents a wealth of data, much of which is based upon the RAND Corp.’s intensive study of a sample of 100 NFIP communities. NFIP coverage is greatly concentrated in the Gulf Coast states—38 percent of NFIP policies are in Florida alone. That is partly due to the heavy concentration of buildings in SFHAs in the South. For example, based on the 100-community sample, nearly 60 percent of single-family residences within SFHAs are located in the South. But even after taking that into account, the Midwest and the Northeast are two regions in which the market penetration within SFHAs is markedly low. The paper examines many factors that affect market penetration, including the cost of flood insurance, differences between inland and coastal communities, differences in enforcement of the mandatory purchase requirement, size of communities, number and percentage of communities’ structures located in SFHAs, value of homes, and years since the last major flooding event. Because of the RAND study’s limited sample size, the fact that it was limited to 75 76

http://www.fema.gov/business/nfip/nfipeval.shtm (last visited on June 29, 2011). The Evaluation of the National Flood Insurance Program – Final Report, p. ix.

American Academy of Actuaries

26

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

single-family homes, and the difficulty in obtaining valid data about the number of structures in SFHAs, solid inferences about the drivers of market penetration cannot be made. But the study demonstrates that the NFIP would be financially improved with a more geographically diversified policy base. The study suggests a number of areas worthy of further inquiry, including the lower rates of penetration in inland communities, collection of more data on compliance with mandatory purchase, and extension of the study to other types of structures and occupancies beyond single-family homes. Relevant findings from the Evaluation’s Final Report include: In general, the Evaluation concludes that the NFIP is moving towards achievement of its goals. The progress made to date is impressive compared with the state of knowledge about and management of floodprone areas in 1968, although it has perhaps been slower than had been anticipated at the outset. Two notable trends have contributed to this progress: there is more widespread acceptance by local governments of the need for land use management to minimize flood damage; and there is broader support for various measures aimed at protecting and preserving natural resources, including streams, wetlands, and other floodplain features. In the face of the considerable accomplishments of the NFIP, noted below, it is nevertheless clear that the future will require even more strenuous efforts to combat flood losses. Past strategies are unlikely to remain adequate to the challenge of the increased losses expected to occur as a result of population growth and movement and the pressure to build in even more hazardous and sensitive areas, such as the coastal zone…77 When analyzing the future of the NFIP, changes in orientation, differing perceptions, and the management of the program must be considered. Many specific and detailed recommendations are made in the Evaluation’s Final Report and in the 13 evaluation substudies. Some of those recommendations include: 

   

77

Revision of the NFIP flood-hazard-mapping criteria to identify natural functions worthy of preservation, high hazard areas that should be avoided, areas protected by flood control structures, and areas of known flood hazard, as well as to reduce the need to revise the maps over time; Revision of the NFIP flood plain management criteria by adding stronger provisions that have been proven to be effective and by encouraging local programs to adopt other, higher regulatory standards; Devotion of more resources to improving state and local flood plain management programs; Refinement of the tools for, and full funding of, a comprehensive strategy to reduce losses to existing buildings; Revision of insurance procedures to encourage greater coverage and take steps to increase compliance with the mandatory purchase requirement;

Ibid., p. x; http://tinyurl.com/692h4rg (last viewed on July 6, 2011).

American Academy of Actuaries

27

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

Implementation of known techniques that protect natural functions while also reducing damage, offering a variety of resource-protection incentives, and coordinating more closely with other federal and state resource protection programs; and Gathering and maintenance of needed data, use of it to measure progress towards the goals of the program, and sharing of it with Congress and the NFIP's other stakeholders.78

16. Potential Congressional Reforms As mentioned above, the NFIP depends upon action by the legislative and executive branches to remain operational past the current sunset date of Sept. 30, 2011. As a result of unprecedented loss activity in 2004 and 2005, some members of Congress advanced proposals during that and subsequent legislative sessions, intended to ensure the ongoing viability of the NFIP. The proposed reforms have attempted to address key issues and considerations from the perspectives of a variety of major stakeholder groups, including: Adequacy and Availability of Funds    

Increase FEMA’s borrowing authority to pay claims from the 2005 catastrophic hurricanes as well as other ongoing obligations and/or forgive the debt that the NFIP has incurred as a result of these storms. Phase out subsidized rates for flood insurance policies on certain vacation homes, second homes, and non-residential buildings constructed before applicable maps went into effect. Increase annual limitation on premium increases from 10 percent to 20 percent.79 Require that expenses of the program for WYO companies be studied in detail (as discussed in Section 7 above).

Without a change in law, the NFIP might not be able to pay future flood claims promptly. FEMA, WYO companies, and policyholders face uncertainty about the availability of funds for future claims and program expenses. As of the time of publication, Congressional action was uncertain about possible re-authorization of and future substantial borrowing by NFIP and subsequent forgiveness of such borrowing. Adequacy of Coverages  

Increase existing flood insurance coverage limits, which have not changed since 1994. Currently they are $250,000/$100,000 (structure/contents) for residential buildings and $500,000/$500,000 for non-residential buildings. Introduce coverage for additional living expenses, business interruption, basements, and replacement cost of contents.

78

Ibid., p. xiii. http://www.gpo.gov/fdsys/pkg/BILLS-112hr1309ih/pdf/BILLS-112hr1309ih.pdf (last visited on June 29, 2011).

79

American Academy of Actuaries

28

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

   

Notify tenants of availability of contents coverage. Clarify replacement cost provisions, forms, and policy language. Reduce the required waiting period from 30 days to 15 days before flood insurance takes effect. Remove the waiting period for the effective date of policies on a home being purchased or transferred when purchase of coverage is made within 30 days.

Many policyholders would like higher coverage limits and/or broader coverage options. But these changes may necessitate higher premiums or further financial strains on the program. WYO companies are concerned that changes be administratively manageable. WYO companies also expect to be compensated for additional costs resulting from any changes. Some members of Congress have expressed concerns about their constituents’ desire for broader coverage availability, the effect on their constituents of higher rates, and additional exposure to the program. Wind/Flood 

Direct the NFIP to offer optional coverage for wind events. Requirements for the availability of wind coverage could include risk-based premiums and the agreement of local governments to adopt and enforce building codes designed to minimize wind damage, in addition to the existing NFIP requirements for flood plain management. Require NFIP participation in state mediation in claims involving both a flood component and a wind component (in situations in which wind coverage is secured in the private market).

A proposal to add wind coverage to the NFIP has been controversial. The Flood Insurance Subcommittee of the American Academy of Actuaries commented on H.R. 3121, the 2007 NFIP reform bill, in a letter to the Speaker of the House of Representatives. The Academy Subcommittee identified several financial issues that its members believed should be taken into account when assessing that bill. Its concerns included:     

The increased potential for further large losses in excess of available funds; Political pressure on Congress to suppress rates; Effects of high loss severity and loss volatility on pricing, particularly the need for a large contingency load; Potential issue of cross-subsidies between properties immediately along the coast and those inland, and the effect of cross-subsidization on collected premiums; and A requirement that the NFIP cease issuing and renewing policies if the NFIP borrows funds from the U.S. Treasury and until that loan is repaid.

Increased Participation in the NFIP 

Establish a grant program for communities that sign up homeowners for nonmandatory purchase of flood insurance.

American Academy of Actuaries

29

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

  

Request GAO study of methods to increase flood insurance program participation among low-income families. Request GAO study to review mandatory flood insurance purchase requirements, their effects on properties in areas protected by levees, and the effects of expanding the purchase requirements to non-federally regulated lenders. Increase penalties for lender noncompliance with mandatory purchase requirements.

Many property owners are unaware of their flood exposure and the availability of coverage. NFIP’s financial health would be improved by having a wider geographic spread of policies across the country.

17. Summary and Conclusions Because the peril of flood was deemed uninsurable by the private sector insurance market, the federal government established the National Flood Insurance Program to provide flood insurance coverage to property owners. The NFIP is a federal government/private sector program. The NFIP has many stakeholders, and the policies associated with it greatly affect certain categories of people in the United States. The NFIP has provided communities and property owners with a better understanding of how to evaluate their flood risks and protect against the volatile nature and potentially devastating consequences of flood events. The NFIP also has focused attention on how to better protect and use the flood plains throughout the United States. It has created a successful working partnership between the federal government and private sector insurance companies in marketing and administering an insurance program for the American public. The complexity of the NFIP, combined with the perception of its selective impact, creates a dilemma. Critical factors such as coastlines, rivers and streams, building construction and the use of land, the scientific understanding of hydrology, and the technologies used to measure and address flood risk constantly are changing, creating opportunities for constructive NFIP reform. The significant losses suffered in the 2004, 2005, and 2008 hurricanes prompted a flurry of proposals to modify the NFIP. It is important for policymakers to understand the issues involved in, and potential consequences of, possible revisions to the NFIP. The authors hope that this monograph helps readers develop a clearer overall view of the NFIP and provides a context for evaluating the complex issues surrounding it.

American Academy of Actuaries

30

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

Appendix A Background—History and Intent of Program80 Early History Several important federal actions preceded the 1968 establishment of the NFIP. In 1917, Congress approved a Flood Control Act.81 By appropriating $45 million for a long-range and comprehensive program of flood control for the lower Mississippi and Sacramento rivers, Congress accepted federal responsibility for flood control.82 Congress then passed the Flood Control Act of 1936, which provided for the construction of approximately 250 projects.83 The Act used funds for work relief. It established a two-pronged attack on the problem of reducing flood damages: the Department of Agriculture would develop plans to reduce runoff and retain more rainfall and the U.S. Army Corps of Engineers would develop engineering plans for downstream projects.84 The Act represented the initial steps toward the development of a national flood-control program. By 1929, the private sector insurance industry had stopped covering flood losses.85 Based on data from floods that occurred between 1951 and 1954, a 1956 study on floods and flood losses for the American Insurance Association provided further reinforcing data to illustrate the insurance industry’s conviction that flood insurance products were not commercially feasible. In response to the 1933 Long Beach, Calif., earthquake, and contrary to its past decisions, Congress passed legislation to provide direct assistance to private citizens suffering from disaster damage by issuing federal loans through the Reconstruction Finance Corp.86 Nearly 20 years later, the Disaster Relief Act of 1950 created the first permanent system for disaster relief.87 Following massive flooding in 1951, President Harry Truman recommended the creation of a “national system of flood disaster insurance.”88 The following year, he submitted a proposal to Congress to establish a national system of flood-disaster insurance.89 80

A Chronology of Major Events Affecting The National Flood Insurance Program, December 2005. Completed for the Federal Emergency Management Agency Under Contract Number 282-98-0029. The American Institutes for Research, The Pacific Institute for Research and Evaluation, Deloitte & Touche LLP. 81 http://www.mvd.usace.army.mil/mrc/history/AppendixD.htm (last visited on June 29, 2011). 82 Id. 83 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 84 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 85 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 86 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 87 The Disaster Relief Act of 1950 created the first disaster relief system that did not require a Congressional act to serve as its trigger. See http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 88 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 89 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011).

American Academy of Actuaries

31

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

Early Federal Actions Despite President Truman’s early efforts, no law providing a federal source of flood insurance was enacted until the Federal Flood Insurance Act of 1956, which directed the Housing and Home Finance Agency to establish a program of federal insurance and reinsurance against the risks of losses resulting from floods and tidal disasters.90 The Housing and Home Finance Agency created the Federal Flood Indemnity Administration to carry out the tasks set forth in the Federal Flood Insurance Act of 1956.91 No technical studies were undertaken to determine the costs of starting a federal program for flood insurance, however, and Congress did not appropriate any funds for the Federal Flood Indemnity Administration.92 As a consequence, it ceased to exist.93 The National Flood Insurance Act of 1968 (Title XII of the Housing and Urban Development Act of 1968) created the NFIP and the Federal Insurance Administration (FIA), within the Department of Housing and Urban Development, to provide flood insurance in communities that voluntarily adopted and enforced flood plain management ordinances that met minimum NFIP requirements.94 The premiums of policyholders of structures in flood-plains were subsidized.95 Occupants of structures built in flood plains after the Act’s passage were to pay actuarially based premiums.96 Section 1302(c) states, “The objectives of a flood insurance program should be integrally related to a unified national program for flood-plain management.”97 The NFIP was authorized to borrow up to $1 billion from the U.S. Treasury to cover losses that exceeded the program’s revenues.98 In late 1968, the industry’s flood insurance pool, the National Flood Insurers Association (NFIA), was created in accordance with sections 1331 and 1332 of the National Flood Insurance Act.99 The NFIA was administered by the Insurance Services Office. Membership in the NFIA was open to all qualified companies licensed to write property insurance under the laws of any state.100 The companies would sell and service policies written as part of the NFIP. Six months later, the Department of Housing and Urban Development and the NFIA signed an agreement for the marketing of flood insurance policies and the adjustment of

90

http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 92 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 93 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 94 http://www.gao.gov/htext/d05532t.html (last visited on June 29, 2011). 95 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 96 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). The law ultimately defined subsidized pre-FIRM buildings as those on which “construction or substantial improvement … started on or before December 31, 1974, or before the effective date of the initial Flood Insurance Rate Map (FIRM) of the community, whichever is later.” [emphasis added] 97 http://codes.lp.findlaw.com/uscode/42/50/4001 (last visited on June 29, 2011). 98 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 99 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 100 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 91

American Academy of Actuaries

32

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

claims.101 Pursuant to the agreement, the NFIA would appoint a servicing company, generally on a statewide basis, to disseminate information to the public and insurance agents about the insurance aspects of the program, to process all insurance policies, and to handle the adjustment of claim payments for losses.102 Throughout the early 1970s, NFIP’s subsidized rates for flood insurance were lowered several times to encourage participation in the program.103 The Flood Disaster Protection Act of 1973 amended the National Flood Insurance Act of 1968.104 Among other provisions, the new Act included the following: 

 

A requirement that property owners in participating communities purchase flood insurance as a condition of the receipt of federal or federally related financial assistance for the acquisition, construction, or improvement of structures in special flood hazard areas (SFHAs). In addition, property owners had to purchase flood insurance to be eligible to obtain federal disaster assistance for construction or reconstruction purposes.105 A requirement stating that, as a condition of future federal financial assistance, states and communities “participate in the flood insurance program and …adopt adequate flood plain ordinances with effective enforcement provisions consistent with federal standards to reduce or avoid future flood losses.”106 A provision for grandfathering, for purposes of determining insurance rates, of structures built in flood-hazard areas before the areas were so identified.107 A mandate that federally regulated lending institutions cannot make, increase, extend, or renew any loan on a property located in an SFHA in a participating community without requiring flood insurance.108

The Housing and Community Development Act of 1974 amended the National Flood Insurance Act to require federally regulated lenders to notify prospective borrowers of a property’s location in an SFHA.109 In January 1978, the federal government assumed the direct insurance writing and claims handling operation of the NFIP, using an NFIP Servicing Agent to handle the sales and servicing responsibilities.110 Prospective policyholders continued to use local agents and

101

http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 103 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 104 http://www.federalreserve.gov/boarddocs/rptcongress/flood/default.htm (last visited on June 29, 2011). 105 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 106 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 107 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 108 http://www.federalreserve.gov/boarddocs/rptcongress/flood/default.htm (last visited on June 29, 2011). 109 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 110 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 102

American Academy of Actuaries

33

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

brokers to obtain their policies.111 The FIA and the NFIP were transferred in 1979 to the newly created FEMA.112 In January 1983, the GAO found that, due to data and methodological weaknesses in determining the rate structure, the NFIP had not collected sufficient premiums to cover the cost of providing insurance to nearly 2 million policyholders.113 As a result, the FIA had to borrow $854 million from the U.S. Treasury between 1970 and 1980.114 In October 1983, some private-sector insurance companies entered into an arrangement with the FIA to sell and service flood insurance under the newly created WYO program.115 During the WYO program’s first year, 48 companies agreed to become WYO participants.116 The first WYO policies were sold in November 1983.117 This dramatically increased insured participation. WYO companies eventually would become the principal source of flood insurance. Later Federal Actions In 1990, the Community Rating System (CRS) was created. It is a voluntary program intended to recognize and encourage NFIP communities that implement flood-prevention and flood plain management measures that go beyond the basic standards required by the NFIP.118 Under the CRS, communities receive credit for more restrictive regulations; acquisition, relocation, or flood proofing of flood prone buildings; preservation of open space; and other measures that may reduce flood damages or protect the natural resources and beneficial functions of flood plains. Flood insurance premium rates for structures located in CRS communities are adjusted to reflect the reduced flood risk resulting from community activities, such as those listed above, that meet the three goals of the CRS:   

Reduce flood losses to insurable property, Strengthen and support the insurance elements of the NFIP, and Encourage a comprehensive approach to floodplain management.

There are 10 CRS classes, ranging from Class 1, which obtains the most credit points and receives the largest premium reductions, to Class 10. A Class 10 community is not very active in CRS and receives no premium reduction. CRS premium discounts range from 45 percent for Class 1 communities to 5 percent for Class 9 communities for structures 111

http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 113 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 114 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 115 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 116 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 117 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 118 http://www.wvdhsem.gov/nfip1_CRS.htm (last visited on June 29, 2011). 112

American Academy of Actuaries

34

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

located within the flood plain. Lower discounts are provided for structures outside the flood plain. The CRS recognizes 18 creditable activities, which are organized into four series: Public Information, Mapping and Regulations, Flood Damage Reduction, and Flood Preparedness.119 About two-thirds of structures covered by the NFIP are within the approximately a thousand participating CRS communities. The Community Development and Regulatory Improvement Act, also known as the National Flood Insurance Reform Act of 1994, included the most comprehensive changes to the NFIP since the Flood Disaster Protection Act in 1973.120 Revised provisions included:      

Non-waiver of the requirement that flood insurance is purchased by recipients of federal disaster assistance.121 Expanded requirements for lenders when making loans, and a requirement that coverage is maintained for the life of the loan.122 Codification of the community rating system, and direction that credits may be given to communities that implement measures to protect natural and beneficial flood plain functions and manage erosion.123 An increase in the maximum coverage amounts available and a requirement to review and assess every five years the need to update and revise Flood Insurance Rate Maps (FIRMs).124 Requirement of an economic impact study on the effect of charging actuarial rates for pre-FIRM structures.125 Prohibition on disaster assistance to individuals in a SFHA who previously received disaster assistance and did not maintain flood insurance.126

The Bunning-Bereuter-Blumenauer Flood Insurance Reform Act of 2004 included reforms to address repetitive loss properties and constituted a reauthorization of the NFIP until Sept. 30, 2008.127 Additional funding mechanisms focused mitigation efforts on “severe” repetitive loss structures that resulted in a disproportionate number of claims to the NFIP.128 The goals of the 2004 Act were to provide people who have experienced serious and repetitive flood damage with financial assistance from the NFIP, communities, and states; to end the abuses by those who misuse the program; and to improve consumer understanding of the rights of NFIP policyholders.129 119

http://www.fema.gov/pdf/nfip/manual200805/19crs.pdf (last visited on June 29, 2011). http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 121 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 122 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 123 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 124 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 125 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 126 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 127 http://www.floods.org/PDF/ASFPM_FIRA2004_070804.pdf (last visited on June 29, 2011). 128 http://www.floods.org/PDF/ASFPM_FIRA2004_070804.pdf (last visited on June 29, 2011). 129 http://www.floods.org/PDF/ASFPM_FIRA2004_070804.pdf (last visited on June 29, 2011). 120

American Academy of Actuaries

35

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

In July 2004, FEMA issued an interim final rule in the Federal Register to amend the FIA, the Financial Assistance/Subsidy Arrangement (the standard insurance contract used by WYO companies), and the related regulations regarding issues of federal jurisdiction (the applicability of federal law to lawsuits involving WYO companies, and the applicability of reimbursement to WYO companies for the cost of litigation).130 In addition, FEMA amended the procedures necessary for companies seeking to obtain or suspend their status as WYO companies.131 Financial Actions The intent of the NFIP was to generate premiums sufficient to cover at least expenses and losses relative to what is called the historical average loss year, which differs from the traditional insurance definition of solvency.132 Throughout the 1980s, FEMA initiated extensive rate increases and coverage changes designed to place the NFIP on sound fiscal ground.133 During fiscal year 1986, no taxpayer funds were required to meet the NFIP’s flood insurance expenses, meaning that, for the first time, the NFIP was selfsupporting.134 In addition, at the beginning of the fiscal year, the NFIP was required, for the first time, to pay all program and administrative expenses with funds derived from insurance premiums.135 Before that time, program costs for administrative expenses, surveys, and studies were financed through congressional appropriations.136 In the Budget Reconciliation Act of 1990, Congress required policyholders to pay for expenses beyond the insurance costs of the NFIP, such as mapping, flood studies, and flood plain management activities.137 This legislation was controversial because the benefits of those activities are enjoyed by all communities and residents in the flood plains, not just NFIP policyholders. In fiscal year 1992, the NFIP experienced losses that were more than twice its historic loss level, and, in 1993, it had to borrow $100 million from the U.S. Treasury.138 This was the first time since 1984 such borrowing had been necessary.139 The borrowed funds were repaid in fiscal year 1994.140

130

http://www.epa.gov/fedrgstr/EPA-IMPACT/2003/October/Day-14/i25905.htm (last visited on June 29, 2011). 131 http://www.epa.gov/fedrgstr/EPA-IMPACT/2003/October/Day-14/i25905.htm (last visited on June 29, 2011). 132

http://www.fema.gov/library/file;jsessionid=E9F9828C0A315EFF418280AE4E944244.WorkerLibrary?typ e=publishedFile&file=rate_rev04.pdf&fileid=e60f7330-abc9-11db-b560-000bdba87d5b (last visited on June 29, 2011). 133 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 134 http://www.fas.org/sgp/crs/misc/R40650.pdf (last visited on June 29, 2011). 135 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 136 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 137 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 138 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 139 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 140 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011).

American Academy of Actuaries

36

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

In September 1996, the NFIP experienced losses that were much higher than its historic loss levels and the NFIP borrowed $626 million in Treasury Department funds.141 The NFIP borrowed an additional $192 million over the next six months.142 In October 1996, Congress approved a supplemental request to increase the NFIP’s borrowing for fiscal year 1997 to $1.5 billion from $1 billion.143 By September 1997, the U.S. Treasury had loaned the NFIP $917 million.144 This had been repaid by June of 2001.145 Tropical Storm Allison made landfall that year, and the borrowing resumed.146 Allison resulted in more than 30,000 claims and approximately $1 billion in claim payments.147 By late 2002, the NFIP had paid the final $10 million installment on the $650 million it had borrowed to pay claims arising from Tropical Storm Allison.148 In 2004, FEMA paid out $1.8 billion in claims, or approximately two and a half times the amount paid out in 2003.149 FEMA used $225 million in NFIP borrowing authority to pay 2004 flood loss claims.150 In July 2005, Hurricane Dennis hit the Florida panhandle in the same area that had been affected by Hurricane Ivan the previous year.151 Ivan had cost the NFIP approximately $1.5 billion, with Dennis adding another $100 million plus.152 Later in the season, Hurricane Katrina struck Louisiana and Mississippi, resulting in floodwall and levee failures that caused up to 80 percent of the city of New Orleans to flood.153 Hurricane Rita then struck the Gulf Coast along the western Louisiana and eastern Texas shores, causing the city of New Orleans to suffer new levee breaches and additional flooding.154 In response to the losses associated with hurricanes Katrina and Rita, President George W. Bush signed H.R. 3669, which increased the NFIP’s borrowing authority from $1.5 billion to $3.5 billion.155 The Congressional Budget Office (CBO) later estimated that FEMA probably would not be able to repay the funds borrowed under H.R. 3669 within the “next 10 years,” that Katrina-related claims would “exceed the total resources that will be available to FEMA under H.R. 3669,” and that “repayments of borrowed funds would not occur until after

141

http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 143 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 144 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 145 http://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_mip_apnd_h.pdf (last visited on June 29, 2011). 146 http://www.fema.gov/news/newsrelease.fema?id=3565 (last visited on June 29, 2011). 147 http://www.fema.gov/news/newsrelease.fema?id=4918 (last visited on June 29, 2011). 148 http://www.fema.gov/news/newsrelease.fema?id=3565 (last visited on June 29, 2011). 149 See http://www.gao.gov/new.items/d06119.pdf (last visited on June 29, 2011). 150 http://financialservices.house.gov/media/pdf/102005dm.pdf (last visited on June 29, 2011). 151 See http://www.cnn.com/2005/WEATHER/07/10/tropical.weather/index.html (last visited on June 29, 2011). 152 http://www.fema.gov/business/nfip/statistics/sign1000.shtm (last visited on June 29, 2011). 153 http://www.dhs.gov/xfoia/archives/gc_1157649340100.shtm (last visited on June 29, 2011). 154 http://www.pbs.org/newshour/updates/rita_09-24-05.html (last visited on June 29, 2011). 155 See http://www.senate.gov/~budget/republican/pressarchive/2005/2005-11-18Hurricanes.pdf (last visited on June 29, 2011). 142

American Academy of Actuaries

37

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

2015.”156 In October 2005, David Maurstad, acting director of the FEMA Mitigation Division and federal insurance administrator, testified before the Senate Committee on Banking, Housing and Urban Affairs on the future of the National Flood Insurance Program.157 Maurstad reported to the committee that the magnitude and severity of flood losses caused by Hurricanes Katrina and Rita were “unprecedented in the history of the NFIP.”158 He predicted that Katrina and Rita-related flood claims would “result in flood insurance claims that significantly exceed the highest number of claims filed from any single event in the NFIP’s history, and well more than triple the total number of claims filed in 2004.”159 He also predicted that Katrina and Rita-related NFIP claims could exceed $22 billion and noted that, in its entire history, the NFIP had paid out only $15 billion total.160 In March 2006, President Bush signed S. 2275, which authorized the NFIP to increase its borrowing authority to $20.775 billion.161 In June 2010, the National Flood Insurance Program Extension Act of 2010 lowered the NFIP’s borrowing authority to $20.725 billion. The most recent large events for the NFIP occurred in September 2008, when hurricanes Gustav and Ike made landfall near the Texas-Louisiana border. As a result of those two hurricanes, the NFIP handled more than 50,000 claims and paid out nearly $3 billion.

156

http://www.cbo.gov/showdoc.cfm?index=6658&sequence=0 (last visited on June 29, 2011). http://banking.senate.gov/public/_files/maurstad.pdf (last visited on June 29, 2011). 158 http://banking.senate.gov/public/_files/maurstad.pdf (last visited on June 29, 2011). 159 http://banking.senate.gov/public/_files/maurstad.pdf (last visited on June 29, 2011). 160 http://banking.senate.gov/public/_files/maurstad.pdf (last visited on June 29, 2011). 161 http://www.thomas.gov/cgi-bin/bdquery/z?d109:SN02275:@@@L&summ2=m& (last visited on June 29, 2011). 157

American Academy of Actuaries

38

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

Appendix B Oversight of the NFIP The NFIP is a large federal program with more than 5 million policies in force162 and annual written premiums of approximately $3.4 billion as of April 2011.163 It has an especially large impact on specific communities, businesses, and individuals in the United States, particularly on those communities and individuals that are located within the flood plains, known as special flood hazard areas (SFHAs). As a result, many different constituencies are invested in the work of the NFIP. Those constituencies include policyholders/homeowners, business owners, community officials, community and state flood plain professionals and managers, insurance agents, insurance industry professionals and managers, commentators, banking and mortgage officials, state and federal regulators, the administration, and, of course, Congress. Congress has an extensive oversight responsibility for the NFIP, primarily through the Senate Banking, Housing & Urban Affairs Committee and the House Financial Services Committee. Those committees and their predecessors were the authors of the original NFIP legislation and all NFIP-altering legislation promulgated since then. Major changes to the NFIP are accomplished by legislation originating from those two committees. Less significant reforms can be made through regulatory modifications sponsored by FEMA. Such regulatory modifications must go through the federal rulemaking process. That process, including public comment periods, generally takes one to two years. FEMA is part of the executive branch of the federal government. It is housed within the Department of Homeland Security (DHS).164 Oversight of the NFIP, therefore, also comes from DHS and the Office of Management and the Budget (OMB). In addition, Congress often assigns studies of the NFIP and its operations to the Government Accountability Office, the Congressional Research Service, and the Congressional Budget Office, which usually respond with detailed analyses and commentary.165, 166 The Inspector General of DHS and FEMA’s Office of the Chief Financial Officer also periodically review various aspects of the NFIP’s operations. In addition, the inspector general conducts an annual financial audit of the NFIP.

162

http://www.fema.gov/business/nfip/statistics/pol.shtm (last visited on June 29, 2011). http://bsa.nfipstat.com/reports/1011.htm (last visited on June 29, 2011). 164 http://www.dhs.gov/xabout/structure/editorial_0644.shtm (last visited on June 29, 2011). 165 See http://www.gao.gov/new.items/d06174t.pdf (last visited on June 29, 2011). 166 See http://www.cbo.gov/ftpdocs/82xx/doc8256/06-25-FloodInsurance.pdf (last visited on June 29, 2011). 163

American Academy of Actuaries

39

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

Appendix C Premium Rate Structure of the NFIP There are a number of classes of risks that include special considerations in their rate determination beyond the standard ratemaking process of using the hydrologic/financial model. 

Preferred Risk Policy

One key and increasingly important class of business is the Preferred Risk Policy (PRP) business. PRP risks are charged full-risk (more commonly known as actuarial) rates, which are based on consideration of the risk involved and accepted actuarial principles. The eligibility qualifications for PRP are primarily (1) that the location of the structure is outside of any special flood hazard area (SFHA) on the current flood insurance risk maps (FIRMs) and (2) the structure has a favorable loss history. PRPs comprise the lowest risks in the NFIP and therefore receive the most favorable premium rates. The PRP program is important for the NFIP in helping to meet one of its current goals—increasing the NFIP market penetration of properties outside the special flood hazard areas (SFHAs). The number of PRP policies has been growing rapidly in recent years. The Preferred Risk Policy (PRP) Eligibility Extension became effective on Jan. 1, 2011. The Flood Map Modernization and the Risk MAP programs have caused many areas to be recategorized from moderate-to-low risk to high-risk. For the properties within those areas, this could mean a significant increase in rates. To help ameliorate any potential hardship to property owners, FEMA has introduced a new process to allow those owners to continue to buy the lower-cost PRP policy for two years, at which time they are required to buy the policy at standard rates. 

Properties with Flood Map Grandfathering

Flood Map Grandfathering is an NFIP administrative procedure that allows properties that have experienced changes to their FIRMs to continue to pay premiums based on their prior (lower-premium) rate classes.167 The NFIP continually makes changes to FIRMs for various reasons, including new development and construction projects within a community that change the flow or retention of flood waters or reconsideration of the risks presented by an existing flood plain. A new FIRM could affect premium rates for buildings within its boundaries. The NFIP has set certain criteria for negatively affected buildings to qualify for grandfathering. The NFIP compensates for the grandfathered policies within each 167

http://www.fema.gov/library/file;jsessionid=412F42857F2B3889B72BDE6C854E8897.Worker2Library?ty pe=publishedFile&file=map_changes_and_insurance_savings.pdf&fileid=66669c00-d232-11db-866c000bdba87d5b (last visited on June 29, 2011).

American Academy of Actuaries

40

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

class of business by aiming to set premium rates for the whole class, including grandfathered and non-grandfathered buildings, at an actuarially adequate level. According to the NFIP, the number of buildings with grandfathered rates is relatively low. As the Risk MAP project continues, however, the number of buildings eligible for grandfathering is likely to increase. 

Pre-FIRM Properties

The rationale for allowing subsidized classes of coverage was to permit the large inventory of structures that were built in SFHAs prior to the general implementation (circa 1974) of FIRMs and flood-related building codes (known as pre-FIRM structures) to obtain flood insurance at “reasonable” rates.168 The subsidized pre-FIRM rates are determined such that, when considered with all other policies in the program, the overall premium level for the program is sufficient to meet or exceed the historical average loss year. Not all pre-FIRM SFHA structures receive subsidized rates. Those property owners have a choice of either subsidized rates or full-risk rates. The full-risk rates for SFHA buildings with reference levels above the base flood elevation (BFE) are often less than the subsidized rates. By providing appropriate documentation, many pre-FIRM property owners are eligible for cheaper full-risk rates. 

Pre-1981 V Zone areas

“Velocity,” or “V” Zones, are primarily coastal areas subject to the risk of wave action in addition to flood waters reaching and exceeding the BFE.169 Prior to 1981, NFIP building standards accounted for still water elevations but not associated wave action. In October 1981, the NFIP promulgated new, more stringent standards based on new engineering developments and studies. Subsequently, a decision was made, however, to grandfather the 1975 to 1981 construction and allow less than full-risk premium rates.170 

Areas protected by flood-protection systems

For the most part, flood-protection systems refer to levees. The flood legislation allows for buildings in such areas to receive X-zone rates, including PRP eligibility, if the flood protection system meets explicit standards promulgated by FEMA and the U.S. Army Corps of Engineers and the structures’ reference levels are at BFE or above.171 The statute also allows X-zone rates in certain cases even for systems not 168 169

http://www.fas.org/sgp/crs/misc/RL32972.pdf (last visited on June 29, 2011). http://www.fema.gov/plan/prevent/floodplain/nfipkeywords/zone_v.shtm (last visited on June 29, 2011).

170

http://www.fema.gov/library/file;jsessionid=E9F9828C0A315EFF418280AE4E944244.WorkerLibrary?typ e=publishedFile&file=rate_rev04.pdf&fileid=e60f7330-abc9-11db-b560-000bdba87d5b (last visited on June 29, 2011). 171 See http://www.fema.gov/plan/prevent/fhm/fq_genin.shtm#in8 (last visited on June 29, 2011).

American Academy of Actuaries

41

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

meeting the standards if there is a project underway to construct or improve the system to meet those standards. Those areas that do not yet meet standards are designated as “A99” or “AR” Zones on the FIRMs.172 Such designations apply to policies that receive rates for their respective classifications as if the levees provided 100-year protection or better. 

The D Zone

The D Zone is a FIRM category for areas within NFIP-participating communities in which the flood hazard has not been determined; i.e., the areas have not been mapped. The mapping process in developing the FIRMs is accomplished in order of priority. There are some large geographical areas with tiny population densities, and in which future development is unlikely. Those areas are considered low priority with respect to mapping. The D Zone premium rates are estimated using average rates across a spectrum of classes. 

Areas of Residual Risk

Certain flood risks, such as levee failure and coastal erosion, are not currently recognized on the FIRMs or within the actuarial rating structure.173 In the future, those areas may be mapped specifically and given premium rates. In addition, there are a number of NFIP special-purpose programs, under which buildings can be insured using rates that are not or cannot be actuarially calculated. These include:

172

The emergency program allows property owners in communities that are in the process of applying for NFIP participation to obtain coverage.174 As a general rule, those communities do not yet have FIRMs in effect. While the premium rates are low, the allowed coverage is limited and temporary. The emergency program generally has accounted for a tiny proportion—currently less than 1 percent—of the in force policy base.

Group flood contracts are issued by the NFIP in response to presidential disaster declarations.175 States may apply for a group NFIP policy, under which property owners who are disaster recipients may apply for a variety of limited coverage options with low premium rates. Group flood contracts have three-year policy terms and are the only non-one-year policy contracts within the NFIP. Such contracts generally cannot be renewed.

http://www.fema.gov/plan/prevent/fhm/fq_genin.shtm#in8 (last visited on June 29, 2011).

173

http://www.fema.gov/library/file?type=publishedFile&file=fema549_apndx_e_ra8.pdf&fileid=143da3a00316-11dc-a1f1-000bdba87d5b (last visited on June 29, 2011). 174 http://www.fema.gov/plan/prevent/floodplain/nfipkeywords/emergency_program.shtm (last visited on June 29, 2011). 175 http://www.fema.gov/business/nfip/19def2.shtm#G (last visited on June 29, 2011).

American Academy of Actuaries

42

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

The mortgage portfolio protection program (MPPP) was introduced on Jan. 1, 1991, as a tool to assist the mortgage lending and servicing industries in bringing their mortgage portfolios into compliance with the Flood Disaster Protection Act of 1973, which established mandatory purchase requirements.176 The MPPP is intended to be used by lenders as a last resort to force-place coverage when a borrower cannot or will not purchase the policy directly.

176

http://www.fema.gov/pdf/nfip/manual200510/10mppp.pdf (last visited on June 29, 2011).

