Page 1

HARVARD C O L L E G E Economics Review SPRING 2011 ¤ VOL IV ISSUE 2


ECONOMICS INSIDE Behavioral Economics Martin Dufwenberg

Libertarian Paternalism and Growing Up James Jamison

Got Altruism? James Andreoni

Behavioral Economics and Interdisciplinary Endeavours Peter R. Blake, Ed.D.

Moral Sentiments & Climate Change Norman Schofield


INTERVIEW Yasheng Huang ’85

Effect of Financial Incentives on Status-based Discrimination in a Marriage Market Haimanti Bhattacharya

An Economic Lesson from the Sports Industry Andrew Zimbalist

HARVARD CO L LE G E Economics Review SPRING 2011 ¤ VOL IV ISSUE 2

Letter From the Editors Dear Reader,


Editors-in-Chief Pamela Ban Marianna Tishchenko President Michael Ding Business Director Kevin Lee Content Directors Han He Casra LaBelle Publication and Layout Director Gary Norris


Regan Bozman Albert Cui Roger Hu Kwon-Yong Jin Billy Kim Yueran Ma William Newberry

In this issue, we take on the topic of Behavioral Economics, a dynamic field of economics that has received a great deal of attention in recent years. The incorporation of psychology and cognitive science into classical economy theory yields exciting and meaningful insights. Behavioral economic models let us examine the decision-making process of individuals with powerful tools. This issue contains articles on a variety of topics. Our Cover Theme section, for example, comprises articles dealing with altruism, strategic interactions, and libertarian paternalism. In providing a variety of articles in the dynamic field of Behavioral Economics, we aim to give the reader an introduction into the field’s workings. We have, as usual, included a series of other articles by professional contributors to continue educating students on topics that they might not encounter in the classroom. We hope that you find this edition of the Harvard College Economics Review insightful, entertaining, and educational. As always, please write to us and/or visit us online at Sincerely,

Best regards,

Michael Ding

E-mail: Website: COPYRIGHT 2010 HARVARD COLLEGE ECONOMICS REVIEW. $7.99. ISSN: 1946-2042. All rights reserved. No part of this magazine may be reproduced or transmitted in any form without written permission of the Harvard College Economics Review. Opinions published in this periodical are those of contributors and do not necessarily reflect those of the editors.

Pamela Ban

Marianna Tishchenko

Libertarian Paternalism and Growing Up

On the Cover:

Behavioral Economics

HARVARD C O LLEG E Economics Review SPRING 2011 ¤ VOL IV ISSUE 2

The HCER brings together academics, practitioners, and students to analyze important findings in behavioral economics—a dynamic field of research that has attracted a great deal of attenB E H AV I O R A L tion in recent years.


INSIDE Behavioral Economics Martin Dufwenberg

Libertarian Paternalism and Growing Up James Jamison

Got Altruism? James Andreoni

Behavioral Economics and Interdisciplinary Endeavours Peter R. Blake, Ed.D.

Moral Sentiments & Climate Change Norman Schofield

Behavioral Economics


INTERVIEW Yasheng Huang ’85

Effect of Financial Incentives on Status-based Discrimination in a Marriage Market Haimanti Bhattacharya

An Economic Lesson from the Sports Industry Andrew Zimbalist


Martin Dufwenberg

Got Altruism?


James Andreoni

Moral Sentiments & Climate Change Norman Schofield



James Jamison

Behavioral Economics and Interdisciplinary Endeavours


Peter R. Blake, Ed.D.

Effect of Financial Incentives on Status-based Discrimination in a Marriage Market


Haimanti Bhattacharya



INTERVIEW Yasheng Huang ’85


An Economic Lesson from the Sports Industry


Andrew Zimbalist ISBN 978-0-9823780-2-1




Behavioral Economics

An introduction to the development and applications of the study of economic phenomena through a psychological lens


conomics has traditionally relied on sophisticated mathematical methods as exemplified for example by general equilibrium theory, price theory, or game theory. However, the traditional approach has largely made relatively simple assumptions about human nature such as that each individual only cares about his own consumption. Research in neighboring social sciences, by contrast, usually relies on mathematical methods to a lesser extent while entertaining a richer description of man that may take into account for example how social relations or emotions influence people. Behavioral economics (BE) combines the strengths of both approaches, incorporating psychological insights into economic analysis typically with continued use of sophisticated analytical tools. BE has recently become the fastest growing field in

economics and in my view the most exciting. The background is a wealth of evidence, often experimental, identifying empirical phenomena that are not well explained by traditional analysis. BE develops new models that incorporate emotions, fairness, reciprocity, social norms, bounded rationality, myopia, etc. BE is developed hand-in-hand by theorists and experimentalists, with many scholars having a foot in each camp. By its nature, BE is cross-disciplinary. Behavioral economists are often inspired by researchers in fields such as psychology, sociology, philosophy, and the neurosciences. In return, BE offers analytical tools and modes of thinking that can inspire scholars outside economics. To illustrate a particular BE exercise, consider the “trust game” described in the following figure:

Player 1 /\ / \ Not/ \Hire / \ / \ Player 2 (1, 1) /\ / \ Low/ \High / \ / \ (0, 3×(1-g)) (2, 2) The figure depicts a “game tree” which describes a situation where an employer (player 1) and an employee (player 2) interact. Think of the tree as growing upside down, with its branches corresponding to choices. The interaction starts with player 1 who decides whether to Hire or Not player 2, as indicated via the label-


ing of branches. The game ends if player 1 chooses Not (to hire). If player 1 chooses Hire, however, then player 2 must choose to exert Low or High effort; again these choices correspond to branches in the game tree, as indicated. The expressions (1, 1), (0, 3×(1-g)), and (2, 1), finally, specify the players payoffs (or how much they like the corresponding outcomes) associated with the indicated sequence of choices involved. Player 1’s payoff is given first in each pair. These payoffs should be interpreted as money payoffs, with the exceptions of player 2’s payoff 3×(1-g) following choices Hire and Low which should be thought of as a money payoff (=3) tempered by 2’s feelings of guilt. Player 2 feels guilt because he hurts player 1 by choosing Low. This feeling is as captured through g which is a number between 0 and 1; the higher is g the lower is 3×(1-g) so the more player 2’s enjoyment of money is reduced if he exerts Low effort. I will consider how these guilt feelings of 2’s, as captured through g, may influence the interaction between the players. Below an assumption will be made such that g depends on the players’ beliefs, in a particular way, but we start with a much simpler benchmark case: Assume that g = 0. That case corresponds to a traditional economic analysis where feelings of guilt are not accounted for. In this case, player 2 would choose High, since 3×(1-g) = 3×(1-0) = 3 > 2. If player 1 anticipates this she would choose Not (to hire), since 1 > 0. The outcome is inefficient in the sense that each player gets a payoff of only 1, whereas each player would have gotten 2 had they instead chosen Hire and High. Now assume instead that 2 is prone to suffer from guilt, captured via the possibility that g > 0. To be specific, assume (in line with some experimental as well as survey evidence) that g depends on 2’s beliefs about the probability with which 1 believes 2 will choose High. Specifically, assume that player 1 assigns probability p to player 2 choosing High. The number g, then, is player 2’s best guess about p (or more precisely, for the mathematically inclined, assume that 2 has a probability measure describing her beliefs about p; g

