Women's Summit 2008

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Prepared by: Harrison S. Campbell, Jr. Associate Professor of Geography The University of North Carolina at Charlotte Charlotte, North Carolina 28223

Prepared for: The University of North Carolina at Charlotte Urban Institute And Charlotte-Mecklenburg Women’s Summit

August 2007

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Acknowledgements This report is part of a larger conversation about the social and economic status of women in Charlotte-Mecklenburg. Initiated by Liz Hair and Jennifer Roberts, Lisa Lewis Dubois Chaired the Steering Committee and, with support from the John S. and James L. Knight Foundation, set the course to more closely study key indicators of women’s status and the economic impact of gender-based inequality. My involvement in this discussion would not have come about had it not been for the leadership, advice, assistance and collegial support of several individuals who deserve acknowledgement. Pilar Zuber initiated my involvement in this endeavor, for which I am grateful. Nelse Grundvig and Pam Davenport at the North Carolina Employment Security Commission were generous with their time and ideas; our many useful conversations substantially improved the report’s overall quality.

Of course, without the talented work of many people at the UNC Charlotte Urban Institute this work would never have seen the light of day. Most of the graphics, especially the maps, were developed by the capable hands of Jonathan Kozar. Anne-Marie Mills and Lisa Shepard not only did a fine job with production but unflinchingly endured my incessant questions. Project Manager Elizabeth Fenner guided production of the report and the Women’s Summit project. Linda Shipley, Senior Associate Director of the UNC Charlotte Urban Institute, was the liaison between the University and the Women’s Summit. Of course, this report is just one example of the leadership provided by the Urban Institute’s Director, Jeff Michael, without whom, none of this would be possible. As usual, any errors or omissions are solely my responsibility.

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment / Acknowledgments

Table of Contents

Acknowledgments Executive Summary.......................................................................................................... i I. Introduction .................................................................................................................1 Caveats .........................................................................................................................2 About the Data ............................................................................................................3 Outline of the Report ..................................................................................................3 II. A Brief Profile Of Working Women in Mecklenburg County..............................4 Women’s Earnings ......................................................................................................4 Industry and Occupations .........................................................................................6 Women’s Labor Force Participation.........................................................................9 Women in the Workforce ........................................................................................11 Educational Attainment ...........................................................................................12 Poverty and Education.............................................................................................14 Poverty, Work and Children ...................................................................................17 Profile Summary .......................................................................................................18 III. The Economic Impact of Gender Inequality .........................................................19 The Earnings Gap......................................................................................................19 The Labor Force Participation Gap ........................................................................20 Economic Impact to Mecklenburg County ...........................................................21 Economic Output Impacts .......................................................................................22 Employment Impacts ...............................................................................................23 Household Earnings Impacts ..................................................................................23 Tax Impact..................................................................................................................23 Incremental Improvements .....................................................................................23 Impact Summary.......................................................................................................24 IV. Summary and Conclusions......................................................................................25 Appendix Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment Table of Contents

Figures and Tables

FIGURES Figure 1. Median Earnings for Full‐Time Workers, 2005......................................................4 Figure 2. Median Earnings for Women, 2005 price levels (Map) ........................................5 Figure 3. Earnings Ratio, 2000 (Map).......................................................................................5 Figure 4. Womenʹs Full‐Time Earnings by Race/Ethnicity, Mecklenburg 2005.................6 Figure 5. Women in Managerial Positions, 2000....................................................................7 Figure 6. Women in Managerial Positions, 2000 (Map) ........................................................8 Figure 7. Womenʹs Occupational Concentration ...................................................................8 Figure 8. Labor Force Participation, 2005................................................................................9 Figure 9. Womenʹs Labor Force Participation (%), 2000 (Map) .........................................10 Figure 10. Womenʹs Labor Force Participation by Race/Ethnicity, Mecklenburg 2005....10 Figure 11. Percent of Population Age 25+ with Bachelorʹs Degree of Higher ..................12 Figure 12. Women with Bachelorʹs Degree of Higher (%), 2000 (Map) ..............................12 Figure 13. College‐Educated Women Age 25+ in Mecklenburg by Race/Ethnicity (%), 2005 ............................................................................................................................13 Figure 14. Poverty Rate of the Population Age 16+, 2005 ....................................................14 Figure 15. Women in Poverty (%), 2000 .................................................................................15 Figure 16. Mecklenburg Poverty Rate by Educational Attainment, 2005 (%) Population Age 25+ .................................................................................................16 Figure 17a. Educational Attainment of Men in Poverty, Mecklenburg 2005 .....................16 Figure 17b. Educational Attainment of Women in Poverty, Mecklenburg 2005................16 Figure 18. Cumulative Commute Times, Mecklenburg 2005...............................................18

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment Figures and Tables

Figures and Tables

TABLES Table 1.

Average Monthly Earnings by Selected Industry, Charlotte‐Gastonia‐Concord NC‐SC MSA (NC Part), 2Q 2006...........................7

Table 2.

Educational Attainment of the Population Age 25+

by Sex, Mecklenburg 2005 (%) ...............................................................................13

Table 3. Poverty Rate of Population Age 16+, 2005 (%).....................................................14 Table 4. Work Status of Women in Poverty Age 16+, 2005 (%)........................................17 Table 5. Children in Poverty, 2005 .......................................................................................17 Table 6.

The Gender Wage Gap in Mecklenburg County, 2005.......................................20

Table 7. Labor Force Participation Gap, Mecklenburg County 2005...............................21 Table 8. Economic Impact of Closing the Gender Gap in Mecklenburg County ..........23 Table 9.

Economic Impact of Incremental Gender Gap Progress ....................................24

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment Figures and Tables

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Executive Summary The primary purpose of this report is to estimate the economic impact of equalizing malefemale workforce differences in Mecklenburg County, North Carolina. The report addresses two major questions: •

If women living in Mecklenburg County received the same earnings from work as men, how much more earnings would they realize and what economic impact would it have on the county? And

•

If women living in Mecklenburg County participated in the labor force at the same rate as men, how much more additional earnings would accrue to them, and what impact would it have on the county economy?

Motivated and patterned after a similar statewide report performed in South Carolina, we find that although women in Mecklenburg fare better in the labor market than state or national averages, male-female differences in the labor market persist. Removing these barriers and inequalities would have a significant impact on women and the Mecklenburg economy.

A brief profile of working women in Mecklenburg County reveals: •

Among full-time wage earners in 2005, women had median earnings of $34,171 compared with men’s median earnings of $45,048. Therefore, women’s earnings were 76 percent of men’s median earnings (i.e. 24 percent lower).

•

Women’s labor force participation, though rising, is still well below that of men. In 2005, women in Mecklenburg had a labor force participation rate of 66.9 percent versus 81.8 percent for men.

•

Overall, women, especially African American women, had lower rates of educational attainment than men which contribute to higher overall rates of poverty in the female population.

•

Women’s lower earnings contribute to poverty which is highly correlated with educational attainment. Over 60 percent of women in poverty have no education beyond high school.

Highlights of the economic impact associated with eliminating gender-based differences in the workforce suggest: •

If women currently holding full-time jobs had comparable earnings to men, they would receive $1.7 billion more in earnings each year.

•

If women’s labor force participation were equal to that of men, there would be 26,112 more women employed full-time earning a total $1.2 billion annually. Further, an additional 16,910 women would be working part-time earning an additional $212 million. Total new earnings from closing the participation gap would amount to nearly $1.4 billion.

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment / Executive Summary

Page i

•

After accounting for taxes and out-of-county spending, new consumption expenditures fueled by women’s new earnings would expand the county’s economic output by $3.1 billion, create 28,887 new jobs and generate nearly $1.1 billion in additional earnings to Mecklenburg households.

•

Based on these estimates, complete closure of the earnings and participation gaps would generate about $160 million in tax revenue to the city of Charlotte and Mecklenburg County annually.

•

Even modest, incremental improvements in the status of working women would generate significant economic benefits.

The report concludes that: •

Closing the gender gap in Mecklenburg County is feasible but policy makers in the public and private sectors need to be involved.

•

In the short-run, private sector initiatives might focus on policies that provide on-thejob training, skill acquisition, flexible work schedules, on-site or near-site day care and before-tax contributions to child care funds. Employers might realize benefits from worker training cost savings, added productivity and engendered worker loyalty.

•

As part of a long-run solution, public policies might focus on programs that help women gain more formal education and training, especially in high-wage occupations and industries where women are currently underrepresented.

•

Interruptions in work history disproportionately affect women to the detriment of their seniority and accumulated work experience. Hence, employers might also explore strategies to smooth the transition back to work after periods of leave. Such strategies might include workplace re-entry seminars or periodic workshops that prepare returning workers after a period of leave.

•

The greatest and most equitable approach to closing the gender gap would be to target policies and outreach to African American and Hispanic / Latino women who experience the largest discrepancies in the workforce.