American Academy of Actuaries

43

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

Reprinted from the NFIP Actuarial Rate Review Supporting Oct. 1, 2010 Rate Changes

American Academy of Actuaries

44

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

Reprinted from the NFIP Actuarial Rate Review Supporting Oct. 1, 2010 Rate Changes

American Academy of Actuaries

45

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

Appendix D WYO and Direct Programs The operations of the WYO companies’ flood businesses are governed primarily by Code of Federal Regulations Title 44, Part 62, Subpart C.177 Subpart C contains the terms and conditions for property and casualty companies participating in the WYO program. Additional elements of the WYO program are contained in Part 62’s Appendix A (the Financial Assistance/Subsidy Arrangement), which spells out the expense allowances and responsibilities of the WYO companies,178 and Appendix B (the Financial Control Plan), which lists the financial, audit, examination, and data requirements for WYO companies.179 The direct program’s operations are governed by Subpart B of the above regulation. Qualifications as a WYO Company The Code of Federal Regulations (Title 44, Part 62, Subpart C) sets forth the requirements of the WYO program. To qualify as a WYO company, a company must:  Be a licensed property insurance company.  Have at least five years of history writing property coverage.  Disclose any legal proceedings or other formal proceedings regarding the company’s business practices to which it has been subjected with any state or federal governmental agencies in the past five years.  Submit its most recent annual statement.  Show that it meets or exceeds the National Association of Insurance Commissioners’ standards for risk-based capital and surplus.  Submit its last audit, which should contain no negative findings. The company must provide evidence that it can process flood insurance and meet the requirements of the financial control plan. The company also must submit its plans for producer training, marketing plans, sales targets, claims handling, and plans for handling disasters. Financial Transactions180 WYO companies collect flood premiums separately and place these funds, less the companies’ expenses as discussed below, into an NFIP-specific restricted account. Any 177

http://edocket.access.gpo.gov/cfr_2008/octqtr/44cfr62AppB.htm (last visited on June 29, 2011).

178

http://www.fema.gov/library/file;jsessionid=182C99427CAE7C33E557E4E3A6D9D48E.WorkerLibrary?t ype=publishedFile&file=wyoarrg2010_rev.pdf&fileid=ddf19180-b5d2-11df-97ce-001cc4568fb6 (last visited on June 29, 2011). 179 http://edocket.access.gpo.gov/cfr_2010/octqtr/pdf/44cfr62AppA.pdf (last visited on June 29, 2011). 180 Federal Regulations (Title 44, Part 62, Subpart C).

American Academy of Actuaries

46

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

excess over the amounts required for the administration of their NFIP policies is then remitted to the U.S. Treasury. WYO companies may withhold operating, administrative, and production expenses from the premium collected. The amount to be withheld for operating and administrative expenses is determined by FEMA based on average industry expense ratios, as detailed in the regulation. WYO companies also may retain 15 percent of written premium for commissions to producers. An additional amount, up to two percentage points, may also be awarded by FEMA based on a company’s achievement of the marketing goals of the NFIP for the year. Loss payments are made from federal funds retained in the account. Loss-adjustment expenses (allocated and unallocated) also are drawn from this account. Unallocated lossadjustment expenses and allocated loss-adjustment expenses are reimbursed subject to a fee schedule.181 If the funds in the account are not sufficient to pay all losses, a company may draw from the federal flood fund using a letter of credit (LOC) procedure. 182 As described in the financial control plan and the WYO Accounting Procedures Manual, a company can request funds by providing specific required documentation in an LOC application.183 LOC applications are reviewed on a daily basis in the Risk Insurance Branch, which then authorizes specific amounts to be placed in WYO companies’ “accounts” at the Treasury. The WYO companies then can draw from those accounts as needed and deposit the funds in their restricted accounts. Financial Controls184 To ensure that taxpayer funds are spent appropriately, WYO companies are subject to the financial control plan.185 WYO companies are also subject to audits, examinations, and the regulatory controls of the states in which they operate, as well as internal company audits. FEMA may use findings from such activities to monitor WYO companies. Under the financial control plan, a WYO company will be subject to an audit of flood insurance financial statements every two years by a CPA firm at the company’s expense. The financial audits are intended to evaluate each WYO company’s financial statements, internal controls, and compliance with laws and regulations. In addition, transactional records are reconciled monthly with financial statements. FEMA also conducts reviews of each WYO’s claims, underwriting, customer service, marketing, and litigation activities at least every three years. None of these reviews limit the authority of the GAO or FEMA to further review a WYO company’s activities at any time. 181

http://edocket.access.gpo.gov/2008/pdf/E8-18176.pdf (last visited on June 29, 2011). http://edocket.access.gpo.gov/cfr_2002/octqtr/44cfr62.24.htm (last visited on June 29, 2011). 183 http://bsa.nfipstat.com/manuals/acctproc_200510.pdf (last visited on June 29, 2011). 184 Federal Regulations (Title 44, Part 62, Subpart C); The Write Your Own Financial Control Plan Requirements and Procedures. 185 http://bsa.nfipstat.com/manuals/fcp99jc.pdf (last visited on June 29, 2011). 182

American Academy of Actuaries

47

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

Appendix E Glossary Actuarial rate Term frequently used to describe NFIP full-risk rates. Actuarially sound premium rate Premium rates calculated such that they return to the insurer the expected value of all future costs associated with an individual risk transfer, plus an additional margin or contingency loading. Additional living expenses Extra charges covered by homeowners’ policies above policyholders’ customary living expenses. Additional living expenses apply, where applicable, to situations in which the insured requires temporary shelter because damage by a covered peril has rendered the home temporarily uninhabitable. Examples include costs for hotel or motel, costs for restaurant meals, and costs for clothes-laundering service. Borrowing authority Statutory authority that permits a federal agency to incur obligations and make payments for specified purposes with money loaned by the U.S. Treasury. Business interruption coverage Commercial coverage that reimburses a business owner for lost profits and continuing fixed expenses during the time that a business must remain closed while the premises are being restored after physical damage from a covered peril. Business interruption insurance also may cover financial losses incurred when civil authorities limit access to an area after a disaster and such actions prevent customers from reaching the business premises. Claims A demand for payment for a loss incurred from a potentially insured peril under the terms of a plan or insurance contract. Coastal Barrier Resource System Coastal areas, e.g., certain barrier islands, designated by Congress in the Coastal Barrier Resources Act (Pub. L. 97- 348) and the Coastal Barrier Improvement Act of 1990 (Pub. L. 101-591). These federal laws were enacted on Oct. 18, 1982, and Nov. 16, 1990, respectively. The laws were implemented as part of a Department of the Interior initiative to minimize loss of human life by discouraging development in high-risk areas, reduce wasteful expenditures of federal resources, and preserve the ecological integrity of areas designated by statute as Coastal Barrier Resources Systems and Otherwise Protected Areas. The laws provide protection by prohibiting all federal expenditures or financial

American Academy of Actuaries

48

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

assistance, including flood insurance, for residential or commercial development in areas so identified.186 Contingency provision A provision for the expected differences, if any, between estimated costs and average actual costs that cannot be eliminated by changes in other components of the ratemaking process. Federal disaster assistance Money or direct assistance provided by agencies of the federal government (notably FEMA) to individuals, families, and businesses in an area in which property has been damaged or destroyed and for which losses are not covered by insurance. It is meant to help with critical expenses that cannot be covered in other ways. This assistance is not intended to restore damaged property to its condition before the disaster. Federally regulated lending institutions Loans issued by the VA and FHA are subject to the supervision of federal institutions and are federally backed. Mortgage loans bought by Fannie Mae and Freddie Mac are considered federally backed, too, since the two organizations are federally chartered. These two entities buy more than 40 percent of all the conventional mortgages issued every year. Flood Insurance Rate Map (FIRM) Produced by FEMA, an official map of a community that delineates both the special hazard areas and the risk premium zones applicable to the community. Flood plain Any land area susceptible to being inundated by flood waters from any source. For NFIP purposes, flood plains are equivalent to Special Flood Hazard Areas (SFHAs). Government Accountability Office The Government Accountability Office (GAO) is the nonpartisan audit, evaluation, and investigative arm of Congress and an agency in the legislative branch of the U.S. government. Historical average loss year (HALY) A concept used by the NFIP beginning in the 1980s to establish a benchmark by which to judge the level of premium rates for subsidized policies. The HALY concept was developed to determine how much of a discount subsidized policies should receive. HALY is the mean of all the NFIP annual losses and loss-related expenses for a specific period of years (e.g., 1978 to the present), after trending (for inflation and flood policies’ distributional changes) each year’s losses to the present. HALY, therefore, is the average of the estimate for each historical loss year of what those storms would produce in losses in current dollars and assuming a current distribution of policies by rating class. While premiums for post-FIRM full-risk rated policies have always contemplated the full range 186

Flood Insurance Manual, May 2007, Page: CBRS 1.

American Academy of Actuaries

49

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

of loss probabilities, the concept of HALY has been used to assure that the NFIP’s aggregate premium (the sum of both full-risk and subsidized premiums) generates sufficient income after operating expenses to pay for the typical non-catastrophic-loss year that had been the NFIP’s experience prior to Katrina. The NFIP incorporates Katrina experience in the HALY average by weighting the 2005 accident year one percent and all other years 99 percent. HALY has helped determine a minimum for subsidized premiums. Subsidized premiums through the years, however, have reached levels well above that minimum. While even the non-catastrophic-loss years of the NFIP vary greatly, HALY is the center around which those loss years will vary. (See the discussion of the “NFIP Actuarial Rate Review Memorandum in Support of the May 1, 2010 Rate and Rules Changes” on Page 6 for a more complete description of issues surrounding the HALY concept.187) Insurance Services Office A corporation headquartered in Jersey City, N.J., that provides data, analytics, and decision-support services for professionals in several fields, including insurance. Its services include the calculation of property and liability insurance loss costs and the development of insurance policy forms. Mandatory purchase Pursuant to the provisions of the Flood Disaster Protection Act of 1973, individuals, businesses, and others buying, building, or improving property located in identified areas of special flood hazards within NFIP-participating communities are required to purchase flood insurance as a prerequisite for receiving any type of federal financial assistance (e.g., any loan, grant, guaranty, insurance, payment, subsidy, or disaster assistance) when the building or personal property is the subject of or security for such assistance. Mitigation practices In the context of flood risk emergency management, mitigation efforts attempt to prevent hazards from developing into disasters and to reduce the effects of disasters when they occur. Mitigation focuses on long-term measures for reducing or eliminating risk. The implementation of mitigation strategies can be considered a part of the recovery process if initiated after a disaster occurs. Mitigative measures can be structural or non-structural. Structural measures use technological solutions, such as flood levees; nonstructural measures include legislation and land-use planning. Profit provision The provision for underwriting profit in an actuarially developed rate, typically expressed as a percentage of the rate.

187

http://www.fema.gov/library/file?type=publishedFile&file=actuarial_rate_rev2009.pdf&fileid=55182cb094ea-11df-a6e7-001cc4568fb6 (last visited on June 28, 2011).

American Academy of Actuaries

50

www.actuary.org


FLOOD INSURANCE SUBCOMMITTEE MONOGRAPH

Return on capital A financial measure that quantifies how well a company generates profit relative to the capital it has invested in its business. It is defined as net operating profit divided by invested capital and usually is expressed as a percentage. Special Flood Hazard Area An area having special flood, mudflow, or flood-related erosion hazards. Each such area is shown on a Flood Insurance Rate Map as Zone A, AO, A1-A30, AE, A99, AH, AR, AR/A, AR/AE, AR/AH, AR/AO, AR/A1-A30, V1-V30, VE, or V.188 Under NFIP mapping standards, those zones comprise areas having a “100-year flood risk,� i.e., their probability of inundation in any year is 1 percent. Subsidized rates NFIP subsidized rates are national rates set by broad occupancy type classifications, which produce a premium income less than the expected expense and loss payments for the flood insurance policies issued on that basis. The difference between the full-risk premiums for these policyholders and the subsidized premiums they actually pay is revenue foregone by the National Flood Insurance Fund.189 Sunset provision In public policy, a provision in a statute or regulation that terminates or repeals all or portions of the law after a specific date, unless further legislative action is taken to extend it. Not all laws have sunset clauses; in the absence of a sunset clause, the law remains in effect until repealed. Underwriting Examining and accepting or rejecting insurance risks and classifying the ones that are accepted to charge appropriate premiums for them. Underwriting gain or loss The difference between the premium income and the claims and expenses incurred; excludes any investment gains or losses.

188 189

NFIP Flood Insurance Manual, May 2007. Page DEF 8. NFIP Actuarial Rate Review Memorandum in Support of the May 1, 2006 Rate and Rules Changes, p.

9.

American Academy of Actuaries

51

www.actuary.org


Selected Provisions of Various Social Security Reform Proposals Bipartisan Policy Center (Domenici-Rivlin) November 17, 2010 http://www.ssa.gov/OACT/solvency/BipartisanTaskForce_20101117.pdf The Domenici-Rivlin plan would, based on the intermediate assumptions of the 2010 Trustees Report, keep the Old-Age, Survivors, and Disability Insurance (OASDI) trust fund solvent for the next 75 years. The plan includes the following eleven provisions that have significant effects on the OASDI program’s financial operations and actuarial status: 1. Increase the OASDI contribution and benefit base over 38 years starting in 2012, so that 90 percent of covered earnings will be taxable for 2049 and later. 2. Change the OASDI cost of living adjustment (COLA) to be based on a chained version of the Consumer Price Index for Urban Wage and Clerical Workers (CPI-W) starting for December 2012. 3. Cover earnings of all state and local government employees hired in 2020 and later under OASDI. 4. Eliminate the excise tax on premiums for employer sponsored group health insurance and make all such premiums subject to OASDI payroll tax, phased in between 2018 and 2028. 5. Reduce the 15-percent Primary Insurance Amount (PIA) formula factor to 10 percent, phased in gradually between 2023 and 2052. 6. Starting in 2012, enhance the special minimum benefit to provide a PIA level at benefit eligibility equivalent to 133 percent of the 2009 Federal Aged Poverty threshold for individuals with earnings of at least 20 percent of the “old-law taxable maximum” in at least 30 years. The minimum would be reduced for fewer qualifying years, to zero for less than 20 years. The target poverty level would be wage indexed from the 2009 level to two years before the year of initial benefit eligibility. Up to eight child-care creditable years would be allowed if caring for a child under age six. The earnings requirement and 1

2013 Summer Summit Materials | American Academy of Actuaries


number of child care creditable years allowed would be scaled for workers becoming disabled or dying before attaining age 62. 7. Subject contributions to all voluntary salary reduction plans to OASDI payroll tax in the same manner as for 401(k)s. 8. Index the PIA formula for Old-Age and Survivors Insurance (OASI) benefits to longevity by reducing the factors starting in 2023 by the change in the ratio of (a) the period life expectancy at 67 for 2018 to (b) the period life expectancy at 67 for the 4th year before initial benefit eligibility. For disabled workers at conversion to retirement at normal retirement age, the reductions would apply based on the proportion of years not disabled from 22 to 61. 9. Increase benefits gradually between ages 81 and 85 reflecting an increase in PIA equal to five percent of the average retired worker PIA in the year age 80 is reached. 10. Tax Reform for Business: Establish a value added tax (VAT) of 3.0 percent for 2012 and 6.5 percent for 2013 and later. Reduce the corporate income tax from 35 to 27 percent. 11. Tax Reform for Individuals: Modify the personal income tax to make two brackets with marginal rates of 15 and 27 percent. OASDI benefits are included as regular income with no thresholds. Capital gains are included as regular income. A non-refundable credit for low income tax filers age 65 and older is established. A non-refundable credit of 7.5 percent of OASDI benefit is established. Thus, revenue to Old age, Survivors, Disability and Health insurance (OASDHI) is based on 7.5 and 19.5 percent marginal rates on all OASDI benefits.

National Commission on Fiscal Responsibility and Reform (Bowles-Simpson) December 1, 2010 http://www.ssa.gov/oact/solvency/FiscalCommission_20101201.pdf The Bowles-Simpson plan would allow the OASDI program to meet the requirements of sustainable solvency under the intermediate assumptions for the 2010 Trustees Report. The plan includes the following eight basic provisions that have significant direct effects on the OASDI program’s financial operations and actuarial status: 1. After 2022, index the Normal Retirement Age to maintain a constant ratio of (a) life expectancy at Normal Retirement Age to (b) potential work years (Normal Retirement Age minus 20 years). Maintain the Earliest Eligibility Age at Normal Retirement Age minus five years. Increases in the Early Eligibility Age (EEA) and Normal Retirement Age would be limited. Consistent with this intent, the following provision is included as a potential limitation. For individuals who have earned four quarters of coverage in 25 years before age 62, retain the early eligibility age and normal retirement age at 62 and 67, respectively, if average indexed monthly earnings (AIME) is under 250 percent of aged poverty (wage indexed from 2009), with this limitation phased out completely if AIME is over 400 percent of aged poverty.

2

2013 Summer Summit Materials | American Academy of Actuaries


2. Create a new bend point in the PIA formula at the AIME for the 50th percentile of new retired worker awards. Over the period 2017-2050, gradually reduce the 32 percent PIA factor that applied below the new bend point to 30 percent, the 32 percent PIA factor that applies above the new bend point to 10 percent, and the 15 percent PIA factor to five percent. 3. Change the OASDI COLA to be based on a chained version of the Consumer Price Index (CPI-W) starting for December 2011. 4. Beginning in 2017, increase the special minimum benefit by making the following changes. (a) A year of coverage is defined as a year in which four quarters of coverage are earned; (b) The minimum PIA for 30 years of coverage is equal to 125 percent of the monthly poverty level (indexed by chained CPI from 2009 to 2017 and by average wage thereafter, for successive cohorts); (c) The minimum PIA is zero for 10 or fewer years of coverage, and increases linearly for 11 through 30 years of coverage; (d) Scale year of coverage requirements for disabled workers based on years of potential work. 5. Effective for all beneficiaries in 2011 and later, provide a five percent increase in PIA phased in over the 20th through 24th years after initial benefit eligibility. The total increase in PIA is five percent of the PIA for a worker of the same age with earnings at the average (AWI) at ages 22 through 61. 6. Increase the OASDI contribution and benefit base (taxable maximum) by an additional two percent each year starting in 2012, until 90 percent of covered earnings are taxable. Additional increases are expected for 38 years, reaching 90 percent taxable for 2049 and later. Establish starting in 2013 a new PIA bend point at the monthly equivalent of the taxable maximum that would be determined without regard to this provision, with a benefit formula factor of five percent for AIME above this new bend point. 7. Allow retirees to start receiving up to one-half of their benefits at age 62 with the remainder not available until EEA. Actuarial reduction applies accordingly. 8. Cover earnings of all state and local government employees hired in 2021 and later.

Sen. Harkin – Strengthening Social Security Act of 2013 March 29, 2012 http://www.ssa.gov/oact/solvency/THarkin_20130318.pdf The Office of the Chief Actuary (OACT) estimates that Sen. Harkin’s bill would extend full solvency of the OASDI from 2033 to 2049, based on estimates using the intermediate assumptions of the 2012 Trustees Report. The bill includes the following three provisions with direct effects on the OASDI program: 1. Eliminate the taxable maximum, fully effective 2018. Phase in the elimination over five years by taxing all earnings above the current law taxable maximum at a rate of 2.48 percent in 2014, 4.96 percent in 2015, …, and 12.40 percent in 2018 and later. Credit the additional earnings for benefit purposes by: (a) calculating a second average indexed monthly earnings (“AIME+”) reflecting only earnings for each year that are above the 3

2013 Summer Summit Materials | American Academy of Actuaries


current law taxable maximum1, (b) applying a five percent factor on this newly computed “AIME+” to compute a second primary insurance amount, (c) adding this second primary insurance amount to the current-law primary insurance amount. 2. Increase the first PIA bend point 15 percent above the current law level for newly eligible beneficiaries, fully effective 2033. Phase in by increasing the first bend point one percent above the current level for newly eligible beneficiaries in 2019, two percent for newly eligible beneficiaries in 2020, …, and 15 percent for newly eligible beneficiaries in 2033 and later. 3. Compute the cost-of-living adjustment (COLA) using the Consumer Price Index for the Elderly (CPI-E), effective December 2014. OACT estimates this new computation will increase the annual COLA by about 0.2 percentage point, on average.

Sen. Sanders - Keeping Our Social Security Promises Act March 19, 2013 http://www.ssa.gov/oact/solvency/BSanders_20130319.pdf Sen. Sanders’ bill would extend the full solvency of the OASDI program from 2033 to 2061, based on estimates using the intermediate assumptions of the 2012 Trustees Report. The bill would subject a worker’s OASDI covered earnings in excess of $250,000 in any calendar year after 2013 to the combined OASDI payroll tax rate of 12.4 percent. This is the same tax rate that is applied, under current law, to OASDI covered earnings up to the contribution and benefit base ($113,700 for 2013). Under present law, the contribution and benefit base is scheduled to increase in the future based on increases in the average wage in the U.S. economy. However, the threshold of $250,000 would be constant after 2014 until the contribution and benefit base exceeds this level (in the year 2033), at which point the threshold would be set equal to the contribution and benefit base for that and all subsequent years. Earnings subject to tax above the threshold would not be included in earnings credited for the purpose of OASDI benefit computation.

Sen. Graham/Paul/Lee - The Social Security Solvency and Sustainability Act April 13, 2011 http://ssa.gov/oact/solvency/GrahamPaulLee_20110413.pdf The bill is projected to make the OASDI program solvent for the next 75 years, under the intermediate assumptions of the 2010 Trustees Report. The bill includes the following two provisions: 1. The Normal Retirement Age will increase three months each year starting with individuals reaching age 62 in 2017 and stopping when the Normal Retirement Age reaches age 70 for individuals reaching age 62 in 2032. Thereafter, the Normal Retirement Age will be indexed to maintain a constant ratio of expected retirement years to potential work years, about one month every two years. The Earliest Eligibility Age will be increased by three months per year starting with individuals reaching age 62 in 2021 and will stop when the earliest eligibility age reaches age 64 for individuals reaching age 62 in 2028 and later.

4

2013 Summer Summit Materials | American Academy of Actuaries


2. The PIA formula will be modified between 2018 and 2055 to gradually reduce benefits on a progressive basis for workers with career-average earnings above the 40th percentile of new retired workers. The reduction for a steady maximum earner will be equivalent to that of replacing the current wage indexed PIA formula from one generation to the next, with a CPI-indexed formula across generations. Disabled worker beneficiaries will be held harmless, as will children of deceased workers and surviving spouses with a child in care.

President Obama’s FY2014 Budget Proposal April 10, 2013 http://www.whitehouse.gov/omb/budget/factsheet/chained-cpi-protections The President’s FY 2014 Budget called for switching to chained-CPI, but only if: 1. The change is part of a balanced deficit reduction package that includes substantial revenue raised through tax reform. 2. It is coupled with measures to protect the vulnerable and avoid increasing poverty and hardship. This includes benefit enhancements for the very elderly and others who rely on Social Security for extended periods of time. a. The benefit enhancement would be equal to 5% of the average retiree benefit, or about $800 per year if the proposal were in effect today. b. It would phase in over 10 years, beginning at age 76, or (for other beneficiaries, such as those receiving Disability Insurance) in the 15th year of benefit receipt. c. The benefit enhancement would begin in 2020, phasing in over 10 years for those 76 or older (or in their 15th year of eligibility or beyond) in that year. d. Beneficiaries who continued to be on the program for an additional 10 years would be eligible for a second benefit enhancement, starting at age 95 in the case of a retired beneficiary. The president’s budget would not apply chained-CPI to means tested programs like the Supplement Security Income (SSI) program, which provides benefits to low income elderly, blind, and disabled Americans.

SSA Summary of Provisions that Would Change Social Security http://ssa.gov/oact/solvency/provisions/summary.pdf

5

2013 Summer Summit Materials | American Academy of Actuaries


ISSUE GUIDE

A Guide to Analyzing Social Security Reform


TABLE OF CONTENTS

Preface................................................................................................................... 1 Introduction......................................................................................................... 2 Changes to Financing......................................................................................... 3 Changes to the Benefit Formula........................................................................ 4 Changes to the Taxation of Benefits.................................................................. 5 Means Testing...................................................................................................... 6 Raising the Retirement Age............................................................................... 7 Individual Account Proposals............................................................................ 9 Public Statements on Social Security by the Academy................................. 11

The American Academy of Actuaries is a 17,000-member professional association whose mission is to serve the public and the U.S. actuarial profession. The Academy assists public policymakers on all levels by providing leadership, objective expertise, and actuarial advice on risk and financial security issues. The Academy also sets qualification, practice, and professionalism standards for actuaries in the United States. Copyright Š2012 American Academy of Actuaries. All rights reserved.


PREFACE

The American Academy of Actuaries’ Social Security Committee has written this guide for the general public, policymakers, and journalists, to use as they evaluate Social Security reform proposals. The Academy is a nonpartisan, nonprofit association with a mission to serve the public and the United States actuarial profession by providing independent and objective actuarial information, analysis, and education for the formation of sound public policy. The Academy and its Social Security Committee strongly believe Social Security is facing serious financial challenges that threaten the long-term sustainability of the program. Congress and the administration should take action sooner rather than later to address Social Security’s fiscal sustainability and retirement security for current and future retirees.

▲ ISSUE GUIDE: SOCIAL SECURITY REFORM

WWW.ACTUARY.ORG

1


INTRODUCTION

The Social Security Board of Trustees issues an annual report on the Old-Age and Survivors, Insurance and Federal Disability Insurance (OASDI) trust funds that provides a detailed assessment of the financial condition of Social Security program for the next 75 years. The 2012 trustees’ report1 shows that benefits and administrative expenses exceeded payroll tax income in 2012. That report projects that, based on the trustees’ intermediate or “best estimate” assumptions, the deficits will continue indefinitely into the future. As a result, the Social Security trust funds that have been built up by past surpluses of taxes over expenses are projected to run out of money in 2033. Once the trust funds are depleted, benefit payments will be limited to what can be funded by tax revenues at that time. The Academy’s Social Security Committee has published a monograph entitled Social Security Reform Options, which provides a comprehensive overview of Social Security reform options, and a series of issue briefs analyzing recent and current proposals for reforming Social Security, including their impacts on benefits and solvency. This guide applies the concepts from these publications to proposed changes to the Social Security system that policymakers have raised and poses questions that should be asked of these policymakers on the topic of Social Security solvency. We encourage you to access these documents at http://www.actuary.org/briefs.asp#soc for a more indepth analysis. It is important to keep in mind that addressing Social Security’s long-term financial challenges may involve a combination of changes from some or all of the types of reform presented in this guide. When evaluating proposals, be sure to consider the effect of any proposed changes to the Social Security program with respect to the program’s projected deficit, the trust fund income and outgo, and the program’s overall financial solvency. You also should consider how the changes will affect various demographic groups (i.e., women2 compared with men, low income compared with high income, or young compared with elderly).

The Social Security Committee publishes an annual issue brief on the trustees report. For more information, please read An Actuarial Perspective on the 2012 Social Security Trustees’ Report (May 31, 2012) http://www.actuary.org/ files/SSC_IssueBrief_2012TrusteesReport_120501.pdf 2 For more information about how the Social Security program impacts women, please read Women and Social Security (July 2007 issue brief) http://www.actuary.org/pdf/socialsecurity/women_07.pdf 1

▲ ISSUE GUIDE: SOCIAL SECURITY REFORM

WWW.ACTUARY.ORG

2


CHANGES TO FINANCING

Social Security is financed primarily by dedicated payroll taxes, shared equally, in most circumstances, by covered workers and their employers. The tax is a flat percentage of earnings up to a maximum amount, called the Social Security earnings base, that is indexed each year to increases in the national average wage. Social Security also receives interest income from the investment of trust fund assets in U.S. Treasury securities and smaller amounts from other sources, such as income taxes levied on Social Security benefits. According to the intermediate estimates in the 2012 trustees’ report, current assets in the trust fund as well as future revenues are projected to fall short of future benefits and expenses during the next 75 years by $6.5 trillion on a present value basis, or approximately 0.9 percent of gross domestic product (GDP) on the same basis. This shortfall can be addressed by increasing the current payroll tax rate of 12.40 percent by 2.61 percentage points, to a rate of 15.01 percent—or through other taxation increases equal to 0.9 percent of GDP.

Questions to Ask When Evaluating Proposals to Change the Financing of Social Security n

Should the Social Security payroll tax rate be raised to strengthen system finances? If so, when?

n

Should the Social Security payroll tax rate be indexed so that taxes increase automatically when projections indicate the system is inadequately financed?

n

Should the Social Security payroll tax be graduated (according to income brackets) like income tax?

n

Should general revenue, or revenue from another source such as a value-added tax,3 be used to supplement the Social Security payroll tax?

n

Should the Social Security taxable earnings base be increased beyond the level specified by the current indexing formula, or even eliminated altogether as with Medicare, so that all earnings are subject to the payroll tax?

n

If all earnings were subject to the payroll tax, should all earnings also be used when calculating Social Security benefits? If all earnings are taxed but only earnings up to a limit are used to calculate benefits, would the program lose support among workers with earnings above that limit?

n

Should a portion of trust fund assets be invested in asset classes other than Treasury securities with higher potential investment returns but greater investment risk?

A value-added tax is a form of consumption tax. From the perspective of the buyer, it is a tax on the purchase price. From that of the seller, it is a tax only on the “value added” to a product, material or service, from an accounting point of view, by this stage of its manufacture or distribution. The manufacturer remits to the government the difference between these two amounts, and retains the rest for themselves to offset the taxes they had previously paid on the inputs. http://en.wikipedia.org/wiki/Value_added_tax 3

▲ ISSUE GUIDE: SOCIAL SECURITY REFORM

WWW.ACTUARY.ORG

3


CHANGES TO THE BENEFIT FORMULA

Social Security historically has provided benefits directly related to workers’ earnings during their careers. The benefit formula4 is indexed to wage inflation before age 62 and to the cost of living after age 62. Benefits calculated in the current Social Security formula replace a much higher portion of lifetime average earnings for lower-paid workers than for higher-paid workers. For example, the replacement rate (i.e., the percentage of a worker’s pre-retirement earnings that is replaced by Social Security) at normal retirement age of a worker whose earnings have always equaled the national average wage ($41,674 in 2010) is about 55–60 percent higher than for a worker whose earnings have always equaled the Social Security earnings base ($106,800 in 2010). This progressive benefit formula is the primary method through which the program addresses adequacy of benefits for workers with low earnings. Proposals that change the benefit formula or the cost-of-living adjustment also may affect the adequacy of benefits for workers with low earnings.

Questions to Ask When Evaluating Proposals That Change Social Security’s Benefit Formula n

How would a proposed benefit change affect the standard of living during retirement for workers and their family members at different income levels?

n

How would a proposed change affect the standard of living for disabled workers and beneficiaries of deceased workers at different income levels?

n

How would a proposed change affect the benefits of divorced spouses?

n

If benefit payments or cost-of-living adjustments were reduced when projections show the system is inadequately financed, how should those cuts be allocated?

n

If savings come primarily from reducing benefits for high-income workers, would the program retain support among those workers?

Social Security Reform: Changes to the Benefit Formula and Taxation of Benefits (June 2010 update) http://www.actuary.org/pdf/socialsecurity/Social_Security_Reform_Issue_Brief_6-15-10.pdf 4

▲ ISSUE GUIDE: SOCIAL SECURITY REFORM

WWW.ACTUARY.ORG

4


CHANGES TO THE TAXATION OF BENEFITS

Annuities generally are included in income for tax purposes except for portions of the annuity that already have been taxed. Before 1984, no portion of a recipient’s Social Security benefits was included in income for tax purposes. Today, if a recipient’s adjusted gross income, plus nontaxable interest income, plus one-half of their Social Security benefit, exceeds a specified threshold, then a portion of the Social Security benefit is included in taxable income. That threshold is $25,000 for a single person and $32,000 for a married couple filing jointly, and the portion of the Social Security benefit included in taxable income could be up to 85 percent. Unlike most dollar thresholds in Social Security and income tax formulas, these are not indexed for inflation in prices or wages. Revenue from taxation of Social Security benefits does not go into the federal government’s general fund like other income tax receipts; instead it is used to help finance Social Security and Medicare.

Questions to Ask When Evaluating Proposals That Change the Taxation of Social Security Benefits n

Should Social Security benefits continue to be taxed using the current formula? If not, should they be taxed like other forms of annuity income? Should the amount taxed be based on the specific history of taxable contributions made by each recipient?

n

How would taxing benefits in this way affect the relative benefits of lower-paid workers versus higher-paid workers?

n

If the current method for taxing benefits is retained, should the income thresholds be updated and/or indexed for inflation?

n

Should revenue from the taxation of Social Security benefits continue to be split between Social Security and Medicare, go entirely to Social Security, or go to the federal government’s general fund like other income tax receipts?

n

To what extent would changes to benefit taxation influence workers’ retirement decisions?

▲ ISSUE GUIDE: SOCIAL SECURITY REFORM

WWW.ACTUARY.ORG

5


MEANS TESTING

One of the proposed Social Security reforms is a reduction or elimination of benefits for wealthier and/ or higher-income participants and beneficiaries, and is generally referred to as means testing.5 Advocates of means testing say that reducing or eliminating benefits for those whose income or assets exceed certain thresholds would reduce Social Security’s financial deficit while helping to preserve Social Security as a safety net for those who truly need it. An underlying belief for those who support means testing is that government funds should not be used to aid those who are not in financial need. Social Security benefits, on the other hand, currently are based on a worker’s covered wages. This link between the wages that have been taxed during a worker’s career and the benefits the worker receives after retirement creates the perception of an “earned right” to program benefits. Since Social Security’s inception, the program has paid benefits to all retired workers who have worked in covered employment for a sufficient period, and to their family members and beneficiaries, without regard to income or wealth. Because the great majority of workers today are covered by Social Security, another important feature of Social Security is its universality. The principles of earned right and universality contribute to the program’s broad popular support. There are a number of philosophical and practical concerns associated with means testing. Of pri­ mary concern is the possible erosion of popular support for the system if the earned right and universality principles are modified or abandoned. Behavioral finance suggests that if the accumulation of savings for retirement were to reduce workers’ Social Security benefits, workers may not save as much outside Social Security. Means testing could lead to the manipulation of income and assets to reduce the effect of testing, which would defeat the purpose while creating economic inefficiencies. It also would increase the administrative and compliance burdens associated with the Social Security program.

Questions to Ask When Evaluating Proposals for Means Testing n

Should means testing be based on income or assets, or both?

n

How would income and/or assets be measured?

n

How would the appropriate income and/or asset threshold for benefit reductions be determined?

n

How would means testing be administered?

n

Should Social Security be modified to resemble government safety net programs?

n

Would this change in philosophy weaken public support for the program?

n

Would direct savings from means testing be offset by indirect costs, such as reduced incentives to work or save for retirement, legal or illegal avoidance of benefit reductions, and increased administrative or oversight costs?

n

Could alternatives that avoid direct means testing, such as changing the benefit formula or taxation of benefits, achieve similar results within the current program structure?

Means Testing for Social Security (Dec. 2012; updates a 2004 issue brief) http://www.actuary.org/files/Means_Testing_SS_IB.pdf

5

▲ ISSUE GUIDE: SOCIAL SECURITY REFORM

WWW.ACTUARY.ORG

6


RAISING THE NORMAL RETIREMENT AGE

When the Social Security program began paying monthly benefits in 1940, workers could receive unreduced benefits beginning at age 65—the normal retirement age (NRA). The law was changed in 1983 to increase the NRA gradually, beginning in 2000, from age 65 to age 67, recognizing, at least in part, the marked increase in longevity. Raising the retirement age further could improve significantly Social Security’s financial status.6, 7 The cost of Social Security is increasing partly due to the fact that workers now are living longer, which means they receive benefits for a longer period. Since Social Security began paying monthly benefits, life expectancy at age 65 for both men and women has increased by about five years. Life expectancy is expected to increase by about five more years during the next 75 years according to the trustees’ projections. To further complicate this situation, studies8 have shown that the average age of retirement in the United States decreased through the mid-1980s, though more recent studies appear to show that the average age is now increasing. The combination of living longer and retiring earlier means that Social Security must pay benefits for a longer period of time, while payroll taxes are collected for a shorter period. Proposals calling for an increased normal retirement age include: ad hoc increases to the NRA as deemed necessary, indexing the NRA to keep the average benefit payment period the same, indexing the NRA to keep the ratio of retirement years to working years the same, and adjusting the NRA as necessary to maintain actuarial balance. Workers are required by law to wait until the early eligibility age—currently 62—to receive benefits, which are reduced for retirement before the NRA. Some proposals would raise the early eligibility age in addition to the NRA.