is the mean of that probability measure). The psychological sentiment modeled here is that the stronger 2 believes that 1 trusts him to choose High, the more guilty 2 will feel by choosing Low. What will happen in this situation? Since g may take any value between 0 and 1 it may seem that we can’t tell. However, I will now argue that it is plausible that the players will choose Hire and High and thus manage to get two money units each. The reason is as follows: If 1 chooses Hire he must hold belief p≥1/2; if he didn’t he would be better off choosing Not. Player 2 should be able to figure that out. Hence when 2 faces the choice between Low and High she 2 should realize that p is at least 1/2, which means that her guess g should be at least 1/2. It follows that 3×(1-g)<2 (since g≥1/2), so 2 should thus choose High. Player 1 should figure all of this out, and so chose Hire to start with. The illustrated phenomenon – how 1’s choice of Hire forces 2 to impute certain beliefs to 1 such that 2 will act in a way favorable to 1 – has been labeled psychological forward induction. I find it an intriguing illustration of an insight that BE may offer. Incorporate a little bit of guilt and there will be no need to worry about an inefficient outcome anymore (and no one suffers from guilt)! Why has BE recently become such a popular field? I asked that question to my students in an experimental economics class I teach. They came up with four answers: (i) BE has been inspired by the turmoil on financial markets in recent years. (ii) The interest in BE derives from economists increased use of experimental methods. (iii) Economists already did all they could using traditional approaches. (iv) Economists at last realized that traditional theories do not work well. While each of these answers may make some sense, I was actually after a completely different one. It explains the increased recent interest in BE as a side effect of a parallel development the cause of which itself had little connection to BE. My answer puts the field of game theory at center stage. Game theory is a mathematical toolbox which allows for analysis of strategic interaction, meaning situations

in which multiple decision makers influence each other. One example would be the above trust game, where the payoffs to 1 may depend on what 2 does and the payoffs to 2 depend on what 1 does. Looking at the history of game theory one sees that the 60s and early 70s was largely a slow period for the subject; during this time most economists did not pay much attention to game theory and were often not trained in the relevant techniques. In the mid70s and increasingly onwards this started to change. For example, the game-theory based study of economic situations with asymmetric information (which is relevant, for example, in auctions) took off in the 70s. Bargaining theory and industrial organization are two other game theorybased field which became popular. New exciting game-theoretic results were derived concerning for example interactive epistemology (when players reason about each others’ reasoning), repeated games (where players interact over and over again), and the impact of communication in games. Alongside with these contributions, interest in game theory grew and students were increasingly taught game theory. This happened because people were excited about asymmetric information, bargaining, industrial organization, and foundational topics in game theory. It did not originally happen because people wanted to enrich economics psychologically. However, it just so happens that game theory also furnishes the right tools for conducting exercises of that sort. The above analysis of guilt in employer-employee relationship, which is game-theoretic in nature, can exemplify. So when it happened that a new younger generation of economist was brought up on game theory, at some point people in this generation realized that they were well positioned to explore topics in behavioral economics. Eventually this lead to a boom in experimental and theoretical research addressing psychological aspects of game play and, more generally, to a boost of interest in behavioral economics. H - Martin Dufwenberg is affiliated with the University of Arizona





Got Altruism? For decades scientists of all kinds have been asking the same question, “Why are people nice to each other when it appears to violate their self interest?” Selfishness seems to follow sensibly from evolutionary biology notions of self-preservation. Altruistic behavior, on the other hand, needs some explaining. Biologists have linked altruistic choices to the survival of

one’s genes through mutual benefit, such as kin-selection, norm adherence, or reciprocity. What about economists? How does economics—the science based on the premise of self-interested choice—explain unselfish behavior? The challenge for economists is to find a way in which it is selfish to be unselfish. That is, there must be a good reason

why behaving like a nice person makes you better off. Here the biologists, psychologists, sociologists and even linguists can help. Humans evolved as a social species in close proximity to their kin. There may be natural instincts for feeling good about yourself when you behave like a nice person—the warm glow you feel in your heart—that the human psyche may


have carried forward from our ancestors. Indeed, as shown in my work with John Miller of Carnegie Mellon University, when you give one person $20 and tell them that they can share some of the $20 with another person who had been given nothing, people are willing to share significant amounts. In fact, sharing seems to have the qualities of any other good. If I give you more money or lower the price of pizza, you usually choose to eat more pizza. The same is true for giving. People will give more when there is more to share and when the cost of sharing goes down. So knowing that giving shares sensitivity to price and income like other goods is a nice start, but giving may differ from pizza in other ways. For instance, when you are eating pizza your stomach signals your brain when it’s hungry and when it’s full. So, does your heart send signals to your brain to say it’s hungry for giving or it’s had it’s fill of altruism? How do we know when to give, when to stop, and whom to spurn? My current research agenda on altruism is about finding answers to these questions. The answers we’re finding are telling us that giving is a lot more complicated than pizza. Even though the decision of how much to give resides in the mind of the giver, we’re finding that what the giver thinks about what’s going on in the mind of the recipient is tremendously important. I’M ON YOUR MIND What kind of person do you think I am? I’d hope you think I’m a good and moral person, and that you think I would do the right thing. Imagine this experiment, which I conducted with Douglas Bernheim of Stanford University. Anne is introduced to Bob and both are informed that Anne has been given $20, that Bob was given nothing, and that Anne will be have an opportunity to share whatever she wants to Bob. In this case, most people in Anne’s position choose to share 50-50 and nobody gives just $1. No surprise. But, is that because people like Anne are nice, or because they are afraid of what people like Bob will think of them? To answer this, we next told people that after the sharing decision, there is a 75% chance that this decision will be ignored

and instead will be replaced with a “forced choice” in which the recipient gets just $1. So, no matter what Anne chooses, there will be a 3 in 4 chance Bob will get just $1 and Anne $19. Both the Anne and Bob know the odds, but if Bob sees that he got only $1, only Anne knows whether she was truly selfish or was forced by the evil economists to act selfishly. The result? Now 50-50 is no longer very popular at all, but giving just $1 is the runaway favorite. It looks like givers aren’t so afraid to be selfish, but their afraid to be seen as selfish. In other words, when there is a chance for people to behave selfishly without others being able to blame them or think less of them, they take that chance in a snap. I’m on your mind and part of why I give so you to think I’m altruistic, even if I’m not. YOU’RE ON MY MIND Linguists, anthropologists, and evolutionary biologists agree: Humans developed language to help them cooperate. Talking means we share a joint focus, and if you truly understand my words then I’m getting you to read my mind and to see the world and I see it. In life, most gifts are the product of a social interaction—someone asks you for something and you decide to give or not give. Asking expresses a desire, and if language is successful, asking has the consequence of making the giver see the recipient’s point of view. By the definition of communication I have “read your mind” and am “in your shoes.” This, it turns out, is a key part of giving. Justin Rao, now doing research at Yahoo! Labs, and I explore communication altruism. Imagine again that Anne has $20 to share with Bob. This time we add three twists. First, we sometimes let the recipient Bob make a request from Anne. Second, at other times we keep the recipient silent but let the allocator Anne explain her choice. Third, at other times we let both Anne and Bob have a turn at talking. Allowing recipients to ask had a big impact on sharing. Most people asked for 50% and said “it’s fair,” and giving rose by 60%. The bigger shock, however, was that allowing allocators to explain their choices sent giving tumbling – allocators