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment / Executive Summary

Page ii

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment I. Introduction The primary purpose of this report is to estimate the economic impact of equalizing malefemale workforce differences in Mecklenburg County, North Carolina. For as long as official records have been kept, women in the civilian workforce have earned less than men and experienced lower rates of labor force participation. Consequently, this report addresses two major questions: 1. If women living in Mecklenburg County received the same earnings from work as men, how much more earnings would they realize and what economic impact would it have on the county?

2. If women living in Mecklenburg County participated in the labor force at the same rate as men, how much more additional earnings would accrue to them? What impact would it have on the county economy?

The combined impact associated with these two questions yields an estimate of the total impact to Mecklenburg County.

In many ways, the economic analysis contained in this report is hypothetical.

What if… women and men had identical earnings?

What if… labor force participation were equalized?

In truth, there are many reasons for the inequalities—education, training, work force experience, occupation, industry of employment, family responsibilities, etc.—and an analysis of the reasons for inequality are beyond the scope of this report.

The bare truth, supported by the raw data, is that women and men are not equals in the labor force. This report does not query all the reasons for inequality, but as shown in the next section, the differences are sometimes striking.

This report has been motivated by a similar statewide report performed in South Carolina (Schunk and Teel, 2005)1 and is largely patterned after it. Though there are significant differences between the two reports, the research methodologies are essentially the same.

It is worth noting here, that in many ways, the working status of women in Mecklenburg County compares favorably to the status of working women in other parts of the state and the nation as a whole. That is, in Mecklenburg County, women earn relatively high wages.

The difference between men and women in terms of labor force participation, educational attainment, poverty, etc. are not as wide as they are in North Carolina or nationally. We suspect that much of women’s success in the local labor market is related to the underlying condition and structure of the local economy. Mecklenburg County has a healthy economy, with a large proportion of jobs in relatively high-wage service sectors that are not too vulnerable to swings in the business cycle. A relatively large public sector also contributes to relative labor force equity. However, as shown below, differences exist and equalizing those differences would have a significant impact on the local economy, as well as the lives of working women in the county.

1 Donald L. Schunk and Sandra J. Teel (2005). The Status of South Carolina’s Women, Moore School of Business, University of South Carolina.

Page 1 Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Caveats Just as important as the questions addressed here are the issues not addressed in this report.

Some of these include:

2

•

When examining differences in earnings or labor force participation, the report does not control for differences in educational attainment or work experience. That is, some portion of the wage difference between women and men is due to differences in accumulated human capital. The purpose of this report is not to ask why these differences exist; rather, we simply note the differences and analyze the impact of neutralizing those differences.

•

It is likely that some of the differences between men and women could disappear over time with appropriate human capital investments. However, the cost of those investments and the means by which they are delivered are policy questions beyond the scope of this report.

•

One would also expect that women’s advances in education and training would have more than an immediate pay-off. Indeed, such acquisitions would yield benefits well into the future. Typically, the future value of current education and training investment is assessed via “age-earnings” profiles, stratified by educational attainment.

Unfortunately, current data at the local level are insufficient for such an analysis.

However, it is safe to say that the estimates contained in the report are probably rather conservative given that no account is made for the future earnings trajectory of women with added human capital.

•

A variety of family issues are mentioned, but not analytically addressed. While data reflecting work force and marital status are presented along with poverty rates for men, women and children, these issues are not systematically addressed in an analytical fashion. There is no doubt, however, that one’s marital status and family circumstances affect earnings and labor force participation (Rose and Hartmann, 2004)2. For women, this frequently implies work histories punctuated by periods of labor force inactivity.

•

If, according to our scenarios, women attach themselves to the labor force at the same rate as men, there must be jobs for them, lest higher rates of unemployment will result.

In this report we do not indicate how these new jobs would be created, in what sectors of the economy they would be found, or how women would go about filling those new jobs. As shown below, the local economy would have to expand by about 8 percent to absorb the additional labor force implied in our results. Interestingly, in the aggregate, the number of new jobs required would be roughly equal to three years of normal growth in Mecklenburg County. However, further expansion of the local economy would be driven by the spending of new wages accruing to women. While achieving this kind of growth is entirely plausible, this report does not indicate what types of jobs they would be, or where they would be found.

Stephen J. Rose and Heidi I. Hartmann (2004). Still a Man’s Labor Market: The Long Term Earnings Gap, Washing-

ton, DC: Institute for Women’s Policy Research.

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 2

•

Finally, there are important supply-demand issues that are not considered. Specifically, if women’s wages rise to equal those of men, the overall “price of labor” would rise.

This might have two effects in the labor market. First, we would expect female labor force participation to rise, increasing the supply of labor in the county as the returns to their labor services increase.3 Working against this, however, might be the long-run response from local employers. As wages rise, local businesses might reduce demand for all types of labor (women and men) in response to increasing labor costs. The forces of supply and demand work in opposite directions and the final outcome is difficult to predict with certainty.

About the Data The data used in this report are the most current and reliable available at this time. They have been collected from a variety of sources but the most heavily used data come from the 2005 American Community Survey published on-line at the U.S. Census Bureau. Most of the analytical/ impact discussion is based on these data as is much of the descriptive discussion contained in the following section. The 2005 American Community Survey provides estimates of many Census variables for off-census years. However, reliable estimates are not currently available for all counties in all states. We are fortunate, however, that reliable data are available for the largest of places, including Mecklenburg County. Data from the 2005 American Community Survey are available for the U.S., North Carolina and 37 of the state’s 100 counties.

Whenever possible these data have been used. Because data for nearly two-thirds of the state’s counties are not available for 2005, the maps generally show data collected from Census 2000, some of which have been price-updated to 2005. In other cases, specific data items have been collected from the North Carolina Employment Security Commission; the Bureau of Economic Analysis; the Office of Management and Budget; and the Census of Governments, Local Government Finances of various years. Each data source is thoroughly referenced in the report.

Outline of the Report The remainder of this report is organized as follows: Section II contains a profile of women and work in Mecklenburg County. This descriptive discussion presents data that provide both a context for describing the attributes of women in the work force and illustrate differences from their male counterparts. In general, the workforce situation in Mecklenburg is compared to national and state averages. A somewhat different picture is revealed in a variety of maps that compare Mecklenburg County to other counties in the state. Similarly, many variables are disaggregated by race and ethnicity to show how important workforce characteristics vary across the population of working women. Some of these data form the basis of the economic impact analysis presented in Section III which answers the primary questions this research seeks to address. The analytical section not only presents the economic impact of reducing male-female gaps in the labor force, but defines a variety of terms to clarify the analysis. The modeling procedure is briefly discussed while a more complete and technical description of impact analysis can be found in the appendix to this report. The report concludes with a brief summary of the findings. While it is not the intent of this report to make policy recommendations, a few policy areas are highlighted that policymakers might consider if gender-based disparities are to be redressed.

3 Other things equal, such a supply response might actually serve to lower wages.

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 3

II. A Brief Profile Of Working Women in Mecklenburg County To provide some context for analysis, this section presents a profile of Mecklenburg County women in the work force. We start with a discussion most pertinent to this analysis, a comparison of earnings of labor force participation. From there, the discussion moves to some of the implications of gender-based inequalities focusing on education, work and poverty.

Women’s Earnings The differences between women and men in their 2005 median annual earnings in Mecklenburg, the state and the U.S. are shown in Figure 1. Immediately obvious is that while local earnings are higher than state or national averages, a significant a wage gap exists at all levels. In 2005, median earnings for women working full-time and living in Mecklenburg County were $34,171 while men earned $45,048. Consequently, earnings for the typical woman working full-time were 75.9 percent of men’s earnings—very close to the national average of 76.7 percent. Across the state, full-time women and men tend to earn less overall though women’s earnings fared slightly better when compared to men (79.4 percent). Differences in the earnings gap can be attributed to men’s success in the local labor market. For example, local women working full-time earned 14.9 percent more than women statewide, and 6.2 percent more than women nationally. The same figures for men are 20.3 percent and 7.3 percent.

Figure 1. Median Earnings for Full-Time Workers, 2005 $50,000 $40,000

$45,048 $34,171

$41,965 $37,441 $29,729

$32,168

$30,000

Female Male

$20,000 $10,000 $0 Mecklenburg

NC

US

Source: US Census, 2005 American Community Survey, Table B20017

Compared to other counties, women in Mecklenburg receive relatively high full-time earnings (Figure 2).4 The clear pattern to immerge from this map is that earnings in the urban areas tend to be higher than non-urban earnings, especially those non-urban areas in the most remote western parts of the state and the state’s Inner Coastal Plain near Virginia’s Tidewater. Bertie and Tyrell Counties, for example, have the state’s lowest full-time earnings for women at just $20,796 and $20,892, respectively. Similarly, women’s full-time earnings are relatively low in some of the state’s agricultural counties like Sampson, Duplin and Robeson. As shown elsewhere,5 income in many of these lower earnings counties is disproportionately derived from transfer payments such as Social Security and disability payments. On the other hand, women’s median full-time earnings 4

Note, data in Figure 2 are from Census 2000 and have been price-updated to 2005.