Questions to Ask When Evaluating Proposals That Raise the Retirement Age n

Do improvements in older workers’ health and longevity justify increasing the age requirement for full benefits? Would this be true even if there were no projected shortfall?

n

If workers must wait longer to receive full Social Security benefits, would jobs be available for them? Is the answer the same for men and women? For all income levels?

n

Would raising the retirement age place a particular burden on workers in physically demanding occupations?

Raising the Retirement Age for Social Security (October 2010; updates an October 2002 issue brief) at http://www.actuary.org/pdf/socialsecurity/Social_Sec_Retirement_Age_IB_FINAL_10_7_10_2.pdf 7 The Academy issued an advocacy statement in August 2008 in support of raising the Social Security retirement age. http://actuary.org/pdf/socialsecurity/statement_board_aug08.pdf 8 http://www.tiaa-cref.org/institute/research/briefs/pb_earlyretirement0711.html 6

▲ ISSUE GUIDE: SOCIAL SECURITY REFORM

WWW.ACTUARY.ORG

7


n

How would raising the normal retirement age for workers affect the benefits of family members?

n

If the age requirement for full benefits is increased, should the early eligibility age be increased as well?

n

To what extent would cost savings be offset by additional disability benefits?

n

How would raising the retirement age affect Medicare and employer-sponsored health and retirement plans?

â–˛ ISSUE GUIDE: SOCIAL SECURITY REFORM

WWW.ACTUARY.ORG

8


INDIVIDUAL ACCOUNT PROPOSALS

Social Security currently is a defined benefit program. This means that the contributions of all workers are pooled and available to pay benefits to any covered worker or family member. Some reform proposals include converting a portion of the system to a defined contribution program.9 Under this type of program, workers would have individual accounts based directly on a worker’s own contributions plus investment earnings, and funds would be available only to pay benefits to that particular worker and his or her family members. As employer-sponsored retirement plans are increasingly shifting from defined benefit to defined contribution type plans, some believe Social Security should move in the same direction. Others believe the shift toward defined contribution plans makes preserving Social Security as a defined benefit system even more important. Individual accounts would redirect a portion of the payroll taxes previously used to support benefits provided under the current formula, so that the formula would have to be reduced (in addition to any reductions otherwise required to achieve long-term solvency). If the accounts have investment earnings above a specified target level, assets in the accounts at retirement could purchase annuity benefits sufficient to offset these reductions. Proponents of individual accounts say that this would allow the current level of benefits to be paid without raising taxes.

Questions to Ask When Evaluating Proposals for Individual Accounts n

Will an individual account proposal do a better job of providing retirement security than the guaranteed benefits in place under the current Social Security program?

n

Would workers’ individual accounts be mandatory or voluntary? If they are voluntary, how would the program protect itself from becoming an annuity program for low-income workers and an individual account program for high-income workers?

n

How would the program protect (and pay for) benefits already earned, especially for current older workers and retirees?

n

Would the program continue to provide a basic level of support for older workers and retirees who would not have an opportunity to build substantial account balances?

n

What level of investment return would a worker need to achieve to receive the same benefits as under the current formula? What happens if investment returns are inadequate?

n

If the current progressive formula is eliminated, would higher-income workers subsidize the accounts of lower-income workers? If not, how will the program address the possibility that benefits would be inadequate for lower-income workers?

Social Security Individual Accounts: Design Questions (October 2003) http://www.actuary.org/pdf/socialsecurity/ ssia_1003.pdf 9

▲ ISSUE GUIDE: SOCIAL SECURITY REFORM

WWW.ACTUARY.ORG

9


n

How would the program provide adequate benefits to workers who become disabled early in their careers, or to the survivors of workers who die early in their careers?

n

Would the transition to individual accounts require financing from general revenue? If so, how much?

n

Would individual accounts be managed and invested centrally, or would workers be allowed to choose their own investment managers and their own investments?

n

How would those investment vehicles affect the relevant financial markets, and what is the likely affect on the value of current private (and public) investments?

n

What type of oversight would there be for the investment alternatives to protect workers?

n

Under what circumstances, if any, would workers be allowed access to their accounts before retirement?

n

Would payout of benefits by lifetime annuities be mandatory or voluntary? If lifetime annuities are voluntary, how will the program address the risk that employees who do not elect annuities will outlive (or outspend) their retirement accounts? How would payout annuities be designed, priced, and administered?

â–˛ ISSUE GUIDE: SOCIAL SECURITY REFORM

WWW.ACTUARY.ORG

10


PUBLIC STATEMENTS ON SOCIAL SECURITY BY THE AMERICAN ACADEMY OF ACTUARIES The following publications are available on the Academy website at http://www.actuary.org/briefs. asp#soc. Means Testing for Social Security (December 2012; updates a 2004 issue brief) http://www.actuary. org/files/Means_Testing_SS_IB.pdf Understanding the Assumptions Used to Evaluate Social Security’s Financial Condition (May 31, 2012) http://www.actuary.org/files/SSC_IssueBrief_Assumptions_120501.pdf Significance of the Social Security Trust Funds (May 31, 2012) http://www.actuary.org/files/SSC_IssueBrief_TrustFund_120501.pdf An Actuarial Perspective on the 2012 Social Security Trustees’ Report (May 31, 2012) http://www. actuary.org/files/SSC_IssueBrief_2012TrusteesReport_120501.pdf Automatic Adjustments to Maintain Social Security’s Long-Range Actuarial Balance (August 2011; updates a 2002 issue brief) http://www.actuary.org/pdf/pension/IB_Auto_Adjustments_FINAL.pdf Raising the Retirement Age for Social Security (October 2010; updates an October 2002 issue brief) http://www.actuary.org/pdf/socialsecurity/Social_Sec_Retirement_Age_IB_FINAL_10_7_10_2.pdf Social Security Reform: Changes to the Benefit Formula and Taxation of Benefits (June 2010 update) http://www.actuary.org/pdf/socialsecurity/Social_Security_Reform_Issue_Brief_6-15-10.pdf Academy recommendation to raise the Social Security retirement age (August 2008) http://actuary.org/pdf/socialsecurity/statement_board_aug08.pdf Social Security: Evaluating the Structure for Basic Benefits (September 2007 issue brief) http://www. actuary.org/pdf/socialsecurity/structure_sept07.pdf Women and Social Security (July 2007 issue brief) http://www.actuary.org/pdf/socialsecurity/ women_07.pdf Investing Social Security Assets in the Securities Markets (March 2007 issue brief) http://www.actuary.org/pdf/pension/sec_0307.pdf Social Security Individual Accounts: Design Questions (October 2003) http://www.actuary.org/pdf/ socialsecurity/ssia_1003.pdf

▲ ISSUE GUIDE: SOCIAL SECURITY REFORM

WWW.ACTUARY.ORG

11


Mary Downs, Executive Director Craig Hanna, Director of Public Policy David Goldfarb, Pension Policy Analyst Members of the Social Security Committee: Janet Barr, MAAA, ASA, EA, chairperson; Eric Klieber, MAAA, FSA, EA, vice-chairperson; Robert Alps, MAAA, ASA; Eric Atwater, MAAA, FSA, FCA, EA; Raymond Berry, MAAA, ASA, MSPA, EA; Douglas Eckley, MAAA, FSA; Timothy Leier, MAAA, FSA, EA; Timothy Marnell, MAAA, ASA, EA; John Nylander, MAAA, FSA; Brendan O’Farrell, MAAA, EA, FSPA, FCA; Zenaida Samaniego, MAAA, FSA; Bruce Schobel, MAAA, FSA, FCA; Mark Shemtob, MAAA, ASA, EA; P.J. Eric Stallard, MAAA, ASA, FCA; Ali Zaker-Shahrak, MAAA, FSA.

For further information, contact us at:

American Academy of Actuaries 1850 M Street NW, Suite 300 Washington, DC 20036-5805 Telephone 202 223 8196 Facsimile 202 872 1948 WWW.ACTUARY.ORG

▲ ISSUE GUIDE: SOCIAL SECURITY REFORM

WWW.ACTUARY.ORG

12


American Academy of Actuaries

MARCH 2009

MAY 2012

Significance of the Social Security Trust Funds

Key Points Since the enactment of the Social Security program in 1935, dedicated payroll tax revenues have been placed into the trust funds and benefit payments have flowed only out of those trust funds. n The trust funds have several functions, including: tracking resources dedicated to Social Security, keeping payroll tax rates level over time, connecting the benefits to taxes paid and requiring their long-run balance. Notably, the trust funds are not allowed to borrow. n Social Security revenue exceeded cost between 1984 and 2009, leading to a buildup of $2.7 trillion in trust fund assets. Under current law, these assets are projected to be redeemed over the next two decades. The anticipated redemption of those assets has led to an ongoing debate over whether the assets are “real.” n The perceived significance of the trust fund assets largely depends on the context in which they are viewed: the Social Security system, the unified federal budget, or the whole economy. n

S

ocial Security is designed primarily as a pay-as-you-go system. After an initial start-up period, its trust funds1 generally have contained only modest contingency reserves to cover short-term fluctuations in income and outgo.2 However, in each year from 1984 until 2009, tax revenues exceeded program costs.3 As a result, the trust fund assets grew to a historically high level of $2.7 trillion at the end of 2011. Sizeable trust fund balances are a temporary feature of the program and were never intended to be an important source of the program’s long-term economic viability. Program costs exceeded tax revenues in 2010 and negative net cash flows are projected to continue in all future years. These annual deficits will be offset by redeeming trust fund assets until the trust fund assets are eventually exhausted. In anticipation of the redemption of trust fund assets, an ongoing debate has intensified over whether those assets are “real”—whether they are a store The “trust funds” considered in this issue brief are the Old-Age and Survivors Insurance (OASI) trust fund and the Disability (DI) trust fund. They are two separate funds—the OASI trust fund was established 1937 and the DI trust fund in 1957—but for many purposes they can be analyzed together as the trust funds supporting the entire OASDI system. Except when referring to years prior to the introduction of DI, this brief makes no distinctions between the two trust funds and all stated amounts refer to the combined OASDI trust funds. 2 The combined OASI and DI trust funds held assets equal to less than one year’s program cost in all years from 1971 to 1992 and less than one-and-a-half year’s program cost in all years from 1962 to 1996. 3 “Costs” when used in this brief, has the same meaning as in the Trustees Report and includes scheduled benefit payments, administrative expenses, net interchange with the Railroad Retirement program, and payments of vocational rehabilitation services for disabled beneficiaries paid from OASDI trust funds. 1

The American Academy of Actuaries is a 17,000-member professional association whose mission is to serve the public and the U.S. actuarial profession. The Academy assists public policymakers on all levels by providing leadership, objective expertise, and actuarial advice on risk and financial security issues. The Academy also sets qualification, practice, and professionalism standards for actuaries in the United States. ©2012 The American Academy of Actuaries. All Rights Reserved.

1850 M Street NW, Suite 300, Washington, DC 20036 Tel 202 223 8196, Fax 202 872 1948 www.actuary.org Mary Downs, Executive Director Mark Cohen, Director of Communications Craig Hanna, Director of Public Policy Don Fuerst, Senior Pension Fellow Jessica Thomas, Senior Pension Policy Analyst


of value that can be drawn on to pay future benefits or whether they are an accounting device without significance. Since the program’s enactment, dedicated payroll tax revenues have been placed into the trust funds and benefit payments have flowed only out of the trust funds. The trust funds are legal entities—established, in part, to preclude political interference with the program. The trust fund structure also serves other functions: n Tracking—serves as a record of claims to a share of resources dedicated to Social Security. n Smoothing—enables Social Security to be financed as a level percentage of payroll across generations. n Social contract—enhances the legitimacy of beneficiaries’ claims to benefits. n Governance—places constraints on the benefits Congress can promise by limiting the funds available to pay those benefits. The economic and fiscal consequences of the trust funds can be analyzed from various perspectives, such as: n Social Security system perspective— Social Security holds financial assets backed by the full faith and credit of the U. S. government. Those assets are the only legal source of benefit payments, so the trust fund assets have real and tangible consequences.

Unified budget perspective—The federal government is one fiscal entity. The securities in the Social Security trust funds are both assets of the trust funds and liabilities to the Treasury’s general fund.4 The assets and liabilities offset each other, and therefore have no effect on the government’s net balance sheet. n The whole economy perspective— Whether the trust funds have macroeconomic consequences depends primarily on whether they affect national savings. They may also affect the distribution of after-tax income as they shift the tax burden between payroll and income taxes over time. Whether the trust funds are seen as “real” depends largely on the perspective from which this question is approached. The debate over the nature of the trust funds can become a distraction from the long-term fiscal issues facing Social Security. In an effort to lay this debate to rest, this issue brief explains how the Social Security trust funds can be understood from each of these three perspectives. n

Background The Old-Age, Survivors, and Disability Insurance (OASDI) trust funds, also known as Social Security trust funds, are accounts managed by the Department of the Treasury and designated to be used only for payment of Social Security benefits and administrative costs.

“The general fund of the Treasury” is the common term for the funds held by the Treasury of the United States, other than receipts collected for specific purpose (such as Social Security) and maintained in a separate account for that purpose. 4

Members of the Social Security Committee who participated in revising this issue brief include: Robert Alps, MAAA, ASA; Eric Atwater, MAAA, FSA, FCA, EA; Janet Barr, MAAA, ASA, EA - chairperson; Raymond Berry, MAAA, ASA, EA; Michael Callahan, MAAA, EA, FSPA; Eric Klieber, MAAA, FSA, EA - vice chairperson; Timothy Leier, MAAA, FSA, EA; Timothy Marnell, MAAA, ASA, EA; John Nylander, MAAA, FSA; Brendan O’Farrell, MAAA, EA, FSPA, FCA; Steven Rubenstein, MAAA, ASA; Bruce Schobel, MAAA, FSA, FCA; Mark Shemtob, MAAA, ASA, EA; P.J. Eric Stallard, MAAA, ASA, FCA; Ali Zaker-Shahrak, MAAA, FSA 2

ISSUE BRIEF MAY 2012


Social Security taxes are required by federal law to be deposited into the trust funds and invested in interest-bearing securities backed by the full faith and credit of the U.S. government.5 In recent years6, all trust fund investments have been made in special-issue Treasury securities. Special-issue securities differ from Treasury securities issued to the public in two ways: they can be redeemed at par7 at any time, and they cannot be traded in the financial markets.8 The law stipulates a formula that sets the interest rate applicable for securities issued in a given month to the average market yield (rounded to the nearest eighth of a percent) on marketable interest-bearing securities of the federal government which are not due or callable until after four years from the last business day of the prior month. Trust fund assets are invested and redeemed regularly to cover fluctuations in cash flows as revenues come in, old securities mature, and benefit payments and other expenses become due. Most of these trust fund operations are carried out with little public attention, but the net change in the trust funds from year to year is usually much more notable. When Social Security tax revenues and interest on the trust funds exceed benefits and administrative costs over a given period, the trust funds grow, and when costs exceed income, the trust funds are drawn down (redeemed). The trust funds are not authorized to borrow, so their net assets cannot be less than zero.

Social Security was designed to be essentially a pay-as-you-go system, with a trust fund9 required to maintain a modest contingency reserve because the program is not allowed to borrow. When the program was new, with few people receiving benefits, the trust fund grew large relative to the annual benefit cost, but small compared to the gross domestic product (GDP).10 As the program matured, in the 1960s and 1970s, the trust funds only held assets sufficient for approximately one year of benefit payments and administrative costs. However, the legislative changes implemented by the 1977 and 1983 Social Security amendments reduced Social Security benefits and increased the Social Security retirement age and tax revenues in anticipation of a greatly increased number of beneficiaries relative to workers when members of the baby boom generation began retiring, which was then about 25 years in the future. In particular, tax rates were scheduled to increase in advance of the expected baby boomers’ retirement wave, and to remain constant thereafter. For the next 26 years, from 1984 to 2009, tax income consistently exceeded program expenses. These sustained surpluses have led, as of December 2011, to the accumulation of $2.7 trillion in the trust funds—enough to fund the system for nearly four years even in absence of any further tax revenue. Now that the baby boomers have begun to retire, expenses have begun to overtake tax in-

Social Security Act, Title II, Sec. 201(d) [42 U.S.C. 401] It has been Treasury Department policy for over 20 years to invest trust fund assets only in special-issue Treasury securities. The DI trust fund still contains some publicly-issued marketable government bonds. For more information, please see this link: http://www.ssa.gov/OACT/NOTES/note2000s/note145/note145.html: 7 Par Value is the value printed on the face of a security. For both public and special issues held by the trust funds, par value is also the redemption value at maturity. 8 The trust funds can redeem securities before maturity only when needed to pay program costs, not to reinvest in securities earning a different interest rate. Even when cash flows are positive, short term fluctuations in revenue can require that bonds be redeemed before maturity. The Secretary of the Treasury determines the need for asset redemption. 9 Initially only the OASI trust fund was established (1937) and the DI trust fund was later established (in 1957). 10 For example, in 1944, trust fund assets were 20 times the annual cost of the program, but less than 3 percent of GDP. Also, program cost was rising rapidly in its early years, so those assets could have financed only the next seven years of actual benefit costs. For comparison, trust fund assets at the end of 2010 were about 18 percent of GDP. 5 6

ISSUE BRIEF MAY 2012

3


come, leading to a gradual drawdown of the trust funds and their eventual exhaustion over the next few decades.11 Without further tax increases or reductions in scheduled benefits, the trust funds are projected to eventually run out of assets under most projection scenarios.12 As the focus has shifted to the drawdown of the trust funds, their nature and significance have become a contentious issue in public policy debates. In a 2000 presidential debate, Vice President Al Gore promised to “keep Social Security in a lockbox,” while Governor George W. Bush responded that “what he’s doing is loading up I.O.U.’s for future generations.” This exchange illustrates the politically charged policy debates that have developed over the past few years. Versions of these arguments have been repeated many times, revealing deep differences in views about the nature of the trust funds. Some perceive the assets in the trust funds as a source of advanced funding for future benefits, while others see those assets as nothing more than accounting illusion. Social Security is treated as an “off-budget” item for the purpose of federal budget reporting—that is, it is considered an entity separate from the “on-budget” part of the government, which comprises the bulk of government agencies (other than Social Security and the U.S. Postal Service). This separation means that Social Security’s past surpluses have not offset the government’s reported deficits, nor will benefit payments in excess of dedicated payroll tax income in the future add to the government’s deficits. In addition, the Treasury securities held by the trust funds are treated as part of the federal government’s debt. The Office of Management and Budget also

publishes, separate from its officially reported budget, a “unified budget,” which encompasses all government programs, including Social Security. The unified budget is used primarily for long-range planning purposes. When Social Security was running a surplus, the trust funds relieved the Treasury of the need to cover a portion of the government’s deficit through the sale of debt securities to investors outside the government. When benefit payments exceed payroll tax income in the future, the need for outside securities sales will increase as the securities held in the trust funds are redeemed. Regardless of accounting conventions, the accumulation of Social Security trust funds reflects the fact that, from 1984 to 2009, some part of revenues dedicated to Social Security was invested in government securities, that is, loans to the Treasury. As those loans were available to finance other government expenditures, actual on-budget government spending in those years required less revenue from other taxes (mainly individual and corporate income taxes) and/or less borrowing from the public than would have been the case without the Social Security surpluses. The securities held in the trust funds are certificates of that lending/borrowing relationship between Social Security and the general fund of the Treasury. All lending/borrowing arrangements involve a reallocation of resources over time. Early on, the borrower consumes more, and the lender less, than their respective incomes; later, the reverse occurs. As a result, during the period the securities in the trust funds are redeemed, other sources of Federal revenue will finance part of Social Security benefit

Expenses overtook tax income in 2010 and are projected to remain higher than tax income indefinitely under the Trustees’ intermediate assumptions. The trust fund assets, however, are projected to keep increasing until 2023 due to interest income. 12 The Social Security trustees project the finances of the system under three sets of assumptions: intermediate, high-cost, and low-cost. Under the low-cost assumptions, the trust funds are not projected to run out of assets over the 75-year projection period. Under the other sets of assumptions, as well as under the Congressional Budget Office (CBO)’s assumptions, the trust funds are projected to run out of assets. 11

4

ISSUE BRIEF MAY 2012


payments. This will not be an unprecedented situation. From 1957 to 1965, redemptions of trust fund assets financed a portion of Social Security costs in every year. In 1959, interest and asset redemptions paid for more than 17 percent of the program’s cost. Assuming no changes in the law, as the ratio of benefit recipients to workers increases over the next two to three decades, the proportion of benefits not financed from current Social Security taxes will increase, reducing the trust fund assets until they are eventually exhausted. Just before that happens, approximately one-fourth of the benefit payments will come from redemptions of trust fund securities by the general fund of the Treasury. Once the trust funds are exhausted, Social Security will not have authority to pay more than it collects in dedicated taxes, and the program will face a shortage, initially equal to the amount of the latest repayments from the general fund.

The Trust Funds’ Many Purposes and Consequences The original designers of Social Security had specific reasons for financing the program by a dedicated tax and segregating the income from that tax into trust funds separate from the government’s other assets. Over the years, the trust funds came to serve additional purposes. Those who are focused on the different functions of the trust funds, or their consequences in different areas, may easily reach different conclusions about the trust funds’ relevance. From one perspective, the trust funds are an accounting mechanism serving a useful tracking function—as a record of claims to a share of resources dedicated to Social Security through cumulative Social Security taxes. This is necessary because dedicated revenues do not exactly match the program cost in each year, so the

Social Security Non-Interest Income and Cost as a Percentage of GDP, 1957–2011 6.00%

5.00%

4.00%

3.00%

Non-Interest Income

2.00%

Cost

1.00%

2011

2009

2007

2005

2003

2001

1999

1997

1995

1993

1991

1989

1987

1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

1959

1957

0.00%

ISSUE BRIEF MAY 2012

5


trust funds lend the excess revenue (surplus) to the general fund of the Treasury in some years, and require repayments of such loans in other years when there is revenue shortfall (deficit). From another perspective, the buildup of the trust funds serves a distributional or smoothing function. The trust funds enable Social Security to be financed as a level percentage of payroll across generations. For a quarter century ending in 2009, Social Security taxes continually exceeded program outgo and, on a unified budget basis, Social Security surpluses were adding to government financial resources. Going forward, that flow will reverse and Social Security will reduce government financial resources. The trust funds provide a direct connection between the payroll taxes that finance Social Security and the benefits provided by the system. This connection fosters a widely held perception that the benefits are earned and thus makes rescinding or reducing them politically difficult. In this way the trust funds can be seen as an element of a social contract between the workers and the government. Finally, an important feature of the trust funds is that they are not allowed to borrow. Thus they set a limit on the amount the program can spend. In this manner, the trust funds perform a governance function by providing a brake on any temptation for Congress to raise benefits to levels not supported by commensurate taxes. There are at least three perspectives from which economic consequences of the trust funds can reasonably be examined. First, one can focus specifically on the Social Security (OASDI) system. This is the most common perspective of actuarial analysis and particularly useful when analyzing the program’s finances, especially its solvency. Second, one can take the perspective of the federal government as a whole. This is consistent with the focus on overall government revenues and outlays, and 6

ISSUE BRIEF MAY 2012

the resulting unified budget surplus or deficit. Finally, one can take the very broad perspective of the entire economy, focusing on aggregate national savings, capital accumulation, and output (gross domestic product). The following sections elaborate on these three perspectives.

The OASDI System Perspective When focusing on the ability of Social Security to pay scheduled benefits, the trust funds’ role in receiving and holding dedicated tax revenue and disbursing benefits is highly important. Under current law, the trust funds are the only source available to the program for benefit payments, and if they became exhausted, the Social Security program would have no legal authority to pay benefits until the funds are replenished by additional tax revenue. If revenue were then not sufficient to pay all benefits as they become due, the Social Security Administration would have to delay payments until the necessary resources became available. If this situation were to continue, benefit payments would fall further and further behind. From this perspective, the trust fund assets have very real and tangible consequences. Social Security usually is thought of as a defined benefit pension program because its benefits are determined by a formula based on individual earnings and years of work. However, its financing has more in common with defined contribution pensions, since contributions are a fixed percent of workers’ earnings rather than an amount derived from an actuarial calculation of the program’s annual cost. Such a combination of defined benefits and defined contributions can be sustained for a long period if the ratio of workers to beneficiaries is stable (i.e., its age distribution and other characteristics do not change over time). When the population ages and benefit


costs increase relative to contributions, this arrangement produces imbalances between contributions and payments. Thus, Social Security is viewed as storing its positive imbalances (surpluses) as trust fund assets, to be used as a temporary supplement to payroll tax revenue when the imbalances between benefits and contributions turn negative (deficits). Such use of trust fund assets was expected and intended, as well as significant for the program’s ability to pay benefits regularly when due. Like all Treasury securities, the specialissue securities held by the trust funds are backed by the “full faith and credit of the United States”. There has never been a substantive default on U. S. government obligations, and investors seem confident that one will not happen in the future, as evidenced by yields on U.S. Treasuries that are regularly lower than on any other debt securities of comparable maturity. The securities in the Social Security trust funds are considered safe, like other Treasury securities. It could be argued that, even though the securities in the trust funds are backed by the full faith and credit of the U. S. government, the trust funds themselves were created by statute, and that new laws can change their claims on any property or even conceivably abolish the entire Social Security program. While this is true, such radical acts of legislation are extremely unlikely and, in any case, a similar scenario would be equally possible if the trust funds held different types of assets. Assuming that Social Security worked entirely through individual accounts, for example, new laws could drastically reduce net benefits by raising taxes on distributions from those accounts.

The Unified Budget Perspective Trust Fund Investments as Both Assets and Liabilities Regardless of how the federal budget is reported, Social Security is a part of the federal government. The unified budget perspective considers the totality of government finances rather than distinguishing between those parts that are on-budget and off-budget, or viewing one government program in isolation from all other government programs. The federal government can only spend as much money as it collects in taxes and borrows from the public.13 When one part of the government borrows from another part, the government’s debt to outside parties does not change and as a result, the aggregate amount that it can spend is unchanged. When focusing on the finances of the government as a whole, it is easy to see how the trust funds might appear as an accounting artifact. The government owes to itself, so its capacity to pay for its programs is unaffected. When Social Security redeems a Treasury bond, the general fund of the U.S. government pays cash to Social Security, and that amount becomes unavailable for other programs. If the government was going to spend that money on something else, it now has to forgo that spending, or borrow from the public, or raise taxes. In that sense, whether the trust funds exist makes no difference, since the government can effectively “decide” to pay scheduled Social Security benefits. In addition, if increased taxes are used to raise cash for redeeming the securities held in the trust funds, it would make no difference to the government’s unified budget whether the Social Security tax or other taxes were raised. No

Strictly speaking, that is an incomplete list, because the federal government can also finance its spending by seignorage (“printing money”). However, in the US, seignorage has never been a large source of government finances. Central banks of developed countries tend to restrain seignorage because excessive seignorage can lead to hyperinflation. 13

ISSUE BRIEF MAY 2012

7


matter how the internal accounting works, for each dollar of spending, the government has to collect a dollar through some combination of taxes and/or borrowing. Conversely, when considering the trust funds’ buildup phase from a unified budget perspective, Social Security taxes during that period may be viewed as not really dedicated to financing of Social Security benefits. They instead were partly financing other government programs. From the unified budget perspective, the trust funds have no effect on the government’s aggregate finances in the short run. While some trust fund features can have a significant effect on the Social Security program—if there is no money in the trust funds, benefits cannot be paid even should the rest of the government be flush with surpluses—that is because federal law prevents the use of the government’s general fund to pay Social Security benefits. From a unified budget perspective, however, this law can always be changed or, if not changed, the constraint can be avoided by an ad hoc transfer of assets from the general fund to the trust funds.14

Trust Funds’ Budgetary Effects in the Long Term The dynamics of the trust funds may affect other government budgetary choices. At one extreme the on-budget deficit (or surplus) may be assumed to be completely independent from Social Security finances so that every dollar of Treasury securities issued to the

trust funds means one dollar less borrowed from the public. Based on this point of view, it would follow that the trust fund accumulation has reduced public debt dollar for dollar. Few economists believe that this scenario is fully correct. An opposite view assumes borrowing from the public to be fixed—limited by investors’ willingness to buy government securities in the market—so that the ability of the general fund to borrow from the trust fund enables the government to run larger onbudget deficits, dollar for dollar. Most analysts believe that the reality is somewhere between those two assumptions.15 Although researchers have tried to estimate the effect, if any, that the trust funds have had on the federal budget, such measurement is complicated by the dynamic nature of the legislative process and a legitimate difference in opinion exists among experts. The selection of assumptions about economic, legislative, and political changes the government would have employed will lead to different outcomes.

The Whole Economy Perspective National Savings An argument often heard in debates about Social Security is that saving for retirement should be real saving that would contribute to total national savings. According to standard macroeconomic theory, sustained higher national savings leads to more investment, which in turn results in more capital and, hence, higher future economic output or GDP.16 Fo-

This analysis abstracts from another important legal matter: The Budget Enforcement Act of 1990 actually requires that Social Security be treated as off-budget, which means that the unified view of whole-government finances could be at odds with legal definitions of budgetary terms. Nevertheless, abstractly speaking, if one takes a view that “the government” is a black box such that only outside cash flows are observable, and all inner workings—including legislative and regulatory action—can be assumed to be hidden inside, such abstractions are a consistent part of the model. Simplifications of this kind are common, and often useful, in economic models. As with any assumptions, their justification in any specific case is open for discussion, but such discussion is beyond the scope of this brief. 15 Even the two idealized cases may not span the full range of opinions. Smetters (AER, 2004) and Nataraj and Shoven (NBER, 2004) have suggested that running surpluses in Social Security trust funds may have increased even the unified deficit. The theoretical idea is that the added layer of accounting complexity enables legislators to game the public by avoiding casting unpopular votes and picking the set of numbers that make them look good to their partisan base. 14

8

ISSUE BRIEF MAY 2012


cusing on the macroeconomic aspects of retirement savings, the most important question about the trust funds is whether they have an effect on the overall level of national savings. National savings equals private savings minus government deficits.17 Therefore, for national savings to increase, private savings must increase or government deficits must decrease. The savings levels of governments and the private sector often change in opposite directions. For example, if the government increases income taxes without increasing spending, deficits will decrease. But if individuals react to their lower after-tax income by saving less, national savings will increase less, or not at all. A tax cut, conversely, is likely to increase private savings, but it also increases government deficits—and the effect on national savings is the net of the two effects. If the Social Security trust funds affect government deficits (a possibility discussed in the previous section), the private sector response is likely to be partially offsetting. The trust funds’ effect on national savings, therefore, is likely to be equal in direction and smaller in magnitude than the effect on the whole of government finances. Any effect on national savings is likely to be small. Another way to think about national savings is that, with a given level of economic output, aggregate savings can be increased only by reducing aggregate consumption. The only way a society can save more is if it spends less now. It does not matter whether the saving is done primarily through government or private actions. This makes it clear that the trust funds do not directly change national savings. They may create expectations that induce individu-

als to change their savings behavior, but they do not directly contribute to national savings.

Distributional Aspects Social Security payroll tax is a level percent of earnings up to the taxable maximum. Income tax, in contrast, is progressive—lower– income individuals pay lower taxes not only in absolute dollars, but also as a percent of their incomes. Consequently, overall taxation is less progressive when Social Security is running a surplus—and the excess payroll tax income finances a portion of other government expenditures—than it would be if those expenditures were financed by higher income taxes. Conversely, overall taxation would tend to become more progressive when all payroll tax income must be used immediately to pay benefits, especially if income taxes were raised to cover all or part of Social Security’s annual funding shortfalls. Changes in the level of government borrowing from the public can have similar distributional effects, but the analysis is more complicated. Individuals at different income levels react to changes in their after-tax incomes by different relative adjustments to their consumption and savings. While the net effect of those differences on national savings is likely to be small, the effect on individuals may be important, particularly to those individuals who perceive themselves adversely affected by any resulting redistribution of wealth. To the extent the build-up and subsequent drawdown of the Social Security trust funds influence how the burden of taxation is distributed among taxpayers at different income levels, affected individuals may find the trust funds very real indeed.

This view that higher savings leads to more investment is more generally applicable to the long term. In the short term, investment is less directly related to savings. For example, in a recession, with businesses having a pessimistic outlook because of depressed demand, additional savings might have no positive effect on investment and even may have the opposite effect by further depressing aggregate demand. 17 A textbook definition would be the sum of private savings and government surplus, but deficit is simply negative surplus, and governments more often run deficits than surpluses so we rephrase the identity in more familiar terms. 16

ISSUE BRIEF MAY 2012

9


No Net Wealth: Not Unique to the Trust Funds Since the trust funds do not represent net economic wealth to the society, it would be understandable to conclude that, from the perspective of the entire economy, they are not “real.” By the same criteria, however, Treasury securities held by the public would also not be “real” and, moreover, corporate bonds

and other debt securities would equally not be “real”. This is because any debt security represents an asset to one party of the transaction, in this case the federal government, and a liability of exactly offsetting value to the other, that is, the bondholders. The conclusion that trust fund assets do not represent net wealth for society therefore would not change if the trust funds were invested in many private securities.18

Some Common Questions and Answers About the Trust Funds Q: Why have the trust funds grown to such a large size? A: When the trust funds were originally created, they were intended to hold only a small amount of assets – enough to pay benefits for a few months – to act as a buffer against short-term fluctuations in tax receipts and benefit payments. This is called “pay-as-you-go” or “PAYGO” financing. As part of a package of reforms enacted in 1983 to place the system into 75-year actuarial balance (through 2058), Congress set a level tax rate beginning in 1990, even though it was well known that benefit payments would rise steeply when the baby boom generation began retiring. This policy decision is understandable given the political liabilities involved with raising taxes. The accumulation of a large trust fund is an unintended consequence of this policy decision. Q: Have Social Security trust funds been raided and spent on other government programs? A: The trust funds hold Treasury debt securities. This means their excess revenue has been lent to the general fund of the Treasury. The federal government has used the borrowed revenue to finance its expenditures, as any borrower does and, like any borrower, is legally obligated to repay the trust funds when needed for benefit payments. The question of the government’s overall spending is unrelated to the purpose and use of the Social Security trust funds. Q: Would the trust funds be more real if they were invested in private-sector stocks and bonds? A: All securities are financial claims on a share of real resources. Payments in any given year must come from total resources available in the economy that year. This fact does not depend on whether those payments are financed in that year with current taxes, government borrowing, or sale of private securities. The means of financing would make some difference in who bears how much of the cost, but not in what the total cost is.

10

ISSUE BRIEF MAY 2012

Q: Why can’t Social Security save for its future beneficiaries like a family saves for a future expense like a car or college tuition? A: Analogies between individuals or companies on one hand and governments or nations on the other are often of limited use because the differences are too great. In a reasonably stable economy, families or businesses can save and expect to use their savings in the future because they are small relative to the capital markets and the economy. But in a national context—the only way to secure future resources is by investing in productive activities that lead to economic growth. Social Security, of course, is not the whole economy, so it can “save” to the extent of accumulating calls on future revenue from the Treasury (and does so, as evidenced by the accumulated trust funds), but it is a large enough fraction of GDP that both the accumulation and dissipation of its savings have an impact on the rest of the economy. Q: How do the trust funds fit into policy decisions about Social Security’s future? A: The past cannot be undone, so any policy decisions about Social Security’s future must take into account the trust funds as they exist today. Beginning in about 2050, the cost of the system is expected to level off at just under 6 percent of GDP. At that point, the system could theoretically return to PAYGO financing with a tax rate about a third higher than the current tax rate. The transition from now to then could be financed by a combination of drawing down the trust funds while gradually increasing the tax rate. Alternatively, benefits could be reduced so that the current tax rate, in combination with the trust funds, would be sufficient to keep the system solvent indefinitely. A multitude of intermediate solutions involving both tax increases and benefit reductions is available. The ultimate level of Social Security taxes and bene­ fits is a policy decision which will be based on what share of the nation’s resources we as a society are willing to devote to Social Security. In any likely scenario, the impact of the trust funds will be temporary and not critical to this decision.