mostly gave nothing and simply said “I’m sorry” or “tough luck.” When people in Bob couldn’t make a request, Bob didn’t seem to get on Anne’s mind. Even worse, Anne appeared to be using language to put here perspective (“tough luck”) into Bob’s mind. So what happened when they both could speak? Something in between? Nope. The recipients, as before, asked for 50-50 and said “it’s fair,” but now the allocators said “yes, you’re right,” and generosity rose to record levels! How could this be? Why didn’t the allocators respond to the 50-50 request with “I’m sorry”? Is it that by asking, Bob got in Anne’s mind and she couldn’t get him out? In a final twist to our study we explored this forcing our subjects to imagine themselves in both roles—Anne imagine making request as if she were Bob, and Bob imagined explaining his sharing choice as if he were Anne. The result was conclusive. Giving in this “induced empathy” experiment was identical to the experiment in which both people got to speak. This tells us that talking works the same as simply imagining what you would say in the other’s position. That is, language by its very nature triggers empathic thinking and, as a result, altruistic choices. OUT OF MIND, OUT OF SIGHT What if you see someone is about to ask for something, and you know that if they do then a primitive part of your brain will be triggered by empathy and you will feel an urge to give. What do you do? In a field experiment conducted with Hannah Trachtman as part of her senior honors thesis at Harvard, we stationed Salvation Army bell-ringers outside one of two entrances to a supermarket and measured how it affected which door shoppers chose to go in and out of. We found people systematically avoided the door with the bell-ringers. So, people are sophisticated. They know that avoiding being asked is going to keep the recipient out of their minds and avoid the empathy triggered by language, which controls that natural urge we humans have for altruism. So, why did I agree to take a day out of my busy schedule to write this article? Simple. I was asked. H





Moral Sentiments & Climate Change 1 KNOWLEDGE OF SCIENCE AND SOCIETY There have been speculations that it is no accident that the most important cosmologist after Ptolemy was Nicolaus Copernicus (1473 – 1543), born only a decade before Martin Luther. Both attacked orthodoxy in different ways.1 Copernicus formulated a scientifically based heliocentric cosmology that displaced the Earth from the center of the universe. His book, De revolutionibus orbium coelestium (On the Revolutions of the Celestial Spheres, 1543), is often regarded as the starting point of the Scientific Revolution. Moreover, in1526 Copernicus also wrote a study on the value of money, Monetae cudendae ratio. In it Copernicus formulated an early version of the theory, now called Gresham’s Law, that bad (debased) coinage drives good (non-debased) coinage out of circulation, Margolis (2002) noted that something very significant occurred in the years after Copernicus. His ideas influenced many scholars: the natural philosopher, William Gilbert, who wrote on magnetism in De Magnete (1601); the physicist, mathematician, astronomer, and philosopher, Galileo Galilei (1564 –1642); the mathematician and astronomer, Johannes Kepler (1571 –1630). Philosophiæ Naturalis Principia Mathematica (1687), by the physicist, mathematician, astronomer and natural philosopher, Isaac Newton (1642 – 1726) is considered to be the most influential book in the history of science.2 Margolis (2002) argues that, from about 1600, scholars learnt to look at scientific and social problems from different angles, and that within the next two hundred years this habit of mind became quite common, and was, in fact, the reason why the technological/ industrial revolution gathered apace in the eighteenth and nineteenth centuries. After Newton, a few scholars realized that the universe exhibits laws that can be precisely written down in mathematical form. Moreover, we have, for some mys-

terious reason, the capacity to conceive of exactly those mathematical forms that do indeed govern reality. We believe that this mysterious connection between mind and reality was the basis for Newton’s philosophy. While celestial mechanics had been understood by Ptolemy to be the domain most readily governed by these forms, Newton’s work suggested that all reality was governed by mathematics. We shall call the underlying hypothesis entertained by these scholars the universality of mathematics. Major universal mathematicians include the Scot, James Maxwell (1831–1879), the Frenchman, Henri Poincar (1854–1912), the German, Albert Einstein (1879–1955), and the Englishman, Stephen Hawking (born 1942).3 Hawking and Mlodonow (2010) argue for this universal principle, citing its origins in Pythagoras ( 580 BCE to 490 BCE) Archimedes (287-212 BCE) and Euclid, and the recent developments in mathematical physics and cosmology. They present a strong form of this principle, called modeldependent realism, arguing that it is only through a mathematical model that we can properly perceive reality. Without the application of this universal mathematics, our society would be quite different and much poorer. Jardine (2008) discusses the scientific innovations by Hooke and Huygens in the period round the glorious revolution, while Appleby (2010) discusses the technological changes wrought by Arkwright, Hargreaves, and Crompton soon after. There is still controversy over whether the rapid technological and economic transformations that we experience today are the consequence of the development of science itself, or the result of the institutional changes in the political economy that started in Great Britain in the 1600’s.4 Ferris (2010) argues that the political and economic innovations of the time were linked to these developments in mathematics and science. The influence of Newton can perhaps be detected in the work of the philoso-

pher, mathematician, and political scientist, Marie Jean Antoine Nicolas de Caritat, Marquis de Condorcet (1743 – 1794), known as Nicolas de Condorcet. His work in formal social choice theory (Condorcet ([1785],1994) was discussed in Schofield (2006) in connection with the arguments about democracy by Madison and Jefferson. This work also discussed the influence of the work on Moral Sentiment by the Scottish Enlightenment writers, Francis Hutcheson (1694–1746), David Hume (1711–1776), Adam Smith (1723–1790) and Adam Ferguson (1723–1816). Between Copernicus and Newton, the writings of Thomas Hobbes (1588 – 1679), Ren Descartes (1596–1650), John Locke (1632–1704), Baruch Spinoza (1632– 1677), and Gottfied Liebnitz (1646–1716) laid down foundations for the modern search for rationality in life.5 Hobbes was more clearly influenced by the scientific method, particularly that of Galileo, while Descartes, Locke, Spinoza, and Liebniz were all concerned in one way or another with the imperishability of the soul.6 Liebniz in particular was concerned with an [E]xplanation of the relation between the soul and the body, a matter which has been regarded as inexplicable or else as miraculous. Without the idea of a soul it would seem difficult to form a general scheme of ethics.7 Indeed, the progress of science and the increasing secularization of society over the last century led Ferguson (2003) to note that [l]oss of faith in [the British Empire] often went hand in hand with loss of faith in God. 1.1 BELIEFS In the 1920’s and 1930’s, after World War I and the devastation wrought by the application of science and technology, a general fear become prominent that civi-