5 See Harrison S. Campbell, Jr. (2003). “Unearned Income and Local Employment Growth in North Carolina: An Economic Base Analysis,” Southeastern Geographer, 43(1): 89-103.

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 4

in the urbanized areas of Charlotte, Greensboro and Raleigh-Durham are high relative to state averages.

Figure 2. Median Earnings for Women, 2005 price levels

Higher median earnings for women do not always translate into smaller disparities between women and men as shown in Figure 3. While total earnings for women are relatively high in the state’s urban areas, they are higher still for men, exacerbating the earnings dierential in larger counties such as Guilford, Wake and Mecklenburg. It is interesting to note that earnings appear more equal in some of the state’s lower earnings counties such as Onslow, where a high proportion of all workers are found in the retail sector.

Figure 3. Earnings Ratio, 2000

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 5

Within Mecklenburg County median wages vary noticeably by race and ethnicity (Figure 4).

Non-White women realize significantly lower wages. African American women, who earn 17 percent less than the female average, earn 37 percent less than White women and the disparity is even greater among Asian and Hispanic/Latino women.

$23,499

$23,194

$26,123

$30,000

$29,212

$40,000

$34,171

$50,000

$40,073

Figure 4. Women’s Full-Time Earnings by Race/Ethnicity, Mecklenburg 2005

All White African American Asian

$20,000

Other Hispanic

$10,000 $0 Mecklenburg Source: U.S. Census Bureau 2000, Table PCT86.

Industry and Occupations Women’s earnings are, in part, related to the industries and occupations in which they work.

Even when women are employed in high-wage industries and occupations their earnings do not always equate to those of men. Using a different set of statistics, Table 1 shows average weekly earnings for women and men by selected industry in the Charlotte metro area during the second quarter of 2006. Though not directly comparable to the earnings data presented above, these data show that the industry in which women find work directly affects their earnings. For example, women receive their highest average monthly earnings in the Finance and Insurance industry ($4,747) and their lowest earnings in the Retail sector ($1,789) with average monthly earnings of $3,001 across all sectors. However, even in the high wage sectors of Finance and Insurance or Health Care and Social Assistance women’s earnings are half (or less) that of men and overall, women’s earnings are only 62 percent of men’s.

Turning to the occupational characteristics of women, we note that according to one source,6 99 percent of all secretaries and 97% of all childcare workers are women while 88 percent of non-physician health care workers and 66 percent of food services workers are women which is thought to contribute to the “feminization” of poverty. However, taking women as a group, and examining their occupational characteristics locally, is somewhat more encouraging. In 2000, for example, 41.5 percent of employed Mecklenburg women held Management, Professional and Related occupations, which compares favorably to state and national averages (Figure 5).

Across the state (Figure 6) we see that counties in which a relatively large number of women hold managerial positions roughly correlates with median full-time earnings shown in Figure 2. In fact, if we examine the concentration of women in some of the top managerial positions, as shown in Figure 7, we see that when women in Mecklenburg are compared to the national average, they are 70 percent more likely to hold top executive positions and twice as likely to be financial managers.

6

J. W. Harrington and Barney Warf (1995). Industrial Location: Principles, Practice & Policy, London: Routledge.

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 6

This somewhat mixed picture of women’s equity in the local economy helps reveal some of the factors affecting their earnings. Clearly, the rewards from work affect one’s labor force participation, a subject to which we now turn.

Table 1. Average Monthly Earnings by Selected Industry, CharlotteGastonia-Concord NC-SC MSA (NC Part), 2Q 2006 Monthly Earnings Women Men W/M Ratio $3,001 $4,845 0.62 $3,348 $4,874 0.69 $1,789 $3,002 0.60 $4,747 $9,385 0.51

Industry Total Manufacturing Retail Finance and Insurance Professional, Scientific and Technical Services Health Care and Social Assistance Public Administration

$3,696

$6,271

0.59

$3,028 $2,876

$7,494 $3,275

0.40 0.88

Source: North Carolina Employment Security Commission and US Census Bureau Local Employment Dynamics Program (LED), QWI Online, accessed 08/11/07 See http://lehd.did.census.gov/led/datatools/quiapp.html

Figure 5. Women in Managerial Positions, 2000 41.5

42

Percent of Women

40 38 36.2 35.1

36 34 32 30 US

NC

Mecklenburg

Source: U.S. Census Bureau 2000, Table PCT86.

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 7

Figure 6. Women in Managerial Positions, 2000 (Map)

Figure 7. Women’s Occupational Concentration

1.7

Top executives

0.8

1.2

Operations specialtists

Mecklenburg NC

1.0

2.0

Financial managers

0.9

0

0.5

1

1.5

2

2.5

Concentration Ratio (US = 1.00) Source: U.S. Census Bureau 2000, Table PCT86.

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 8

Women’s Labor Force Participation Labor force participation and the labor force participation rate are important indicators of labor force “attachment.” To be considered part of the labor force, one must be (a) age 16 or older and (b) employed, or unemployed and actively seeking work. Those who are not employed and not seeking work (e.g. retirees, housewives, the severely disabled, or those discouraged from work) are not counted as part of the labor force. The number of work force “eligibles” are those 16 years old and above. Thus, the labor force participation rate is simply the percentage of work force eligibles that are part of the active labor force.

While women generally have lower rates of labor force participation than men, the difference has narrowed steadily in the last 50 years. Schunk and Teel (2005) indicate that in 1948 “… the male participation rate was 86.6 percent while the female participation rate was 32.7 percent.

By 2004, the male participation rate had fallen to 73.3 percent while the female rate had risen to 59.2 percent” (p. 36). There are many reasons for the increase in women’s labor force participation including rising levels of educational attainment, smaller average family sizes and better employment prospects overall. Women in North Carolina have long been especially active in the labor force (see Inset). This is particularly true in urban areas. As shown in Figure 8, labor force participation in Mecklenburg (for both men and women) is higher than state and national averages. Though the gender gap persists, in 2005, labor force participation rates for women and men in Mecklenburg County were 66.9 percent and 81.8 percent, respectively.

These are particularly high rates of labor force participation, especially in light of state and national averages. For example, in 2005 the comparable rates for women and men in North Carolina were 59.8 percent and 72.7 percent, very close to the national averages of 59.0 percent and 72.6 percent.

Figure 8. Labor Force Participation, 2005 81.8

Precent in Labor Force

90 80 70

72.7

66.9 59.8

72.6 59.0

60 50

Women

40

Men

30 20 10 0

Mecklenburg

NC

US

Source: Author calculations based on US Census, 2005 American Community Survey, Table 23001

In Figure 9 we see the familiar urban-non-urban pattern emerge. In urban counties where populations are younger and the returns to work are greater, labor force participation among women is stronger. Still, in no county does the labor force participation rate of women exceed that of men. Participation rates also vary by race and ethnicity which is shown for Mecklenburg County in Figure 10. While White women comprise the largest segment of the local labor force, their rates of participation (62.1 percent) are slightly below the county-wide average. African Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 9

American women, on the other hand, show the highest rates of participation at 74.2 percent; Hispanic and Latino women register the lowest rates of participation at 56.0 percent. Thus, the overall welfare of women not only depends on their (expected) earnings from work, but on their propensity to engage in the formal labor market. Such variations, by sex and race/ethnicity, have important implications for women’s economic welfare.

Because participation in the labor force is partly conditioned on expected earnings, and earnings are known to increase with educational attainment, it makes sense to examine the educational characteristics of the population.

Figure 9. Women’s Labor Force Participation (%), 2000

Figure 10. Women’s Labor Force Participation by Race/Ethnicity, Mecklenburg 2005

Percent in Labor Force

80 70 60

74.2 66.9

62.1

57.0

56.0 All

50

White

40

African American

30

Other

20

Hispanic

10 0 Mecklenburg

Source: US Census Bureau, 2005 American Community Survey, Tables B23002, B23002A-

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 10

Women in the Workforce One of the major changes that has occurred in the national labor force has been the increasing number of women who work outside the home. This is expressed in the female labor force participation rate, the percentage of adult women who are active in the labor force. Female participation in the national labor force has risen from 31.5 percent in 1970 to 58.8 percent in 1994, compared with the male participation rate of 51.2 percent in 1970 and 75.1 in 1994. The working woman is not a new trend in North Carolina, however. In 1976, fully 54.3 percent of adult North Carolina women were in the labor force and that proportion has risen steadily, reaching 60.4 percent in 1994. Thus, the difference between North Carolina and the nation has narrowed in recent years. Put another way, in 1976, among the 50 states, North Carolina ranked fourth, trailing only Nevada, Alaska and Hawaii in terms of female labor force participation. By 1994, while still ahead of the national average (58.8 percent), North Carolina ranked 27th among the 50 states and the District of Columbia.