Even though debt creates both liabilities and assets, it can, of course, lead to wealth creation if it is put to productive use. Capital for productive economic development generally can be increased only through investment based on higher current savings and, hence, lower near-term consumption. With government expenditures, it is not always easy to tell what is consumption and what is investment. For example, road construction and other government-financed infrastructure can increase productive capital of businesses, and education can increase future productivity of the labor force. It is in this respect—with the complexities of evolving technology, economic cycles, international flows of consumption, and investment patterns; political trends; and tax policy—that government policies must balance operating to meet public needs with maximizing long-term sustained economic growth. Whether Social Security contributes to economic growth or not, it can reasonably be argued that it is not an appropriate role for Social Security to function as an instrument of growth. On the other hand, it would be difficult to ignore that economic growth is a crucial factor in the ability of Social Security or, indeed, any retirement program, to provide income to participants after they have stopped working. This is because most goods, and practically all services, can only be consumed when they are produced or shortly thereafter, so current consumption of goods and services cannot deviate greatly from current production for a long period. In the near term, government policy—or even market forces—can only change how the existing pie is sliced. In the longer term, however, policies that encourage a reasonable balance among

consumption, savings, and investment can promote economic growth and a larger pie for the future.

Conclusion Because its income exceeded outgo from 1984 to 2009, Social Security has accumulated much larger trust funds than was usual in its history. In recent years, the nature of those trust funds has become a contentious issue in policy debates, with one side viewing the assets in the trust funds as a tangible store of value and the other describing them as an accounting illusion. When the special Treasury securities held by the Social Security trust funds are redeemed, the U. S. Treasury will need to pay for those securities from the general fund. In those years, the federal government will need to finance those redemptions by borrowing more from the public, spending less or collecting more in taxes. Whether that poses a difficult budgetary dilemma will depend on whether the rest of the federal government is running a sufficient surplus. The existence of the trust funds, per se, is not responsible for this circumstance, as the increase in program cost relative to tax revenue over the next 25 years will be due to the reduction in the ratio of workers to beneficiaries. This downward trend is due to demographic changes within the population, including a declining birth rate and an increasing number of retirees from the baby boom generation. In many instances when debate ensues over whether the Social Security trust funds are “real”, differences can be semantic. The significance of the trust funds depends on the context in which they are viewed. From the viewpoint of Social Security’s ability to pay benefits,

To some extent, the same argument may also apply to equities; for example, merely buying or selling a stock does not create wealth. However, the change in the value of equity over time represents the market’s valuation of the real underlying assets. There is no counterparty whose wealth decreases as a result of equity gaining value. 18

ISSUE BRIEF MAY 2012

11


given the existing design of the program and the legal context in which it operates, the trust fund assets have concrete and easily recognizable consequences. From the viewpoint of total government finances, trust fund assets may be considered irrelevant, although that would require an assumption that scheduled benefits would be paid regardless of the ability of the trust funds to finance the payments. From the viewpoint of the entire economy, trust fund assets represent no net wealth, but the same is true for many other securities, public and private. More importantly, the trust funds are unlikely to have a significant effect on national savings and hence on economic growth. Even so, there are ways in which trust fund dynamics may affect the distribution of income and wealth—so in some contexts the trust funds can also be relevant from the whole economy’s perspective. The simple question, “Are the trust funds

real?” is likely to lead to misunderstanding because its meaning is so dependent on the often unstated context. A better understanding of the trust funds would probably stem from a more productive discussion focused on specific functions and consequences of the trust funds. Even if there were a universal agreement on the nature of the trust funds, differences in matters of policy would remain, since, under current projections, the trust funds provide only a temporary buffer against future increases in program cost, and benefit cuts or tax increases will ultimately become necessary if the system is to remain solvent. Finally, regardless of the financing mechanism, each future year’s benefits ultimately are provided out of that year’s total economic output. A society always has to decide how to allocate resources, including the trade-offs between work and leisure and the effects on both tangible and intangible standard of living.

1850 M Street NW Suite 300 Washington, DC 20036 Tel 202 223 8196 Fax 202 872 1948 www.actuary.org

12

ISSUE BRIEF MAY 2012


Statement of Donald E. Fuerst, MAAA, FSA, FCA, EA Senior Pension Fellow American Academy of Actuaries

To the Committee on Ways and Means Subcommittee on Social Security U.S. House of Representatives Hearing on Proposed adjustments to Social Security benefits, as included in the President’s Fiscal Year 2014 Budget, the report by the National Commission on Fiscal Responsibility and Reform, and the report of the Bipartisan Policy Center’s Debt Reduction Task Force.

May 23, 2013

The American Academy of Actuaries is a 17,000-member professional association whose mission is to serve the public and the U.S. actuarial profession. The Academy assists public policymakers on all levels by providing leadership, objective expertise, and actuarial advice on risk and financial security issues. The Academy also sets qualification, practice, and professionalism standards for actuaries in the United States.

1


Chairman Johnson, Ranking Member Becerra – and distinguished members of the subcommittee. Thank you for the opportunity to appear before you today to assist in your examination of bipartisan proposals to adjust Social Security benefits and their impacts on the program’s finances, beneficiaries, workers, and the economy. I appear before you today on behalf of the American Academy of Actuaries, where I am the Senior Pension Fellow. The Academy is the nonpartisan professional association representing all actuaries in the United States. Our mission is to serve the public and the actuarial profession by providing independent and objective actuarial information, analysis, and education to help in the formation of sound public policy. Background Americans are living longer today than they did in the past, thanks in part to improvements in public-health systems, the quality and quantity of our water and food supply, healthcare, and medical technology. These improvements have led to substantial increases in life expectancy, and in particular, the life expectancy of the elderly. A longer life creates some obvious and numerous benefits for individuals, but brings with it not only personal challenges, but a societal challenge in how to prepare for and manage financial security in retirement. Social Security is a major component of financial security for the elderly. However, the financial sustainability of the program is itself facing challenges. As the Social Security actuaries remind us each year, the program is not in actuarial balance. At some point in the near future —2033 according to the 2012 Trustees Report—absent corrective measures, the program will no longer generate enough revenue to pay full benefits in a timely fashion. Addressing the program’s solvency now would allow Congress to have a fuller range of options to consider, many of which could be more modest in their adjustments, such as slow phase-ins over many years. Deferring efforts to address the solvency of the program to the next decade or beyond will more profoundly affect beneficiaries and the taxpaying public. Social Security’s Old-Age and Survivors Insurance (OASI) challenges stem from our population demographics: Partly from lower birth rates and immigration levels, and in part from Americans living longer. Simply put, the longer someone lives, the more benefits Social Security must pay. In 1940, when the new Social Security Administration began paying monthly retired-worker benefits, the “full retirement age” was 65. At that time, workers who survived to age 65 had a remaining life expectancy of 12.7 years for males and 14.7 years for females. In 2011, life expectancy at age 65 was 18.7 years for males and 20.7 years 1 for females, an increase of six full years for males and females. 2 Actuaries expect the trend to continue into the foreseeable future. Under the Social Security actuaries’ intermediate projection, future life expectancy is projected to increase about one year 1

2012 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds, Table V.A4.—Cohort Life Expectancy. 2 For more information, please view the table located in the appendix.

2


per decade. In 20 years, life expectancy at age 65 for males is expected to be more than 20 years and more than 22 years for females. The Social Security Amendments of 1983 3 The 1983 Social Security Amendments made increases in the full retirement age partially reflect the improvements in life expectancy since 1940. These scheduled increases were part of a package of changes adopted to fend off near-term program insolvency. Under the 1983 adjustments, the full retirement age has gradually increased to age 66 for workers born in 1943 (who reached age 66 in 2009). Still to come as a result of the 1983 amendments, the full retirement age gradually increases to age 67 for workers born in or after 1960. These increases are summarized in the table below. Year of Birth

Current Law Social Security Full Retirement Age

1943—1954 1955 1956 1957 1958 1959 1960 & older

66 66 and 2 months 66 and 4 months 66 and 6 months 66 and 8 months 66 and 10 months 67

From today’s perspective, however, the 1983 schedule of increases in the full retirement age account for only two of the additional six years of life expectancy that we’re experiencing today. Congress should enact additional raises to the retirement age to keep healthy older Americans productively working and to keep Social Security affordable. The 1983 amendments solved what was then an immediate crisis—full benefits could not have been paid later that same year without action. But valuations done at the time of the 1983 amendments projected solvency for 75 years, and did not project sustainable solvency beyond that time frame. Today, the Social Security actuaries’ projections anticipate the crisis arriving sooner than projected in 1983. Congress should act now to provide sustainable solvency for future generations. An important component of a change that will promote sustainable solvency should be raising the full retirement age with a provision for additional raises if life expectancy after full retirement age continues to increase.

3

Public Law 98-21.

3


Raising the Full Retirement Age Social Security defines the full retirement age as the earliest age an individual can receive unreduced old-age benefits. For 60 years, starting in 1940, the full retirement age was 65. Actuaries and demographers project increases in life expectancy to continue, although the rate of increase is the subject of ongoing debate. Regardless, any increase in longevity among the elderly population without a corresponding change in the full retirement age actually constitutes an increase in the amount of benefits paid, and this has a profound effect on the system. First, although the monthly amount a retiree receives remains unchanged under current law formulas, the number of payments a retiree will receive obviously increases with longer life spans. These additional payments lead to a higher value of benefits, increasing the cost of the system. Second, as life expectancy increases, one’s lifetime proportion spent in retirement increases, while the proportion spent working decreases. That means that individuals receive benefits for a greater portion of their lives and pay Social Security taxes for less, even if the number of years they pay into the system remains the same. To address the effect of longevity increases on Social Security, increases in the retirement age could be structured at least three ways: 1) a modest increase to merely restrain the growth in the value of future benefits, 2) an increase designed to keep benefits approximately the same in value, or 3) more significant increases that would reduce overall benefits. It should be noted that the third method goes beyond addressing longevity increases and would actually reduce the lifetime value of benefits. Based on the assumptions in the trustees report, the full retirement age would need to increase by about one month every two years in order to offset the effects of increasing life spans on the system. The Actuarial and Public Policy Case for Raising the Full Retirement Age The shifting balance between working years and retirement years has contributed to the system’s long-term actuarial imbalance. As actuaries, we see this as a demographic problem that demands a demographic solution. That said, the Academy does recognize that an increase in the retirement age is not a solution to address the entire imbalance in the system. It is but one component, though a necessary one, of restoring Social Security’s long-term financial health. In particular, raising the full retirement age will: • Compensate for Increases in Longevity – Raising the full retirement age connects Social Security directly to covered workers living longer. Even large proposed increases in the full retirement age would still provide for a lengthier retirement on average than workers enjoyed during most of the time their full retirement age was 65. Current retirement surveys consistently show that, in order to maintain one’s current standard of living, most workers expect they will need to or want to work longer to accumulate the needed retirement savings in employer-sponsored plans and personal savings accounts. 4


Preserve the current benefit formula. The current benefit formula has existed for more than 30 years and provides a certain balance to the circumstances of the many demographic groups among Social Security’s covered population. Raising the retirement age addresses the financial challenges associated with longevity without having to modify the current formula.

Increase labor force participation. Raising the full retirement age can induce workers to remain in the labor force longer. Part of this effect is behavioral: to the extent Americans consider a higher full retirement age the “normal” retirement age, workers will view delaying retirement as the “new normal.” More significantly, workers could continue full-time employment in order to retire later with an adequate Social Security benefit or switch to part-time employment to supplement a lower benefit. Making greater use of older workers increases total economic output and raises the living standard for both active and retired workers.

Preserve disability benefits. Raising the full retirement age will not reduce disabledworkers’ benefits, while a direct reduction in Social Security’s benefit formula would reduce these benefits.

Considerations When Raising the Retirement Age Possible negative consequences of raising the full retirement age include: • Disproportionate effect on low-wage workers. Not all demographic groups will experience the same increases in longevity. Lower-wage workers and those with lower levels of education generally have experienced smaller increases in longevity compared to more highly compensated and more educated workers. Also, lower-wage workers often may not be able to work to a higher full retirement age because they tend to work in more physically demanding jobs. An increase in the full retirement age can adversely affect these workers who may have a need for claiming benefits early. Notably, lowerwage workers also rely most heavily on Social Security for income, sometimes as their only income source. Policy options that might address these problems include liberalizing the current “vocational factors” that are used to define disability for workers at age 40 or older, or revising the early retirement factors to provide a lesser reduction on benefits below the first bend point. 4 •

Jobs may not be available. Barriers exist to keeping elderly workers in the labor force, especially if younger workers are readily available. If the economy cannot provide jobs for older workers, raising the retirement age will constitute a financial hardship for many

4

Social Security’s bend points, where the factors in the formula change, are dollar amounts indexed over time by increases in average wages. The 2013 bend points, for example, are $791 and $4,768.

5


who may have to retire with reduced benefits. The seriousness of this concern depends on how the labor market responds over the coming decades to the gradual aging of the population. Options for Increasing the Full Retirement Age While the American Academy of Actuaries advocates for inclusion of an increase in full retirement age in efforts to restore Social Security’s long-term actuarial balance, it has not endorsed any one proposal as there are several approaches that can make this adjustment. Increasing the retirement age can contribute significantly to stemming the impact of the program’s inadequate financing as a result of the demographic trend of increased longevity and help put the program back on track toward actuarial balance. Some approaches for increasing the Social Security full retirement age include the following: •

Fixed-schedule increases to full retirement age. Various methods exist to increase the retirement age. The Social Security Administration’s Chief Actuary Steve Goss and his staff have developed eight examples, including the most accelerated – beginning the increase in the full retirement age from age 66 to age 67 immediately, followed by increases by one month (in retirement age) every two years (in birth-date years) until the full retirement age reaches age 70. This would reduce the long-range actuarial deficit by about one-third. Further reductions of the long-range deficit would require a rate of increase more rapid than one month every two years.

Index based on years in retirement. This method would index the full retirement age in a manner that keeps life expectancy at the full retirement age constant over time. For example, life expectancy at age 65 (weighted between males and females and rounded to the nearest whole year) is now nearly 20 years. To keep a 20-year life expectancy at full retirement age, the full retirement age would have to increase about one month every year or two for life based on the expected increases in longevity from the trustees report. This method would decrease system costs over time because the payout period for benefits would remain the same, while the period over which payroll taxes increases, although the savings from this change alone would not restore actuarial balance. Another, perhaps more accurate, method would index retirement age based on demographic trends as they develop, given that experts disagree on the long-term rate of mortality improvement.

Index based on ratio of retirement to working years. This indexing would change the period from workforce entry age to the full retirement age to increase at the same rate as life expectancy at the full retirement age. This method, which was recommended in 1983 by a majority of the members of the National Commission on Social Security Reform, would increase the full retirement age a little more slowly than maintaining a constant life expectancy at full retirement age and, therefore, would reduce program costs to a lesser degree. But by using this method, policymakers may intend that some portion of the increase in life expectancy at full retirement age may reflect years of unhealthy life during which workers could not continue working and that extra years of life expectancy 6


should be split in some manner between work and retirement. •

Index to maintain actuarial balance. If actuarial balance of the OASI Trust Fund occurred through reforms, Congress could put in place a mechanism that would automatically adjust the full retirement age as necessary to restore actuarial balance going forward. An adjustment of this nature also could be combined with automatic adjustments to the payroll-tax rate or benefit amounts to maintain actuarial balance. The Academy’s issue brief, Automatic Adjustments to Maintain Social Security’s Long-Range Actuarial Balance 5, discusses this topic in great detail.

Similar Modifications The National Commission on Fiscal Responsibility and Reform (Bowles-Simpson/SimpsonBowles) report includes a proposal to index both the full retirement age and the early retirement age to increases in longevity after 2027, the first year workers receive unreduced benefits at age 67 under current law. The proposal also contains a directive to the Social Security Administration to create a “hardship exemption” for workers who cannot continue working past age 62 but who do not qualify for disability benefits. It is estimated the proposal would reduce the long-range actuarial deficit by 18 percent. The proposal would not fully reflect the improvement in longevity since 1940. This could be accomplished by ad hoc increases to the full retirement age instead of or in addition to indexed increases. However, this could be done by increasing in accelerated fashion the full retirement age enough to eliminate Social Security’s long-range actuarial deficit entirely. This still would not take into account the full extent of improvements in longevity, but few major proposals eliminate more than approximately one-third of the deficit through adjustment to the full retirement age. This is due in large measure to concern for workers in strenuous jobs, who might not be able to continue working beyond the current full retirement age. The Bipartisan Policy Center Debt Reduction Task Force (Domenici-Rivlin) proposed a way to adjust benefits for longevity by decreasing the 90 percent, 32 percent and 15 percent factors used in calculating the Primary Insurance Amount as people live longer. The factors would be multiplied by the ratio of life expectancy of someone reaching age 67 in 2018 to the life expectancy of someone reaching age 67 in the fourth year before benefit eligibility. The Task Force proposal also would apply to disabled workers at the time of conversion to disabled worker status, with the ratio only applying to the proportion of the benefit earned while not disabled. The Task Force’s method could be designed to produce essentially the same effect as any particular change to the full retirement age and thus may have a similar financial benefit. However, it does not provide similar signaling to American workers as an increased retirement age. An increased retirement age signals American workers that they can and should continue to 5

http://www.actuary.org/files/publications/IB_Auto_Adjustments_FINAL.pdf

7


participate in the workforce, thereby having greater opportunity to save more for retirement and earn a greater Social Security benefit. Mitigating the Effect of a Higher Retirement Age There are ways to lessen the impact on working Americans and certain segments of the workforce who could be particularly affected by any increase in the Social Security retirement age: •

Gradually phase in any change over an extended period of years, even decades, to allow for more time for society to adapt to the new work-life reality.

Modify the progressivity of the benefit formula in conjunction with retirement age changes or introduce progressivity into the early retirement reduction factors. This may address the disparate distribution of longevity gains across the populationwith wealthier socioeconomic groups recently showing more longevity improvements than poorer socioeconomic groups.

Additional occupational bridge pensions, perhaps combined with revisions to existing disability programs, could address people having difficulty in continuing to work in occupations that involve physical labor.

Measures that facilitate employment at older ages (such as reductions to the payroll tax at older ages) to address the greater difficulties that older workers sometimes face in finding jobs.

Early Eligibility Age The Early Eligibility Age for Social Security has been age 62 since 1956 for women and 1961 for men. The 1983 amendments that raised the full retirement age did not change the early eligibility age, but the amount of benefit at the early eligibility age was further reduced for those workers with a full retirement age above 65. Early benefits are reduced 30 percent at age 62; that is 6.7 percent per year for the first three years and 5 percent for the remaining years to full retirement age (for persons born in 1960 or later). Some proposals to change the full retirement age also suggest changing the early eligibility age, for example to keep the earliest eligibility age five years before the full retirement age. From a financial as well as societal perspective, raising the early eligibility age has the beneficial result of encouraging most individuals to work longer. Individuals who remain in the labor force are a productive part of our economy and are able to save more for their deferred retirement. Raising the early eligibility age also helps prevent payment of benefits that may prove inadequate. The 1983 amendments increased the maximum number of years one can retire early and draw early benefits from three to five years prior to full retirement age, and increased the reduction in the benefit from 20 percent at three years to 30 percent at five years. Increasing the 8


full retirement age beyond age 67 without also raising the early eligibility age would mean that benefits at the early eligibility age would be reduced more than the current 30 percent. This reduction applies to the benefits over the full lifetime of the individual and may as a result prove to be inadequate, potentially causing pressure for benefit increases. Raising the early eligibility age in concert with the full retirement age would maintain the maximum reduction at 30 percent. Raising the early eligibility age does not significantly change Social Security’s financial position because early retirement benefits are already reduced to the approximate actuarial equivalent payments. Partial Claiming of Benefits at Age 62 Another proposal that mitigates the effect of higher early eligibility age is to allow individuals to claim a portion of their benefit, say up to 50 percent, at age 62 or later, even if before early eligibility age. This portion of the Primary Insurance Amount would be reduced for early commencement accordingly. This proposal would provide partial income to individuals who have difficulty working but do not qualify for disability benefits. Again, because early retirement benefits are approximate actuarial reductions, there would be relatively small effects on the financial status of Social Security. Currently, the age 62 early retirement age is the single most popular age for electing receipt of benefits. Despite the larger benefits that are available at later ages, many people continue to elect to receive benefits at their first opportunity. Retaining age 62 as the early retirement age or allowing partial benefits at age 62 is likely to result in many people selecting commencement at this early age with a smaller lifetime benefit. This may result in lifetime benefits that prove to be inadequate. Studies indicate the proportion of jobs that are physically demanding has shrunk to less than 8%, and less than 20 percent of workers who retire early do so for health reasons 6. Perhaps programs other than Social Security’s OASI could be designed to provide adequate lifetime retirement income for the percentage of workers needing such assistance. Summary In closing, I again thank the subcommittee for the opportunity to present some ideas on behalf of the American Academy of Actuaries on these important issues facing Social Security. The increased longevity of Americans is a benefit to us in many ways, but it also increases the cost of financing a secure retirement. Delaying action to fix Social Security can only reduce the options available to us. By dealing with this issue sooner rather than later, we can ensure that the system that has benefited several past generations will continue to provide retirement security for generations to come. Thank you. I would be happy to answer any questions you might have at the appropriate time. 6

Munnell and Sass, “Working Longer,” and Lakdawalla, Bhattacharya and Goldman, “Are The Young Becoming More Disabled.”

9


Appendix Life Expectancy at Age 65

Year (age 65) 1940 1950 1960 1970 1980 1990 2000 2010 2011 2035 2060 2085

Male 12.7 13.1 13.2 13.8 14.7 16.0 17.5 18.6 18.7 20.3 21.7 22.9

Female 14.7 16.2 17.4 18.5 18.8 19.3 20.0 20.7 20.7 22.3 23.4 24.6

Source: 2012 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds http://www.ssa.gov/oact/tr/2012/tr2012.pdf

10


American Academy of Actuaries

MARCH 2009

MAY 2012

Key Points Since the 1980s, the Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds consistently has indicated that, in the absence of corrective legislation, assets currently in the trust funds plus future payroll tax income will not be sufficient to finance all scheduled benefits over the 75-year valuation period. n The trustees are not the only ones making projections about Social Security’s future. Within the federal government, the Congressional Budget Office makes its own projections. Outside experts from think tanks and academia also weigh in with their opinions. n All of these projections rely on assumptions about future demographic and economic trends because the future cannot be known with any certainty. The selection of assumptions affects the results of any projection and, hence, the policy prescriptions of anyone relying on such a projection. The trustees report describes in detail the assumptions used by the trustees and the rationale behind these assumptions. n It is important that any report about Social Security’s future include a description of the assumptions used in the calculations. And that anyone citing the report understands how differences in assumptions affect the results. Facilitating such an understanding is the purpose of this issue brief. n

Understanding the Assumptions Used to Evaluate Social Security’s Financial Condition

E

very recent annual report from Social Security’s Board of Trustees projects that, under the board’s intermediate (best-estimate) assumptions and in the absence of corrective legislation, assets currently in the trust funds plus future payroll tax income will not be sufficient to finance all scheduled benefits over the 75-year valuation period. The trustees report uses long-term financial projections the results of which depend on assumptions adopted by the board. In addition, Social Security reform proposals introduced in Congress or developed by outside experts sometimes are evaluated for their potential effect on the program’s financial condition using the same or similar projection methods and assumptions. The Congressional Budget Office (CBO) makes its own projections of Social Security’s financial condition. CBO uses the demographic projections produced by Social Security’s actuaries, but applies its own economic assumptions. CBO projections have yielded a long-range deficit somewhat smaller than the long-range deficit that results from the trustees’ intermediate assumptions. Experts outside the government have also performed independent analyses of various reform proposals. These experts also use assumptions in their projections of Social Security’s financial future, which may differ from those used by the trustees. Because small changes in assumptions can have large effects on cost estimates over long periods, even when the assumptions used in these

The American Academy of Actuaries is a 17,000-member professional association whose mission is to serve the public and the U.S. actuarial profession. The Academy assists public policymakers on all levels by providing leadership, objective expertise, and actuarial advice on risk and financial security issues. The Academy also sets qualification, practice, and professionalism standards for actuaries in the United States. ©2012 The American Academy of Actuaries. All Rights Reserved.

1850 M Street NW, Suite 300, Washington, DC 20036 Tel 202 223 8196, Fax 202 872 1948 www.actuary.org Mary Downs, Executive Director Mark Cohen, Director of Communications Craig Hanna, Director of Public Policy Don Fuerst, Senior Pension Fellow Jessica Thomas, Senior Pension Policy Analyst


analyses appear to closely match those used by government actuaries, it is possible to skew the results, intentionally or unintentionally, to favor one proposal over another. The nature and extent of any changes designed to resolve the program’s financial shortfall depend, of course, on the magnitude of the problem. Although the projection based on the trustees’ intermediate assumptions generally is quoted when discussing Social Security reform proposals, the range of alternative assumptions used by the trustees illustrates the considerable uncertainty about the future. This issue brief describes the major assumptions used in projections of Social Security’s financial condition and how variations in the assumptions affect the results. This issue brief also encourages policy advocates to disclose the assumptions underlying their reform proposals and to apply assumptions consistently.

Background Since Social Security’s earliest days, its Board of Trustees has reported annually to Congress on the projected long-range financial status of the system. The trustees base their projections on actuarial assumptions. The actuaries at the Social Security Administration make initial recommendations for these assumptions, which then are modified as deemed necessary by the trustees and their staffs. The final assumptions selected by the trustees are subject

to review by the chief actuary of the Social Security Administration, whose Statement of Actuarial Opinion in the report includes an opinion as to whether the assumptions are reasonable. Based on these assumptions, the actuarial staff of the Social Security Administration prepares the projections that are presented by the trustees. The trustees evaluate the program over a 75-year long-range projection period to view the adequacy of financing over the lifetime of virtually all current program participants. The actuaries typically use year-by-year assumptions about a number of critical economic and demographic parameters for the first 25 years of the projection period and then apply “ultimate” rates over the remainder of the 75-year period. The trustees report describes in detail the assumptions and methods used. Each year, the Social Security program gains another year of actual experience that can affect the projections in two ways. First, everything else being equal, if experience is more favorable than projected, the system’s financial forecast improves, and, if less favorable, the forecast worsens. Second, emerging experience constitutes additional evidence that can be used for setting assumptions. For example, if mortality improves more rapidly than expected, then future mortality expectations might be adjusted to reflect that trend. The normal process provides for monitoring experience to detect any differences between actual experience and past projections and for fine-tuning assumptions based on the results

Members of the Social Security Committee who participated in revising this issue brief include: Robert Alps, MAAA, ASA; Eric Atwater, MAAA, FSA, FCA, EA; Janet Barr, MAAA, ASA, EA - chairperson; Raymond Berry, MAAA, ASA, EA; Michael Callahan, MAAA, EA, FSPA; Eric Klieber, MAAA, FSA, EA - vice chairperson; Timothy Leier, MAAA, FSA, EA; Timothy Marnell, MAAA, ASA, EA; John Nylander, MAAA, FSA; Brendan O’Farrell, MAAA, EA, FSPA, FCA; Steven Rubenstein, MAAA, ASA; Bruce Schobel, MAAA, FSA, FCA; Mark Shemtob, MAAA, ASA, EA; P.J. Eric Stallard, MAAA, ASA, FCA; Ali Zaker-Shahrak, MAAA, FSA 2

ISSUE BRIEF MAY 2012


of this analysis. The actuaries and trustees must use their own judgment about the reliability of the past for projecting the future. When a change occurs in some demographic or economic factor, no one can determine immediately whether the change represents a short-term fluctuation or a long-term trend, just as no one can know if a week without rain is the beginning of a drought. For this reason, changes in assumptions generally lag behind changes in the underlying demographic and economic experience. Every four years since 1999, the Social Security Advisory Board has appointed a technical panel composed of leading economists, demographers, and actuaries from outside the Social Security Administration to review the trustees’ assumptions. The technical panel provides independent analysis of the trends affecting Social Security’s finances. In the past, these panels have concluded that the trustees’ assumptions are reasonable. The technical panels, however, frequently recommend specific changes to the assumptions. The trustees weigh these recommendations carefully and often make changes to their assumptions along the lines of these recommendations— although they sometimes choose not to follow some of the recommendations. In the end, the trustees have the final say regarding the assumptions. The trustees report presents three projections: intermediate, low-cost and high-cost. The intermediate, or “best-estimate,” projection is the one usually cited by policymakers and the news media. The low-cost and highcost projections show how the results of the projection would change under alternative sets of assumptions. Although these alternative assumption sets differ substantially from the best estimate assumptions, the trustees believe

they represent reasonable possible scenarios for a future either more or less favorable to Social Security’s finances than that predicted by the best estimate assumptions. The trustees report also includes sensitivity analyses that show how the results of the projection would change if each major assumption is changed one by one to its value under the low-cost or high-cost assumption set while the other assumptions remain at their intermediate-cost values. Finally, the trustees report includes an analysis of the results from a stochastic model of the system. In this analysis (as described in the Academy’s 2005 issue brief, A Guide to the Use of Stochastic Models in Analyzing Social Security), the projection is run multiple times under different sets of assumptions and the results analyzed statistically to draw conclusions about the probabilities that actual longterm system performance will lie in different ranges. In addition to the projection for the trustees report, the Office of the Chief Actuary (OCACT) regularly provides analyses of legislative proposals for changing Social Security submitted by members of Congress and, sometimes, by experts outside the government. To the extent possible, these analyses use the same assumptions as the most recent trustees report. When a proposal requires introduction of an assumption not required for the trustees report, that assumption is chosen by OCACT consistent with the best estimate assumptions. For example, proposals that involve investing some or all of the trust fund assets in private sector securities require adding an assumption regarding the future investment returns from such securities. This issue brief describes the assumptions that must be made in any actuarial projection of Social Security’s finances and explains how ISSUE BRIEF MAY 2012

3


different assumptions affect the projections. Its purpose is not to describe the specific assumptions used in the trustees report. After publication of the trustees report each year, the Social Security Committee of the American Academy of Actuaries updates its issue brief, An Actuarial Perspective on the Social Security Trustees Report, which describes the specific assumptions the trustees used in their most recent report and any major changes since the previous report.

Assumptions The assumptions used for Social Security’s financial projections fall into two broad categories—demographic and economic. Demographic assumptions are used to project the future population of Social Security participants and provide a basis for estimating the number of workers paying into the system, the number of retired- and disabled-worker beneficiaries, and the number of family members and survivors receiving benefits. Economic assumptions are used to project wages and the resulting taxes paid into the program, benefit payments, and the investment income on the system’s accumulated assets. Together, these factors are used to calculate the system’s projected annual income and expenses. Although the assumptions are described one by one, they are not independent of each other. Factors underlying the various economic assumptions tend to move together as the economy experiences short-term cyclical ups and downs and longer-term trends. For example, real wage growth, interest rates, and labor force participation rates all tend to be higher and unemployment rates lower when

the economy is vigorous. Factors underlying many of the demographic assumptions also respond to changes in the economy. For example, birth rates and immigration rates tend to be higher and disability rates lower when the economy is vigorous. In all these examples, the effect is the opposite when the economy falls into recession. The trustees take these relationships into account when setting the intermediate assumptions. When setting the lowcost and high-cost assumptions, however, the assumptions that yield the lowest and highest costs are grouped together even though the resulting combinations may not yield a plausible scenario.1

Major Demographic Assumptions FERTILITY: As workers retire, they are replaced by new entrants into the labor force, most of whom were born in this country. The fertility rate, or average number of children born to a woman during her lifetime (if she survives the child-bearing years), is the primary determinant of whether the number of new workers will be sufficient to pay for the benefits promised older workers, assuming currentlaw tax rates. A higher fertility rate increases the number of workers coming into the system, improving overall finances. The fertility rate fell from 3.7 in 1957 to an all-time low of 1.74 during the mid-1970s, but has increased somewhat since then to slightly above 2.0. Recent trustees reports project the fertility rate will stabilize near this level under the intermediate assumptions. When the fertility rate is adjusted to exclude children who do not survive to age 10, and who therefore never participate in Social Security, the rate stays generally constant at

There is one exception to this rule: The inflation assumption is higher in the high-cost estimate and lower in the low-cost estimate, although higher inflation improves the actuarial balance. 1

4

ISSUE BRIEF MAY 2012


approximately 3.0 from the early 20th century up to the 1960s, except for a period of low fertility during the depression and World War II and a period of high fertility during the baby boom of the late 1940s and 1950s. With improvements in health care, sanitation, and nutrition, the adjusted fertility rate today is only slightly lower than the unadjusted rate. The decline in the adjusted fertility rate from 3.0 to 2.0, which occurred remarkably quickly during the 1960s and 1970s, is one of the principal factors underlying the expected increase in benefit payments as a percentage of gross domestic product (GDP) from the historical range of 4.0 to 4.5 percent to around 6.0 percent by the middle of the 21st century. IMMIGRATION: Immigration also accounts for some new entrants into the labor force. Indeed, if the fertility rate remains at or below the replacement level (approximately 2.1 births per woman), then any long-term population growth must come from net immigration (i.e., immigration less emigration). Most immigrants are young and have all or most of their working lifetimes ahead of them when they enter the country, while emigrants are more likely to be in the older part of the age spectrum. As a result, a higher net immigration rate, like a higher fertility rate, tends to improve overall system finances. Social Security projections take into account both legal and other than legal immigration. (The latter includes those who entered the country legally but overstayed or otherwise violated the terms of their visas.) Legal immigration has increased substantially since World War II, driven primarily by legislative increases in immigration quotas. Under the intermediate assumptions, net legal immigration levels off at approximately the current rate. Rates of net other than legal immigration

are subject to much uncertainty. A decline in the number of individuals apprehended attempting to cross into the United States illegally as well as anecdotal evidence indicates that the rate of other-than-legal immigration declined during the recent recession. In recent reports, the trustees expect a return to the prerecession level in the immediate future followed by a long-term gradual decline. Before the 2008 report, the actuarial projection took into account individuals other than legal permanent residents only on a net basis, so that the assumed age profile of immigrants and emigrants was effectively the same. Beginning in 2008, the trustees have made separate assumptions for other-than-legal immigration and emigration, with a younger age profile for immigrants. This was the major factor in the reduction of the projected long-range actuarial deficit in that year from 1.95 percent to 1.70 percent of taxable payroll. MORTALITY: The mortality assumptions are perhaps the most publicly debated of the demographic assumptions. The mortality assumptions are used to estimate, among other things, how long retired and disabled workers and their survivors are projected to receive benefits. Mortality assumptions also determine how many workers are expected to die before retirement, often resulting in payments to survivors. Although reductions in pre-retirement mortality reduce the cost of survivor benefits, they also increase the number of workers who will reach retirement age. Reductions in post-retirement mortality result in longer lifetimes for those receiving benefits and generally have a much greater impact on the total cost of benefits. Increases in longevity accelerated greatly in the 1970s, leading the trustees to update frequently the mortality assumptions used for Social SecuriISSUE BRIEF MAY 2012

5


ty projections. Since 1982, however, longevity has increased more slowly, and the projected reduction in mortality rates has changed less dramatically than in the past. The rate at which longevity will continue to increase is the subject of much debate. There certainly is potential for more rapid decrease in mortality based on medical advances that slow disease development or allow better management of chronic conditions, such as heart disease, cancer, and stroke. But it is also difficult, if not impossible, to anticipate new diseases that may surface in the coming decades, the effect of lifestyle changes ( e.g., less smoking but more obesity), how rapidly medical breakthroughs will be accessible to the general population, and whether new treatments will be affordable. There is widespread agreement that death rates will continue to decline in the future—the issue is the pace at which these declines will occur. DISABILITY: The disability-incidence assumption is the most important determinant of the projected cost of the disability insurance (DI) portion of Social Security. Social Security law provides objective criteria for determining when covered workers become eligible for disability benefits, although some degree of subjectivity is inevitable in applying the law. Partly for this reason, disability-incidence rates tend to be cyclical, depending on the health of the economy and, to some extent, political and social attitudes toward disability. The trustees set the disability incidence assumption initially by looking at past trends and making projections about the future without regard to the increases in the Social Security normal retirement age (SSNRA) or the age at which workers can receive unreduced benefits scheduled under present law. These rates then are adjusted upward to reflect the additional workers who are expected to file for disability 6

ISSUE BRIEF MAY 2012

benefits because of the scheduled increases in the SSNRA.