lization would fall, just as the Ottoman, Russian and Habsburg empires had fallen, soon to be followed by the British Empire.8 These fears were exemplified first by Spengler’s Decline of the West (1918, 1922) and later by Toynbee’s Study of History (1934). Ferguson (2006) quotes Spengler to the effect that The masses will accept with resignation the victory of the Caesars. Mead (2007) suggests, in contrast, that “it is to a dynamic religion rather than secularization that we must look for explanations of the Anglophone ascendency [of the American empires].”9 Indeed, much recent work substantiates the ideas of the Scottish moral philosophers, and the later suggestions of Darwin (1871), proposing that we all have an innate sense of moral values. Ober and Macedo (2006) suggest that moral goodness is something real, and does not need to be based on the notion of a trancendent soul.10 The last twenty years has seen a growing literature on a game theoretic analysis of the evolution of social norms to maintain cooperation in prisoners’ dilemma like situations. Gintis (2000, 2003), for example, provides evolutionary models of the cooperation through strong reciprocity and internalization of social norms.11 The anthropological literature provides much evidence that, from about 500KYBP years ago, the ancestors of homo sapiens engaged in cooperative behavior, particularly in hunting and caring for offspring and the elderly.12 On this basis we can infer that we probably do have very deeply ingrained normative mechanisms that were crucial, far back in time, for the maintenance of cooperation, and the fitness and thus survival of early hominids.13 These normative systems will surely have been modified over the long span of our evolution. A related literature deals with various detailed aspects of how these norms may have evolved.14 Some of this literature is also based on evolutionary theory15, some from neuroscience16, some from child development17 and some from the study of primates .18 Hauser (2006) argues that there is a deep structure to moral values, akin to the notion of a template in language19 while Deacon (1997) argues instead that lan-

guage and the brain co-evolve.20 Since language evolves very quickly (McWhorter, 2001; Deutcher, 2006), we might also expect moral values to change fairly rapidly, at least in the period during which language itself was evolving. In fact there is empirical evidence that cooperative behavior as well as notions of fairness vary significantly across different societies.21 While there may be fundamental aspects of morality and “altruism,” in particular, held in common across many societies, there is variation in how these are articulated. Gazzaniga (2008) suggests that moral values can be described in terms of various modules: reciprocity, suffering (or empathy), hierarchy, in-group and outgroup coalition, and purity/ disgust. These modules can be combined in different ways with different emphases. Currently this literature lacks a fundamental unifying theoretical structure, although the earlier work by George Price ([1971], 1995) gave a formal stochastic model relating fitness to traits that can be used to study selection in any evolving process, including economic development22 Binmore (2005, 2007) makes a number of very relevant comments on norms and culturally determined values, on the basis of notions from evolutionary game theory.23 The most important point is that norms can be seen as particular kinds of equilibrium selection mechanisms that are generated by the nature of the technology that the society has developed, and the enviroment in which it is located. So hunter-gatherer societies will tend to exhibit equity or egalitarian share and effort norms.24 Agricultural, or limited access societies, of the kind discussed North, Wallis and Weingast (2009), will focus on norms associated with hierarchy, power, honor and obedience. Open access societies will focus on norms of freedom, fair play and merit.25 The industrial development that occurred in Britain and the US in the past brought these equity norms into contest with economic principles of “efficiency” and the free market.26 The recent technological changes have exacerbated economic inequality, particularly in the US. These different normative beliefs about the proper balance between efficiency and equity are just as important as preferences in affecting political choice.

In any polity the underlying moral beliefs can be fairly heterogenous, reflecting these different emphases on efficiency, equality, freedom, and hierarchy.27 There is still no generally accepted theory about how these beliefs are propagated and transformed in a society. It has been suggested that they can be regarded as “memes,” acting like genes, mutating and multiplying under selection pressure.28 Indeed scientific notions, such as that of “meme” itself, as well as moral principles can be thought of memes.29 Bikchandani et al.(1992) write about fads and information cascades. Schofield (2006) introduced the notion of belief cascades in an attempt to capture the idea that such changes of political beliefs can be the result of new theories about how the world works, constructed in order to deal with the quandaries that the society faces.30 As we have suggested above, political beliefs will be affected by expectations about the future, as well as interpretation of the past. The collapse of the Soviet Empire in 1989 first brought about a sense of relief, as exemplified by the notion of the triumphant “end of history” of Fukayama (1992), and a period of stability and globalization. Below, we refer to this as the holocene, lasting most recently from about 1990 until 2001. However, American hegemony was short-lived. The “Clash of Civilizations” (Huntington, 1998) after 2001, the recent recession of 2008/9, and the current fears over the effects of climate change and international disorder, remind us of the earlier fears, in the inter war years, about the over-rapid development of science and the possibility of civilization’s collapse through war. In hindsight, these earlier fears in the 1930’s over future war were entirely justified. 2 UNCERTAINTY Many authors, from Paul Kennedy (1987) on, have discussed the similarities and differences between the Roman, British and American empires in terms of military over-reach and hubris.31 Indeed, Ferguson (2005) uses an interesting typology of empire, distinguishing between those that are autocratic, aristocratic, oligarchic or democratic, and whether they are based on the prncipal factors of land, labor or capital. While there are obvious differences between these empires, Ferguson


(2010) also suggests that the American empire, like earlier ones, may collapse in a chaotic fashion, possibly bringing about catastrophe. He notes that the total US federal debt increased from $5 trillion in 1992 to $7 trillion (about 70% of GDP) in 2000, to $17 trillion (about 117% of GDP) in 2010. In fiscal year 2000 there was a federal surplus of $236 billion, which by 2004 had become a deficit of about $520 billion, partly because of the Bush tax cuts.The estimated federal deficit for the fiscal year ending Sept. 30, 2010, is $1.47 trillion, over 10% of GDP. Stiglitz and Bilmes (2008) laid part of the blame for the increasing federal deficit on the Iraq war, citing a total estimated past and future cost of $3 trillion.32 Bacevich (2010) develops this theme of military over-reach, suggesting that the US has become wedded to permanent war. Like Kennedy and Ferguson, Bacevich (2010) argues that the recent crisis, and the problem of debt, has made this imperial military and economic strategy impossible for the US to maintain. In terms of economic decline, the trade deficit of the United States with China increased from $103 billion in 2002 to $268 billion in 2008, though it dropped to $227 billion in 2009. China now holds about $900 billion in US debt, followed by Japan with $800 billion. The rest of the US debt is spread between OPEC, Brazil, Hong Kong and Taiwan. We seem to be entering a new type of multipolar world, with no hegemon, and potential conflicts between regional powers such as China, India, Brazil, as well as the oil rich states of the Middle East and Russia.33 The 1990’s may, in the future, seem like an economic holocene, maintained by the economic and military hegemony of the United States. An important aspect of this dominance lay in the belief in the “soft power” of the US, namely the validity of the principles of democracy and capitalism.34 The double shock of 2001 and the crisis of 2008/9 has brought this period to an end, and it may well be that without such a hegemon, political and economic instability will be exacerbated. In recent years, fears over an uncertain future have been have been compounded by changes in our understanding about how the world, and society really work. First, at the level of the logic of science, Prigogine (1996) has used chaos theory to argue against the predictability and determinism of a Newtonia universe. Hayek