The high level of working women has its roots in the nature of the state’s dominant industry, textile manufacturing. From the outset in the late 19th century, most mills located in largely agricultural areas and entire families sometimes left the farm to work in the mills. Men, women, and children were given jobs in the mills, but women were soon channeled into sex-typed jobs. As something of a carry-over from the farm, men were given jobs involving heavy work and authority. They were also expected to be long term workers, justifying the expense of longer training for more complex, higher wage tasks. Women were considered to be more patient and neat workers who were better suited for working on fast-moving, repetitive machines. Winding and spinning operations were almost always limited to females, for example. An early “glass ceiling” existed also. Some low level managerial positions opened up in the 1930’s when new management techniques called for more precise record keeping. But one report indicated that there was only one known female overseer in the entire South in 1935. Typically, women’s wages were only 60 percent of those paid to men.

Initially, many female mill hands were young, unmarried, or widows, but this changed after child labor was outlawed. In 1930, less than half of the female mill workers in North Carolina were married but this proportion rose to 72 percent by 1940. The average age also rose and by 1940 nearly 64 percent were between 25 and 44 years of age. The primary reason that so many women worked was that wages were low for men and women. Everybody who could had to work in order for the family to make ends meet. According to Jacquelyn Down Hall’s history of the North Carolina textile industry, Like A Family, “As long as mill work paid Just a niff to Keep Sole and Body to gather, families could survive only by pooling the wages of everyone over the legal working age of sixteen.” And women also were expected to run their households and nurture their children even after working long hours at the mill. Thus, having a career and a family is nothing new to the women of North Carolina.

Source: Alfred W. Stuart (2000). The North Carolina Atlas: Portrait for a New Century, Chapel Hill: UNC Press, p. 129.

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 11

Educational Attainment Directly related to work, earnings and labor force participation is the educational attainment of the population. For decades, levels of education throughout North Carolina have lagged behind national levels. As shown in Figure 11, the state still lags behind national averages, but such is not the case in Mecklenburg County. In 2005, 37.2 percent of women and 41.0 percent of men attained at least a college-level education, compared to the national rates of 26.0 percent and 28.5 percent. In North Carolina, 24.5 percent of women and 25.8 percent of men have earned a Bachelor’s degree or higher. As a basic form of human capital acquisition, and an important ingredient to economic growth, educational attainment is among the top priorities of policymakers in virtually every state in the nation. Just as gender differences in labor force participation have narrowed, so too have differences in educational attainment. Today, women are more likely to both attend and graduate from college, causing the education gap to narrow over time. Just as there are gender-based differences in education, there are urban-rural differences (Figure 12). Generally, the highest rates of educational attainment among women are found in urban counties with substantial high-order service, government, and education sectors.

Figure 11. Percent of Population Age 25+ with Bachelor’s Degree of Higher 41.0

Mecklenburg

37.2 Men

25.8

NC

Women

24.5 28.5

US

26.0 0

5

10

15

20 25 Percent

30

35

40

45

Source: Author calculations based on US Census Bureau, 2005 American Community Survey, Table B17001.

Figure 12. Women with Bachelor’s Degree of Higher (%), 2000

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 12

A slightly more detailed look at educational attainment in Mecklenburg County is shown in Table 2. The table reflects a higher propensity for males to drop out before completing high school, and an elevated tendency for women to attain Associate’s degrees from community colleges (included in “College 1–3 Years”). However, men, on average, are more likely to attain the highest levels of education (Bachelor’s degrees and advanced, professional degrees including the Master’s). While recent trends in college matriculation appear to work in women’s favor, the present population of women is over-represented at the Associate’s degree level and somewhat under-represented at higher levels. These patterns are especially pronounced when we account for race and ethnicity (Figure 13). While 44.5 percent of White women have earned at least a Bachelor’s degree, only 22.8 percent of African American women have done so. Recall, that African Americans also have the highest labor force participation rate among women and relatively modest full-time earnings. This indirect evidence suggests that many African American women have significant need (as evidenced by high labor force participation) but insufficient means (from low educational attainment) to put their earnings on par with White women, much less White men.

Table 2. Educational Attainment of the Population Age 25+ by Sex, Mecklenburg 2005 (%) Female < High School High School Graduate College 1-3 Years Bachelor’s Degree Advanced

Male 10.6 22.1 30.1 27.1 10.1

11.7 21.2 26.2 28.1 12.8

Source: Author calculations are based on US Census Bureau, 2005 American Community Survey, Table B15002. Note: “College 1–3 Years” includes those with some college and those with Associate degrees. “Advanced” includes Master’s, Professional and PhD Degrees.

Figure 13.

College-Educated Women Age 25+ in Mecklenburg by Race/Ethnicity (%), 2005

37.2

All Women

44.5

White African American

22.8

Asian

38.8

Other

14.3

Hispanic

18.1

0

10

20

30

40

50

Percent of Pop. Age 25+ Source: Author calculations are based on US Census Bureau, 2005 American Community Survey Tables 15002, 15002A-I

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 13

Poverty and Education The observations above are supported in Figure 14 and Table 3 that present poverty rates for men and women, and womenâ&#x20AC;&#x2122;s poverty rates by race/ethnicity. Nationally, 13.6 percent of women over the age of 16 were in poverty in 2005 and were 40 percent more likely than men to experience poverty. At the state level, women were 46 percent more likely to live in poverty. The good news from Table 3 is that poverty rates among women in Mecklenburg are substantially lower than state or national averages; but Mecklenburg women are still nearly 37 percent more likely to live in poverty than men in the county. Further, most susceptible to poverty conditions are African American women whose poverty rates are approximately three times that of White women.

Figure 14. Poverty Rate of the Population Age 16+, 2005 20

15.6 13.6

Percent

15

11.2

10.7

9.7

8.2

10

5

0 US

NC

Male

Mecklenburg

Female

Source: Author calculations based on US Census Bureau, 2005 American Community Survey, Table B17001.

Table 3. Poverty Rate of Population Age 16+, 2005 (%) US Male Female

9.7 13.6

NC 10.7 15.6

White African American Asian Other Hispanic

11.1 24.9 11.7 23.3 22.7

11.5 26.3 14.6 36.5 31.8

Mecklenburg 8.2 11.2 6.8 18.9 NA NA 17.3

Source: Author calculations are based on US Census Bureau, 2005 American Community Survey, Table B17001A-I

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 14

Once, again, relative to other women in the state, women in Mecklenburg generally face lower poverty rates (Figure 15). Lower poverty rates are seen in large swaths of the Piedmont that hold the states most industrialized and urbanized counties. Small pockets of low poverty rates found south of Asheville reflect affluent retiree communities in Henderson, Polk and Transylvania Counties, while a disproportionate number of women live in poverty on the Inner Coastal Plain. So, while Mecklenburg women are more likely than men to live in poverty, their poverty rates are not as alarming as in other parts of the state where as many as one-quarter of all women live below the poverty line.

Figure 15. Women in Poverty (%), 2000

The connection between poverty and education is made explicit in Figure 16. Dropping out of high school is disastrous for the earnings potential of men and women in Mecklenburg. Simply finishing high school reduces the chances of living in poverty by about 50 percent. Among those women who dropped out of high school, 25.5 percent live in poverty while only 13.6 percent of high school graduates live below the poverty level. For men in Mecklenburg, there is an interesting feature revealed in Figure 16—with each increase in educational attainment level up through the Bachelor’s degree, poverty is reduced by about 50 percent. For women, the major jump comes between having some college education (including an Associate’s degree) and having a degree from a four-year institution. Among those with community college educations or 1–3 years of college, 9.5 percent of women live in poverty. With a Bachelor’s degree, that rate drops to 2.5 percent. This is perhaps the most significant result of this study.

A community college education, or simply attending a four-year college (without graduating), is a necessary but insufficient condition for improving the economic status of women in Mecklenburg County. It is not enough just to finish high school. It is not enough just to attend community college. Even modest levels of economic security require a college education. On average, those women obtaining Bachelor’s or Graduate degrees faced poverty rates of only 2.5 and 2.1 percent—lower rates of poverty than is found comparably educated men, and poverty rates that are 4.5 times lower than the average for women overall.

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 15

Figure 16. Mecklenburg Poverty Rate by Educational Attainment, 2005 (%) Population Age 25+ 2.1 2.9

Advanced

2.5 2.6

BA Coll. 1-3 HS Grad

Female

9.5

4.9

Male

13.6

8.7

>HS

25.5

15.0

0

5

10

15

20

25

30

Percent

Looking at only those in poverty, Figures 17a and 17b tell a supporting story about poverty and education: Very few poverty-stricken women have high levels of education.

Figure 17a. Educational Attainment of Men in Poverty, Mecklenburg 2005 6% Less than high school graduate

12%

29% High school graduate (includes equivalency) Some college or associate's degree

Bachelor's degree

21%

Graduate or professional degree

32% Source: Author calculations based on US Census Bureau, 2005 American Community Survey, Table B15004

Figure 17b. Educational Attainment of Women in Poverty, Mecklenburg 2005 7%

2% 29%

Less than high school graduate

High school graduate (includes equivalency)

30%

Some college or associate's degree

Bachelor's degree

Graduate or professional degree

32% Source: Author calculations based on US Census Bureau, 2005 American Community Survey, Table B15004

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 16

Poverty, Work and Children It comes as no surprise that women who work full-time, year-round earn more each year and are less prone to poverty. Shown in Table 4, only 9.4 percent of women in Mecklenburg County who work full-time live in poverty. It also comes as no surprise that women who do not work, especially single women, face much higher poverty rates (44.8 percent in Mecklenburg). More disturbing is that women working part-time are more likely to live in poverty (45.8 percent) than women who do not work at all and the rate of poverty among Mecklenburg women working part-time is higher than state or national averages.