Major Economic Assumptions WAGE INCREASES: The nominal (i.e., without adjustment for inflation) increase in wages earned by workers from year-to-year affects both the revenue received and benefits paid by Social Security. As wages increase, taxes on those wages go up proportionately, raising revenue immediately. The formula for determining initial benefits is indexed to wage increases, however, so higher wages gradually result in higher benefits. Wage increases are made possible by increases in worker productivity. Productivity is defined as the ratio of real GDP to hours worked by all workers. Since production is the ultimate source of workers’ compensation, it should not be surprising that increases in productivity give rise to higher compensation. Wage increases, however, do not exactly track increases in productivity due to the following factors: n Change in Average Hours Worked: Over the past 40 years, the average annual hours worked has declined at an average rate of 0.3 percent per year, partly because the work force has included an increasing proportion of women and older workers, who work fewer hours on average. This trend has offset some of the effect of improvements in productivity on workers’ compensation. The trustees assume the average hours worked will level off at approximately the current rate for the indefinite future. This reflects their assessment that factors underlying the past trend will not continue into the future. n

Wages as a Percent of Compensation: Social Security benefits are based only on cash compensation, i.e., wages and self-


employment income. From 1969 to 2009, the portion of total compensation paid to employees as wages declined on average 0.2 percent per year, due largely to increases in the cost of employer-provided health insurance. This trend further offsets the effect of productivity improvements on annual wage increases. Before 2010, the trustees expected the 0.2 percent per year trend to continue. Due to the passage of the Patient Protection and Affordable Care Act of 2010 (PPACA), however, the trustees now expect growth in the cost of employer-provided health insurance to moderate somewhat, and have changed the assumed rate of decline in earnings as a percent of compensation from 0.2 percent to 0.1 percent per year once the PPACA becomes fully effective. n GDP

Price Index: The nominal value of work-

er production also increases due to inflation, which is measured by the price index for gross domestic purchases (also known as the GDP deflator). This is different from price inflation measured by the consumer price index (CPI), because it applies to goods produced in the United States, while the CPI applies to goods consumed in the United States, including imports but excluding exports. There are other technical reasons why the two indices differ. The GDP deflator generally is a few tenths of a percent less than the CPI. CONSUMER PRICE INDEX: Legislation enact-

ed in 19732 provides for annual cost-of-living adjustments (COLAs) in Social Security benefits. These benefit adjustments are intended to keep pace with inflation. COLAs are calculated based on increases in the CPI for urban wage earners and clerical workers (CPI-W)3, which are calculated on a monthly basis by the Bureau of Labor Statistics. A COLA effective for December (applicable to the following January benefit payment) of a given year is equal to the percentage increase (if any) in the average CPI-W for the third quarter (July, August, and September) of that year over the average CPI-W for the third quarter of the last year in which a COLA was effective. If there is an increase, it is rounded to the nearest tenth of one percent. If there is no increase, or if the rounded increase is zero, there is no COLA.4 The assumed annual increase in the CPI affects projected future benefit payments. Since 1975, when automatic adjustment of benefits began, the annual rate of increase in the CPI has varied widely, from double digits in 1979– 1981 to 0.1 percent in 2008. INCREASES IN REAL WAGES: The increase in nominal wages minus the increase in the CPI is called the real-wage differential—the increase in the buying power of wages after adjustment for price increases. If wages were used for indexing benefits after commencement, as well as for calculating initial benefits, then the increases in revenue and benefits resulting from real-wage increases would offset each other. But because benefits after eligibil-

Public Law 93-66 enacted in 1973 provides for cost-of-living adjustments, or COLAs, determined annually. Effective in 1983, the increases were determined each December. 3 The CPI-W is strictly an index and no single monthly amount is of any value. It is the changes in these index values over time that is used to determine COLAs. 4 For example, prior to 2011, the last year in which a COLA became effective was 2008. The average CPI-W for the third quarter of 2008 was 215.495, which is used as the base from which the increase (if any) in the average CPI-W effective December 2011 is measured. For the third quarter of 2011, the average CPI-W was 223.233. Because this average exceeds 215.495 by 3.6 percent (rounded to the nearest one-tenth of 1 percent), the COLA effective for December 2011 was 3.6 percent. Benefits were increased effective January 2012 by this percentage. 2

ISSUE BRIEF MAY 2012

7


ity are indexed to the CPI, any excess of wage increases over CPI increases causes the program’s cost to be lower than would be the case if benefits after eligibility rose at the same rate as wages. The average future rate of increase in real wages is one of the most important factors affecting the financial health of Social Security. INTEREST RATES: Social Security’s assets are invested in special-issue Treasury securities, the interest rates of which are pegged to the rates on securities issued to the public. The interest-rate assumption approximates the yields on intermediate-term Treasury securities. Interest rates affect Social Security in two ways. First, higher interest rates raise the return on the system’s accumulated assets and thus improve the financial condition of the program; lower rates have opposite effect. Second, higher interest rates reduce the present value of the program’s long-term actuarial deficit. Real interest rates (i.e., nominal interest rates less inflation) have varied widely over the past several decades. In the mid-1980s, the real interest rate rose to 9 percent. It has declined since then and has been particularly low since the 2008-2009 recession. But over the longer periods, it generally has averaged around 3 percent. LABOR FORCE PARTICIPATION RATES: Labor force participation rates measure the proportion of the working-age population that is employed, self-employed, or looking for paid work. The labor force includes workers with earnings covered by Social Security, those not in covered employment, and the unemployed. Everything else being equal, a higher labor force participation rate improves the program’s financial condition for two reasons. First, it increases tax revenue earlier than it increases the resulting benefits, which improves the actuarial balance due to the time value of 8

ISSUE BRIEF MAY 2012

money. Second, it increases tax revenue more than it increases benefits, primarily because the proportion of two-earner married couples increases, and the additional payroll tax paid by the lower earning spouse provides additional benefits only to the extent that worker benefits based on that spouse’s own wage record exceed spouse benefits based on the higher-earning spouse’s wage record. An important consideration for Social Security is labor force participation rates at ages when old age benefits are payable, i.e., beginning at age 62. When workers leave the labor force at these ages, they generally are considered to have retired. Participation in the labor force among potential workers at these ages therefore varies according to patterns of retirement—earlier retirement leads to lower participation rates. Labor force participation rates at ages 60 through 64 have changed considerably for both men and women. Before 1985, the labor force participation rate for men at ages 60 through 64 had been decreasing dramatically, from more than 80 percent in 1962 to 56 percent in 1985. The rate then leveled off for a period before beginning a slow increase, due in large part to improved health and the need to work longer to save for a longer period of retirement. The pattern for women has been steadily increasing labor force participation rates at all ages since the early 20th century, with particularly dramatic increases from the late 1960s until about 1980. Since then, the rates for women have leveled off at rates somewhat lower than for men. Increased labor force participation rates among older women reflect this long-term trend. The trustees have concluded that the incentives for remaining longer in the labor force are permanent and, as a result, have increased the assumed labor force participation rates at middle and higher ages in recent reports.


Possible changes in labor force participation rates in response to demographic changes predicted for the next several decades are among the greatest uncertainties in projecting the future financial condition of Social Security. With expected slower growth in the population at traditional working ages, will older workers want to work longer? And will their employers want to maintain an older workforce? UNEMPLOYMENT: The unemployment rate measures the proportion of workers in the labor force unable to find work. Higher rates of unemployment reduce projected future income. Unemployment also generally reduces benefits, but the effect is much smaller and is largely deferred. High unemployment therefore adversely affects the program’s financial health. But unemployment does not have as significant an impact on system finances as do some of the other factors discussed here. The spike in the unemployment rate due to the recession that began in 2008 caused benefit payments to overtake payroll tax income about five years earlier than predicted before the recession hit—but the spike in the unemployment rate did not have a large effect on the system’s long-range finances. GDP GROWTH: The trustees do not directly make an assumption regarding the growth of GDP, which is the total dollar value of all goods and services produced in the United States. The trustees indirectly arrive at their estimate of GDP growth by estimating growth in the labor force and growth in productivity (which is closely related to growth in real wages), both of which are discussed above. GDP growth was high in the 1960s and 1970s, due primarily to the large increases in the labor force. But if the retirement of the baby boomers leads to a shortage of workers, the labor force component of GDP growth could decrease dramati-

cally. If the labor force growth rate was to slow and productivity did not rise to compensate, GDP growth would decline significantly. Long-range GDP growth will depend on a variety of factors, such as whether workers retire at a different rate than projected, whether future workers will be more or less productive than assumed, and whether a shortage of workers will lead to a change in immigration law. At present, a wide divergence of views exists on these questions. Taken together, these assumptions underlie the projections of the program’s short-term and long-term financial condition. These projections provide policymakers with an indication of whether reform is needed.

Social Security Reform and the Stock-Yield Assumption Some Social Security reform proposals would invest all or a portion of the assets accumulated to fund future benefits in private-sector securities, particularly stocks. Some of these proposals would allow workers to set up individual accounts; others would continue the current arrangement in which the government directly invests all of the system’s accumulated assets. Advocates for these reform plans assert that investing payroll taxes in common stocks would provide a better return than the special U.S. government securities used by the current program. This claim is based on historical data showing that stocks have consistently outperformed U.S. government interest-bearing securities over long periods—20 years, for example. Although the annual real yield on stocks is not an assumption used in the annual report, such an assumption must be made to evaluate any reform proposals involving stock investments. It is not surprising that the higher the assumed real yield on stocks, the more ISSUE BRIEF MAY 2012

9


proposals for investing Social Security assets in stocks appear to be favorable. In its formal analyses of legislative proposals that include investment of trust fund assets in private sector securities, OCACT chooses yield assumptions that are consistent with the best-estimate assumptions from the most recent trustees report. Many economists question whether the past superior long-term performance of stocks over other investment alternatives will continue. In addition, recent volatility in the securities markets has focused investors’ attention on the greater risks inherent in equity investments. These issues are explored in depth in the 2007 Academy issue brief Investing Social Security Assets in the Securities Markets. Given the high degree of uncertainty regarding the future performance of the securities markets, it is important when evaluating any reform proposal that changes the way Social Security assets are invested to use a range of possible yields to illustrate this uncertainty.

than he or she pays into the system. It is not surprising that, at the current payroll-tax rate, the OASDI program cannot sustain itself in this situation. It seems unreasonable, however, to argue that workers will not extend their working years longer than currently projected, based on extended years of ability to work and the need to save more (beyond Social Security benefits) for the lengthened period of retirement. Mortality improvement by itself has a major impact on Social Security’s projected financial status and presents great difficulties when making long-range projections. The controversy surrounding the assumed rate of mortality improvement in the 75-year projection already has been described. Given these sharp disagreements among experts over projecting mortality for 75 years, the futility of reliably projecting mortality over an infinite time horizon becomes apparent.

Assumptions over an Infinite Time Horizon

As Yogi Berra once observed, “It’s tough to make predictions, especially about the future.” Reasonable people can and do disagree about economic and demographic conditions 25, 50, or 75 years into the future. Yet making such assumptions is critical for evaluating the current status of the Social Security program and the various proposals for reforming it. There always have been some observers who have questioned whether the Social Security trustees’ assumptions are the best basis for evaluating the financial condition of Social Security and the impact of various reform proposals. Other assumptions certainly may be reasonable. And even small changes in assumptions over a 75-year period can lead to large changes in the projections. Any projection over a 75-year period is subject to a high

Since the 2003 report, the trustees have included the program’s unfunded obligations and actuarial balance over an infinite time horizon. Given the uncertainty of projections 75 years into the future, extending these projections into the infinite future can only increase the uncertainty— so that the results can have only limited value for policymakers. This is due largely to anomalies and incongruities that inevitably arise from extending any set of long-range actuarial assumptions to infinity. For example, extending to infinity the assumptions used for labor force participation rates and mortality improvement leads ultimately to a situation in which the typical worker is expected to receive benefits for a period longer 10

ISSUE BRIEF MAY 2012

Conclusion and Recommendations


degree of uncertainty. The trustees’ intermediate assumptions are what they are described to be—a best estimate of future demographic and economic trends based on careful study and analysis of all available data. A number of different proposals for Social Security reform are before the public. When evaluating these plans, policymakers should be aware of the demographic and economic assumptions that underlie the analyses. In some cases, the potential advantages of a particular reform proposal may depend as much on the assumptions used as on the proposal’s actual provisions. In addition, policymakers should take care that assumptions are being used consistently across all proposals that are being compared.

To remove some of the uncertainty about the effects of Social Security reforms, we offer the following recommendations: 1. All analyses of Social Security reform proposals that include financial projections should also disclose the assumptions used. 2. Any such analysis of proposals should use assumptions that are internally consistent. 3. In cases in which substantial uncertainty exists as to the appropriate level of a critical assumption, sensitivity analysis or a range of assumptions should be provided. 4. When calculations for competing reform proposals use different sets of assumptions, comparisons of these proposals should recognize the effects of the differing assumptions.

1850 M Street NW Suite 300 Washington, DC 20036 Tel 202 223 8196 Fax 202 872 1948 www.actuary.org

ISSUE BRIEF MAY 2012

11


A PUBLIC POLICY SPECIAL REPORT

Actuarial Soundness May 2012 American Academy of Actuaries Actuarial Soundness Task Force


A PUBLIC POLICY SPECIAL REPORT

Actuarial Soundness May 2012

Developed by the Actuarial Soundness Task Force of the American Academy of Actuaries

The American Academy of Actuaries is a 17,000-member professional association whose mission is to serve the public and the U.S. actuarial profession. The Academy assists public policymakers on all levels by providing leadership, objective expertise, and actuarial advice on risk and financial security issues. The Academy also sets qualification, practice, and professionalism standards for actuaries in the United States.


ACTUARIAL SOUNDNESS SPECIAL REPORT

2012 Actuarial Soundness Task Force Shawna Ackerman, FCAS, MAAA, Chairperson

Steve Alpert, FSA, EA, MSPA, MAAA, FCA Donna Novak, ASA, MAAA, FCA John Pedrick, FCAS, MAAA

Lee Barclay, FCAS, MAAA Arthur Panighetti, MAAA, FSA Kevin Russell, FSA, MAAA

1850 M Street N.W., Suite 300 Washington, D.C. 20036-5805

© 2012 American Academy of Actuaries. All rights reserved.


ACTUARIAL SOUNDNESS SPECIAL REPORT

TABLE OF CONTENTS

Introduction..........................................................................................................................1 1. Health...............................................................................................................................1 NAIC Model Laws .......................................................................................................2 Actuarial Standards of Practice ....................................................................................2 Practice Notes ...............................................................................................................3 2. Life...................................................................................................................................9 Individual State Regulation ........................................................................................10 3. Pension...........................................................................................................................11 Individual State Regulation ........................................................................................13 4. Property/Casualty...........................................................................................................16 Actuarial Literature Review .......................................................................................16 Statements of Principles .............................................................................................17 Individual State Regulation ........................................................................................18 NAIC Model Laws .....................................................................................................20 Catastrophe Insurance Programs ................................................................................20 5. Conclusion .....................................................................................................................24

Appendix............................................................................................................................25

American Academy of Actuaries

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

INTRODUCTION The terms “actuarially sound” and “actuarial soundness” have appeared in actuarial literature since the early 1900s. They appear in historical, active, and proposed state and federal statutes and regulations. In these statutes and regulations, the terms have been applied to rates, reserves, funding levels, and solvency. They have been applied to forprofit entities and governmental programs. Discussions among several committees and groups of the American Academy of Actuaries prompted questions regarding the use of terms such as actuarially sound and actuarial soundness. This paper catalogs the use of these and similar terms, providing examples where the terms are defined and where they are not. This paper is not an exhaustive search of every use and definition related to actuarial soundness. This paper is not meant to produce a single definition or to provide a direction for the use of these terms. It is meant to give general background and to assist in possible future efforts to provide actuaries and the public with specific direction on the use of these terms. The concept of actuarial soundness is becoming more visible in public discourse, particularly in the context of existing federally funded programs like the National Flood Insurance Program. As a result, a robust examination of this issue by the actuarial community may be helpful. In addition to searching the actuarial literature, the task force reviewed the Actuarial Standards of Practice (ASOPs), National Association of Insurance Commissioners (NAIC) model laws and regulations, and selected state statutes and regulations for each of the four actuarial practice areas: health, life, pension, and property/casualty. This approach produced different results by practice area. Sections 2–5 of the issue brief provide an overview of the historical definitions and use of the terms as well as some of the issues that arise for each of the four actuarial practice areas. Additional documents specific to a particular practice area also are referenced where relevant. We begin our discussion with a dictionary definition of the word “sound”: “Based on truth or valid reasoning; accurate, reliable, judicious, sensible; agreeing with established views or beliefs; showing good judgment or sense.” 1

1. HEALTH Health actuaries long have utilized some form of the term actuarially sound in conducting their actuarial work or in describing a statutory or regulatory requirement. In this section, we look first to the NAIC model laws, in which we found one use of the term actuarially sound. We then identify instances in which the term actuarially sound is used in a health-

1

http://www.yourdictionary.com/sound (last visited on March 9, 2012).

American Academy of Actuaries

1

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

related ASOP. Our final subsection includes several practice notes that utilize the term often. NAIC MODEL LAWS One model law related to health issues uses the term actuarially sound. Small Employer Health Insurance Availability Model Act (Prospective Reinsurance With Or Without An Opt-Out) In Section 6. Restrictions Relating to Premium Rates, this model law states: (1) Each small employer carrier shall maintain at its principal place of business a complete and detailed description of its rating practices and renewal underwriting practices, including information and documentation that demonstrate that its rating methods and practices are based upon commonly accepted actuarial assumptions and are in accordance with sound actuarial principles. (2) Each small employer carrier shall file with the commissioner annually on or before March 15, an actuarial certification certifying that the carrier is in compliance with the Act and that the rating methods of the small employer carrier are actuarially sound. Such certification shall be in a form and manner, and shall contain such information, as specified by the commissioner. A copy of the certification shall be retained by the small employer carrier at its principal place of business. (emphasis added) ACTUARIAL STANDARDS OF PRACTICE (ASOPs) ASOPs identify what the actuary should consider, document, and disclose when rendering actuarial work in the United States. In the ASOPs, there is only one place in which actuarial soundness is defined—ASOP No. 26, Compliance with Statutory and Regulatory Requirements for the Actuarial Certification of Small Employer Health Benefit Plans. ASOP No. 26 states: Actuarial Soundness—Small employer health benefit plan premium rates are actuarially sound if, for business in the state for which the certification is being prepared and for the period covered by the certification, projected premiums in the aggregate, including expected reinsurance cash flows, governmental risk adjustment cash flows, and investment income, are adequate to provide for all expected costs, including health benefits, health benefit settlement expenses, marketing and administrative expenses, and the cost of capital. (emphasis added) The published comments on the exposure draft of ASOP No. 26 from 1995 state that the issue of whether and how to describe actuarial soundness of small group premium rates was a significant portion of the work performed by the committee that drafted the ASOP. That committee noted that “many of the applicable laws … require the actuary to address actuarial soundness,” so the committee found it appropriate to address the issue. Note, however, that the definition of actuarial soundness in ASOP No. 26, like all of the definitions in all of the standards, is specific to that standard and does not purport to

American Academy of Actuaries

2

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

provide a definition of actuarial soundness or actuarially sound for all areas and types of actuarial practice or in any other context. PRACTICE NOTES Practice notes are published by the American Academy of Actuaries and describe various methods actuaries may use to follow the guidance provided by ASOPs or legal or regulatory requirements. Practice notes, however, are not in themselves guidance, do not purport to codify generally accepted practice and are not binding on actuaries or any other parties. A number of health practice notes use terms related to actuarially sound or actuarial soundness. Actuarial Certification of Restrictions Relating to Premium Rates in the Small Group  Market (Dec. 2009)  This practice note uses the term actuarial soundness in answering a number of questions. Consider the following examples: Q2. What is an actuary certifying to when a statement of compliance with small group legislative and regulatory requirements is made? The repealed NAIC Model Act, Premium Rates and Renewability of Coverage for Health Insurance Sold to Small Groups (Premium Rates Model Act), defines an actuarial certification in Section 2(A) as a “written statement that a small employer carrier is in compliance with section 4 (Restrictions Relating to Premium Rates) of this act, based on a review of methods, actuarial assumptions, and appropriate records.” Furthermore, the NAIC Model Act, Small Employer Health Insurance Availability Model Act (Small Employer Model Act), defines an actuarial certification similarly in Section 3(A). Both of these NAIC Model Acts require that the certification be done annually and that the rating methods of the carrier be actuarially sound. (emphasis added) When the actuary certifies compliance, it generally means that the actuary has conducted appropriate tests and reviews and has determined that the carrier complies with the state’s definition of compliance. Using the NAIC Model Act as a guide to preparing opinions on compliance, the actuary may review the following: 1. Classes of business (defined in Q10) have been established in accordance with applicable laws. 2. Index rates (defined in Q6) have been calculated as required by law. 3. Premium rates (defined in Q9) for groups within a class do not vary from the index rate for that class by more than is allowed by the law, taking into account any differences in case characteristics

American Academy of Actuaries

3

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

4. 5. 6.

7.

8.

9.

(also defined in Q5), except for groups where transition period allowances are applicable and permitted by law. The index rate for any class does not exceed the index rate for any other class by more than is allowed by law. Rate increases from the prior rating period do not exceed the percentage increases allowed by law. Rating restrictions associated with permitted case characteristics have been met and only allowable case characteristics have been used in adjusting the rates for compliance testing. Rates have been calculated in compliance with applicable laws, and in compliance with any regulations established by the commissioner to implement the law. Differences in rates for plan design are reasonable, reflect objective design differences, and do not include differences in the nature of groups assumed to elect a plan, to the extent permitted by law. Rating methods and practices are in accordance with sound actuarial principles, to the extent permitted by law.

Note that the above text refers to laws and regulations in effect in 2012, but the Affordable Care Act (ACA) and the health care exchanges that will be created as a result of the ACA will result in significant changes to the small group health insurance market beginning in 2014. Q17. What tests are performed to demonstrate compliance? The actuary usually performs the tests necessary to prove and document compliance with the applicable small employer laws and regulations for which the certification is being made and, if required, to determine that the rating methods are actuarially sound. The level of testing required generally will vary with both the specific certification requirements of the particular state and the complexity of the rating practices employed by the small employer carrier. For example, for a carrier that uses a pure community rating approach, a thorough review of rating and underwriting practices may constitute a sufficient level of testing. On the other hand, group specific calculations may be required of a carrier that incorporates all allowable rating parameters in its rating structure. (emphasis added) Generally, tests are performed that demonstrate that the underwriting methods and premium rates charged are established according to the following: 1. The rates are based on generally accepted actuarial methods and in accordance with sound actuarial principles, to the extent permitted by law;

American Academy of Actuaries

4

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

2. Rates are calculated using allowed case characteristic factors, with the range of these factors within the limits allowed by law; 3. Rates do not use any prohibited separate policy fees or charges, similarly, they do not include any prohibited rebates, refunds, or discounts; 4. The index rate for any class of business does not exceed the index rate for another class by the prescribed percentage; 5. The premium rates for small employers with similar case characteristics within a class of business do not vary from the index rate of that class by more than the prescribed percentage; and 6. The percentage increase in renewal premium rates has not exceeded the sum of the following: a. the percentage change in the new business premium rate measured from the first day of the prior rating period to the first day of the new rating period; b. an adjustment, not to exceed a prescribed annual percentage (e.g., 15 percent) adjusted prorata for periods of less than one year, due to the claims experience, health status, or duration of coverage of the employees or dependents of the small employer; and c. any adjustment due to a change in coverage or changes in the case characteristics of the small employer, as determined from the carrier's rate manual for the class of business. The actuary typically will wish to determine if the state has put forth testing procedures that must be followed or if specific policy data must accompany the certification. In the absence of prescribed testing procedures, the actuary usually will wish to be satisfied that the tests performed are sufficient to support the certification. The complexity of the testing method called for generally depends upon the rating practices employed by the carrier. One approach that is generally appropriate for most small employer carriers is to base the testing on the rate manual that must be maintained for each class of business. The requirement to test that the rating practices are based on generally accepted actuarial methods and are in accordance with sound actuarial principles can often be satisfied with a review of the various rating factors included in the rate manual. The actuary typically confirms that only allowable and permitted case characteristics are being used, that the factors associated with these case characteristics are within the limits allowed by law, and that such factors are uniformly applied. If not involved with the development of such factors, the actuary generally reviews the reasonableness of the range of values being used. A familiarity with the underwriting and renewability rules of the carrier and a review of

American Academy of Actuaries

5

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

the supporting data or actual experience on which the rates or most recent rate changes are based are also usually desirable to support the actuary’s opinion. Q23. What testing should be done if the actuary is required to attest to the rates being actuarially sound? (emphasis added) Some states require the actuary to attest to the soundness of the rates charged. This should be relatively straight forward if the actuary attesting is the same actuary who derived the rates. In that case, the methods and assumptions used would be known. But if the actuary signing the certification is required to certify to the soundness of the rates when he/she did not participate in their determination, a review of the pricing methods and assumptions and plan experience may be in order. Rates should be such that they are not inadequate or excessive, and premiums should be reasonable in relationship to the benefits covered. If the actuary relies on the certification made as part of a recent rate filing, that fact should be disclosed. Actuarial Certification of Rates for Medicaid Managed Care Programs (Aug.  2005)  This practice note was developed by the Academy’s Medicaid Rate Certification Work Group in the course of its review of the CMS regulations that require certification of the actuarial soundness of Medicaid managed care premium rates. The work group was asked to:   

Review the CMS regulations that require certification of the actuarial soundness of Medicaid managed care premium rates; Determine the extent to which the Academy has addressed the term actuarial soundness in any public statements; and Make a recommendation to the Academy’s Health Practice Council regarding the best way to proceed on this issue. The work group’s recommendation was to publish a practice note. The Health Practice Council approved this recommendation and directed the work group to proceed with the drafting of the practice note.

The federal requirements, as stated in 42 CFR Section 438.6(c), are as follows: (2) Basic requirements. (i) All payments under risk contracts and all risk sharing mechanisms in contracts must be actuarially sound. (emphasis added)

American Academy of Actuaries

6

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

(ii) The contract must specify the payment rates and any risk sharing mechanisms, and the actuarial basis for computation of those rates and mechanisms. (3) Requirements for actuarially sound rates. In setting actuarially sound capitation rates, the State must apply the following elements, or explain why they are not applicable (emphasis added): (i) Base utilization and cost data that are derived from the Medicaid population, or if not, are adjusted to make them comparable to the Medicaid population. (ii) Adjustments are made to smooth data and adjustments to account for such factors as medical trend inflation, incomplete data, MCO [managed care organization], PIHP [prepaid inpatient health plan], or PAHP [prepaid ambulatory health plan] administration, and utilization (iii) Rate cells are specific to the enrolled population, by— (A) Eligibility category; (B) Age; (C) Gender; (D) Locality/region; and (E) Risk adjustments based on diagnosis or health status (if used). (iv) Other payment mechanisms and utilization and cost assumptions that are appropriate for individuals with chronic illness, disability, ongoing health care needs, or catastrophic claims, using risk adjustment, risk sharing, or other appropriate cost neutral methods. Section 438.6(c)(1)(i) defines actuarially sound capitation rates as capitation rates that: (A) Have been developed in accordance with generally accepted actuarial principles and practices; (B) Are appropriate for the populations to be covered and the services to be furnished under the contract; and (C) Have been certified as meeting the requirements of this paragraph (c), by actuaries who meet the qualification standards established by the American Academy of Actuaries and follow the practice standards established by the Actuarial Standards Board.

American Academy of Actuaries

7

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

Section 438.6(c)(5) also specifies what is not actuarially sound under special contract provisions. (See Sections III and IV of this practice note for additional information.) For example: ii. If risk corridor arrangements result in payments that exceed the approved capitation rates, these excess payments will not be considered actuarially sound to the extent that they result in total payments that exceed the amount Medicaid would have paid, on a fee-for-service basis, for the State plan services actually furnished to enrolled individuals, plus an amount for MCO, PIHP, or PAHP administrative costs directly related to the provision of these services (emphasis added). iii. Contracts with incentive arrangements may not provide for payment in excess of 105 percent of the approved capitation payments attributable to the enrollees or services covered by the incentive arrangement, since such total payments will not be considered to be actuarially sound (emphasis added). Section 438.6(c) requirements for actuarial soundness thus are a combination of two types of requirements. The first is the broad requirement of being developed in accordance with generally accepted actuarial practices and principles. The second is the potentially more restrictive requirement that CMS may impose on fiscal arrangements. This practice note concentrates on issues concerning the former. For issues concerning the latter, it is acknowledged that CMS or the states may impose additional restrictions, and this practice note, therefore, addresses only the potential areas of conflict between these requirements and generally accepted actuarial principles and practices. For the purposes of the practice note, the work group developed the following proposed definition of actuarial soundness to apply to Medicaid managed care rates developed on behalf of a state for submission to CMS (based on the description in ASOP No. 26, discussed earlier): Actuarial Soundness—Medicaid benefit plan premium rates are “actuarially sound” if, for business in the state for which the certification is being prepared and for the period covered by the certification, projected premiums, including expected reinsurance and governmental stop-loss cash flows, governmental risk adjustment cash flows, and investment income, provide for all reasonable, appropriate and attainable costs, including health benefits, health benefit settlement expenses, marketing and administrative expenses, any state-mandated assessments and taxes, and the cost of capital. (emphasis added) This definition is only applicable for the purposes of this practice note and is not guidance. It is not applicable to any actuarial practice other than suggested use for actuarial certification of rates for Medicaid managed care programs and does not have the binding authority such as may be found in a definition found in an ASOP.

American Academy of Actuaries

8

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

There are some differences between the proposed definition above and the language in ASOP No. 26. “Governmental stop-loss” is included in the practice note description of actuarial soundness in recognition of noninsured stop-loss programs funded by states to cover certain costs in excess of specified amounts, or for certain types of services, or for treatment of certain medical conditions. The words “reasonable, appropriate, and attainable” clarify that the costs of the Medicaid benefit plan do not normally encompass the level of all possible costs that any managed care organization (MCO) might incur, but only such costs that are reasonable, appropriate, and attainable for the Medicaid program. In addition, all expected costs directly related to the Medicaid benefit plan normally would be included. An actuary may be asked to assist an MCO by providing an opinion as to whether the rates bid by the MCO or offered by a state are actuarially sound for that particular MCO. The analysis forming the basis of such an opinion usually would include expected costs specific to that MCO. This is a separate and distinct analysis compared to the analysis performed by the actuary who, on behalf of a state, forms an opinion concerning the actuarial soundness of rates to be offered to MCOs and for submission to CMS. The paragraph above uses the words “actuarially sound” in the context of a particular MCO. There is no federal regulatory requirement that rates be actuarially sound for a particular MCO. Some states, however, may require MCOs that make rate bids or that accept offered rates to provide the state with an opinion as to the actuarial soundness (or an opinion addressing acceptability without using the term actuarial soundness) of the rates for that particular MCO. An MCO reasonably could decide to accept rates for a particular year, knowing that it expects an underwriting loss in that year. Such a decision may be a reasonable business decision, given that the MCO is entering a new market or expects underwriting gains to emerge in future periods. As a final note, the term actuarial soundness does not appear in the literature regarding health insurance financial reporting in either U.S. Generally Accepted Accounting Principles (GAAP) or International Accounting Standards Board (IASB) insurance contracts literature.

2. LIFE The use of the term “actuarial soundness” historically has not been used often in the life insurance practice area. Life insurance reserves and policy minimum nonforfeiture value assumptions have utilized prescribed assumptions that were accepted by regulators. With the advent in the 1990s of asset adequacy analysis, and continuing through today’s principle-based reserves and capital development, there may be additional development of the concept and usage of actuarial soundness in the life insurance arena. Since codification in 2001, life insurance statutory accounting, including reserve development, has been centralized in the NAIC process of developing assumptions and

American Academy of Actuaries

9

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

methodologies. 2 A review of the NAIC Life and Health Valuation Law Manual yields only a few uses of the term actuarially sound, related to variable life investments and accelerated life insurance benefits (terminal illness). In addition, the asset adequacy statement of actuarial opinion requires language stating that the reserves are in accordance with sound actuarial principles. These references do not contain language defining actuarially sound. For example, the Variable Life Insurance Model Regulation—Section 4.C.3 under Mandatory Policy Benefit and Design Requirements states: “The insurer shall demonstrate that the reflection of investment experience in the variable life insurance policy is actuarially sound.” Another reference to actuarially sound can be found in the Accelerated Benefits Model Regulation. Section 10.A.1 states: “The insurer may require a premium charge or cost of insurance charge for the accelerated benefit. This charge shall be based on sound actuarial principles.” Section 10.A.2 and 10.A.3 each state: “The interest rate or interest rate methodology used in the calculation shall be based on sound actuarial principles and disclosed in the contract or actuarial memorandum.” Actuarial Guideline XXVII—Accelerated benefits—Section II.B states: “No additional reserves need be held as long as the actuary is convinced that the method used to discount the death benefit reflects sound actuarial principles.” Section 8 of the Actuarial Opinion and Memorandum Regulation, Statement of Actuarial Opinion Based On an Asset Adequacy Analysis, contains the following requirement: In my opinion the reserves and related actuarial values concerning the statement items identified above: (a) Are computed in accordance with presently accepted actuarial standards consistently applied and are fairly stated, in accordance with sound actuarial principles;

INDIVIDUAL STATE REGULATION New York State Regulation 126, Regulations Governing an Actuarial Opinion and Memorandum, Section 95.8, The Statement of Actuarial Opinion Based On an Asset Adequacy Analysis, Section 6, states: The opinion paragraph shall include a statement such as the following: “In my opinion the reserves and related actuarial values concerning the statement items identified above: (i) Are computed in accordance with those presently accepted actuarial standards of practice which specifically relate to the 2

Neither the U.S. GAAP nor IASB insurance contracts literature apply the term actuarial soundness to life insurance.

American Academy of Actuaries

10

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

opinion required under section 95.8 of New York Insurance Department Regulation 126 to the extent not inconsistent therewith and in accordance with the requirements of such regulation, and which are consistently applied and are fairly stated, in accordance with sound actuarial principles;”

3. PENSION With the exception of its use in the context of governmental plans, the term actuarial soundness does not have a significant presence in pension programs. It is not mentioned at all in the funding rules for tax-qualified pension plans under the Employee Retirement Income Security Act (ERISA), promulgated by the Internal Revenue Service (IRS), nor does the term appear in the accounting literature for nongovernmental organizations for which the Financial Accounting Standards Board (FASB) or the IASB is responsible. All those standards require that valuations be based on best-estimate assumptions, which are widely interpreted as being central, expected-value assumptions without adjustment for or discussion of degree of risk. Statutory accounting principles for pension plan accounting promulgated by the NAIC generally follow U.S. GAAP and similarly do not mention actuarial soundness. Actuarial soundness similarly does not appear in any of the ASOPs applicable to the pension area, nor is it included in any Academy practice note in the pension area. The 2010 Social Security Trustees Report 3 does not refer to actuarial soundness per se, although the chief actuary’s certification notes that “the techniques and methodology used herein to evaluate the financial and actuarial status of the Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds are based upon sound principles of actuarial practice and are generally accepted within the actuarial profession.” In the academic literature, there is a 1953 paper, Pension Plans—the Concept of Actuarial Soundness, 4 in which the author attempts to provide a definition related to the present value of accrued benefits on plan termination. This article was written well before the development of ERISA and current accounting standards, however, and is neither part of the actuarial exam syllabus nor considered particularly relevant to current actuarial practice. Actuarial soundness does appear with regularity in one particular segment of pension practice: pension plans sponsored by state and local governments, referred to here as

3

http://www.ssa.gov/OACT/TR/2010/tr2010.pdf (last visited on Jan. 23, 2012). Pension Plans—The Concept of Actuarial Soundness, Dorrance C. Bronson (FSA), Journal of the American Association of University Teachers of Insurance, Vol. 20, No. 1, Proceedings of the 17th Annual Meeting (March 1953), pp. 36-47.

4

American Academy of Actuaries

11

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

governmental plans. These plans generally are controlled by state laws and the Governmental Accounting Standards Board (GASB). Under the relevant governmental accounting standards, the term appears generally in reference to the funding of a plan, usually in the context of the broad notion of having a rational pattern of funding that is anticipated to accumulate sufficient assets in a plan to make pension payments when they come due—a period that can extend many years into the future and long after an employee ceases working. The following citations are instructive in this regard. It should be noted that GASB 27, the relevant standard for pension plans, is in the process of being substantially revised. As of August 2011, GASB has published an invitation to comment (2009), a preliminary views document (2010), and an exposure draft of a proposed standard (2011). GASB is expected to publish a final revised accounting standard in 2012. How many of the existing concepts will survive the standard setting process is uncertain. 