(1974) made a somewhat similar point, in his Nobel lecture, about the difference between economics and what he considered to be the way natural science works: While in the physical sciences it is generally assumed, probably with good reason, that any important factor which determines the observed events will itself be directly observable and measurable, in the study of such complex phenomena as the market, which depend on the actions of many individuals, all the circumstances which will determine the outcome of a process... will hardly ever be fully known or measurable. For Milton Friedman (1953) on the contrary, it was irrelevant whether economic theory made unrealistic assumptions, as long as it worked. But the recent recession strongly suggests that economic theory just does not work. The collapse of belief in the logic of economic theory is exemplified by the confession of Alan Greenspan, former chairman of the Federal Reserve, to Congress in 2008, when asked whether his ideology about market equilibrium was right, and working, replied that he was shocked to learn that it was wrong.35 He has also commented that “our current understanding of the future is extremely limited.” In the face of this uncertainty about the future, we argue that it behoves us to attempt to create an ethical basis for our actions when they have such possibly dire consequences. There may be disagreements about an ethical foundation for society, with a pure free market orientation at one pole, and an extremely egalitarian focus at the other. Almost all people believe in some version of propinquity , my family, my neighborhood, my country. On the other hand, there is belief that the future, our children, and future generations, should be protected from our greed. As an illustration, both Jefferson and Condorcet argued that debt or other liabilities should not be incurred if they could not be paid off in a generation. (Their argument was that in about a generation of 20 years, half the population would have changed through birth and death.) This is a version of “intergenerational utilitarianism” proposed by Collier (2010). This principle asserts that we should be “fair” towards the future, by taking into

account the expected overall utility of future generations.36 A natural consequence of this principle is that we should avoid destroying the world we live in for short term gain. Note that this is a utility principle, not an income principle. If climate change is expected to have greatest impact on the poor, in Africa say, then this principle implies that costs should be borne in the developed economies to offset the likely enormous utility costs of the poor in the future. One aspect of this calculation is the appropriate discount parameter to use. Collier suggests that if we do choose to burden the future, then we should lay aside assets to cover the anticipated future costs. Relatively risk free assets such as U.S.Treasury bonds give about 3 to 4% return, so this can be used to infer the appropriate transfer to the future. Posner (2005) estimates that the cost of climate change could reach about $8 trillion a year, so discounted at 3%/annum would give a total cost of about $65 trillion. If we follow Collier (2010) and do not discount the future then the total cost would be astronomical. An ancillary calculation made by Collier is that when we deplete non-renewable natural resources, oil, minerals etc. then we should also lay aside economic assets, namely investment capital, to cover the fact that these resources will not be available to the future. Finally, carbon, generated by our own economic activity, is a burden, a negative externality, that will affect the future, through its impact on climete. One way to cover the transfers to the future would be through a carbon tax. Since the developed economies currently produce the bulk of CO2,a carbon tax would have the beneficial effect of somewhat reducing consumption, in these economies, of carbon based fuels, and this would make non carbon fuels more viable. Collier suggests a tax of $40/ton of carbon emitted.37 Such a tax has the advantage that if estimated costs to the future rise, then tha tax rate can be adjusted. One further aspect of this way of dealing with the externality is the matter of uncertainty. There is a great deal of uncertainty at present, over the effects of economic activity. Even with the mathematical models of climate change that we discuss below, this uncertainty will persist. If our activities cause even more uncertainty over the consequences of our actions, then we should further compensate the future. Stern (2007,2009) argues that we


should be extremely risk averse over climate effects. Since future generations will face the costs of our decisions, we too should be uncertainty averse, and devote resources to the attempt at gauging these costs. One of the problems with dealing with climate change is that it concerns decision making in what are known as “large worlds.” Models of decision making work well in “small worlds” where probabilities can be estimated. Chilichilnisky (2009, 2010a,b) provides the beginning of a theory of decision-making in such “large worlds”involving uncertain, potentially disasterous “black swan” events.38 In our opinion, uncertainty about the future resides in the possibility that the dynamic systems that will determine our future are, in fact, chaotic. From the time of Newton to Laplace, the dominant notion in science was determinism. In the developing social sciences and economics, statistics provided a way of interpreting and controlling events. But the efforts to extend the simple Newtonian model of celestial mechanics by Poincar in the late nineteenth century showed that apparently deterministic physical systems could be deeply chaotic or nonpredictable.39 An essentially mathematical theory that has been developed in the few decades or so is complexity or chaos theory, dealing with the non-deterministic properties of dynamic systems.40 This theory is only a few years old but it already forces us to rethink habits of mind about how the world and society work. One area where this theory has proved of use is in understanding the complex positive and negative feedback mechanisms that govern climate and its effect on human evolution. Celestial mechanisms to do the Earth’s orbit interact with geological processes on the planet to affect the CO2 level. For example, the uplift of the Tibetan plateau has acted to remove CO2 over the last 40 million years, inducing oscillations between glacial and interglacial periods. The current ice age, the PlioceneQuaternary glaciation, started about 2.58 million years ago during the late Pliocene. The planet generally became drier during this ice age, and the ancestors of our species, Homo habilis (from 2.5 MYBP) and Homo erectus (from 1.8 MYBP), adapted to the new savanah conditions in Africa. Remains of Homo erectus have been found in Java dating to 1.6 MYBP. Above, we have

mentioned the extensive literature on the evolution of these early hominids. It has been argued that Homo erectus included meat in its diet, and used fire, thus increasing the energy available to became an efficient and cooperative predator.41 Mitochondrial analysis from modern humans suggests a common ancestor in Africa about 200KYBP.42 Equipped with language, a system of moral values, associated with cooperation, and a technology of increasingly sophisticated tools, this early hunter gatherer spread throughout the planet. It is thought that there were two conduits out Africa, about 70KYBP, one from the Horn of Africa and one across the Sinai peninsula into Asia. From about 90 to 10 KYBP, climate became highly unstable. Without our ancestors’ braininess, language and culture, the uncertainty induced by climatic chaos could have driven Homo sapiens to extinction. Indeed, it has been argued that an eruption in Sumatra about 70KYBP induced an instant ice age and almost killed off all H.sapiens. It may well have finished off H. erectus.43 The human population is estimated to be between 250,000 and 500,00 in 62KYBP, slowly increasing to about 6 million in 12KYBP, at the end of the ice age. Climatic amelioration at this beginning of the Holocene in 12KYBP meant warm, wet conditions over much of Eurasia allowing for the transformation of hunter gatherer society to agricultural communities in the Middle East. After the transition, human population increased to about 60 million in 3KYBP (the beginning of the bronze age) and then to about 240 million in 2KYBP. Farming appeared in the Fertile Crescent about 11.5KYBP with wheat, barley, then peas and lentils. It spread to Egypt by 9.5KYBP, and had independent origin in China and India about the same time, but much later in the New World (Diamond, 1997). Pastoral agriculture appeared about the same time: goats were tamed in Iran by 12KYBP, sheep in Iraq by 9KYBP, and various breeds of cattle in the middle east and India by 8KYBP. Cochran and Harpending (2009) argue that this population explosion was coupled with both cultural and genetic transformations. In particular, the change from hunter gatherer society to agriculture and “closed access society” was associated not