Table 4. Work Status of Women in Poverty Age 16+, 2005 (%)

Female

US

Worked full-time, year-round Worked part-time or part-year Did not work

6.4 36.6 57.0

NC

Mecklenburg

7.5 39.1 53.4

9.4 45.8 44.8

Source: Author calculations are based on US Census Bureau, 2005 American Community Survey, Table B17004

The impact of poverty among men and women is also felt among their children. As shown in Table 5, a higher proportion of children live in poverty than the population as a whole. While Mecklenburg children are less likely to live in poverty than children at the state or national levels, the impact of family structure and marital status is clear. Nationally, 33.5 percent of children in poverty live in two-parent households. In Mecklenburg County the comparable figure is 29.9 percent. These are children of intact families of the working poor. However, children of single mothers (â&#x20AC;&#x153;female householder, no husband presentâ&#x20AC;?) are twice as likely to experience poverty. Single mothers are obviously not insulated from poverty by the earnings of a husband, frequently work part-time, or find themselves under-employed. Ironically, the choices some women face are lose-lose choices: Work longer hours to raise the family out of poverty and risk neglecting their children; or reduce work hours and earnings to provide the care and guidance their children require and risk living in poverty.

Table 5. Children In Poverty, 2005 US Number In married-couple family (%) Male householder, no wife present (%) Female householder, no husban present (%) Overall Child Poverty Rate (%)

NC

Mecklenburg

13,008,489 33.5

438,097 28.9

32,769 29.9

7.4

6.7

6.2

59.1

64.3

63.9

18.2

21.0

15.8

Source: Author calculations based on US Census Bureau, 2005 American Community Survey, Table B17006

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 17

How women make these choices is a personal matter, but women’s travel time from home to work suggests some of the conditions they face. Figure 18 shows cumulative commuting times for all men and women working outside the home in Mecklenburg County. The steeper slope of the curve describing women’s travel time to work reflects their shorter drive times overall. As discussed in Hanson and Pratt (1988),7 shorter commute times do not reflect their great fortune to work close to home but rather reflects the necessity for many women to be near home, school, aging parents, etc. to take care of domestic responsibilities. Indeed, shorter commute times for women reflect constraints on their activities including work place—constraints imposed by their traditional role as primary caregiver to their children.

Figure 18. Cumulative Commute Times, Mecklenburg 2005

Cumulative %

100% 75% 50%

Male Female

25%

60-80

40-44

30-34

20-24

10-14

<5

0%

Minutes Source: U.S. Census Bureau, 2005 American Community Survey, Table B08412

Profile Summary Relative to women in other regions, the economic status of women in Mecklenburg is reasonably positive. Like their counterparts in other regions, there exist significant differences between women and men in the Mecklenburg workforce. The discussion above reveals that: • • • •

Among full-time wage earners in 2005, women had median earnings of $34,171 compared to men’s median earnings of $45,048. Consequently, women’s earnings were 76 percent of men’s median earnings (i.e. 24 percent lower).

Women’s labor force participation, though rising, is still well below that of men. In 2005, women in Mecklenburg had a labor force participation rate 66.9 percent versus 81.8 percent for men.

Overall, women, especially African American women, had lower rates of educational attainment than men which contribute to higher overall rates of poverty in the female population.

Women’s lower earnings contribute to poverty which is highly correlated with educational attainment. Over 60 percent of women in poverty have no education beyond high school.

7 Susan Hanson and Geraldine Pratt (1988). “Reconceptualizing the Links Between Home and Work in Urban Geography,” Economic Geography, 64: 299–321.

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 18

III. The Economic Impact of Gender Inequality Our discussion thus far indicates that while the economic position of Mecklenburg women is relatively strong, there remain significant gender gaps in the local work force. From an economic impact perspective, two important dimensions of these gender-based gaps are differences in annual earnings and differences in labor force participation. Closing these “gaps” would raise the economic status of women. Equalization of earnings and labor force participation would also produce broader benefits to the Mecklenburg economy. As a direct result of closing labor force gender gaps, women’s consumption expenditures would rise, placing new demands on local firms and creating yet more income and jobs in the local area.

The purpose of this section is to quantify the magnitude of the combined effects of closing the male-female earnings gap and the gap in labor force participation.

Below, two scenarios are outlined—an earnings gap scenario and a participation gap scenario.

The earnings gap scenario eliminates male-female differences in earnings by making the median annual earnings of existing full-time female workers equal to men’s earnings, thereby closing the earnings gap. That is, holding all other factors constant, we estimate the added earnings to existing female full-time workers if their earnings were made identical to full-time earnings of men. The participation gap scenario is applied to the earnings of new full-time female workers. In this scenario, we estimate the magnitude of additional earnings to new full-time female workers if their labor force participation equaled that of men. In this second scenario, we also equalize male-female wages for the new female members of the labor force. The sum of added earnings from both scenarios (a) represents all the added earnings that would accrue to women if all gender-based labor force gaps were eliminated and (b) forms the input information to be used in the economic impact analysis.

The Earnings Gap As shown in Figure 1 above, median annual full-time earnings for Mecklenburg women were $34,171 in 2005. In 2005, median earnings for men were $45,048. Thus, in 2005, women earned about 75.9 percent as much as men. While there are many reasons for this gap, it is safe to say that if this gap in earnings were eliminated, the median woman working full-time could expect to earn an additional $10,877 each year. As shown in Table 6, there were 159,632 men in Mecklenburg working full-time, who, in the aggregate, earned nearly $7.2 billion in 2005. The 117,054 full-time working women earned nearly 4.0 billion in total. However, if these women received the same median earnings as men, they would have earned about $5.5 billion. Thus, at current employment rates, women’s earnings trail those of men by $1.7 billion solely because their earnings were 24 percent lower than earnings for men. We confine our analysis of the earnings gap to full-time workers because such gaps do not exist among part-time workers.8

8 According to the Bureau of Labor Statistics, for example, in 2006 median weekly earnings for men age 16 and above were $192; women aged 16+ earned $213. For the bulk of the part-time labor force, age 25 and above, the difference was even smaller with men earning $255 per week compared to $253 for women. See Bureau of Labor Statistics, Median weekly earnings of part-time wage and salary workers by selected characteristics. Available at http://www.bls.gov/cps/cpsaat38.pdf Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 19

Table 6: The Gender Wage Gap in Mecklenburg County, 2005 Men Full-time Median Annual Earnings Full-time Employment Total full-time Earnings Total full-time Earnings @ $45,048 Aggregate Earnings Gap

$45,048 159,632 $7,191,102,336 $7,191,102,336

Women $34,171 117,054 $3,999,852,234 $5,454,785,072 $1,736,317,264

Source: Author calculations are based on US Census Bureau, 2005 American Community Survey, Table B17004

The Labor Force Participation Gap Equalizing earnings between men and women would provide a substantial boost to womenâ&#x20AC;&#x2122;s economic welfare. In the aggregate, however, lower overall earnings for women are not only related to the wage gap, but lower rates of labor force participation (the percentage of the civilian, non-institutionalized population age 16 and above who are either employed or unemployed but actively seeking work; see Section II). While the male-female gap in labor force participation has been converging over the last several decades, there remain significant gaps in labor force participation between men and women. Even without removing the earnings gap, womenâ&#x20AC;&#x2122;s economic status would improve with increasing attachment to the labor force.

Table 7 shows that womenâ&#x20AC;&#x2122;s rate of labor force participation in Mecklenburg County is 66.9 percent compared to 81.8 percent for men. If women participated in the labor force at the same rate as men, the female labor force would expand to 248,072 (an addition of 45,277). Holding full-time employment rates constant, the number of employed full-time women in the county would expand by 26,112 while the number of women employed part time would increase by 16,910 with the remaining 2,255 unemployed.9 If each new full-time working woman earned the same as men currently earn ($45,048), total earnings would increase by $1.176 billion. Further, increased earnings from additional parttime working women would amount to another $212 million, holding their median earnings at their current level. Thus, aggregate earnings for all new women in the labor force would increase by nearly $1.4 billion if the participation gap were closed.

9

The number of unemployed assumes the same 2005 county-wide unemployment rate of 4.9%.

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 20

As shown in Table 7, the combined effect of closing the wage gap and the labor force participation gap would elevate the earnings of Mecklenburg women by $3.1 billion.