“[T]he measurement of the employer’s pension expenditures/expense for an accounting period is similar to the employer’s required contributions for that period, in accordance with an established and actuarially sound funding policy” (GASB 27, Paragraph 1, Objective) (emphasis added) “Many respondents, however, including actuaries, general and financial administrators of plans and employer entities, and auditors, pointed out that some of the parameters for measuring pension expenditures/expense would defeat the Board’s objective because they were incompatible with actuarially sound practices commonly used in determining funding requirements or were inappropriate for governmental plans for other reasons.” (GASB 27, Basis for Conclusions, Paragraph 78) (emphasis added) “In order to enhance stability in the employer’s contribution rates and simplify the calculations, many plans use practices that are acceptable under recognized funding methodologies but would be precluded for accounting under the parameters of the 1990 ED. Stability of contribution rates is a common funding objective for governmental plans and frequently is required by statute. For example, the statute or policy may prohibit increases in benefits unless they can be funded without a significant increase in the contribution rates and without jeopardizing the actuarial soundness of the plan.” (GASB 27, Basis for Conclusions, Paragraph 80). (emphasis added) “Some respondents to the 1990 ED thought that the existing accounting maximum of 40 years [to amortize changes in unfunded liability, due to plan changes or actuarial gains and losses] had worked well for many years and should be retained. Some respondents also questioned the appropriateness of any reduction in amortization periods and the resulting increase in required contributions, when many plans are well funded, funding excesses are not uncommon, and many employers are having difficulty balancing their budgets without curtailing services. However, the maximum period most frequently recommended in the responses was 30 years. Several reasons were given for recommending that period, including, for example, it is a reasonable approximation of total service life for many employee groups and is consistent with an entry age approach to

American Academy of Actuaries

12

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

cost allocation, it is acceptable from a sound funding perspective, and it is frequently used in practice as a maximum period.” (GASB 27, Basis for Conclusions, Paragraph 108). “The Board notes that both a closed and an open approach are acceptable under recognized actuarial funding methodologies. The approach selected depends on many factors and should be appropriate to the circumstances of the plan, participating employers, and their operating environments. Either approach can produce satisfactory results from a sound funding perspective.” (GASB 27, Basis for Conclusions, Paragraph 115). “When the funding methodology is soundly conceived and appropriately applied, the results are monitored through frequent valuations and appropriate adjustments are made, and the employer pays the required contributions, the plan will progress to full funding, whether the amortization approach is open or closed.” (GASB 27, Basis for Conclusions, Paragraph 116). “A large majority of governmental plans use the level percent method, combined, most typically, with the entry age actuarial cost method. The level percent method reflects traditional principles of sound funding which require a level contribution design—that is, a design whereby future citizens are not expected to contribute more than present citizens. That concept is sometimes referred to as intergenerational equity in the burden on taxpayers. The concept is implemented by establishing a contribution rate which, expressed as a percentage of active member payroll, is expected to remain level over time. The contribution rate includes normal cost and an amount, computed as a level percentage of projected covered payroll, that is designed to amortize an unfunded actuarial liability over a specific period of future years. Although inflation is likely to cause the absolute dollar amount of contributions to increase over time, contributions expressed in dollars adjusted for inflation (real dollars) are expected to be constant. Therefore, the burden on citizens does not increase relative to the payroll on which pension contributions are based.” (GASB 27, Basis for Conclusions, Paragraph 124).

INDIVIDUAL STATE REGULATION At the legislative level, many states also mention actuarial soundness in the context of the funding of their pension plans. In many of these cases, the references presume that the concept is widely understood and generally accepted, without further elaboration. In some cases (such as California), the legislature puts the onus on the independent actuary to certify the actuarial soundness of the funding requirement. The following excerpts from selected state laws highlight some of the attempts to refine, define, or elaborate on the term. An exhaustive search and commentary on individual states’ use of the term actuarially sound is outside the scope of this document. Instead, we selected several states to review the usage of the term as it relates to state pension laws and regulations. The selected states represent those with which members of the task force are most familiar. Perhaps the most detailed of the samples reviewed is the Guidelines for Actuarial Soundness proposed by the Texas Pension Review Board for its May 2, 2011, Actuarial Committee Meeting:

American Academy of Actuaries

13

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

1. The funding of a pension plan should reflect all plan obligations and assets. 2. The allocation of the normal cost portion of the contributions should be level or declining as a percent of payroll over all generations of taxpayers, and should be calculated under applicable actuarial standards. 3. Funding of the unfunded actuarial accrued liability should be level or declining as a percent of payroll over the amortization period. 4. Funding should be adequate to amortize the unfunded actuarial accrued liability over a period not to exceed 40 years, with 15-25 years being a more preferable target. Benefit increases should not be adopted if all plan changes being considered cause a material increase in the amortization period and if the resulting amortization period exceeds 25 years. 5. The choice of assumptions should be reasonable, and should comply with applicable actuarial standards. These guidelines appear to establish objective criteria that would allow the Texas pension board to assess the state of its plans and the recommended funding pattern and do not specifically involve the services of an actuary. Other legislated definitions of actuarial soundness provide less detail and, in some cases, require or imply the participation of an actuary in making a determination. For example: RCW 41.26.710 [pertains to retirement systems in the state of Washington] (14) “Actuarially sound” means the plan is sufficiently funded to meet its projected liabilities and to defray the reasonable expenses of its operation based upon commonly accepted, sound actuarial principles. (emphasis added) RCW 41.44.020 [pertains to city employees’ retirement systems in Washington] The purpose of this chapter is to provide for an actuarially sound system for the payment of annuities and other benefits to officers and employees and to beneficiaries of officers and employees of cities and towns thereby enabling such employees to provide for themselves and their dependents in case of old age, disability and death, and effecting economy and efficiency in the public service by furnishing an orderly means whereby such employees who have become aged or otherwise incapacitated may, without hardship or prejudice, be retired from active service. (emphasis added) RCW 41.16.060 [pertains to tax levy for firefighters’ pension fund in Washington ] … if a report by a qualified actuary on the condition of the fund establishes that the whole or any part of said dollar rate is not necessary to maintain the actuarial

American Academy of Actuaries

14

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

soundness of the fund, the levy of said twenty-two and one-half cents per thousand dollars of assessed value may be omitted … (emphasis added) New York State Retirement Law and Social Security, 316-b(2) … the actuarial value of assets shall be calculated using the five year smoothing method that was used for the fiscal year commencing April first, nineteen hundred eighty-seven which method has been determined to be actuarially sound. (emphasis added) Iowa, CH 1077 Sec. 117 5 Sec. 117. Judicial Retirement System — Legislative Intent—Notification— Report 1. It is the intent of the general assembly that once the judicial retirement system attains fully funded status based upon the benefits provided for judges through July 1, 2001, the employer and employee contribution rates established to fund the judicial retirement system should be adjusted to reflect the ratio of employer and employee contribution rates required under the Iowa public employees’ retirement system. 2. … In conducting the study, the state court administrator shall consider, and make recommendations concerning, the appropriateness of funding the judicial retirement system by establishing employer and employee contribution rates which shall maintain the actuarial soundness of the system and which shall reflect the intent of the general assembly as contemplated in subsection 1. (emphasis added) Connecticut, Chapter 66, State Employees Retirement Act Sec. 5-156a. Funding of retirement system on actuarial reserve basis. (a) The state employees retirement system shall be funded on an actuarial reserve basis…. the amount necessary on the basis of an actuarial determination to gradually establish and subsequently maintain the retirement fund on such determined actuarial reserve basis, and make such other recommendations with regard to such fund and its administration as the commission deems appropriate. (b) The Retirement Commission shall determine on an actuarial basis (1) a normal rate of contribution which the state shall be required to make into the retirement fund in order to meet the actuarial cost of current service and (2) the unfunded past service liability. For the first sixteen years, the funding program for the actuarial reserve basis shall consist of the following percentages of the sum of normal cost and the amount required for a forty-year amortization of unfunded liabilities …: 5

Iowa Acts 2000 (78 G.A.) Ch. 1077, § 117.

American Academy of Actuaries

15

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

… provided said state payments shall not be reduced or diverted to any purpose other than the payment into the retirement fund until the foregoing schedule of payments has been completed and said fund is determined to be actuarially sound. (emphasis added) California Government Codes, Section 31454.1 (c) The intent of the Legislature, in enacting this section, is to insure the solvency and actuarial soundness of the retirement systems governed by this chapter by preserving the independent nature of the actuarial evaluation process. (emphasis added)

4. PROPERTY/CASUALTY As is the case with the actuarial practice areas addressed in prior sections, property/casualty practice long has used some form of the term actuarially sound to direct actuarial work or to describe a statutory or regulatory requirement. In this section, we look first to the actuarial literature as the oldest source of the term for property/casualty actuarial work. We then examine Statements of Principles adopted by the Casualty Actuarial Society (CAS) before moving on to selected state regulations and NAIC model laws. We close the section with a review of several catastrophe insurance programs. The term actuarial soundness does not appear in literature for property/casualty insurance financial reporting in either U.S. GAAP or IASB insurance contracts literature. ACTUARIAL LITERATURE REVIEW The term “actuarially sound” occurs in the first volume of the Proceedings of the Casualty Actuarial Society (“the Proceedings”) published in 1914. An extensive search through the proceedings and the CAS journal Variance reveals many instances of the terms actuarially sound and actuarial soundness. The Appendix contains selected sections in which these terms appear. Some sections have been reproduced more extensively due to their general importance and the light they shed on discussions within the CAS. A concise definition of the term actuarially sound in the CAS literature could not be found. For example, in the Proceedings Volume XLI, from 1955, Nathaniel Gaines writes, “In general, actuarial soundness implies an orderly arrangement for financing obligations under a benefit program. Precise formulations of what constitutes actuarial soundness have been adequately developed elsewhere, so that there is no need for further discussion here.” (emphasis added) Mr. Gaines did not provide further details on where that development could be found. One might argue that the entire library of actuarial literature provides an ever-growing definition, culminating in the discussions provided in the CAS Statements of Principles for ratemaking and reserving, which are discussed below.

American Academy of Actuaries

16

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

In another example from the Proceedings, Volume LXVII, from 1980, James MacGinnitie writes, “The problem, of course, is in the elusive nature of the concept of actuarial soundness. How do you determine whether an actuary’s analysis or recommendation is sound?” (emphasis added) He identifies “a false dichotomy between actuarial soundness and business judgment.” (emphasis added) He then concludes, “If it’s actuarially sound, then it should be good business judgment; and it clearly is poor business judgment to implement something that is actuarially unsound.” (emphasis added) The synopsis found in the appendix illustrates the applicability of the concept of actuarial soundness within the actuarial community. It provides a brief history of the relevant casualty actuarial discussions, some of which contain profound and timeless advice. STATEMENTS OF PRINCIPLES Unlike the Health actuarial practice example, the Actuarial Standards of Practice applicable to property/casualty practice do not directly define the term actuarially sound or actuarial soundness. Several of the ASOPs refer to “sound actuarial practices” or “soundly thought out analysis” (see, for example, ASOP No. 12, Risk Classification (for All Practice Areas) or ASOP No. 17, Expert Testimony by Actuaries.) The CAS published two Statements of Principles, however, that address the concept of actuarially sound. Statement of Principles Regarding Property and Casualty Ratemaking 6 The CAS Board of Directors adopted the Statement of Principles Regarding Property and Casualty Ratemaking in May 1988. The statement puts forth four principles that provide a foundation for the development of actuarial procedures and standards of practice. Section II, Principles, states that ratemaking produces cost estimates that are actuarially sound if the estimation is based on the first three principles, which provide that a rate is the expected value of all future costs associated with an individual risk transfer. This particular definition of an actuarially sound cost estimate is used in several states’ regulations, including Washington, as noted below. Statement of Principles Regarding Property and Casualty Loss and Loss Adjustment Expense Reserves 7 The CAS board of directors also adopted the Statement of Principles Regarding Property and Casualty Loss and Loss Adjustment Expense Reserves in May 1988. In this statement, an actuarially sound loss (or loss adjustment expense) reserve is defined as a provision based on estimates derived from reasonable assumptions and appropriate actuarial methods for the unpaid amount required to settle all claims and associated claims expenses. 6 7

http://www.casact.org/standards/princip/sppcrate.pdf (last visited on January 24, 2012) http://www.casact.org/standards/princip/sppcloss.pdf (last visited on January 25, 2012)

American Academy of Actuaries

17

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

INDIVIDUAL STATE REGULATION An exhaustive search and commentary on individual states’ use of the term actuarially sound is outside the scope of this document. Instead, we selected several states to review the usage of the term as it relates to property/casualty insurance. The selected states represent those in which members of the task force that prepared this document live. Washington Like many other states that adopted the All-Industry Bills 8 in the late 1940s, Washington state’s rate standard requires that rates for property and casualty insurance “not be excessive, inadequate, or unfairly discriminatory” (RCW 48.19.020). In 1990, the Washington Office of the Insurance Commissioner looked to the CAS’ statement of principles to clarify the meaning of that statutory standard. Using Principle 4 and the concept of rates as expected costs, Washington adopted a regulation that states: A rate is reasonable and not excessive, inadequate, or unfairly discriminatory if it is an actuarially sound estimate of the expected value of all future costs associated with an individual risk transfer. Such costs include claims, claim settlement expenses, operational and administrative expenses, and the cost of capital.” (WAC 284-24-065(1)) (emphasis added) This regulation provides a framework under which rate regulation in Washington could move beyond just allowing the traditional 5 percent underwriting profit provision. In 2008, when the Washington State Legislature enacted title insurance reform, wording nearly identical to that of WAC 284-24-065(1) was incorporated into the legislation (RCW 48.29.143) as the basis for a future prior approval system for the regulation of title insurance rates. Ohio  Ohio’s laws regarding property/casualty insurance rates echo the language of the AllIndustry Bills referenced above, stating: “Rates shall not be excessive, inadequate, or unfairly discriminatory” (Ohio Revised Code [ORC] Sections 3935.03[B] and 3937.02[D]). Neither of these sections of the Ohio law links this requirement with actuarial soundness, which possibly is because these laws predate the CAS Statement of Principles. In practice, the wording of the CAS Principle of Ratemaking is used to link the statutory requirements to actuarially sound rates. 8

Prompted by the passage of McCarran-Ferguson, in the 1940s, the NAIC sponsored the creation of an “all-industry” committee comprised of 19 insurance trade organizations. In 1946, the NAIC collaborated with the all-industry committee to develop the so-called All-Industry Bills, which were adopted by the NAIC as model regulations designed to guide states in regulating insurance in accordance with McCarranFerguson. The All-Industry Bills required that rates be “reasonable and adequate,” that they not “unfairly discriminate,” and that past and future loss experience, as well as a reasonable underwriting profit, be considered. See Lemaire, Jean, Automobile Insurance: Actuarial Models (1985).

American Academy of Actuaries

18

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

Ohio has an exclusive workers’ compensation fund, 9 which is managed by the Administrator and Board of the Ohio Bureau of Workers’ Compensation. The requirements for setting rates are provided in ORC Section 4123.34: The administrator … shall fix and maintain, with the advice and consent of the board, for each class of occupation or industry, the lowest possible rates of premium consistent with the maintenance of a solvent state insurance fund and the creation and maintenance of a reasonable surplus… (B) … and the administrator shall adopt rules, with the advice and consent of the board, governing rate revisions, the object of which shall be to make an equitable distribution of losses among the several classes of occupation or industry… (C) The administrator may apply that form of rating system that the administrator finds is best calculated to merit rate or individually rate the risk more equitably, predicated upon the basis of its individual industrial accident and occupational disease experience, and may encourage and stimulate accident prevention. The administrator shall develop fixed and equitable rules controlling the rating system, which rules shall conserve to each risk the basic principles of workers’ compensation insurance. The term “not excessive” arguably can be seen here as “the lowest possible rates”; “not inadequate” as “consistent with the maintenance of a solvent state insurance fund”; and “not unfairly discriminatory” as “to make an equitable distribution of losses” and “rate the risk more equitably.” While not using the word “actuarial,” the “… rules shall conserve to each risk the basic principles of workers’ compensation insurance.” California In California, the term “actuarially sound” most often is applied in the context of rates and premiums. For example, the California Automobile Assigned Risk Plan (CAARP) provides the following: Premium charges for the plan shall not be excessive, inadequate, nor unfairly discriminatory, and shall be actuarially sound so as to result in no subsidy of the plan. In no event shall the commissioner be required to approve a plan rate that includes a provision for operating profits greater than zero dollars. The commissioner shall not be required to allow a contingency provision with respect to a plan rate if the commissioner takes final action on an application for a rate change within 180 days from the date the application is submitted to the commissioner by the plan’s advisory committee. 10 (emphasis added)

9

In an “exclusive” workers’ compensation fund, the state develops its own rates and experience using inhouse actuaries or actuarial firms. See http://www.aascif.org/public/1.1.3_types.htm (last visited on January 25, 2012). 10 California Insurance Code (CIC) Section 11624 (e)

American Academy of Actuaries

19

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

The Fair Access to Insurance Requirements (FAIR) Plan contains similar language and provides additional instruction for rates and premiums as follows: Rates for the FAIR Plan shall not be excessive, inadequate, or unfairly discriminatory, and shall be actuarially sound so that premiums are adequate to cover expected losses, expenses and taxes, and shall reflect investment income of the plan. If the plan returns premiums to members annually, the rates shall not include any component relating to surplus enhancements. 11 (emphasis added) For the voluntary market, California adopted a prior approval system in 1988; the statute governing the system states that “no rate shall be approved or remain in effect which is excessive, inadequate or unfairly discriminatory.” 12 The term “actuarially sound” does not appear in the statute, but it is used in the regulations that the commissioner adopted in 2008 to implement the statute. Unlike the CAARP and the FAIR Plan, however, the term is not applied to rates or premiums; it rather is applied to the components or process of ratemaking. California Code of Regulations (CCR) Section 2642.8 defines a criterion of most actuarially sound as follows: The “most actuarially sound” choice is the most appropriate choice within the range of permissible actuarially sound choices, considering both the relative likelihood of all choices within the range and the context in which the choice will be employed. (emphasis added) This criterion is applied to several component selections of the ratemaking process, such as loss development factors, credibility standards, and trend periods. NAIC MODEL LAWS NAIC model laws relating to property/casualty insurance contain relatively few references to the concept of actuarial soundness. In a drafting note, the Model Risk Retention Act suggests that “an analysis of actuarial soundness of rates charged” could be useful to a regulator in determining the financial condition of a risk retention group. The Improper Termination Practices Model Act refers to “sound underwriting and actuarial principles” in its section on unfair discrimination. When an insurer terminates a policy because of the insured’s age or disability, or because of the geographic location or age of the insured risk, the action must be “the result of the application of sound underwriting and actuarial principles related to actual or reasonably anticipated loss experience.” CATASTROPHE INSURANCE PROGRAMS State and national catastrophe insurance programs present additional points of discussion regarding what constitutes actuarially sound cost estimates or rates. Rates may be, by design, subsidized. The programs are generally not designed to generate a profit, and

11 12

CIC Section 10100.2 (a)(1) CIC Section 1861.05 (a)

American Academy of Actuaries

20

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

large losses may be funded by alternative mechanisms. A review of several of these programs highlights additional uses of the term actuarially sound. National Flood Insurance Program (NFIP)  The National Flood Insurance Program was established in 1968 to identify flood-prone areas, make flood insurance available to property owners living in communities that joined the program, encourage mitigation, and reduce federal expenditures for disaster assistance. In a 2001 report, the Government Accountability Office defines an actuarially sound program as one in which overall revenues from insurance premiums are sufficient to cover expected losses from claims and the program’s expenses. The report offers the following conclusion: The program is not actuarially sound … Because the program does not collect sufficient premium income to build reserves to meet the long-term future expected flood losses, including catastrophe losses, it is inevitable that losses from claims and the program’s expenses will exceed the funds available to the program in some years and, cumulatively, over time. 13 (emphasis added) In explaining the lack of actuarial soundness of the NFIP, two conditions are identified as contributing factors. The report addresses individual risk transfer, noting that the program is not actuarially sound by design because Congress authorized the availability of subsidized insurance rates for policies covering certain structures to encourage communities to join the program. As of 2000, approximately 30 percent of the policies in force were subsidized, resulting in an estimated $500 million shortfall. 14 The other contributing factor is that the annual target for the program’s overall premium is at least the amount of losses and expenses in an average historical year. At the time of the report, this value was estimated using the average annual losses experienced under the program since 1978 and thus did not include consideration of the potential for catastrophic flood losses. With regard to the first condition, the program fails to meet the definition of actuarial soundness as laid out in the CAS Statement of Principles for Ratemaking because the rates for some risks do not provide for the expected future costs of those classes of insured losses. With regard to the second condition, the estimation process for determining average annual losses does not consider a sufficiently large range of catastrophic loss potential. When the NFIP experiences losses in excess of its capital and reserves, it borrows from the U.S. Treasury, as it did following losses from Hurricane Katrina in 2005. 13

Flood Insurance, Information on the Financial Condition of the National Flood Insurance Program, U. S. General Accounting Office, July 19, 2001, Page 3. Available at http://www.gao.gov/new.items/d01992t.pdf (last visited on January 26, 2012). 14 Ibid., Page 7

American Academy of Actuaries

21

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

California Earthquake Authority (CEA)  In California, all insurers that sell residential property insurance must offer to sell a policy that covers the peril of earthquake. Following the 1994 Northridge earthquake, and in the context of a severely restricted homeowners’ insurance market, the California legislature in 1996 established the CEA as a publicly managed, largely privately funded entity. Companies that sell residential property insurance in California can choose to offer their own privately funded earthquake insurance product or they can become a participating insurance company of the CEA. Only participating insurance companies can offer CEA earthquake insurance policies. The CEA enabling statutes require that the rates established by the authority be actuarially sound so as not to be excessive, inadequate, or unfairly discriminatory. 15 This statement echoes the CAS statement of principles. The statutes contain the term actuarially sound in two additional sections. A minimum retrofit discount of 5 percent is required for homes meeting specified conditions. A larger discount or credit may be applied, provided that it is determined to be actuarially sound. 16 This requirement, again, is consistent with the statement of principles as it recognizes the costs of the individual risk transfer (in this case, the estimated costs associated with a retrofitted home versus one that is not). As a final note, the CEA enabling statutes provide for the establishment of a mitigation fund, which is funded by a portion of the CEA’s investment income. CIC 10089.37 states: The board shall set aside in each calendar year an amount equal to 5 percent of investment income accruing on the authority's invested funds, or five million dollars ($5,000,000), whichever is less, if deemed actuarially sound by a consulting actuary employed or hired by the authority, to be maintained as a subaccount in the California Earthquake Authority Fund. The authority shall use those funds to fund the establishment and operation of an Earthquake Loss Mitigation Fund. In the event a set-aside of mitigation-related funds may impair the actuarial soundness of the authority, the board may delay the implementation of this section. Any delay shall be reported to the Legislature and the commissioner and reported publicly. In this section, actuarial soundness is used as a measure of the solvency of the program. Florida Citizens Property Insurance Corporation  Florida Citizens was established in 2002 as a not-for-profit, tax-exempt government corporation to provide state-backed insurance coverage, including wind damage coverage, for homeowners who cannot get coverage in the private market.

15 16

CIC 10089.40 (a) CIC 10089.40 (d)

American Academy of Actuaries

22

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

Florida Citizens rates initially were required to be noncompetitive with the voluntary market, using a formula dependent on the highest rate offered in the voluntary market for specific areas. After several legislative changes in the ensuing years, in 2009, Florida enacted House Bill 1495, which requires Florida Citizens to implement rate increases until the implementation of actuarially sound rates. Beginning on Jan. 1, 2010, the rate increases were limited to 10 percent for any single policy issued, excluding coverage changes and surcharges. The limitation is to be removed once actuarially sound rates are implemented. The term actuarially sound is not specifically defined in the legislation; within it, however, conditions are provided, including the following: 

After the public hurricane loss‐projection model … has been found to be accurate and reliable by the Florida Commission on Hurricane Loss Projection Methodology, that model shall serve as the minimum benchmark for determining the windstorm portion of the corporation’s rates. The rates are generally subject to Florida statutes for rate standards (Section 627.062) which contain the standard prohibition against rates that are excessive, inadequate, or unfairly discriminatory as well its own listing of considerations, which are to be made in accordance with generally accepted and reasonable actuarial techniques, including: o Investment income o The cost of reinsurance o Past and prospective expenses

ASOPs recognize that actuaries might reasonably differ in their preferred methods and choices of assumptions and might reasonably reach differing opinions, even when faced with the same facts. Two actuaries could apply a particular ASOP, both using reasonable methods and assumptions, and reach appropriate results that could be substantially different. In the context of ratemaking for insurance companies, for example, disputes over whether a rate is an actuarially sound cost estimate tend to arise due to differences in opinion over the methods used to estimate future costs; the inclusion, exclusion, or limitation of certain costs; and how the rates are distributed to the individual classes of insureds. This result should not be a surprise given that ratemaking is a prospective exercise. In the context of ratemaking and the actuarial soundness of catastrophe programs, evaluations of what constitutes an actuarially sound rate and/or program often are focused on the estimation of losses and/or the cost of financing large losses. For example, the prospective estimation of catastrophic losses might utilize a complex computer model rather than long-term historical averages. As another example, the NFIP has been considered by many as actuarially unsound because, in addition to the issues noted above, there is no provision in the rates for the cost of capital. As noted above, NFIP losses above its capital or reserve levels are funded by borrowing from the U.S. Treasury and are intended to be repaid over time by policyholder premiums. While not all publicly based catastrophe programs rely on outside sources of funding (e.g., taxpayer dollars or assessing a broader policy base), when they do, additional examination is needed to evaluate actuarial soundness. Instructions in the enabling legislation are necessary to

American Academy of Actuaries

23

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

address the level of funding that is expected from premium income and the level that is intended to come from non-premium sources.

5. CONCLUSION Terms such as actuarially sound or actuarial soundness appear in each of the actuarial practice areas discussed above. In some instances, the term is specifically defined. ASOP No. 26, Compliance with Statutory and Regulatory Requirements for the Actuarial Certification of Small Employer Health Benefit Plans, defines actuarially sound small employer health benefit plan premiums. The CAS statement of principles provides a description of an actuarially sound rate and a list of considerations that are not inconsistent with ASOP No. 26. The definitions, descriptions, and discussions surrounding actuarially sound generally are consistent across practice areas when applied in similar circumstances. While the term is defined specifically in some circumstances, it more often is used as a general term, assumed to be understood to mean reasonable and consistent with generally accepted actuarial principles and practices. In applying the term as a description of actuarial work, it becomes incumbent upon the actuary to provide the support and documentation necessary to show users that the work has been done with skill and care by a qualified practitioner.

American Academy of Actuaries

24

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

APPENDIX Use of the terms actuarially sound and actuarial soundness in CAS literature (in the excerpts below, emphasis added): Proceedings Vol. I, 1914-15, p. 196., reviews of publications dealing with workmens’ compensation, review of Three Years under the New Jersey Workmen's Compensation Law, Report of an Investigation by the American Association for Labor Legislation, New York. 1915. “Neither the report nor the Commission suggests that the interests of beneficiaries in fatal cases require that commutation should be based upon tables of mortality and remarriage as well as upon compound interest, in order to be actuarially sound.” Proceedings Vol. II, 1915-16, p. 371., papers presented at the May 1916 meeting, “Valuation of Pension Funds, with Special Reference to the Work of the New York City Pension Commission.” “Practically all of the 228 cities in the United States with more than 25,000 inhabitants have pension systems of some kind. The eighteen cities with a population of over 300,000 pension their firemen, policemen and teachers; and seven of these cities have additional funds for other branches of the municipal service. The majority of these cities, like the City of New York, have failed to exhibit forethought in providing for the actuarial soundness of their system. The time is approaching when either faith cannot be kept with their employees or the cities themselves will be overburdened by a financial strain for which they have not made adequate preparation. The size of the New York system and the longer period of its establishment have resulted in a more imperative need for reorganization than elsewhere. The fact that New York is a pioneer in this field gives peculiar value to the results of its experience.” Proceedings Vol. III, 1916-17, p. 286., reviews of books and publications, review of Report of Illinois Pension Laws Commission, Chicago, Illinois, December 1916. “Some conclusions to be drawn from the brief survey of pensions systems in effect in foreign countries are ‘that the systems vary from those operating loosely without much regard for the probable future cost, to those kept actuarially sound on the theory that a class of persons of given age and service should be accumulating a sufficient fund to pay their own pensions; that the age of retirement is generally 65 years; that the amount of the pension is rarely based on final salary but is generally a per cent. of average salary multiplied by years of service." Proceedings Vol. IV, 1917-18, p. 173, discussion of papers read at previous meetings, “Revision of Workmen’s Compensation Rates,” by Harwood E. Ryan, discussion by Ralph H. Blanchard. “The recognition of the principle by the actuarial committee and the adoption of a resolution calling for further actuarial and statistical study are forward steps. They are evidence of a growing puropse [sic] to begin preparation for further rate revision sufficiently in advance to preclude the familiar explanation that changes proposed in the interest of actuarially sound rate-making were admirable but that practical necessity and a lack of time prevented their adoption.”

American Academy of Actuaries

25

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

Proceedings Vol. XVIII, 1931-32, p. 260, papers presented May 20, 1932, “Criticisms and Answers,” by Gustav F. Michelbacher. I.

Actuarial science has been practiced in the field of casualty insurance for less than twenty-five years. In this comparatively brief period, actuaries have labored valiantly to overcome all manner of difficulties. They have made progress; but, speaking frankly, their accomplishments are not to be compared with those achieved in the field of life insurance, where such problems as rate-making and the establishment of reserves have been reduced to definite formulae which have universal sanction. This failure to produce unequivocal results has irked some executives, who have expressed their exasperation in no uncertain terms. In fact, a feeling seems to exist in certain quarters that the business would be infinitely better off today if actuaries had not invaded it with their clumsy attempts to master problems which might have been solved more satisfactorily by persons endowed with "common sense" rather than a penchant for "the scientific method.

VI. What attitude should the casualty actuary maintain under the conditions

which now confront him? The following suggestions are offered for what they may be worth. So far as possible, he should maintain an open mind and be willing to consider any and all suggestions, for many years will elapse before we develop a rating structure that will stand the test of time and, in the interim, every new idea is entitled to its day in court. At the same time, he should constantly strive to perfect his methods and render his materials and equipment more efficient. He should develop a broad interest in all phenomena that even remotely affect the business of casualty insurance, for it is not improbable that the clue to important factors affecting rates will be found in statistical facts outside the usual “experience” which today provides exclusively the materials for rate-making. Particularly should he seek to comprehend and cater to the requirements of supervising officials, agents and policyholders, for a rate that is timely, intelligible and justifiable, as well as actuarially sound, is an achievement to be devoutly desired. He should be willing to accept responsibility for his results and should seek to attain greater accuracy in measuring the hazards of individual risks. His platform may well be that of a scientist like Sir James Jeans, who says in “The Universe Around Us:”

American Academy of Actuaries

26

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

“Science advances ... by providing a succession of approximations to the truth each more accurate than the last, but each capable of endless degrees of higher accuracy ... Guessing has gone out of fashion in science; it was at best a poor substitute for knowledge, and modern science, eschewing guessing severely, confines itself, except on rare occasions, to ascertained facts and the inferences which, so far as can be seen, follow unequivocably [sic] from them.” Thus equipped with a purpose, supplemented by adequate machinery and a proper mental attitude, I venture to prophesy that the casualty actuary will one day place the problem of ratemaking upon a basis which will be beyond criticism. Proceedings Vol. XX, 1933-34, p. 150, discussion of papers read at previous meetings, “Is the Ratemaking Plan the Chief Trouble with Compensation Insurance?”, by Winfield W. Greene, discussion by Clarence W. Hobbs: “The rating system is not perfect. As it stands today, it is the result of a continuous line of experimentation. An endeavor has been made to preserve a foundation of actuarial soundness, and to plug up the leaks as rapidly as possible.” Proceedings Vol. XXIII, 1936-37, p. 92, address delivered at the dinner of the CAS, “Reshaping the Body Politic,” by Clarence W. Hobbs: “Yet I conceive that few actuaries and statisticians worthy of the name fail to grasp some distinct vision of the vibrant and complicated life of the polity into the several parts of which run the lines of underwriting, or fail to catch some reflection of the vast sea of human suffering and death whence emerge their loss statistics. More than one, too has heard the rude and brutal comment, ‘To hell with actuarial soundness: Give me something I can sell!’” Proceedings Vol. XXIV, 1937-38, p. 134, discussion of papers read at previous meetings, “Some Aspects Of Retrospective And Supplementary Rating Plans,” by J. J. Magrath, discussion by S. Bruce Black: “Considering the immediate self-interest of insurance carriers, any form of cost-plus insurance has considerable appeal if, the plan is actuarially sound, and if the carriers are protected against shifting from "retrospective" to "prospective" rating or visa [sic] versa. There is no competitive advantage to any kind of insurance organization, in sound cost-plus insurance.” Proceedings Vol. XXVI, 1939-40, p. 200, discussion of papers read at previous meetings, “State Monopoly of Compensation Insurance, Laboratory Test of Government in Business, Part II, Analysis of The Recent Actuarial Audit of The Ohio State Insurance Fund,” by Winfield W. Greene, discussion by Richard Fondiller: “I have only a scientific interest in the issues drawn between Mr. Greene and the proponents of monopolistic state funds and have prepared this discussion of his paper solely with a view to establishing that my analyses and valuations of the Ohio State Fund were actuarially sound.