only with a dramatic increase in population and “total economic product,” but also in inequality, and the division of society into poorly fed peasants and military and technological elites.44 The induced Malthusian constraint meant that the “real wage” tended to decline except at catastrophic times when population crashed because of plague, as in the fourteenth century.45 From about 1600 our very braininess triggered a scientific explosion in the development of mathematical languages which allowed for the deeper analysis of the world and society. The beginnings of the agricultural and industrial revolutions in the United Kingdom from 1700 on and then later in the United States, and the transition to “open access societies” triggered rapid economic and population growth, but also initially caused an increase in inequality.46 This was reversed from about 1860, and further economic growth induced swift changes in the balance between capital and labor during the late nineteenth and early twentieth centuries. In 1860, GDP/ capita in both the UK and the US was about $2800, rising in parallel to about $5500 by 1914, and staying roughly constant during the times of turmoil until the 1930’s. After World War II, GDP/capita started to rise rapidly from $9.5K in 1950 to $30K by 2003 in the US, and from $7K to $21K in the UK.47 Until about 1970, this pattern of growth seems to have lessened the degree of inequality in the developed economies.48 The period from 1990 to the present can be seen as an economic holocene. In the same way we may see the previous long periods of growth in these two periods from 1860 to 1914 and from 1950 until about 2006 as economic and political holocenes. Both periods were characterizd by hegemonic leadership, first by Great Britain and then by the United States. Maddison estimates that world population grew from a billion in 1820 to about 1.8 billion in 1914. For this most recent period from 1950, world population grew from 2.5 billion to 6.8 billion. The population growth rate increased from about 1.5% in 1950 to over 2% in 1971, and has gradually fallen to 1.1% at present. This recent population growth has induced a number of changes in the world political economy. First, technological development has shifted the balance of economic power

both within developed economies and between the developed and less developed economies. Second, inequality within the developed polities has tended to increase from about 1970, partly because of the premium put on technological skill and partly due to the change in the age distribution of the population. This has been exacerbated by the transfer of manufacturing comparative advantage from developed to less developed countries, particularly China and India. These global changes have made political economic conflict much more difficult to resolve, and have suggested similarities between the present and the end of the last economic holocene in 1914. It is unlikely that we face anything like World War I, but it does now seems that the world and our society are much more complex than implied by the various social theories that were developed to facilitate growth in the past. It is very unclear what triggered the transition to open access society after 1700, to be followed by the disorder of the interwar period and then the astonishing changes after 1950.49 We have suggested that there are elements of the world and society, such as climate and the pattern of economic development, that are chaotic. This presents us with quandaries about how to make decisions with regard to the future. The most mathematical of our theories about society, namely general equilibrium, may also be deeply flawed, and we may need to think again about how to orchestrate our institutions to guard against risk. Since chaos and uncertainty are inextricably linked, a discussion of varieties of chaos can suggest to us why the future is so uncertain, and perhaps provide a better understanding of how to deal with the externalities that we are currently imposing on future generations.50 H Norman Schofield is The William Taussig Professor of Political Economy, Center in Political Economy, Washington University in Saint Louis, 1 Brookings Drive, #1027, Saint Louis MO 63130. Tel: 314935 5630. E-mail: This article is based on work supported by NSF grant 0715929. 3 REFERENCES Acemoglu, D., Robinson, J. (2006). Economic Origins of Dictatorship and Democracy. New York and Cambridge, Cam-

bridge University Press. Appleby, J. (2010). The Relentless Revolution. New York, Norton. Bacevich, A.J. (2008). The Limits of Power. New

York, Henry Holt. Bacevich, A.J. (2010). Washington Rules. New York, Metropolitian. Beinhocker, E.D. (2006). The Origin of Wealth. Boston, MA. Harvard Business School Press. Bikhchandani, S., Hirshleifer, D. and Welch, I. (1992). A Theory of Fads, Fashion, Custom, and cultural Change as Informational cascades, Journal of Politcal Economy, 100, 992-1026. Bikhchandani, S., Hirshleifer, D. and Welch, I. (1998). Learning from the Behavior of Others: Conformity, Fads, and Informational Cascades. Journal of Economic Perspectives 12, 151-170. Binmore, K. (2005). Natural Justice. Oxford, Oxford University Press. Binmore, K. (2007). Game Theory: A Very Short Introduction. Oxford, Oxford University Press. Binmore, K. (2009). Rational Decisions. Princeton, Princeton University Press. Blackmore, S. (2000). The Meme Machine. Oxford, Oxford University Press Bloom, P. (2004). Descartes’ Baby: How the Science of Child Development Explains What Makes us Human. New York, Basic Books. Bloom, P. (2010). How Pleasure Works: The New Science of why we like what we like. New York, Norton. Bowles, S. (2006). Group Competition, Reproductive Leveling and the Evolution of Human Altruism. Science 314, 1569–1572. Bowles, S. et al. (2003). The Co-evolution of Individual Behaviors and Socal Institutions. Journal of Theoretical Biology 223, 135-147. Boyd, J. and Richerson, P.J. (2005). The Origin and Evolution of Culture. Oxford, Oxford University Press. Calvin, W.H. (2003) The Ascent of Mind. New York, Bantam. Calvin, W.H. (2008). Global Fever: How to Treat Climate Change. Chicago, Chicago University Press. Cann, R.L. et al. (1987). Mitochondrial DNA and Human Evolution. Nature 325, 31-36. Chichilnisky, G. (1996). An Axiomatic Approach to Sustainable Development. Social Choice and Welfare 13, 231-257. Chichilnisky, G. (2009). The Topology of Fear. Journal of Mathematical Economics 45, 807-816. Chichilnisky, G. (2010). The Foundations of Statistics with Black Swans. Mathematical Social Science 59, 184-192. Choi, J.-K., and Bowles, S. (2007) The Coevolution of Parochial Altruism and War. Science 318, 636–640. Chomsky, N. (1972). Language and Mind. New York, Harcourt. Clark, G. 2007. A Farewell to Alms. Princeton, NJ, Princeton University Press. Cochran G. and Harpending, G. (2009). The 10,000 year Explosion. New York:Basic. Collier P. (2010). The Plundered Planet. Oxford, Oxford University Press. Condorcet, N. ([1795]. 1955) Esquisse d’un tableau historique des progr s de l’esprit humain: Sketch for an historical picture of the progress of the human mind. London, Weidenfeld. London. Darwin, C. (1982.[1871]). The Descent of Man, and Selection in relation to Sex. Princeton, Princeton University Press. Dawkins, R. (1976). The Selfish Gene.Oxford, Oxford University Press. Dawkins, R. (2008). The God Illusion. New York, Mariner. De Waal, F. (1997). Good Natured: The Origins of Right and Wrong in Humans and Other Animals. Cambridge, MA, Harvard University Press. De Waal, F. (2006). Primates and Philosophers: How Morality Evolved. Princeton, Princeton University Press. Deacon, T.W. (1997). The Symbolic Species. New York, Norton Dennett, D. (1995). Darwin’s Dangerous Idea. New York, Simon and Schuster. Dennett (2003). Descartes, R. [1637]. Discourse on the Method of Rightly Conducting One’s Reason and of Seeking Truth in the Sciences (trans. J. Veitch). New York, Doubleday. Deutscher, G. (2006). The Unfolding of Language. New York, Holt. Fagan, B. (1999). Floods, Famines and Emperors.