Table 7. Labor Force Participation Gap, Mecklenburg County 2005 Male Civilian Noninstitutional Population, Age 16+ Labor Force Labor Force Participation Rate Full-time Employment Full-time Employment for Women if Participation Rate = 81.8 New Female Full-time Workers New Female Full-time Earnings if New Workers Earned $45,048 New Female Part-time Workers New Female Part-time Earnings if New Workers Remain at $12,545 Total New Wages from Closing the Participation Gap Total New Wages from Closing the Wage Gap (from Table 6) Total New Earnings from Closing Both Wage Gap and Participation Gap

Female

287,704 235,305 81.8 159,632

303,267 202,795 66.9 117,054 143,166 26,112 $1,176,286,480 16,910 $212,136,001 $1,388,422,482 $1,736,317,264 $3,124,739,746

Source: Author calculations are based on US Census Bureau, 2005 American Community Survey, Table B17004

Economic Impact to Mecklenburg County The economic impact to Mecklenburg County does not stop with the addition of $3.1 billion in earnings to women. Much of these earnings will be spent in the local economy, placing new demands on local businesses. In turn, as local firms experience growing demand, they will put new demands on their suppliers, who will place new demands on their suppliers, etc. stimulating an economic chain reaction, or “ripple effect” throughout the local economy. The strength of economic ripples reflect the connections between firms and their suppliers (first tier, second tier, etc.). These connections are linkages, or inter-industry transactions, and economic models called input-output models are designed to capture the various tiers, or “rounds,” of spending that emanate from an initial stimulus.10 The local economy will expand to meet the demands of an initial stimulus. The amount of this expansion is summarized in the multiplier effect.

10

In this report, we use the IMPLAN input-output model calibrated for Mecklenburg County.

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 21

The most important step in conducting an economic impact analysis is the accurate identification of the direct effect. In this case, the direct effect is the initial stimulus to the Mecklenburg economy that results from the added earnings of women. We should note, however, that not all of the added earnings will be spent in the local economy. Some earnings will be taxed by local, state and federal governments; some earnings will be spent outside the county. Thus, the $3.1 billion should be reduced to reflect these amounts (see below).

Once the direct effect is clearly defined, the input-output model captures the additional effects.

These additional effects are called indirect effects and induced effects. Indirect effects represent all the production-related “spin-off ” economic activity that results from the spending of women’s added earnings. Because businesses throughout the county are connected to each other through inter-industry linkages (purchases and sales of supplies, etc.), economic impacts in one sector of the economy ripple through other sectors. This creates further rounds of spending, generating yet higher volumes of local business sales, stimulating employment, and expanding wages and salaries paid to employees in the county.

As production-related impacts of the direct and indirect effects work their way through the economy, employers expand payrolls and pay wages and salaries to their employees, further increasing household income. Consumption-related household spending of wages and salaries represents another source of new demand known as the induced effect. The sum of direct, indirect and induced effects equals the total effect, or total economic impact of eliminating gender-based differences in the workforce.

All the economic effects above can be expressed in terms of industrial output, employment or earnings. Industrial output is a broad measure of the value of all goods and services produced in the region. It includes production for final consumers as well as production used by other firms. Employment impacts are simply expressed as the number of jobs (full-time and parttime) that result from the expanded output of firms. Of course, new jobs generate wages and salaries and these comprise the earnings impacts.

Economic Output Impacts As noted above, new earnings accruing to women by closing the gender gap could reach $3.1 billion. However, not all of these wages will be available for consumer spending. First they must be reduced to account for taxes, yielding an estimate of disposable income. Then, we must account for some portion of disposable income that will be spent outside the county. To account for taxes, we multiply total new earnings of women by a disposable income factor of 0.803 which yields disposable income of $2.5 billion.11 According to the IMPLAN model, about 25 percent of household consumption expenditures are spent outside the county. Thus, disposable income available for local consumption is reduced by an in-county consumption factor of 0.767 which yields a grand total of $1.924 billion in added earnings that would enter the local economy via closing the gender gap.

11 According to the Census of Governments, the City of Charlotte and Mecklenburg County levy combined taxes of $0.038 per dollar of income and the state of North Carolina levies $0.068 per dollar of income. According to the Office of Management and Budget, average federal taxes per dollar of income are $0.091.

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 22

The direct effect of expanding Mecklenburg consumption expenditures by $1.924 billion would generate an additional $595 million in production-related indirect effects, and $612 million in induced effects (Table 8). In total, direct, indirect and induced economic output would increase by $3.1 billion. That is, local businesses should expect to see an increase in sales totaling $3.1 billion if both the wage gap and labor force participation gap were eliminated.

Table 8. Economic Impact of Closing the Gender Gap in Mecklenburg County

Output($) Jobs Earnings ($)

Direct 1,923,626,570 19,444 692,758,788

Indirect 594,981,090 3,853 165,111,129

Induced 611,919,582 5,590 206,574,730

Total 3,130,527,262 28,887 1,064,444,644

Note: Monetary values at 2005 price levels

Employment Impacts To accommodate the increase in business output, many firms would hire new employees.

Quite apart from the 43,022 new jobs associated with women’s increased labor force participation, 19,444 new jobs would be directly created by the initial spending of $1.924 billion.

Indirectly, an additional 3,853 jobs would be created through production-related inter-industry linkages while the spending of employee wages would generate 2,484 jobs through induced effects. In total, 28,887 jobs would be generated by closing the gender gap. These jobs are created by the spending of higher women’s wages and are in addition to the number of new jobs associated with increased female labor force participation.

Household Earnings Impacts Along with the jobs created by women’s new consumption spending are wages and salaries.

The direct creation of 19,444 new jobs is estimated to generate about $693 million earnings beyond those associated with the gender gap. Indirect and induced effects amount to $371.7 million for a total impact on Mecklenburg wages and salaries of $1.064 billion. As above, these impacts are derived from the spending of women’s new earnings but do not include the new earnings themselves. Adding new earnings of $3.1 billion associated with closing the gender gap to those generated by their spending in the local economy ($1.064 billion) raises total household earnings in Mecklenburg County by $4.2 billion or about 12 percent over 2005 levels.

Tax Impact The Census of Governments reports detailed expenditure and revenue information for both Mecklenburg County and the city of Charlotte. As noted above (see footnote 11), total tax revenue for the two jurisdictions combined amounts to $0.038 per dollar of income. Most of this involves local property taxes and general sales taxes. Both tax streams are closely related to income levels. Applying this tax factor to $4.2 billion in total household earnings yields an estimated return to city and county governments of $160 million annually.

Incremental Improvements As noted in the introduction, how gender gaps in the labor force are closed is beyond the scope of this study. However, it is safe to say that removing all gaps would require significant expansion in the local economy. For example, in 2005 there were 515,326 non-federal jobs in the county. The addition of 43,022 full-time and part-time jobs associated with closing gender gaps would represent an 8 percent inrease in local employment. In terms of the economic impact discussed above, however, 28,887 would be fueled by the added purchasing power of women if gender-based workforce gaps were eliminated. More plausible are incremental changes that Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 23

move the labor force in the direction of earnings and labor force participation equalization.

Table 9 provides an indication of what incremental improvements in the status of women might imply for the local economy. What is immediately evident in Table 9 is that even small improvements can have a significant impact on women and the local economy. For example, closing gender-based gaps by just 10 percent improves women’s earnings by more than $300 million and total household earnings by more than $400 million. Given that Mecklenburg County generates 10,000-15,000 new jobs in a typical year, closing the gender gap by 25 percent is entirely plausible. A 50 percent closure could be achieved in as little as three years of average growth. Thus, while the impact of equalizing male-female workforce attributes is large, they are not out of reach.

Table 9. Economic Impact of Incremental Gender Gap Progress

Degree of Closure

Women’s Earnings $Mill

10% 25% 50% 100%

Economic Output $Mill

$312.5 $781.2 $1,562.4 $3,124.7

$313.1 $782.6 $1,565.3 $3,130.5

Jobs

Total Household Earnings $Mill

City and County Tax Revenue $Mill

$418.9 $1,047.3 $2,094.6 $4,189.2

$16.0 $40.0 $80.0 $160.0

2,889 7,222 14,444 28,887

Note: Monetary values at 2005 price levels

Impact Summary Removing the gender gap in earnings and labor participation would potentially have a significant impact on working women in Mecklenburg. The added purchasing power associated with gender equality would produce substantial impacts on the local economy.

•

If women currently holding full-time jobs had comparable earnings to men, they would receive $1.7 billion more each year.

•

If women’s labor force participation were equal to that of men, there would be 26,112 more women employed full-time who would be earning $1.2 billion annually. Further, an additional 16,910 women would be working part-time earning and additional $212 million. Total new earnings from closing the participation gap would amount to nearly $1.4 billion.

•

After accounting for taxes and out-of-county spending, new consumption expenditures fueled by women’s new earnings would expand the county’s economic output by $3.1 billion, create 28,887 new jobs and generate nearly $1.1 billion in additional earnings to Mecklenburg households.

•

Based on these estimates, complete closure of the earnings and participation gaps would generate about $160 million in tax revenue to the city of Charlotte and Mecklenburg County.

•

Even modest, incremental improvements in the status of working women would generate significant economic benefits.