American Academy of Actuaries

27

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

Proceedings Vol. XXVIII, 1941-42, p. 222, reviews of publications, Economics of Social Security, by Seymour Edwin Harris, review by Otto C. Richter: “In the chapter entitled Theory of Reserves an attempt is made to define such concepts as actuarial soundness and the reserve system of financing. Unfortunately, the definitions given do not help much to dispel the confusion which frequently surrounded the use of these terms in the recent reserve controversy.” Proceedings Vol. XXX, 1943, p. 102. obituary for John Melvin Laird, 1885–1942: “Although for many years his duties had been of an executive nature, he never ceased to be guided by his actuarial training. He exacted from his actuarial associates rigid standards of performance, but in a friendly and kindly manner that made them eager to live up to those which he set. While never a theorist, he yet insisted that a business decision must be actuarially sound, but without ever losing sight of the fact that the decision had to fit into the day-by-day problems of Company management. He approached business problems with the same logic, clarity and conciseness that characterized his professional papers. He will be missed by the older members of the Society.” Proceedings Vol. XXXIV, 1947, p. 15, papers presented, “Interstate and Overall Rating Plans,” by Seymour E. Smith: “It is believed by the proponents of Retrospective Rating—Plan D that the plan is actuarially sound and will represent a desirable step forward in the rating of sizeable casualty risks. The plan has been so designed as to provide ample safeguards and safety margins so that the integrity of the workmen's compensation rating procedure will in no way be endangered by the combination for rating purposes of workmen's compensation and other third party liability lines.” Proceedings Vol. XLI, 1955, p. 207, papers presented, “Actuarial Aspects of Unemployment Insurance,” by Nathaniel Gaines: “In general, actuarial soundness implies an orderly arrangement for financing obligations under a benefit program. Precise formulations of what constitutes actuarial soundness have been adequately developed elsewhere, so that there is no need for further discussion here.” Proceedings Vol. XLV, 1958, p. 220, papers presented, “The Canadian Merit Rating Plan for Individual Automobile Risks,” by Herbert E. Wittick: “To summarize, the Canadian experience indicates that merit rating of individual automobile risks is not only desirable, but practical. It is actuarially sound and is popular with the great segment of the insuring public who have few, if any, claims. The system keeps rates lower on good business and provides higher rates for the less satisfactory driver. The practical problems are not too difficult and the cost of making the system work is not excessive. A rating plan that does all these things is undoubtedly worthwhile, and represents a real advance over a plan which ignores the claim record of individual risks. In Canada, automobile underwriters generally would not wish to operate without the merit rating plan.” Proceedings Vol. XLV, 1958, p. 255, seminar reports, Current Rate Regulatory Problems, summation by James B. Donovan: “It was suggested that precise

American Academy of Actuaries

28

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

uniformity of opinion among actuaries can never be ascertained; after all, this is an inexact science and we do not expect that a dollars-and-cents formula can be produced as the only actuarially sound answer to many of these complex problems. Nevertheless, to the maximum extent possible, without in any manner interfering with the individual's own sincere opinion, it would be in the best interests of the profession that efforts be made to minimize this kind of contest. Whether the actuary is with the Insurance Department or whether he is with a company, the opinion that he does give should be recognized by all as one that can be accepted as sound and intellectually honest and, to the maximum extent possible, does not present the type of conflict which would be to the detriment of the whole profession. In last analysis, such an endeavor can be an extremely important factor in eliminating many of these industryGovernment disputes and in others could be determinative. To the extent that this goal could be accomplished, without curtailing in any way the intellectual freedom of each individual actuary, it would make not only for the solution of rate regulatory problems but also can only lead to further recognition of the high standards that this society has set for the profession of the actuary.” Proceedings Vol. XLV, 1958, p. 260-02, seminar reports, Standards of Processional Conduct for Actuaries, summation by Winfield W. Greene: “Now my thinking at that time was that the subject ‘A Code Of Ethics For Actuaries’ implied that there should be such a Code. At that particular stage, which was only a few weeks ago, I wasn't convinced that there should be such an animal … Another point that was brought out in our round table discussion was that the more definitely the actuary is regarded as a member of a profession, the more able he is to choose and maintain the actuarially sound position. As somebody said, he should really take the Hippocratic Oath. For example, take the actuary who is employed by a company. His boss wants him to take a certain position. He feels that his actuarial conscience forbids him to do so. The more he is regarded as a member of a profession which is not only just a group of wizards but a group of men dedicated to very high standards of conduct, the better that fellow's chances are of telling his boss ‘Uh,-Uh,’ and still keeping his job. To summarize, I now feel that the objects of our society should be re-stated, that there should be machinery for handling these questions of professional conduct, and that the adoption of a set of guides to professional conduct would be a good thing. The need for such guides has lately been intensified, and this subject merits the utmost serious and conscientious consideration of the society.” Proceedings Vol. XLVI, 1959, p. 228-32, papers presented, OASDI Cost Estimates and Valuations, by Robert J. Myers: “Understandably, the question of the actuarial soundness of the system has provoked much discussion (and confusion, too) over the years. There is not agreement among actuaries as to whether the term “actuarial soundness” can be applied to a national compulsory system with virtually universal coverage. At one extreme, a plan may be said to be ‘actuarially sound’ if the existing fund is at least as large as the value of all accrued benefit rights. This basis is, of course, satisfied by legal reserve life insurance companies but not by many

American Academy of Actuaries

29

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

private pension plans that have assumed considerable liabilities for prior service. Some actuaries define an ‘actuarially sound’ private pension plan as one ‘where the employer is well informed as to the future cost potential and arranges for meeting those costs through a trust or insured fund on a scientific, orderly program of funding under which, should the plan terminate at any time, the then pensioners would be secure in their pensions and the then active employees would find an equity in the fund assets reasonably commensurate with their accrued pensions for service from the plan’s inception up to the date of termination of plan.’ 17 This definition permits a long period before all the past-service credits are fully funded. Other actuaries have a less stringent definition of an actuarially sound system: ‘One which sets forth a plan of benefits and contributions to provide these benefits, so related that the amount of the present and contingent liabilities of the plan as actuarially computed as of any date will at least be balanced by the amount of the present and contingent assets of the plan actuarially computed as of the same date.’ 18 How do these concepts apply to OASDI? The first definition means that it is not actuarially sound, but rather that it is indeterminate from this standpoint; the second definition would say that it is actuarially sound. My personal view is that the second definition can be used and that it is the intent and understanding of Congress that the program has been developed, and should continue, on this basis. Even though it is generally agreed by actuaries that the first and more restrictive definition of actuarial soundness does not apply to OASDI, it may be of interest to compute certain quantities pertinent to it. Such calculation can readily be made, and this has been done on an approximate basis, even though it is recognized that the resulting figures can be misunderstood and misused. One concept of measuring the actuarial condition of a pension plan is to develop the ‘deficit for present members.’ Under this concept, as of the end of 1958, based on the intermediate-cost estimate at 3% interest, the following situation existed for the OASDI program: Item 1. Present Value of Future Benefits and Expenses 2. Present Value of Future Contributions 3. Existing Trust Fund 4. Net Balance, (2) + (3) - (1)

Amount (billions) $544 232 23 -289

Under this concept there was thus an actuarial deficit of almost $300 billion (some 12½ times the amount of the existing trust fund), which, it should be realized, is only of theoretical interest and not of true significance under a long-range social insurance program. 17

Dorrance C. Bronson, “Pension Plans-The Concept, of Actuarial Soundness” Proceedings of Panel Meeting, “What is Actuarial Soundness in a Pension Plan,” sponsored jointly by the American Statistical Association, American Economic Association, American Association of University Teachers of Insurance, and Industrial Relations Research Association, Chicago, Dec. 29, 1952. 18 George B. Buck, “Actuarial Soundness in Trusteed and Governmental Retirement Plans,” ibid.

American Academy of Actuaries

30

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

Still another concept of actuarial soundness applicable to private pension plans may be considered in respect to the OASDI system, namely, the present value of all benefits in current payment status. In a sense, this corresponds to the terminal funding concept of private pension plans. At the beginning of 1959, after the benefit increases provided in the 1958 Amendments had become effective, benefits in current payment status were running at the rate of $760 million a month. These had a present value of about $75 billion, somewhat more than 3 times the then-existing trust fund. But it should be kept in mind that this relationship has no direct bearing on the actuarial soundness of the program, although it is an interesting summary measure of the obligations incurred and does facilitate comparisons with other systems. Although in some quarters there has been considerable criticism of the fact that every two years since 1950 legislative action has liberalized the OASDI system, there is one important point that should be kept in mind. Each time there has been legislative activity, the Congress—particularly, the important, controlling legislative committees concerned—has very carefully considered the cost aspects of all proposed liberalizations. Any changes made have been carefully financed according to the best actuarial cost estimates available. Thus, Congress has attempted to keep the system on a self-supporting basis by keeping benefit costs very closely in balance with contribution income. The Committees have always been anxious to be able to say that the program is ‘actuarially sound.’ In my opinion, this is true under the second, less restrictive definition of ‘actuarial soundness,’ which is fully satisfied by the selfsupporting basis of the system. Certainly, the program can be said to have staunch financial safeguards as long as Congress continues to be cost-conscious, as it has been in the past, and to finance benefit liberalizations adequately.” Proceedings Vol. XLVII, 1960, p. 179, discussion of papers read at previous meetings, OASDI Cost Estimates and Valuations, by Robert J. Myers, discussion by W. Rulon Williamson: “The two illustrative ‘projections’ are set down by Mr. Myers with explanations that show their frailty. One cannot know exactly what the course of evolving history may be. The low and high illustrations are to some extent determined by ‘ideology’ of full employment, the need to check wage inflation, and other political gambits. Marx and Keynes have been well-examined lately. It seems to me that the range used is too narrow in considering ‘the possible’—so that the two prospects might be called low low and low high. But when the mean of the two ‘projects’ is set down, as not any more dependable than the two boundaries of the low low and the low high, and then is quoted as making this highly suspect system ‘actuarially sound,’ ‘reassurance’ has replaced the ‘need for verification.’” Proceedings Vol. XLVII, 1960, p. 195, discussion of papers read at previous meetings, The Compensation Experience Rating Plan—A Current View, by Dunbar R. Uhthoff, discussion by R. M. Marshall: “To the actuarial mind the idea of a credibility greater than unity is unacceptable; it corresponds to the absurdity that the probability of an event happening is greater than certainty. To be actuarially sound the Plan should be corrected so that neither the primary nor the excess credibility can be greater than unity, regardless of whether or not the actual credibility figure may be readily determined.”

American Academy of Actuaries

31

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

Proceedings, Vol. XLVIII, 1961, p. 54-55,presidential address by William Leslie Jr.: “The word ‘chaos’ or its equivalent is being used over and over again to describe one or more problems today facing the insurance industry. Responsible executives with several decades of experience behind them are reporting that today's conditions represent the need to solve problems the like of which they have not seen previously in their careers. There seem many aspects to this report of chaotic conditions. We hear it in discussions of the problems of independent companies viz a viz rating bureaus. A year ago bureaus were alleged to be blocking progress which was sought to be brought to the public by companies operating independently of the bureaus and this year we hear that the bureau companies are ‘walking arm and arm through the marketplace’ leaving a trail of trouble behind; competitively that is. We hear of this chaos being talked of in the broader concept of competition in which there is sincere and open puzzlement as to whether homeowners rates, private passenger automobile rates, surplus lines rates and package policy rates, for example, have not by now departed from the realm of actuarial soundness and represent instead full evidence of a serious rate war. Proceedings, Vol. XLVIII, 1961, p. 186, Footnote 1: Simon, LeRoy J., Myths and Mysteries Concerning the Actuarial Soundness of Merit Rating, paper presented to the Casualty Actuaries of Philadelphia, Sept. 7, 1960 Proceedings, Vol. XLIX, 1962, p. 97, discussion of papers read at previous meetings, Experience Rating Reassessed, by Robert A. Bailey, discussion by Lewis H. Roberts: “An important point is raised by the author to the effect that the parameters of an experience rating plan should be derived from experience. The need for doing so in connection with small risk experience rating, or merit rating, has long been recognized. This may have been because under merit rating plans a small number of classes can be set up to correspond to the several debit and credit groups established under such plans. For other experience rating, however, it would be no less appropriate to tabulate experience by the amount of the modification, and there is no real obstacle to arranging for this to be done. Such a study would provide a valuable check on the actuarial soundness of plans in current use, although it would not guarantee that they are the most efficient of possible plans.” Proceedings, Vol. L, 1963, p. 44, panel discussion during the May 1963 meeting, “An Analysis of the Adequacy of the Various Factors and Rating Values Used in Retrospective Rating,” Chairman, Stephen S. Makgill; panel members Stephen Makgill, James Brannigan, Donald Trudeau, James Boyle: “When the risk's expected losses have been determined, the ratio of these losses to the loss provisions in the minimum and maximum premium are used to obtain the insurance charge percentages directly from Table M. These percentages are, of course, in terms of expected losses and must be converted to premium terms for use in the basic premium. In both the automatic Premium Adjustment Rating Plan and the Premium

American Academy of Actuaries

32

www.actuary.org


ACTUARIAL SOUNDNESS SPECIAL REPORT

Adjustment Plan for Boiler and Machinery risks the tables for the determination of the insurance charge are not labeled Table M as such but the underlying basic data is the same and the use of separate tables for these plans is merely one of mechanical convenience. In the various tabular retrospective rating plans the calculation of the insurance charge has been made in advance and is built into the tabular basic premium ratios. In the various formula type plans the appropriate insurance charge must be calculated on each individual risk. In view of the wide flexibility in the plans, this calculation is somewhat complicated if a high degree of actuarial soundness is to be maintained and as a general rule these calculations are made in the home office of the various carriers and then are checked as to accuracy by the appropriate rating organization.” Proceedings, Vol. LI, 1964, p. 37, previously presented papers, Some Fundamentals of Insurance Statistics, by Harry M. Sarason, discussion by Charles C. Hewitt, Jr.: “The science of statistics is based on similarities and on differences. Similarities lead to classes. Differences lead to sub-classes, to frequency distributions and to individuals—unique individuals, persons. Insurance statistics is in the class of human statistics and in the sub-class of business statistics. Insurance statistics of various kinds have their own distinct characteristics; each statistical study has its individual characteristics. One important characteristic of insurance statistics is change; who knows what tomorrow holds, except change? A sudden change like October 1929, or a mathematically smooth change? The application of insurance statistics to insurance operations involves the vital operations of an insurance business; sales, profit making, and the ability to provide the benefits which have been promised. So important are our statistics and our profession that actuaries, quite as a matter of course, appear before and are a part of boards of directors and governmental committees—so important, that the words ‘actuarially sound’ have been used as part of the presidential vocabulary. Actuaries and other insurance statisticians belong to that class of individuals, purveyors of truth, of whom King Solomon wrote in The Book of Proverbs; ‘Seest thou a man diligent in his business? He shall stand before kings.’” Proceedings, Vol. LIII, 1966, p. 307, papers presented, Underwriting Profit in Fire Bureau Rates, by Laurence H. Longley-Cook: “To justify the use of combined stock and mutual fire insurance loss experience, or as is sometimes suggested experience including independents and direct writers as well, three fallacious arguments are frequently put forward, and these must be reviewed briefly. The first is usually referred to as the ‘broadest possible base’ and the second, less frequently used, I will call ‘a house is a house.’ The third argument is that combined stock and mutual experience is used for workmen’s compensation insurance which, it is generally admitted, is rated on actuarially sound methods.

American Academy of Actuaries

33

www.actuary.org


June 24, 2013 Submitted electronically via Federal eRulemaking Portal: http://www.regulations.gov Michael T. McRaith Director, Federal Insurance Office Attention: Study on Natural Catastrophes and Insurance Room 1319 MT Department of the Treasury 1500 Pennsylvania Avenue NW. Washington, DC 20220 Re:

Study on Natural Catastrophes and Insurance

Dear Director McRaith: On behalf of the American Academy of Actuaries 1 Casualty Practice Council’s Extreme Events Committee, I appreciate the opportunity to provide input to the Federal Insurance Office (FIO) for the purpose of conducting its study of natural catastrophes and the current state of the market for natural catastrophe insurance in the United States, as required by the Biggert-Waters Flood Insurance Reform Act of 2012. 1. The current condition of, as well as the outlook for, the availability and affordability of insurance for natural catastrophe perils in all regions of the United States, including whether a consensus definition of a “natural catastrophe” should be established and, if so, the terms of that definition; Academy Committee response: When participating in a market, insurers determine whether they have a reasonable expectation of an adequate return on the capital exposed to catastrophe risks. In areas of the country with the greatest natural catastrophe risk exposure, the amount of capital required to ensure payment of claims and replacement of capital necessary to remain in business following an event is typically beyond the capacity of the private insurance market. In these cases, it has been necessary to 1

The American Academy of Actuaries is a 17,000-member professional association whose mission is to serve the public and the U.S. actuarial profession. The Academy assists public policymakers on all levels by providing leadership, objective expertise, and actuarial advice on risk and financial security issues. The Academy also sets qualification, practice, and professionalism standards for actuaries in the United States.

1850 M Street NW

Suite 300

Washington, DC 20036

Telephone 202 223 8196

Facsimile 202 872 1948

www.actuary.org


engage additional sources of capital, including federal and state governments, insureds, securities markets, and the public, along with insurers and reinsurers. Rate Adequacy To participate in a market, insurers make a determination that there is a reasonable expectation that they can charge an adequate rate, defined by actuarial principles 2 as a rate that will cover the expected losses to be paid. They also determine whether their rates will provide a reasonable return for their cost of capital. The main challenge for U.S. insurers in funding major catastrophe losses is the enormous amount of capital required to ensure payment of claims and the replacement of that capital after it is depleted. An event that gives rise to $100 billion or more in insured losses is certainly plausible among natural disasters like hurricanes and earthquakes. This level of capital cannot be accumulated quickly from annual premiums. Because exposing that much capital to loss entails a high degree of risk, capital markets require a significantly higher return to justify their investment. The following are critical components of effective regulatory and/or statutory structures governing rate adequacy for catastrophe risks: •

Use of catastrophe models in ratemaking needs to be allowed. These models are the best source of information on projected losses. Actual historical data are often too sparse to adequately estimate losses, so historical loss information must be augmented using information provided by models. Statutory/regulatory allowance of models is necessary to achieve a reasonable loss and cost of capital component in rates. It is also important to recognize the place that the proprietary concerns of modeling companies holds for them to retain an incentive to further refine/enhance their models.

•

Accounting for the cost of capital is essential to obtain an adequate return. Many state laws and regulations do not explicitly address this, or else they explicitly prohibit an appropriate cost of capital on retained risk exposure. In addition, because of the low frequency of catastrophe events, regulated rates often are held below the true costs of the underlying capital required to cover the uncertainty, the volatility, and the risk of impairment or ruin due to the possibility of high losses. The cost-of-capital component is highly significant in high catastrophe areas. Failure to reflect the proper costs of capital, which, in turn, reflect market-clearing prices for capital replacement, ensures that insurance capital will be deployed elsewhere, ultimately increasing the size of the residual market. Accounting for reinsurance costs and other risk transfer funding in rates is essential. Reinsurance is a critical risk transfer tool for insurers with coverages in catastrophe-prone areas. Insurers contractually transfer a portion of their risk to a reinsurer, and, in turn, reinsurers diversify the risk by offering reinsurance products covering many distinct geographic areas and types of business. Given the volatility of the catastrophe risk

2

See CAS Statement of Principles Regarding Property and Casualty Insurance Ratemaking: http://www.casact.org/professionalism/standards/princip/sppcrate.pdf.

2


exposure and the large amount of capital consumed, the cost of capital is typically reflected in catastrophe rates. Increased Private Market Involvement The more companies that participate in a market, the broader the market’s capital base and the more widely available coverage can be, as insurers deploy diverse business models. The risk of ruin is “subadditive”; widely shared among many insurers, a given level of catastrophe risk permits lower total capital requirements than the same level of risk concentrated among a few insurers and perhaps an oversized residual market. Therefore, efforts to increase private market involvement may lead to greater overall insurance availability. Some examples of incentives to increase private market involvement are: •

Premium Tax Credits would allow states to directly encourage more insurance company involvement in the catastrophe market. While this could be marginally helpful, the potential amount of premium tax credit would likely not be large enough on its own to encourage companies to market their policies to coastal areas.

Higher Deductibles can allow insurers to share the costs of losses with insureds, broadening the capital base available to fund losses and allowing insurers to offer coverage to a broader portion of the market. Premium credits that are actuarially commensurate with a higher deductible could make coverage more affordable, though consumers must be educated that they are choosing to retain more risk. This also encourages property owners to take more personal responsibility to implement recommended steps to mitigate property damage to reduce their ultimate costs.

Residual Market Buyout Strategies may incentivize new and existing carriers to enter a market and depopulate residual markets. “Depopulation” programs are only effective when the insurers are carefully vetted for solvency, and reinsurance plans are permitted to charge long-term actuarially adequate rates. Such programs are most successful when insurers aggressively educate consumers about their options and are permitted some flexibility in forms, rates, and coverage to ensure they can remain attractive relative to the residual market.

Minimize Insurers’ Uncertainty by allowing them to use policy coverage language that provides the flexibility they need to successfully manage and price their catastrophe risk. Matching price to risk is essential to the financial solvency of the insurer, and, if policy coverage is made uncertain by regulators or courts, particularly retroactively, that uncertainty could manifest itself in the form of higher prices. Delays and obstacles to quickly adopting needed policy changes are equally disruptive.

Consensus Definition of “Natural Catastrophe”: For internal reporting purposes, an insurer might define an event as a natural catastrophe if it exceeds a certain dollar threshold of loss. For example, Property Claims Services, a unit of Insurance Services Office, Inc. (ISO), defines catastrophes in the U.S., Puerto Rico, and the U.S. 3


Virgin Islands as events that cause $25 million or more in direct insured losses to property and affect a significant number of policyholders and insurers. 3 The term “natural catastrophe” is typically defined, if at all, by example—earthquakes, hurricanes, floods, etc. Outside a specific context, the Extreme Events Committee does not find it essential to establish a consensus definition of the term. Such a definition could be used in a wide variety of contexts: to monitor an insurance company’s experience; to set a threshold for the provision of federal assistance; or to collect data on the catastrophe insurance-related activities of insurers, banks, etc., among many other purposes. The context of a consensus definition would be essential to an evaluation of its usefulness; a definition that might be suitable in one context might be wholly inappropriate in another. 2. The current ability of States, communities, and individuals to mitigate their natural catastrophe risks, including the affordability and feasibility of such mitigation activities; a. The current and potential future effects of land use policies and building codes on the costs of natural catastrophes in the United States; Academy Committee response: Damage prevention begins with proper land use planning (i.e., not building in harm’s way), along with strong, well-enforced building codes, so that what is built can better withstand catastrophic events. Improving land use planning and strengthening building codes are longrange solutions. Additional efforts to mitigate natural catastrophe damage to the current building stock are needed now. Current building stock mitigation efforts should include education. The public must understand the need for and benefits of mitigation before individuals will willingly undertake it. Educational materials on mitigation currently exist for most natural catastrophe perils, but, because much of it has been developed by insurers, some may view it as biased. The public is generally more likely to act on mitigation information if it is disseminated by objective third parties. This could create the potential for an ideal area of partnership among regulators, insurers, and advocacy groups. In addition, financial incentives can reduce the payback period (the number of years it takes for savings to offset the cost). Depending on the jurisdiction, effective incentives may include state income tax credits, direct grants, low-interest loans, or assessments attached to property deeds and identifying the property as improved. Financially sound insurance premium credits are part of the solution, but alone, such credits will not generally create sufficiently short payback periods to aggressively incentivize homeowners to undertake retrofits. Financial incentives must be evaluated for how well they reduce the payback period for incurring mitigation costs. In addition, many property owners evaluate the payback period relative to how long they anticipate owning their property.

3

See http://www.iso.com/Products/Property-Claim-Services/PCS-Catastrophe-Serial-Numbers.html.

4


b. The percentage of residential properties that are insured for earthquake or flood damage in high-risk geographic areas of the United States, and the reasons why many such properties lack insurance coverage; Academy Committee response: According to the California Department of Insurance, approximately 11 percent of homeowners in California carry earthquake insurance. The low take-up rate for earthquake insurance in California may be due primarily to the high price and high deductible; the typical deductible for residential earthquake insurance in California is 15 percent, and the average annual premium in 2011 was $858, as compared to an average residential (fire policy) premium of $752. 4 Another important factor is the fact that, unlike coastal hurricane coverage, governmentsponsored mortgage backers like Fannie Mae and Freddie Mac do not require earthquake hazard insurance. Yet another factor is the psychology of risk; disasters are relatively infrequent occurrences, even in peak zones, so homeowners incorrectly underestimate their probability of disaster loss over the period of their expected occupancy. For example, the probability of a one-in-100-year catastrophe over the term of a 30-year mortgage is approximately one-in-four. According to a RAND Corporation study conducted for the Federal Emergency Management Agency (FEMA), about half the homeowners within Special Flood Hazard Areas (SFHAs), otherwise known as “100-year flood zones,� have flood insurance. 5 However, only about one percent of those outside SFHAs have flood insurance. The low take-up rate in flood insurance seems to be the result of high premiums within SFHAs and underestimation of the hazard outside the areas; again, homeowners doubt that the benefit is worth the cost. 6 c. The role of insurers in providing incentives for risk mitigation efforts; Academy Committee response: The principal lever that insurers use to encourage mitigation is the awarding of premium credits. These credits should reflect the extent to which expected losses are reduced by the presence of mitigation enhancements. Given the low frequency of catastrophic losses to an individual building, the premium savings alone generally does not lead to a payback period short enough to substantially improve take-up rates.

4

See http://www.insurance.ca.gov/0400-news/0200-studies-reports/0300-earthquakestudy/upload/EQ2011_Summary_Revised2.pdf. 5 See http://www.rand.org/pubs/research_briefs/RB9176/index1.html. 6 See http://actuary.org/files/publications/AcademyFloodInsurance_Monograph_110715.pdf.

5


3. The current state of catastrophic insurance and reinsurance markets and the current approaches in providing insurance protection to different sectors of the population of the United States; Academy Committee response: The NAPCO LLC Spring 2013 State of the Market report (herein referred to as the NAPCO report) addresses commercial insurance and reinsurance. 7 According to that report, the current state of the catastrophe market for U.S. risks remains fairly competitive, even after the 2011 global losses and the 2012 U.S. losses caused by Superstorm Sandy. In the U.S., the catastrophe insurance market includes the perils of hurricane, severe thunderstorm (including tornado, hail, and straight-line high wind), earthquakes, wildfires, and drought. The NAPCO report states that, although the market did not shift “sharply … in response to Sandy,” the storm may have temporarily delayed a downward shift in pricing levels. The NAPCO report indicates that “[catastrophe insurance] price increases were higher for those … with heavier losses” and for those located in the Northeast. Current approaches to providing catastrophe risk protection by insurers have combined market forces, client relationships, and technical approaches, along with increasingly widespread use of catastrophe models. Analytical tools continue to evolve to make use of more data from a variety of geographic regions, incorporate more perils, and account for lessons learned from recent loss events. As the NAPCO report states, entering 2013, the catastrophe insurance industry had ample capacity and stable pricing. For example, in early 2013, policies that had insured losses as a result of Superstorm Sandy saw price increases of up to 10 percent, while policies with no loss activity saw price decreases in the mid-single digits. This was very different from the postHurricane Katrina market of 2006, in which risks with Katrina-related loss experience saw price increases in excess of 50 percent, according to the 2007 Wharton report Managing Large-Scale Risks in a New Era of Catastrophes. 8 As 2013 has progressed to date, the NAPCO report states, additional capacity was deployed to catastrophe-exposed areas, and prices to insure nonNortheast exposures decreased nearly 10 percent. As catastrophe loss experience in the Northeast had been relatively benign for decades prior to Sandy, many commercial insurers and reinsurers had become less aggressive about the potential loss exposure and had not priced or underwritten as rigorously as they had in Florida and other Southeast areas at high risk for catastrophe losses. After a few “near misses” (namely Earl in 2010), and having now experienced the significant losses caused by Sandy in 2012, some insurers are considering revising terms and conditions for properties in Northeast coastal areas, according to the NAPCO report. For example:

7 8

See http://www.napcollc.com/articles/NAPCOInsuranceInsights-TheStateOfTheMarket-2013-April.pdf. See http://opim.wharton.upenn.edu/risk/library/Report_on_Phase_I.pdf.

6


“Some insurers have increased their windstorm deductibles for [policies covering some areas of] the Northeast from $10,000 to as much as $100,000,” and some are considering eliminating hurricane deductibles in favor of named windstorm deductibles;

Some insurers are considering imposing “named windstorm deductibles for coastal areas of the Northeast,” where allowed by statute and regulation (some states regulate storm deductibles). These deductibles would be set as a percentage of the asset values at risk in a loss, typically one percent to five percent, depending on the location of the risk;

Some “insurers have expanded their definitions of ‘Tier 1’ counties,” in which windstorm risk is high. “Before Sandy, Tier 1 counties included coastal areas from Chesapeake, Virginia, to Texas.” Since Sandy, one insurer has expanded its Tier 1 definition to include “from Virginia up to Maine, and others are considering whether to make the same change”;

Insurers are also contemplating adding a more significant catastrophe surcharge or a “catastrophe load” for policies in the Northeast, similar to charges currently in place in the Southeast; and

Catastrophe models continue to play “an increasingly important role in insurance pricing,” according to the NAPCO report. “The industry has relied on models for earthquake and windstorms for some time, but new models for other catastrophe risks are in development.” Additionally, since Hurricane Katrina, rating agencies 9 have included a significant risk charge for catastrophe exposures in their capital adequacy models. As global regulators and regulated insurers are adopting economic capital models, they are increasingly focused on the catastrophe risk charge and the cost of the associated capital. 10 This has been the case particularly in Bermuda, where it has been the considered view that the reinsurance market has been aggressive in its analytics and pricing for capital utilization, but best practices are beginning to work their way into the primary pricing function. Catastrophe models continue to incorporate advanced scientific information, industry loss and exposure data, and lessons learned from recent events. To illustrate:

9

Catastrophe-modeling firms are expected to validate their models in the coming months in response to Superstorm Sandy, as they do following any major event;

Also according to the NAPCO report, catastrophe-modeling firms aspire to enhance models for risks like flood, wildfires, and tornadoes. New models may offer a better understanding of these risks and the potential for associated losses. While that heightened understanding could result in higher prices for some consumers, credible new models also encourage price stability in the market.

Nationally Recognized Statistical Rating Organization (NRSRO) See http://opim.wharton.upenn.edu/risk/partners/3_RatingStandards-CatRisks.pdf.

10

7


4. The current financial condition of State residual markets and catastrophe funds in high-risk regions, including the likelihood of insolvency following a natural catastrophe, the concentration of risks within such funds, the reliance on postevent assessments and State funding, and the adequacy of rates; Academy Committee response: Due to the potential variability in magnitude of catastrophic losses, it is virtually impossible to finance all potential losses in any single time period using traditional property insurance. This leaves two extreme choices: prefund all potential losses or use post-loss funding when significant losses occur, along with a practical spectrum of options that combine pre- and post-loss funding. Currently, only states have the power to issue public debt to finance hurricane losses. Additionally, some state insurers have used insurance-linked securities or catastrophe risk bonds (cat bonds) to reduce the financial impact of a single major event. Due to its geography, loss exposure, and long history of elevating property insurance issues in its public policymaking, the Florida market precisely illustrates many of the aspects of these residual markets. Like some other catastrophe-prone states, Florida finances a significant portion of its catastrophic risk exposure through post-loss assessments; these assessments are levied on most property-casualty insurance policyholders. (Other states, like South Carolina, pre-fund nearly all plausible windstorm and earthquake loss scenarios via reinsurance.) The three primary insurance entities in Florida with the power to levy post-loss assessments are state-sponsored insurance or quasi-insurance entities like Citizens Property Insurance Corporation (Citizens), the Florida Hurricane Catastrophe Fund (FHCF), and the Florida Insurance Guaranty Association (FIGA). Citizens is the state’s residual market property insurer. Homeowners have the choice of purchasing coverage from a private insurer or from Citizens, which, thanks to recently-enacted S.B. 1770, 11 will now compete to some degree with private insurers on price. While trying to achieve actuarially sound premium rates, Citizens cannot increase rates by more than 10 percent for any single policyholder in a given year. Given its current rate restrictions and deficiencies, it may be many years before all Citizens policyholder premium rates are actuarially justified. Citizens uses a combination of accumulated premiums and surplus, traditional reinsurance, insurance-linked securities, FHCF reinsurance, and post-event debt secured by assessments (some of which apply to nearly all property/casualty insurers, not only homeowners’ insurers) to finance its probable maximum losses, as explained in greater detail below. Citizens has three accounts: the Personal Lines Account (PLA), Commercial Lines Account (CLA), and the Coastal Account. 12 The Coastal Account has a separate financial identity from the PLA and CLA accounts, and the calculation of deficits and resulting assessments are determined independently for each of the three accounts. 13 According to BusinessWire, 11

See http://laws.flrules.org/2013/60. See https://www.citizensfla.com/about/CitizensAssessments.cfm. 13 See http://www.businesswire.com/news/home/20130531005947/en/Fitch-Affirms-Citizens-Property-InsuranceCorp.s-FL. 12

8


depending on the type of assessment, the current assessment base for Citizens includes all property and casualty lines except the medical professional liability, accident and health, and workers’ compensation lines. When Citizens incurs a certain level of projected financial deficit in any of its three accounts, it has statutory authority, pursuant to Section 627.351(6)(b)2.a., 14 to levy up to three different types of assessments, 15 which are described below: 1. Citizens Policyholder Surcharge. The first type of assessment is the Citizens Policyholders Surcharge, which is levied on Citizens’ policyholders and can be applied to each of Citizens’ three accounts. To the extent that the imposition of this surcharge does not eliminate the deficit, Citizens Regular Assessments and/or Emergency Assessments may then be levied. 2. Citizens Regular Assessment. The second type of Citizens assessment is the Citizens Regular Assessment. Regular Assessments are levied on private, Florida-licensed property and casualty insurance companies and on insureds who buy relevant types of policies from surplus lines insurers. The admitted insurers are permitted to recoup the assessment amount by passing the cost on to their policyholders. Concerned about the effect that these assessments could have on thinly capitalized insurers, in 2012, Florida enacted amendments to FL Statute 627.351. The amendments render the PLA and CLA accounts ineligible for regular assessments and reduces the maximum regular assessment in the Coastal Account from six percent to two percent per account. 16 3. Citizens Emergency Assessment. The third type of Citizens assessment is the Citizens Emergency Assessment. Citizens levies Emergency Assessments on a broad swath of property and casualty insurance policyholders of private and surplus lines companies and on its own policyholders. Citizens may only levy Emergency Assessments once the maximum Policyholder Surcharge and Regular Assessments are imposed and found to be insufficient to cover the deficit, in which case an Emergency Assessment of up to 10 percent of premium or 10 percent of the deficit, whichever is greater, may be levied. Emergency Assessments are collected when policies subject to assessment are renewed or when new eligible policies are issued. Among their other effects, the Statute 627.351 amendments shifted some of the assessments that would have been collected as “Regular Assessments” to the category of “Emergency Assessments.” Created in 1993 in the aftermath of Hurricane Andrew, the FHCF is a state-run entity created to stabilize Florida’s residential property insurance market. 17 All insurers selling homeowners insurance in Florida are required to purchase the mandatory catastrophe loss reimbursement coverage offered by the FHCF. The FHCF mandate applies to Citizens and private insurers alike and is structured much like annual aggregate hurricane-only excess-of-loss reinsurance. Retention level per storm and annual reimbursement amounts are determined by a formula as the “applicable FHCF Retention Multiple” of each insurer’s annual FHCF premium for the contract 14

See http://www.leg.state.fl.us/Statutes/index.cfm?App_mode=Display_Statute&Search_String=&URL=06000699/0627/Sections/0627.351.html. 15 See https://www.citizensfla.com/about/CitizensAssessments.cfm. 16 See http://laws.flrules.org/files/Ch_2012-080.pdf. 17 See http://www.sbafla.com/fhcf/AbouttheFHCF/tabid/278/Default.aspx.

9


year in which the relevant hurricane occurred. Annual premiums are determined by a rating plan applied at the individual company level. 18 (The risk associated with policies in-force as of June 30 of each applicable year determines participants’ FHCF reimbursement premium.) The FHCF is mandated by statute to charge “actuarially indicated” premium rates. 19 However, its rates are not actuarially sound to the extent that actuarially sound rates are generally expected to be sufficient to effectuate the arms-length transfer of risk. The FHCF’s risk load is set at a flat percentage by law and is far below market-clearing levels. As a result, FHCF rates are substantially lower today than private market reinsurance rates. FIGA is a state entity created to settle the claims of insolvent private insurers in Florida. 20 It has the authority to assess nearly all Florida property/casualty insurers in the event of insolvencies related to catastrophic storms. 21 Currently, both Citizens and the FHCF have emergency assessments in place (estimated to end around 2017 22 and 2016, 23 respectively) to pay for post-event deficits stemming from the 2004 and 2005 hurricane seasons. Florida is one of several states with residual market mechanisms in place to address the costs of insuring natural catastrophe perils. Another such mechanism is the Texas Windstorm Insurance Association (TWIA). The TWIA is the insurer of last resort for 266,000 homeowners and businesses. 24 The TWIA relies on premiums and assessments placed on Texas insurance companies because of the high risk of catastrophic losses. 25 Following Hurricane Ike in 2008, some have questioned whether TWIA will be able to provide sufficient coverage when another hurricane strikes, 26 as it has no pre-set funding mechanism for much of its probable maximum loss. 27 To help reduce the financial impact of major catastrophe events, some state funds have begun to rely on alternative forms of risk financing like cat bonds. Cat bonds are insurance-linked securities that transfer a specific risk like that of hurricane or earthquake losses from a sponsor to investors. Investors lose their entire principal in the event of a catastrophe that meets the terms set forth in the bond; consequently, interest rates on these bonds are much higher than on other

18

See http://fhcf.paragonbenfield.com/pdf/13faq.pdf. Florida Rule 19-8.028 defines “actuarially indicated” premium rates as “[p]remiums which are derived according to or consistent with accepted actuarial standards of practice. Actuarially Indicated means an amount determined according to principles of actuarial science to be adequate, but not excessive, in the aggregate, to pay current and future obligations and expenses of the Fund, and determined according to principles of actuarial science to reflect each insurer’s relative exposure to hurricane losses.” See http://www.sbafla.com/fhcf/LinkClick.aspx?fileticket=TwTkAQ3Ntfs%3D&tabid=1305&mid=3529. 20 See http://www.figafacts.com/. 21 See http://www.figafacts.com/media/files/2012AssessmentPacket.pdf. 22 See https://www.citizensfla.com/shared/notices/2005HRAEMRG/faqs4comp.pdf. 23 See http://www.floir.com/siteDocuments/GEARFAQs.pdf. 24 See http://www.khou.com/news/local/Texas-windstorm-insurance-decision-delayed--199953431.html. 25 See http://abclocal.go.com/ktrk/story?section=news/state&id=9039828. 26 See http://www.nbcdfw.com/news/local/Board-Determines-Fate-of-Windstorm-Insurance-Group199845151.html. 27 See http://sciencepolicy.colorado.edu/admin/publication_files/2013.17.pdf. 19

10


types of bonds. Increasingly, state-run entities have sponsored the issuance of cat bonds to manage risk. Through issuer Everglades Re, Florida Citizens sponsored 750 million dollars’ worth of two-year cat bonds in 2012 and an additional 250 million dollars’ worth of three-year cat bonds in 2013. 28 Louisiana Citizens Property Insurance Corporation, another state property insurer of last resort, sponsored a series of three-year cat bonds worth $100 million through the Cayman Islands special-purpose insurer Pelican Re. 29 These bonds provide indemnity coverage, meaning recovery will be triggered by actual losses rather than meteorological determinations or industrywide losses. 5. The current role of the Federal Government and State and local governments in providing incentives for feasible risk mitigation efforts and the cost of providing post-natural catastrophe aid in the absence of insurance; Academy Committee response: A report published in September 2012 by the U.S. Government Accountability Office (GAO), Federal Disaster Assistance: Improved Criteria Needed to Assess a Jurisdiction’s Capability to Respond and Recover on Its Own (Report #GAO-12-838) 30 provides a detailed assessment of the governmental role in post-catastrophe aid over the past decade or so. State and federal governments’ financial role after a natural catastrophe is governed largely by the Stafford Act, through which the federal government provides disaster assistance to state and local governments. According to the GAO report, from Fiscal Year (FY) 2004 through FY 2011, the federal government allocated $90 billion for disaster aid. Additionally, during that period, FEMA received 629 disaster declaration requests and approved 539 of them. Mitigation efforts are an important aspect of any disaster risk management and insurance reform/policy initiative. However, incentives in the form of insurance savings cannot fully reimburse all mitigation efforts, and incentive programs should not be implemented with that expectation. Insurance premium reductions alone cannot be an effective incentive for individuals or governments to undertake mitigation efforts. Pre-event funding by federal and state governments, through tax credits, grants for retrofitting, or other programs that reward mitigation efforts can result in significant benefits through reduced post-catastrophe expenditures by federal and state governments.