New York: Basic Books. Fagan, B. (2008). The Great Warming: Climate Change and the Rise and Fall of Civilizations. New York, Bloomsbury Press. Feingold, M. (2004). The Newtonian Moment: Isaac Newton and the Makingof Modern Culture. Oxford, Oxford University. Ferguson, N. (2001). The Cash Nexus: Money and Power in the Modern World, 1700-2000. New York, Basic Books. Ferguson, N. (2002). Empire: The Rise and Demise of the British World Order. London, Penguin Books. Ferguson, N. (2004). Colossus: The Price of America’s Empire. New York, Penguin Books. Ferguson, N. (2005). The Unconscious Colossus. Daedalus, Spring,18-33. Ferguson, N. (2006). The War of the Worlds. New York, Penguin Books. 13 Ferguson, N. (2008) The Ascent of Money. New York, Penguin. Ferguson, N. (2010). Complexity and Collapse. Foreign Affairs, 89,18-32. Ferris. (2010). The Science of Liberty. New York, Harper. Frank, S.A. (1995). George Price’s Contributions to Evolutionary Genetics. Journal of Theoretical Biology 175, 373-388. Friedman, M. (1953). Essays in Positive Economics. Chicago, Chicago University Press. Fukuyama, F. (1992). The End of History and the Last Man. New York, The Free Press. Galilei, G. ([1610],1992). The Sidereal Messenger (trans. and ed. A.Van Helden). Chicago, University of Chicago Press. Galilei, G. ([1632], 1967). Dialogue Concerning the Two Chief World Sys-tems, Ptolemaic and Copernican ( trans. and ed. S.Drake). Berkeley, University of California Press. Galilei, G. ([1638], 1974). Two New Sciences (trans. and ed. S.Drake). Madison, University of Wisconsin Press. Gaukroger, S. (1996). Descartes: An Intellectual Biography. Oxford, Oxford University Press. Gazzaniga, M.S. (2008). Human. New York, Harper. Gigerenzer, G. (2007). Gut Feelings. New York, Viking. Gintis, H. (2000). Strong Reciprocity and Human Sociality. Journal of Theoretical Biology 206, 169179. Gintis, H. (2003). The Hitchhiker’s Guide to Altruism: Gene-Culture Co-evolution, and the Internalization of Norms. Journal of Theoretical Biology 220, 407-418. Goldstein, R. (2006). Betraying Spinoza: The Renegade Jew who gave us Modernity. New York, Schocken. Hamilton, W. (1964). The Genetical Evolution of Social Behavior I and II. Journal of Theoretical Biology 7, 1-52. Hamilton, W. (1970). Selfish and Spiteful Behavior in an Evolutionary Model. Nature 228, 12181220. Hamilton, W. (1995). The Narrow Roads of Geneland; Volume I. New York, Freeman. Harman, O.S. (2010). The Price of Altruism. New York, Norton. Hauser, M. (2007). Moral Minds. New York: Harper. Hawking, S.W. (1988). A Brief History of Time. New York, Bantam. Hawking, S.W. and Mlodinow, L. (2010). The Grand Design. New York, Bantam. Hayek, F. von, (1945). The Use of Knowledge in Society. American Economic Review 35, 519-530. Henrich, J. et al. (2004). Foundations of Human Sociality. Oxford, Oxford University Press. Hitchens, C. (2007). God is not Great. New York, Hachette. Hobbes, T. (2009 [1651]). Leviathan; or the Matter, Forme, and Power of a Common-wealth, Ecclesiastical and Civil (ed. J.C.A. Gaskin) Oxford, Oxford 14 University Press. Hume, D. (1985 [1739]). A Treatise of Human Nature. London: Collins. Hume, D. 1987 ([1742, 1748]). Essays: Moral, Political and Literary (ed. E. Miller). Indianapolis, IN, Liberty Fund. Hume, D. (1987 [1748]). Philosophical Essays

concerning Human Understanding. Indianapolis, IN, Liberty Fund. Hume, D. (1987 [1758]). An Enquiry concerning Human Understanding. Indianapolis, IN, Liberty Fund. Huntington, S.P. (1998). The Clash of Civilizations and the Remaking of World Order. New York, Simon and Schuster. Jablonka, E. and Lamb,M.J. (2006). Evolution in Four Dimensions: Genetic, Epigenetic, Behavioral, and Symbolic Variation in the History of Life. Cambridge MA, MIT Press. Jacques, M.(2010). When China Rules the World. London, Penguin. James (2006) Jardine, L. (1999). Ingenious Pusuits. London, Little Brown. Johnson, C. (2004). The Sorrows of Empire. New York, Henry Holt.. Kennedy, D. (2001). Freedom from Fear. Oxford, Oxford University Press. Kennedy, P. (1987). The Rise and Fall of the Great Powers. New York, Random House. Krugman, P. (2009). How Did Economists get it so Wrong? New York Times, (September 6). Landes, D.S. (1998). The Wealth and Poverty of Nations. New York, Norton. Maddison A. (2007). Contours of the World Economy 1-2030 AD: Essays in Macro-Economic History. Oxford, Oxford University Press. Margolis H. (2002). It started with Copernicus: How Turning the World inside out Led to the Scientific Revolution. New York, McGraw Hill. Maynard Smith, J. (1972). On Evolution. Edinburgh, Edinburgh University Press. Maynard Smith, J. (1982). Evolution and the Theory of Games. Cambridge, Cambridge University Press. McWhorter, J. (2001). The Power of Babel. New York, Holt. Mead, W. R. (2007). God and Gold. New York. Knopf. Mlodinow, L. (2009). Drunkard’s Walk. New York, Vintage. Mokyr J. (2010). The Enlightened Economy: An Economic History of Britain 1700-1850. New Haven, Yale University Press. Murphy (2007) Newton, I. ([1687],1995) Philosophiae Naturalis Principia Mathematica (trans. and ed. , with an introduction by A.Motte) Amherst, NY, Prometheus Books. North, D. C., Wallis,B. R., and Weingast, B. R.

(2009). Violence and Social Orders: A Conceptual Framework for Interpreting Recorded Human History. Cambridge, Cambridge University Press. O’Rourke, K. H., and Williamson, J. G. (1999). Globalization and History.Cambridge,MA, MIT Press. 15 Ober, J., and Macedo, S. (2006). Introduction, to Primates and Philosophers: How Morality Evolved. Princeton, Princeton University Press. Pinker, S. (1997). How the Mind Works. New York, Norton. Pinker, S. (1999). Words and Rules. New York, Basic. Pinker, S. and Bloom, P. (1990). Natural Language and Natural Selection. Behavioral and Brain Sciences 13, 707–784. Posner, R. A. (2005). Catastrophe: Risk and Response. Oxford, Oxford University Press. Price, G. ([1971], 1995). The Nature of Selection. Journal of Theoretical Biology 175, 389-396. Prigogine (1996) Reich (2007) Ramo, J.C. (2009). Age of the Unthinkable. New York: Little Brown. Ridley, M. (1998). The Origins of Virtue: Human Instincts and the Evolution of Cooperation. New York, Penguin. Ridley, M. (2010). The Rational Optimist. New York, Harper. Robson, A. and Kaplan, H. (2003). The Evolution of Human Life Expectancy and Intelligence in Hunter Gatherer Societies. American Economic Review 93, 150-169. Samuelson, L. (2005). Foundations of Human Sociality: A Review Essay. Journal of Economic Literature 43, 488-497. Schofield,N. (2006) Architects of Political Change:Constitutional quandaries and social choice theory ( Cambridge, Cambridge University Press. Schofield, N. 2008. Modelling Political Economy. Homo Oeconomicus 25(3/4),1-32 . Schofield, N. (2010). Social Orders. Social Choice and Welfare 34, 503-536. Shapiro, R. (2008). Futurecast. London, St.Martins Press. Smith, A. (1981 [1776]). An Inquiry into the Nature and Causes of the Wealth of Nations. Indianapolis, IN, Liberty Fund. Spengler, O. (1961). The Decline of the West: Vol.1. The Downfall of the Occident [1918]. Vol.2. Perspectives of World History [1923] (ed. H.Werner, trans. and abridged, C.F. Atkinson). London, Vintage.