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 24

IV. Summary and Conclusions The preceding analysis indicated that women and men have very different experiences in the labor market. In 2005, the gender gap among full-time working Mecklenburg women and men was 76 percent (women earned 24 percent less than men); female labor force participation was 66.9 percent compared to 81.8 percent for men.12 While there are many reasons for these gaps, there is no doubt that the status of women would improve and the local economy would expand if women’s earnings and labor force participation were equal to that of men.

In many ways, the analyses in this report are hypothetical. Hence, the reported impacts should be considered illustrative rather than definitive. Given various caveats, the impacts of closing the gender gap are profound. Completely equalizing the gender gap in Mecklenburg County would add $3.1 billion to women’s wages, $1.9 billion of which would be spent in the local economy, generating an additional $1.1 billion in wages. In all, total household earnings could expand by nearly $4.2 billion if all gender-based differences where eliminated. As many 28,887 new jobs would be created in Mecklenburg, producing $3.1 billion in gross business sales.

All this assumes, however, that 43,000 additional women could be absorbed into the labor market. How can this happen? First, as illustrated in Table 9, the gender gap does not have to close all at once. Gradual incremental changes, coupled with human capital investments, could markedly improve the economic status of women over time. Second, three to four years of normal growth in Mecklenburg County would be sufficient to absorb 26,000 full-time and 17,000 part-time (women) workers.13 Thus, the magnitude of required change is plausible. The important point is that even a gradual closing of the gender gap would produce significant gains to women and the larger economy. Third, the above analyses have been restricted to Mecklenburg County alone. Inter-county commuting is prevalent in this region and some (as much as half) of increased female labor force participation could be accommodated by job growth in adjacent counties.

However, it is unlikely that the gender gap will close without changes in the labor force itself and labor practices, generally. Section II of this report highlighted many gender-related differences the labor force that can be either directly remedied over time with appropriate public policy, or addressed immediately via workplace policy. Over the long run, there is a clear need to improve the human capital endowment of women. Not only does education and training directly affect earnings, but lack of human capital was shown to be highly correlated with poverty, especially among women. This appears to be especially true for non-White women whose median earnings are 50–60 percent that of men.

There is some good news, however. Women who finish high school and attend college or university now have higher rates of graduation than men. In the long run, this will help solve some of the problems associated with the gender gap. However, in the short run, there are many women currently in the workforce who, for a variety of reasons, cannot return to formal education. For them and others, increasing training options in higher-wage occupations and industries, especially those where women are underrepresented, should be a high priority.

Some training could be provided by the public sector; much of it could be provided by current 12 While this gap is “normal” by national standards, many believe it underestimates the magnitude of difference in earnings between men and women. Ross and Hartman (2004), for example, show that because women’s work histories are often punctuated with periods of labor force withdrawal, the gender gap in wages can drop to 38 percent over a 15-year period.

13 We should also note that employment growth also stimulates population growth and in the short-run, about half of all new jobs are typically filled by newcomers. See Greenwood, M., Hunt, G. and McDowell, J. (1986) A Migration and Employment Change: Empirical Evidence on the Spatial and Temporal Linkage, Journal of Regional Science, 26(2):223-234.

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 25

employers, on-the-job, in the private sector. Why would an employer do this? Employers might find it advantageous to supply training for existing workers because (a) it is expensive to hire and train new workers; (b) on-the-job training enhances earnings prospects and engenders worker loyalty; and (c) added worker productivity plus cost savings might make such a strategy good for the “bottom line.”

Family responsibilities also impose differential effects on women. They are more likely to take primary responsibility for children and, increasingly, aging parents or relatives. This suggests at least three complementary “family friendly” strategies for employers.14 1. To accommodate women’s roles as mothers, on-site daycare or before-tax contributions to a child care fund, should be a high priority. It would also be feasible for several employers to pool resources and provide “near-site” daycare.

2. Implementing flexible work schedules might decrease labor force detachment and increase workforce continuity for many women and their employers. Further, employers could encourage men to utilize family leave for situations that typically fall to women.

3. Strategies to smooth the transitions back to work after a period of leave should be explored so that senority an accumulated experience are not completely lost. Such strategies might include workplace re-entry seminars or periodic workshops that prepare returning workers after a period of leave.

Frequently tied to family responsibilities, women’s geographic labor markets tend to be constrained. Improving women’s earnings would lower the relative cost of commuting and open a wider range of employment possibilities. Given the constrained nature of female labor force experience, there is also a clear need to improve the quality of part-time employment (better wages, access to benefits, training, etc.).

The fact that labor force experience and outcomes for women vary by race and ethnicity is, perhaps, underemphasized in this report. Distinct disadvantages experienced by non-White women in the work force were shown in Section II. While all women face earnings and participation gaps, the gap is particularly acute for African American and Hispanic/Latino women who comprise a large proportion of the local work force. Whether through targeted public/private policies or educational outreach, improving the employment prospects of minority women will achieve the greatest, most equitable progress toward eliminating the gender gap in Mecklenburg County.

14

Ross and Hartman (2004),

Women, Work and Wages in Mecklenburg County: An Economic Impact Assessment

Page 26

APPENDIX: INPUT‐OUTPUT ANALYSIS AND THE IMPLAN MODEL Overview of Input‐Output Analysis

Input‐output analysis and the input‐output model are mathematical representations of an economy. Input‐output analysis provides a means for examining the relationships within an economy—both between business and between business and final consumers. One objective of input‐output models is to capture all the monetary market transactions between industries and final consumers for a specific period of time. The resulting mathematical representation of the economy allows analysts to examine very detailed representations of a region’s industrial structure and to trace how changes in one or more sectors of an economy affect other sectors in the region.

Set‐out in matrix form, each economic activity is represented as both a purchaser of industrial inputs and the seller of its output. The model is typically described as a double‐entry accounting model of a regional economy in which industries are linked to one another via their buying and selling patterns. These linkages, called inter‐industry transactions, represent the dollar flows between industries necessary to produce their goods and services. Because these transactions are recorded in terms of dollar flows, input‐output models are also sometimes called cash flow models of production. Further, because input‐output models allow an analyst to trace all the transactions between sectors of regional economy and represent the economy at a very fine level of detail, input‐output analysis has gained widespread popularity for industrial targeting studies and economic impact assessment. Though results of impact assessments are frequently aggregated to facilitate presentation and discussion, the input‐output models developed for this study represent 528 industrial sectors, the highest level of detail possible.

The detail provided by input‐output models is one reason why input‐output analysis has become the accepted standard for economic impact assessment.

Input‐output models consist of three basic components: A matrix of inter‐industry transactions, vectors of final demand, and vectors of value added. The matrix of inter‐industry transactions, called the transactions matrix, is the heart of an input‐output model. In this matrix all the monetary transactions between businesses in different sectors are recorded. Because all industries are represented as both buyers and sellers to other sectors in the economy, the matrix is square with one row and one column for each industry. Reading across the rows of such a matrix illustrates sales of one industry to all other industries in the region. Reading down the columns of the matrix shows all the purchases an industry makes from other industries to produce its output. In effect, each column of a transactions table represents the production function for that industry. The transactions recorded in this matrix are sales between industries for intermediate use, not sales to end‐users.

Sales to end‐users are recorded in several vectors of final demand. The entries in these column

Women, Work and Wages in Mecklenburg County: an Economic Impact Assessment Appendix, A‐1

vectors represent the sale of finished goods and services. Vectors of final demand include sales to consumption, investment, governments and exports. Value‐added consist of several row vectors that include payments to labor, profits and imports. Because this is an accounting model of the economy, total output from industries must equal total inputs. Total outputs can be determined by summing all entries across the rows of the transactions table (intermediate output) with the entries in the vectors of final demand (final output). Likewise, total inputs can be calculated by summing down the columns of the transactions table and adding the components of the value added vectors. The diagram below illustrates the basic structure of an input‐output model.

Industry T 1, 2, 3, ..., n o 1 t 2 a 3 Inter‐industry Transactions l

.

Final .

O Demand n u

t

p u t

Value Added

Total Input

In order to convert this descriptive model of the economy into an analytical, predictive model several computations are required. First, the matrix of transactions needs to be converted to matrix of coefficients so that a change of any level in production can be traced throughout the economy. If the transaction matrix reflects all the purchases and sales made within the region, then a change in one sector will affect the outputs of other sectors that are linked to it. How can these linkages be traced? We start deriving a table of direct input coefficients from the transactions table by dividing each column entry in the transactions table by its corresponding column total. The result is a table of direct input coefficients that illustrates the cents worth of input purchased by one industry from all other industries in the region to produce $1 dollar of output. In other words, if we call the transactions matrix Xij and divide each entry by its column sum, Xj, a matrix of direct requirements will be produced. These coefficients are

Women, Work and Wages in Mecklenburg County: an Economic Impact Assessment Appendix, A‐2

aij = X ij / X j typically denoted by aij. The collection of them in a matrix is represented by A, or which produces a matrix of coefficients A, with A 0 aij.