28

See https://www.citizensfla.com/shared/press/articles/107/03.29.2013.pdf. See http://www.artemis.bm/deal_directory/pelican-re-ltd-series-20121/. 30 See http://gao.gov/assets/650/648177.txt. 29

11


6. Current approaches to insuring natural catastrophe risks in the United States; a. Current and potential future Federal, State, and regional partnerships that support private, direct insurance coverage; Academy Committee response: Two existing partnerships between the public and private sectors are the National Flood Insurance Program (NFIP) and the California Earthquake Authority (CEA). The NFIP is congressionally authorized but is largely administered by private companies known as WriteYour-Own (WYO) companies, which deal directly with the vast majority of NFIP policyholders. In 1996, the California legislature established the CEA as a publicly managed, largely privately funded entity. Companies that sell residential property insurance in California can choose to offer their own privately funded earthquake insurance product, or they can become a participating insurance company of the CEA. Only participating insurance companies can offer CEA earthquake-insurance policies. A potential new partnership might be a federal reinsurance program that would provide private companies with catastrophe protection for extreme events, priced at an actuarially indicated rate. This could reduce the uncertainty of writing natural catastrophe insurance coverage and thus might encourage involved companies to enter or reenter the market. b. The potential privatization of flood insurance in the United States; and, Academy Committee response: The Biggert-Waters Flood Insurance Reform Act of 2012 requires several studies be done concerning the possible privatization of flood insurance in the U.S. Challenges posed by privatization could include: • • •

Only those with an obvious flood hazard will opt to buy the insurance; The cost to insure only those with an obvious flood hazard would be prohibitive if lowerrisk consumers could not be convinced to join the insurance pool; Even with a larger group of insurers and insureds, private insurers require a premium load to account for the cost of capital. Depending on the region and rating plan, actuarially sound rates could be so high that they discourage the broad participation necessary to ensure the financial soundness of the program.

On the other hand, some problems at the granular level, like claims adjusting for combined wind and water claims, the difficulties of dealing with multiple policies on the same property, and insurer disincentives for broad data collection that might assist in evaluating hazard risks, could be ameliorated in a privatized program that harmonizes the administration of wind and flood coverage for consumers. In addition, the aggregate cost of capital required to fund the combined risk of wind and flood could, at least in theory, be reduced if private insurers handled both risks 12


August 2, 2010 Treasury’s Office of Financial Institutions Policy Attention: President’s Working Group on Financial Markets Public Comment Record Room 1417 MT Department of the Treasury 1500 Pennsylvania Avenue, NW Washington, DC 20220 Re: President’s Working Group on Financial Markets: Terrorism Risk Insurance Analysis To the President’s Working Group on Financial Markets: The Terrorism Risk Insurance Subgroup (Academy subgroup) of the American Academy of Actuaries1 thanks the President’s Working Group on Financial Markets (President’s Working Group) for this opportunity to provide comments in response to the request appearing in the Federal Register of June 17, 2010. Key Factors 1. What are the key factors that determine the availability and affordability of terrorism risk insurance coverage? How are these factors being measured and projected today? What factors will determine the availability and affordability of terrorism risk insurance long-term? The President’s Working Group on Financial Markets discussed various factors in its 2006 report, referenced above; how have these factors changed or developed since then? The primary insurance cost issue affecting the availability and affordability of terrorism risk insurance coverage is the potential that a single terrorist attack, using weapons of mass destruction, could cause a huge aggregate loss from a massive number of individual insurance claims. Since September 11, 2001, insurers have worked to improve their understanding of terrorism risk. Unfortunately, this improved understanding of terrorism risk does not provide easy answers to the complicated questions being asked by insurers or by regulators, legislators, and other policymakers. Rather, we now better understand the magnitude of the tremendous uncertainties and estimation problems that face insurers, reinsurers, and other potential suppliers of capital that could be used to finance terrorism risk. 1

The American Academy of Actuaries (“Academy”) is a 16,000-member professional association whose mission is to serve the public on behalf of the U.S. actuarial profession. The Academy assists public policymakers on all levels by providing leadership, objective expertise, and actuarial advice on risk and financial security issues. The Academy also sets qualification, practice, and professionalism standards for actuaries in the United States. 1850 M Street, NW Suite 300

Washington, DC 20036

Telephone 202 223 8196

Facsimile 202 872 1948

www.actuary.org

1


Without TRIPRA or some other federal framework for terrorism risk insurance, the uncertainties regarding potential attacks make it extremely likely that premiums for terrorism risk insurance will be high and volatile and that availability of terrorism coverage will be limited. Without a federal framework for terrorism risk insurance, coverage such as workers’ compensation, which are required to cover claims made by employees injured in terrorist attacks, will become much riskier for insurers and thus more expensive and/or less available over time. Accordingly, the Academy subgroup has concluded that some federal framework for terrorism risk insurance is necessary for terrorism coverage to be widely and readily available. 2. What are the key factors that determine the amount of private-market insurer and reinsurer capacity made available for terrorism risk insurance coverage? How have these factors changed since 2006, when the President’s Working Group on Financial Markets issued its last report? How will such factors evolve in the long-term and upon what factors will available capacity most depend? One key factor is the availability of a federal backstop. Other key factors that influence capacity include the ability to estimate the potential exposure to loss resulting from terrorism events, the ability to obtain reasonable pricing of insurance to cover those events, and the level of insurer/reinsurer capital. Economic Factors 3. How, in general, has the state of the financial markets and economy, and the financial condition of commercial property and casualty insurers, affected the availability and affordability of terrorism risk insurance; and how does that compare with effects on the availability and affordability of other lines or types of commercial property and casualty insurance? Please comment on potential entry of new capital into, as well as any exits from, the terrorism insurance and reinsurance markets. The Academy subgroup does not believe that the current state of the financial markets and the economy has had any significant impact on the availability and affordability of terrorism risk insurance. Underwriting 4. What changes and improvements have taken place in the ability of insurers to measure and manage their accumulation of terrorism risk exposures, and how (as well as to what extent) are primary insurers using available methods? Has improved risk accumulation management led to more availability? Has there been any improvement in modeling of frequency and terrorist behavior? What has been learned from the near-9 years of experience in managing and assessing terrorism risk since September 11, 2001? Overall, how has modeling improved and/or continued to develop since 2006, when the President’s Working Group on Financial Markets issued its last report? How is modeling expected to evolve further in the long-term?

1850 M Street, NW Suite 300

Washington, DC 20036

Telephone 202 223 8196

Facsimile 202 872 1948

www.actuary.org

2


Estimates of the potential losses from terrorist events rely on quantitative approaches that have evolved from those used for natural disasters. There are models that identify potential targets and target areas. Insurers have used this information, along with stochastic analysis, to attempt to understand their exposure. However, with little or no historical frequency of loss, there is little data on which to model terrorism losses. Many of the modeling outcomes have been provided on a deterministic event basis, i.e., the modeled estimate of damages and losses in the defined event. Since the 2006 report, a notable industry-wide initiative has been undertaken to improve exposure data. Insurers have sought to increase their knowledge of exact location information, construction details, and replacement costs for each property. This initiative is driven by information needs in both terrorism and natural catastrophes. Rating agencies have begun asking insurers for information about exposure data quality, and modelers have introduced tools to provide needed improvements. The firms that develop terrorism models are continuing to improve their estimation of frequency and severity. They have learned about intended, planned, failed, and executed attacks and the impact of counterterrorism measures taken in the U.S. and around the world. Modeling firms use this history and their judgment to estimate future possibilities and their frequencies. Since 2006, additional research has been incorporated into the models of property damage and injuries resulting from bombings in urban environments, potential terrorist targets have been added, and frequency estimates have been updated. Terrorism models continue to be updated to incorporate new findings on weapons effects, new potential target facilities, new attempted attacks, new intelligence, and information regarding attack modes used worldwide. 5. What role do mitigation and loss prevention play in underwriting and pricing terrorism risk insurance? How has mitigation developed since 2002, what improvements have been made since 2006, to what effect has the availability of terrorism risk insurance had on mitigation and vice versa; and, how will mitigation evolve in the long-term? Since 2001, some prevention and mitigation efforts have been undertaken by the private sector. These measures include enhanced private security and screening of visitors, barriers in highprofile buildings, and security cameras. Insurers take mitigation efforts into account when underwriting risks; however, these mitigation efforts are not likely to have much impact on the most extreme terrorism events. 6. What is the state of information sharing between and among the private and official sectors related to terrorism risk: (a) how much reliance is placed on open and private source intelligence; (b) how has it affected the availability and affordability of terrorism risk insurance; and, (c) how will such information processes further develop and affect the availability and affordability of terrorism risk insurance in the long-term? The organizations that create terrorism models consult with experts in the development of the 1850 M Street, NW Suite 300

Washington, DC 20036

Telephone 202 223 8196

Facsimile 202 872 1948

www.actuary.org

3


assumptions underlying the models. This includes weapons-effects models, historical event information data, and unclassified threat assessment information. Modelers also share methods with the government pertaining to terrorism risk modeling. Coverage 7. What changes and improvements have taken place with regard to the types of terrorism risk insurance coverage available in the market? What changes and improvements have taken place since 2006? Have there been improvements and changes in forms, are there special terms or conditions? What is the state of standalone, “TRIA-only” coverage? Is available coverage limited to, or broader than that required to be made available under TRIA? Terrorism coverage changes have, to a large degree, been directly related to specific definitions in federal law. The Insurance Services Office (ISO) provides policy language for many insurers. Prior to the 2007 reauthorization of TRIA, the federal definition of a “certified act of terrorism” was limited to foreign acts. During that time, ISO had two types of terrorism exclusion forms: “certified acts” exclusions and “other acts” exclusions. Domestic acts of terrorism were classified as “other acts.” The certified acts exclusion could be used in a situation in which a policyholder rejected the offer of certified acts of terrorism coverage. Other-acts exclusions could be used at the option of the insurer. With the 2007 reauthorization of TRIA, the federal definition of a “certified act” of terrorism, previously limited to foreign acts, was revised to include domestic acts as well. ISO “certified acts” exclusions were updated to reflect this new definition. Also, “other acts” exclusions were removed from ISO programs for property insurance. Use of certified acts exclusions continues to be limited to situations in which the policyholder rejects the offer of certified acts of terrorism coverage. No post-reauthorization change has been required for coverage provided under a policy issued without terrorism exclusions. Such a policy generally covers both certified acts of terrorism and other acts of terrorism, subject to the same terms and conditions as other events covered or excluded under the policy. In states with a statutory requirement for fire coverage, terrorism exclusions (both prior to and since 2007) do not apply to fire following terrorism. 8. What are the differences in availability and affordability of terrorism risk insurance coverage for foreign and domestic terrorist acts? Generally, the source of the terrorist act is much less important than the severity of the event and its impact on the capital of insurers. The availability and affordability of coverage for domestic acts has probably been enhanced by including domestic acts in the definition of “certified acts” and is now similar to the availability and affordability of coverage for foreign acts. 9. Did the Terrorism Risk Insurance Program Reauthorization Act of 2007’s amendment to the definition of “act of terrorism” lead to more availability due to the requirement that such coverage be made available, or was such coverage available prior to 1850 M Street, NW Suite 300

Washington, DC 20036

Telephone 202 223 8196

Facsimile 202 872 1948

www.actuary.org

4


2007; conversely, did the amendment lead to less coverage due to the broadened scope of “act of terrorism” exclusions, or were exclusions revised to distinguish between coverage of foreign and domestic terrorist acts? Generally, the inclusion of domestic acts under the federal backstop has probably improved the availability of terrorism insurance in the marketplace. Insurers are more likely to have the capacity to provide this coverage without risking insolvency due to an extreme domestic terrorist event now that the federal backstop applies to domestic acts. 10. What are the differences in availability and affordability of terrorism risk insurance coverage for losses at U.S. locations, as compared to such coverage for losses at non-U.S. locations? What are the differences as compared between TRIA-covered locations and non-TRIA locations? The Academy subgroup has no information on this, but the broker community may have some helpful information. Policyholder Demand 11. How has the demand for terrorism risk insurance changed since 2006, when the President’s Working Group on Financial Markets issued its last report? Please comment on take-up by policyholder sector, location, line, and other relevant characteristics. How have any changes in demand influenced the willingness of insurers to allocate capital to terrorism risk insurance? Has there been any impact on the amount of capital allocated to non-terrorism coverage or among lines of insurance? The Academy subgroup does not have specific information on take-up rates for terrorism insurance, but the broker community may have some helpful information. 12. To what extent have businesses used captive insurance companies to provide terrorism risk insurance, and what is the potential for the use of captive insurers to insure against such risk long-term? How have stand-alone terrorism captives developed, and how will these evolve long-term, including after the expiration of the Program in 2014? Captives typically do not have sufficient surplus to cover catastrophic events; however, they generally have enough surplus to provide some level of security. The Academy subgroup is not aware of any captive that has been set up specifically to provide terrorism coverage. TRIA and the Terrorism Risk Insurance Program Reauthorization Act (TRIPRA) require the offer of terrorism insurance on the same terms and conditions as for other perils covered by policies in the lines of insurance subject to these acts. To the extent that a captive is subject to the TRIA/TRIPRA mandatory offer provisions, and their insureds (owners) opt for the coverage, the captive is required to provide such coverage and is covered by the federal backstop. While TRIA/TRIPRA is in effect, a captive that had already been set up could have access to recoveries for terrorism losses at levels considerably lower than they would be had the same premium been written through a standard insurer, because the captive’s direct written premium 1850 M Street, NW Suite 300

Washington, DC 20036

Telephone 202 223 8196

Facsimile 202 872 1948

www.actuary.org

5


pertains solely to the limited exposure. This method offers maximum access to the benefits of the federal program for an otherwise self-insured entity. 13. Have state approaches (such as those applicable to mandatory coverage, permitted exclusions, and rate regulation) made coverage more or less available and affordable? Have there been any changes in state insurance regulation of terrorism risk insurance since the Terrorism Risk Insurance Program Reauthorization Act of 2007 was enacted? To what extent has the availability and affordability of terrorism risk insurance been influenced by state insurance regulation, and what role is state regulation expected to have long-term? Please comment on state-approved terrorism related rate loads. Some states initially disapproved original insurer terrorism rate filings and later approved those filings, once the rates had been reduced. Given the requirement of mandatory offer, such a state action has the effect of making terrorism insurance coverage more affordable. However, if the federal terrorism risk insurance program is permitted to expire, insurers would no longer be required to offer terrorism coverage to every client. In such a scenario, state disapprovals could reduce, perhaps considerably, the availability of terrorism insurance. Certain coverages, such as workers’ compensation, may be defined by state law in a manner that provides for terrorism risk insurance coverage. In that case, an insurer needing to limit its accumulation of terrorism risk exposure would have no means by which to do so, other than by avoiding the underlying exposure. If a state did not approve exclusions for terrorist attacks not covered by TRIA/TRIPRA, such a state action could expose insurers to very large losses and potentially affect the availability of underlying non-terrorism coverage. A state’s failure to approve terrorism exclusions could affect the financial solvency of the insurer and the insurer’s ability to pay claims (on terrorism or other losses). 14. What are the differences in availability and affordability of terrorism risk insurance between the licensed/admitted market and the non-admitted/surplus lines market, and to what degree are those differences attributable to the degree and manner in which each market is regulated? Given the “mandatory offer” provision of TRIA/TRIPRA, there are likely few differences in availability and affordability of terrorism risk insurance between the admitted and non-admitted markets. Price of Insurance 15. What improvements have taken place in the ability of insurers to price terrorism risk insurance? How are rating organizations assisting insurers in pricing, and how have rating factors developed? The Academy subgroup response concerns the ability of insurers to estimate costs associated 1850 M Street, NW Suite 300

Washington, DC 20036

Telephone 202 223 8196

Facsimile 202 872 1948

www.actuary.org

6


with potential terrorist attacks. Costs are one element of an insurer’s pricing decision, but pricing per se is outside the purview of the Academy subgroup’s response. A component of the pricing amount for a policy is the expected loss to the exposure. Probabilistic modeling is and has been used by some organizations as a starting point in the pricing process. One example includes advisory loss costs for terrorism developed by ISO and filed for use by insurers. The ISO values have been refined in terms of increased granularity of territorial definitions and in the number of different values employed across the different territories. These advisory loss costs reflect the detailed probabilistic modeling that differentiates risk by the estimated frequencies attributed to different potential targets and the amount of damages that might result due to terrorist attacks at those targets. Particular attention is placed on the differences in risk in areas exposed to well-known potential targets versus the risk in areas of widespread homogeneity. 16. What have been the trends in pricing of terrorism risk insurance? Please comment on the extent to which such coverage is not priced and charged-for. How has pricing changed since 2006, when the President’s Working Group on Financial Markets issued its last report? To what do you attribute any changes? Again, the Academy subgroup response concerns the ability of insurers to estimate costs associated with potential terrorist attacks. Costs are one element of an insurer’s pricing decision, but pricing per se is outside the purview of the Academy subgroup’s response. Since 2006, model-based loss costs used for pricing have been refined to distinguish among various levels of coverage. This pertains to the application of exclusions that may be in place in the underlying policy for non-terrorism events, including fire following terrorist attacks in standard fire policy states and coverages that may be specifically purchased in an individual policy. 17. How has the recent “soft market” impacted the availability of and affordability of terrorism risk insurance? What would be the impact on the availability and affordability of terrorism risk insurance should the market “harden” in near future? The committee does not have specific information on the availability and affordability of terrorism risk insurance. Generally, “soft markets” are characterized by excess capital in the insurance industry, which provides greater capacity to write insurance. Insurers in soft markets will tend to offer more coverage at lower prices. Conversely, a “hard market” refers to a scenario in which there is inadequate capital in the insurance industry. This reduces insurers’ capacity to write insurance. In “hard” markets, insurers will be less likely to write less profitable or more volatile coverage. Terrorism risk insurance is inherently volatile, relative to other lines, so it is reasonable to expect a reduction in the availability of terrorism risk insurance in harder markets. 18. How were primary insurers’ pricing decisions affected by the Terrorism Risk 1850 M Street, NW Suite 300

Washington, DC 20036

Telephone 202 223 8196

Facsimile 202 872 1948

www.actuary.org

7


Insurance Program Reauthorization Act of 2007, particularly as to the requirement to make available coverage for acts of terrorism being no longer defined as limited to those committed on behalf of any foreign person or foreign interest? The Academy subgroup’s response to this question concerns anticipated costs associated with the provision of terrorism risk insurance, which is an important element of pricing decisions. Individual insurer pricing decisions are outside the scope of the Academy subgroup’s response. Terrorism models have been parameterized to include many aspects of terrorism exposure as well as the federal backstop, including deductibles, co-insurance, event type, and whether an event is domestic or foreign. Reinsurance 19. What is the current availability and cost of reinsurance to cover terrorism risk? Please distinguish by line or type of insurance being reinsured and on what basis (treaty or facultative). How has the terrorism reinsurance market changed since 2006, when the President’s Working Group on Financial Markets issued its last report? To what do you attribute any changes? The magnitude of potential insurance claims due to terrorist events makes permanent federal legislation necessary to make terrorism coverage widely and readily available. The Academy subgroup is not in a position to provide a specific market analysis of reinsurance. We can, however, offer several general observations. First, we have seen no evidence that there is sufficient private reinsurance capacity to address the type of extreme events the Academy subgroup has modeled. Several of those events are an order of magnitude larger than the reported reinsurance capacity even under TRIA or TRIPRA. Without a federal framework for terrorism risk insurance, certain modeled events could be two orders of magnitude greater than reported reinsurance capacity. Second, standard reinsurance contract language often excludes terrorist acts covered by TRIA or TRIPRA and all “biological, chemical, or nuclear pollution or contamination.” Reinsurance markets face the same difficulties as primary insurers in pricing coverage in terms of the state of the art of catastrophe modeling tools. The available terrorism models are subject to great uncertainty. Thus, in the short term, reinsurers face significant challenges in quantifying their exposure to terrorism losses. This uncertainty will continue to limit available capacity. In the long term, the amount of private reinsurance capacity will be related to the confidence that the markets develop in their pricing tools and their understanding of risk. It would require a very significant increase in capacity for the private market to absorb the risk now covered by TRIA, even under TRIA’s $100 billion cap. Given current market conditions, it is difficult to envision the markets being able to generate significant additional terrorism reinsurance capacity in the short term.

1850 M Street, NW Suite 300

Washington, DC 20036

Telephone 202 223 8196

Facsimile 202 872 1948

www.actuary.org

8


20. At what policyholder retention levels are insurance programs being structured by policyholders to cover terrorism risk (e.g., deductibles, self-insurance, captives); and, with regard to insurers, how are reinsurance programs being structured and at what attachment points? Please comment on the availability and affordability of reinsurance for terrorism risk. The Academy subgroup does not have specific knowledge of the terms and conditions of various insurance programs. However, under the federal framework, terms and conditions for terrorism risk insurance need to be the same as the underlying coverage. Regarding the availability and affordability of reinsurance, please see the Academy subgroup’s response to Question 19. 21. Are reinsurers allocating more capital to terrorism risk insurance, and has capacity changed since 2006, when the President’s Working Group on Financial Markets issued its last report? Are insurers willing to pay the cost of terrorism risk reinsurance, and is that a factor affecting the allocation of capital to the risk; how much additional capital could be attracted long-term? Please see the Academy subgroup’s response to Question 19. 22. How have provisions of the Terrorism Risk Insurance Program Reauthorization Act of 2007 affected the terrorism risk reinsurance market? More specifically, how has maintaining and not increasing the insurer deductible percentage applied against direct earned premiums (from Program lines), as well as not decreasing the Federal share of losses above the insurer deductible, affected the provision and development of private reinsurance? The Academy subgroup does not have direct knowledge of the effect on the reinsurance market of the TRIPRA renewal in 2007. 23. To what extent have alternate risk transfer methods (e.g., catastrophe bonds or other capital market instruments) been successfully or unsuccessfully used for terrorism risk insurance, and what is the potential for the long-term development of these approaches? The Academy subgroup’s responses have benefited from the expertise of representatives of AIR Worldwide. AIR Worldwide has directly supported a large portion of the transactions for raising risk capital through catastrophe bonds and has modeled most of the catastrophe bonds issued as services provided to investors. Investors do not generally have the risk analysis expertise for extreme events that exists in insurance and reinsurance companies. Therefore, they look to the practices and risk assessments used by those companies, as well as to rating agencies, for guidance. Rating agencies have not indicated any willingness to use probabilistic terrorism loss models for ratings.

1850 M Street, NW Suite 300

Washington, DC 20036

Telephone 202 223 8196

Facsimile 202 872 1948

www.actuary.org

9


Claiming the same risk uncertainties cited by insurers and reinsurers regarding terrorism, as well as the fact that terrorism catastrophe bonds cannot be rated, investors have expressed little appetite for such investment vehicles to date. Thus, the issues limiting the availability of reinsurance for terrorism risk also limit the use of alternative risk transfer methods. Losses Associated with Chemical, Nuclear, Biological, and Radiological (CNBR) Acts 24. What is the current availability and affordability of coverage for CNBR events? For what perils is coverage available, subject to what limits, and under what policy terms and conditions? Is there a difference in the availability and affordability of coverage for CNBR events caused by acts of terrorism? To what extent have various States allowed insurers to exclude coverage for CNBR events (Please comment on requirements for workers’ compensation and fire-following coverage.)? How have exclusions developed? Response: TRIPRA clearly provides that coverage for terrorism is subject to the terms and conditions of the underlying policy. Thus, under the current federal program, coverage for CNBR events caused by terrorists depends on whether the underlying policy would have covered the peril absent terrorist involvement. Under commonly-used workers’ compensation policies, no exception applies to the applicability of coverage if the loss is due to a CNBR event. Such coverage would be available up to the full limits of the policy. Property policies are more complicated in CNBR scenarios. Commonly-used property policies have various provisions that exclude coverage for nuclear reaction, radiation, or contamination. However, damage from certain perils (fire, for example) resulting from a nuclear reaction may be covered. Property policies often contain specific exclusions that could apply in the event of a terrorist attack involving biological or chemical events. Whether coverage applies would depend on the specific facts associated with a particular loss event and the coverage stipulations included in the policy. If such coverage is found to apply, it would usually be available up to the full limits of the policy. Also, whether liability coverage applies in the event of a CNBR attack would depend on the coverage stipulations included in the policy and the specific facts associated with the event. In a post-TRIPRA environment, insurers would have available specific endorsements to exclude coverage for CNBR events initiated by terrorists. Industry use of such endorsements would reflect each insurer’s evaluation of the risk/reward trade-off associated with coverage of this peril. 25. Is it the case that some insurers appear unwilling to provide coverage for CNBR events caused by acts of terrorism, despite TRIA limits on an insurer’s maximum loss exposure? If so, why?

1850 M Street, NW Suite 300

Washington, DC 20036

Telephone 202 223 8196

Facsimile 202 872 1948

www.actuary.org

10


Individual insurer decisions whether to offer coverage are beyond the scope of the Academy subgroup’s response. However, insurers writing workers’ compensation are currently providing large amounts of coverage for CNBR events. Under TRIA/TRIPRA, individual insurer terrorism deductibles can be very large. Where insurers have the option not to provide CNBR coverage on the underlying policy, they may evaluate the potential premium for providing CNBR coverage as not being commensurate with the level of exposure. Availability of coverage for CNBR events has likely increased since 2006, due to insurer and reinsurer perceptions of no losses to date, low correlation with other catastrophic perils, and the availability of additional capital. 26. In the long-term, what are the key factors that will determine the availability and affordability of terrorism risk insurance coverage for CNBR events? The President’s Working Group on Financial Markets previously reported that there appeared to be little potential for market development. Has anything changed since 2006? The key factor is that the exposure is too great to underwrite. In the absence of a federal framework for terrorism risk, insurance coverage for terrorism risk is likely to be volatile, expensive, and of insufficient quantity. CNBR events can cause the largest losses of all terrorism risks. Given the magnitude of potential claims due to CNBR events and the tremendous uncertainty associated with evaluating the likelihood of such events, there are essentially two long-term scenarios: 1. Absence of a federal framework for terrorism risk insurance: In this case, there is likely to be a limited and volatile market for terrorism coverage for CNBR events. To the extent that state laws and regulations mandate inclusion of coverage for CNBR events caused by terrorists, these requirements are likely to reduce the availability of standard coverages. Even so, a terrorist attack using CNBR weapons in this scenario has the potential to cause massive insolvencies of standard insurers, complicating the task of national recovery from an already-devastating event. 2. Presence of a federal framework for terrorism risk insurance: If properly designed, a federal framework would allow for terrorism coverage to be widely available. While the underlying uncertainty about the frequency and severity of terrorist events would remain, the volatility of premiums for this coverage in a federal framework would be considerably less than its volatility in the absence of such a framework. There is one significant difference between the reinsurance and insurance marketplaces in the terrorism insurance context. In general, reinsurance coverages are not mandated by law or regulation to cover any particular perils. Thus, reinsurers are free to draft contracts that exclude coverage for claims their primary company clients must pay. On the one hand, this allows 1850 M Street, NW Suite 300

Washington, DC 20036

Telephone 202 223 8196

Facsimile 202 872 1948

www.actuary.org

11


reinsurers more power to manage their exposures, as their basic business model tends to attract substantial concentration risk. On the other hand, this means that primary companies cannot rely on casting off risks they may have felt forced to undertake. Again, assuming the lack of a federal framework for terrorism risk insurance, we see no prospect, even long-term, of a significant reduction in the uncertainty associated with estimating terrorism risk exposure. Accordingly, we see no prospect of any rapid increase in the amount of private capital invested in terrorism risk reinsurers. Deductible and Co-Share Levels 27. Under the Program, an insurer’s annual deductible is a percentage of certain direct earned premiums (as defined by TRIA and regulation). TRIA, as originally enacted, graduated the percentage applied for each year. The Terrorism Risk Insurance Program Reauthorization Act of 2007 established a set percentage of 20 percent for each Program year beginning in 2007. Please comment for each year since 2006 as to whether direct earned premiums in TRIA lines and insurer deductibles have increased or decreased? If so, in what amounts? Please provide data as available. The industry-wide deductibles under TRIA as of 2010 are: (1) (2) (3)

TRIA Program Year 2006 2007 2008 2009 2010

Previous Yr’s Earned Premium in 1,000s 202,908,056 175,098,424 176,855,955 168,061,365 157,969,859

Ded. 17.5% 20.0% 20.0% 20.0% 20.0%

Industrywide Ded. in 1,000s (1)*(2) 35,508,910 35,019,685 35,371,191 33,612,273 31,593,972

(4) Percent Change in Ded. (3) from Prior Yr. -1.4% 1.0% -5.0% -6.0%

Beginning in 2007, commercial automobile, burglary, theft, and surety insurance are not covered under TRIA. These totals are based on A.M. Best’s industry-wide, countrywide data. Individual insurer deductibles will vary based on their lines of business and premiums. 28. How might any increases to the insurer deductible level or decreases to the Federal share above such deductible levels, prior to the Program’s expiration in 2014, affect the availability and affordability of terrorism risk insurance? Please comment on the degree, amount or increment of any recommended increase. Insurer reaction to deductible changes and changes in the level of federal participation is unclear. If the changes are relatively modest, there is likely to be little impact on availability and 1850 M Street, NW Suite 300

Washington, DC 20036

Telephone 202 223 8196

Facsimile 202 872 1948

www.actuary.org

12


affordability of terrorism risk insurance. Changes that significantly affect an insurer’s potential exposure to a covered loss could result in increased costs or the reduced availability of coverage. Insurers constantly manage their total exposure based on available capital and reinsurance. Expiration of the Program 29. Describe efforts undertaken by the insurance industry and/or policyholders since 2006, when the President’s Working Group on Financial Markets issued its last report, to ensure the availability and affordability of terrorism risk insurance after 2014 when the Program expires, and long-term? The Academy subgroup knows of no formal efforts by the industry or by policyholders to ensure availability of coverage after expiration of TRIPRA in 2014. 30. Please comment on any anticipated state approaches to ensure the continued availability and affordability of terrorism risk insurance after the Program expires in 2014 (such as those approaches taken by the States after September 11, 2001 and before TRIA was enacted on November 26, 2002). The Academy subgroup knows of no formal efforts by states to develop mechanisms to ensure provision of terrorism insurance after 2014. Over the years, various states have developed mechanisms to help provide coverage for natural catastrophes. Examples include windpools in a number of states (e.g., North Carolina, Texas), state reinsurance pools (e.g., Florida), and direct insurance organizations (e.g., Citizens Property Insurance Corp. in Florida and the California Earthquake Authority). One concern is that states will not have the ability to gather the capital necessary to support these funds. In addition, the federal government has the ability to limit total industry and government liability (e.g., the $100 billion cap in TRIPRA). Without this type of authority, it would be difficult for states to create viable alternatives to TRIPRA. 31. Please comment on any other developments in markets that might affect the continued availability and affordability of terrorism risk insurance. The availability and affordability of terrorism risk insurance could be affected by a number of factors. Some of these include: • Actual pattern of terrorism events in the U.S. and abroad • New information that affects industry opinion about the frequency or severity of insured terrorism events • Natural catastrophe (e.g., hurricane and earthquake) events that are large enough to significantly affect industry capital • Other events that might affect industry capital (e.g., events in financial markets, changes to overall industry profitability) 32. In the absence of the Program, in what forms, at what levels, under what terms and conditions, and at what price might terrorism risk insurance be available; and, at what 1850 M Street, NW Suite 300

Washington, DC 20036

Telephone 202 223 8196

Facsimile 202 872 1948

www.actuary.org

13


duration (i.e., long-term)? Please distinguish from state-mandated coverage, such as workers’ compensation and fire insurance. While it is difficult to know exactly what will happen in the absence of the Program, the expectation is that there will be a significant reduction in the availability of coverage. Without the federal backstop, there will be insufficient capital to support a number of plausible terrorism events. As a result, insurers will need to reduce their exposure to terrorism events, particularly large terrorism events. This could result in a number of actions: • • • •

Changes in deductibles, limits, or available coverage for a number of terrorism events, particularly in locations in which the insurer has a concentration of insureds, or events are more likely or more expensive Non-renewal of accounts to limit insurer exposure to the most expensive events and control concentration In cases of mandatory terrorism coverage (workers’ compensation and fire in certain states), a reduction in the number of insureds in which the insurer provides any coverage at all to control exposure to the most expensive events As a result of the capital requirements set forth by rating agencies, insurers will be discouraged from writing terrorism coverage

The Academy subgroup appreciates the opportunity to provide comments to the President’s Working Group in response to the request appearing in the Federal Register of June 17, 2010. Please do not hesitate to contact us through Lauren Pachman (Pachman@actuary.org), the Academy’s casualty policy analyst, should you have any questions concerning our comments. Sincerely,

William VonSeggern Chairperson, Terrorism Risk Insurance Subcommittee American Academy of Actuaries

1850 M Street, NW Suite 300

Washington, DC 20036

Telephone 202 223 8196

Facsimile 202 872 1948

www.actuary.org

14


simultaneously and recalibrated and re-optimized their portfolios to share the risk with their competitors in the market. 7. Such other information that may be necessary or appropriate for the Report. Academy Committee response: We have nothing further to add at this time. ****************************************

We hope these comments help the FIO in its study of natural catastrophes and the current state of the market for natural catastrophe insurance in the United States. We would be pleased to discuss these issues further and/or answer any questions you have related to this letter. If you have any questions about our comments, please contact Lauren Pachman, the Academy’s casualty policy analyst, at Pachman@actuary.org or (202) 223-8196. Sincerely,

Michael E. Angelina, ACAS, MAAA, CERA Vice President, Casualty Practice Council American Academy of Actuaries Stu Mathewson, FCAS, MAAA, CPCU Co-Chair, Extreme Events Committee American Academy of Actuaries

Jeff McCarty, FCAS, MAAA, CERA Co-Chair, Extreme Events Committee American Academy of Actuaries

13

Summer_Summit_Advance_Materials  
Read more
Read more
Similar to
Popular now
Just for you