The matrix A illustrates the all the direct requirements of each industry from all other industries. Multiplying this matrix by a vector of total industry outputs produces a measure of all intermediate output produced in the region. To determine the level of total output we must also add to intermediate output any sales to final demand or end users. If these sales are represented by Y, then the regional industrial production identity can be expressed as

AX + Y = X

where X is total industry output. If the above equation is rearranged and factored, a solution for X can be derived by

Y = X - AX

= (I - A)X

(I - A )-1 Y = X

where I is an identity matrix of the same dimension as A.

The matrix (I‐A)‐1 is frequently called either the Leontief Inverse, or simply the multiplier matrix. It illustrates how a change in one sector affects other sectors of production. This matrix is important for at least two reasons. First, although not all industries are directly related to each other, they may be indirectly related through other industries. Second, how these relationships affect each other quantitatively can be summarized by adding down the columns of the matrix which provides an estimate of all the direct and indirect effects on production from a one unit change initiated in any of the sectors in the regional economy. Because linkages

Women, Work and Wages in Mecklenburg County: an Economic Impact Assessment Appendix, A‐3

will be of varying strength, each industry will be characterized by different multipliers.

The solution to the above formulation implies that there is some level of final demand that needs to be satisfied by local producers. In other words, input‐output models are demand‐ driven models that seek to capture how changes in the demand for final goods and services affect the economy as a whole. To estimate the economic impact of any new industry or change in final demand, input‐output analysis can be used to capture the interdependencies of the region’s economy with those industries experiencing an initial change in the demand for their goods and services. In such a case, the model can be used for predictive purposes by solving the following equation:

ΔX = (I - A )-1* ΔY

where ΔX is the change in total economic activity stimulated by the change in final demand, ΔY.

Like any model, input‐output models rest on a set of assumptions. These assumptions include:

• Constant returns to scale • Linear and homogenous production functions • Perfectly elastic factor supplies • Constant technology

Through the use of input‐output models, we are able to capture industry linkages and estimate the economic impact of one set of activities on all other industries in the region. Because all industries in a region are, to some degree, linked to one another, a change in one sector of the economy will ripple through other parts of the economy. The estimation of these ripple effects, called multiplier effects, is the main objective of economic impact assessment.

Input‐Output Multipliers

Multipliers are numeric summaries that indicate the total change in economic activity due to a one unit direct change. For the purposes of discussion, the direct effect is the level of economic activity directly attributable to the economic sector(s) experiencing an initial (exogenous) change in demand for their goods or services. “Indirect effects” measure the secondary industrial impacts set into motion by the goods and services demanded by the sector(s) included in the direct economic impact. As direct and indirect industrial effects are initiated, regional firms pay wages to labor who, in turn, spend part of their income in the region causing another “round” of spending in the form of new demands for goods and services produced by regional

Women, Work and Wages in Mecklenburg County: an Economic Impact Assessment Appendix, A‐4

firms. These latter effects are called induced effects. The sum of all three effects yields an estimate of the total economic impact of a change in the local/regional economy. As a short‐ hand means for summarizing the total impact associated with a given direct impact, analysts will make reference to multiplier effects. Multiplier effects can be expressed in terms of employment, industrial output, or income.

Input‐output analysts frequently make a distinction between Type I and Type II multipliers.

The Type I multiplier summarizes the relationship between the direct and indirect effects. They are calculated as the direct effect plus indirect effect, divided by the direct effect. The Type II multiplier includes the impact of consumption‐induced effects. It is calculated as the direct plus indirect plus induced effects, all divided by the direct effect.

The distinction is important, especially to development policy analysts, because the Type I multiplier summarizes the strength of the regional industry linkages in the economy and indicates the extent to which the industry in question is functionally integrated with the rest of the regional economy. The inclusion of the induced effects in the Type II multiplier is also important, but for different reasons. Because Type II multipliers capture the effects of household spending on regional economies, they reflect patterns of consumer demand. The size of the induced effects, i.e. the difference between the Type I and Type II multiplier, also reflects the relative pay scales in industries affected by the economic impacts. In other words, Type II multipliers are apt to be higher when the direct and indirect effects of the impact involve expanding employment levels in high‐wage industries.

In regional impact assessment, it is not uncommon for the impacts from induced household consumption to be larger in magnitude than the production‐related indirect effects. This is especially true in small, rural, or lesser developed economic regions. While most cities and towns have a sufficient number of establishments to accommodate the household demand for everyday, low order goods such as gas stations and grocery stores, not all small towns or rural regions have a diversified industrial base characterized by dense networks of inter‐industry transactions. This means that much of the demand induced from household spending can not be met with local production. However, the economic base in smaller and rural economies tends to be correspondingly smaller and less diversified such that many inputs needed for industrial production need to be imported from outside the region. This causes the indirect production‐related impacts to be small relative to the consumption‐induced impacts in less developed regions. It is for this reason that there is also a general relationship between the size of the region and size of its multipliers, with smaller regions having smaller multipliers and larger regions having larger multipliers. As the size of the region increases, the likelihood that inputs needed for production will come from other producers in the region (rather than through imports) also increases. The greater this likelihood, the more likely income and demand will re‐ circulate in the local economy rather than leaking to other regions.

Women, Work and Wages in Mecklenburg County: an Economic Impact Assessment Appendix, A‐5

The IMPLAN Input‐Output Model

The IMPLAN model is a regional input‐output modeling system developed by the Minnesota IMPLAN Group. The most recent version is maintained in the Department of Geography and Earth Sciences at the University of North Carolina at Charlotte. The IMPLAN modeling system is an interactive, computer‐based modeling system capable of producing input‐output accounts and input‐output models for any region in the United States as small as a single county. The system consists of regional data bases and software that allow users to develop these models for the purposes of describing the structure of regional economies and/or predictive analyses, especially those associated with estimating the economic impacts of a quantifiable change in regional production.

Like most regional input‐output models, the IMPLAN model is stepped down from a set of national input‐output accounts. Combined with local data, IMPLAN relies on national sources for base accounting data such as the national “Use” and “Make” matrices and their associated Absorption and Byproducts coefficient tables. These data sources show, at the national level, which industries produce specific goods and services (Make Table) and the sets of inputs these industries use in their production process (Use Tables). An underlying assumption of stepped‐ down input‐output tables is that the industrial technology implied by the national accounts is applicable to sub‐national regions as well.

What differentiates the national input‐output accounts from the regional accounts produced with IMPLAN are parameters that describe trade flows between the region and the rest of the world. Estimating the volume of trade for a sub‐national area is a critical step in “regionalizing” a national input‐output table. There is a large academic literature on this subject that focuses on several computational strategies for regionalizing input‐output tables such as partial survey techniques, location quotients, bi‐proportional RAS procedures, and supply‐demand pooling. The default procedure employed in the IMPLAN system, one which has gained widespread acceptance, is the Regional Purchase Coefficient procedure, or RPC. According to IMPLAN developers, an RPC represents the proportion of local demand purchased from local producers. For example, an RPC of 0.25 for a given commodity means that for each $1 of local need for that commodity 25 percent will be purchased from local producers. This method is based on the characteristics of the region and describes the actual trade flows for a region mathematically [3].

Each commodity produced in a region has an associated RPC which is determined via a set of econometric equations and used to estimate trade flows (imports and exports) of that commodity. If, as in the example above, a particular commodity has an estimated RPC value of 0.25, then 25 percent of that commodity will be purchased from local establishments, and the remaining 75 percent of commodity demand will be imported from other regions. Trade flow

Women, Work and Wages in Mecklenburg County: an Economic Impact Assessment Appendix, A‐6

estimates are important to regional input‐output models because they are what differentiate the regional model from its national counterpart. Trade flows affect the amount of local commodity production available to industries. This affects the elements of the transactions matrix, which in turn affect the direct requirements matrix, A, that is necessary to compute the Leontief multiplier matrix.

Input‐output models are extremely data‐intensive and IMPLAN makes extensive use of many data sources. For example, industry outputs are derived from economic censuses from the Bureau of the Census and projections from the Bureau of Labor Statistics. Employment data are derived from ES202 employment security data and are supplemented with information from County Business Patterns. Data from the Regional Economic Information System (REIS) are also used supplement ES202 employment information as well as providing a data source for employee compensation (wages and salaries) and proprietor income. The elements of final demand are largely derived from the National Income and Product Accounts (NIPA) and federal procurement and sales reports. The end result of combining reliable data with accepted modeling techniques is an extremely flexible system for generating regional input‐output accounts that can be used for industrial targeting, economic impact assessment, and analysis of regional development policy.

Suggested Reading

[1] Hewings, G.J.D. (1985) Regional Input‐Output Analysis, SAGE Publications: Beverly Hills.

[2] Miller, R. and P. Blair (1985) Input‐Output Analysis: Foundations and Extensions, Prentice‐Hall, Inc.: Englewood Cliffs, New Jersey.

[3] Minnesota IMPLAN Group, Inc. (1996) IMPLAN Pro Users Guide, Analysis Guide, Data Guide: Stillwater, MN.

Women, Work and Wages in Mecklenburg County: an Economic Impact Assessment Appendix, A‐7

UNC Charlotte Urban Institute 9201 University City Boulevard, Charlotte, NC 28223