Spinoza, B. ([1677]). Ethics. (trans. R. Elwes). New York, Doubleday. Stern, N. (2007). The Economics of Climate Change. Cambridge, Cambridge University Press. Stern, N. (2009). The Global Deal. New York,Public Affairs. Stewart, M. (2006). The Courtier and the Heretic: Leibniz, Spinoza and the Fate of God in the Modern World. New York, Norton. Stiglitz, J. and Bilmes, L.J. (2008). The Three Trillion Dollar War. New York, Norton. Thuan, T.X. (2000). Chaos and Harmony. Oxford, Oxford University Press. Toynbee, A.J. (1934-1961). A Study of History. (12 volumes). Oxford, Oxford University Press. Trivers, R. (1971). The Evolution of Reciprocal Altruism. Quarterly Review of Biology 46: 35-56. Trivers, R. (1985). Social Evolution. Menlo Park, Cummings. 16 Von Hayek, F. (2007.[1944]). The Road to Serfdom (ed B.Caldwell). Chicago, The University of Chicago Press and London, Routledge & Kegan Paul. Warsh, D. (2006). Knowledge and the Wealth of Nations. New York, Norton. Weber, M. (1958. [1904]). The Protestant Ethic and the Spirit of Capitalism. New York, Scribners. Westen, D. (2007). The Political Brain. New York, Perseus Books. White, T. et al. (2009). Ardipithecus Ramidus and the Paleobiology of Early Hominids. Science 326,65-86. Wilson, E.O. (1981). On Human Nature. Cambridge MA , Harvard University Press. Wrangham, R. (2010). Catching Fire: How Cooking made us Human. New York, Basic. Wright, R. (1994). The Moral Animal. New York, Pantheon. Wright, R. (2000). Non-Zero. New York, Pantheon. Wright, R. (2009). The Evolution of God. New York, Little Brown. Zakaria, F. (2003). The Future of Freedom. New York, Norton

1. Weber (1904) speculated that there was a connection between the values of Protestantism and Capitalism. It may be that there are connections between the preference for scientific explanation and protestant belief about the relationship between God and humankind.

(2009) for the fears about collapse in the interwar years in Britain.

2. See Feingold (2004).

10. See also Wright (2010).

3. Hawking (2010) writes of being able to read the “Mind of God.”

11. Strong reciprocity means the punishment of those who do not cooperate.

4. See for example, Landes (1998) and Warsh (2006). 5. For Hobbes, see Rogow (1986). For Descartes, see Gaukroger (1995). For Spinoza and Liebnitz see Stewart (2006) and Goldstein (2006). 6. It is of interest that the English word “soul” derives from Old English s wol (first used in the 8th century poem, Beowulf. 7. Hawking and Mlodinow (2010) assert that God did not create the Universe, perhaps implying that the soul does not exist. However they do say that they understand Isaac Newton’s belief that God did create and conserve order in the universe. See other books by Dawkins (2008) and Hitchens (2007) on the same theme, as well as Wright (2009) on the evolution of the notion of God. 8. See Lieven (2002) for a brief history of these empires and Overy

9. A recent Gallup poll found that 70% of Americans regarded religion is an important part of their daily lives, compared with 27% of British.

12. Indeed, White et al.( 2009) present evidence of a high degree of cooperation among a very early hominid, Ardipithecus ramidus, dating back about 4MYBP (million years before the present). The evidence includes anatomical data which allows for inferences about the behavioral characteristics of these early hominids. 13. Gintis cites the work of Robson and Kaplan (2003) who use an economic model to estimate the correlation between brain size and life expectancy ( a measure of efficiencey). In this context, the increase in brain size is driven by the requirement to solve complex cooperative games against nature. 14. “Culture” can be thought of as the social context in which these norms are maintained. See Cavalli-Sforza and Feldman, 1981; Wilson (1978); Lumsden and Wilson (1981).



15. Gigerenzer 2007; Ridley, 1998; Wright 1994, 2000; Boyd and Richerson, 2005; Jablonka and Lamb, 2006. 16. Gazzaniga, 2006, 2008. 17. Bloom, 2004, 2010. 18. De Waal, 1996, 2006. 19. This is derived from the work of Chomsky (1972) and Pinker (1997, 1999), 20. See also Bowles et al. (2003), Bowles (2006), Choi and Bowles (2007) and Pinker and Bloom (1990) who present models of the coevolution of language, institutions and cooperation. 21. See Henrich et al. (2004, 2005), which reports on experiments in fifteen “small-scale societies,” using the game theoretic tools of the “prisoners’ dilemma,” the “ultimatum game,” etc. See also the review by Samuelson ( 2005). 22. Price’s work was used by Maynard Smith (1972, 1982) to develop the idea of an evolutionary stable strategy, and by Hamilton 1970) in a model of spite. See Frank (1995), Hamilton (1995) and Harman (2010) for discussion of Price’s work. 23. See also Binmore (2009). 24. Wrangham (2010), for example, argues that the discovery of fire for cooking enhanced sharing norms.

Shapiro (2008) on the changes in the balance of power as a result of globalization. 34. It seems that full democracy is far more difficult to build than was originally believed. The economic crisis has also led many to infer that the economic model underlying capitalism is completely wrong. 35. The comments by von Hayek and Greenspan are cited in Ramo (2009). 36. See also the argument in Chichilnisky (1996). 37. Total US emissions are about 5.6 million metric tons/ annum. One US gallon of gasoline 38. costs $2.70 and emits about 20lb of CO2 when combusted. If the USA imposed a tax of $40 on every quantity of gasoline that would emit one metric tonne of CO2 during combustion, the carbon tax on this gallon of gasoline would be 22 cents, an 8% increase. An average motorist uses about 400 gallons/annum and so emits less than 4 tons of CO2/ annum. 39. See Mlodinow (2008) for a discussion of chaos and randomness and Thuan (2000) for a discussion of the applicability of the idea of chaos in scientific revolutions. 40. See Prigogine (1997) for a philosophical discussion of the general ideas underlying this theory, and Beinhocker (2006) for a wide ranging application of some of these ideas to economics. 41. Wrangham (2009)

25. Modern fully developed democracies with free markets are termed open access societies by North et al. (2009).

42. Cann et al (1987). We use KYBP to mean thousand years before the present.

26. Mokyr (2010) charts the changes in belief that occurred in Britain in the period of industrialization, 1700-1850. David Kennedy (2001) gives a historical account of the beginning of US dominance in the period up to 1945.

43. Such a catastrophic event would cause a bottleneck in the development of H.sapiens, and may have induced a sudden and very rapid transformation in the evolutionary path.

27. Westen (2007) comments on the influence of moral values on political choice. 28. See Dawkins (1976) and later work by Dennett (1995) and Blackmore (2000). 29. Dennett (2003). 30. Indeed, much of the literature cited above can be seen as part of an extensive effort to construct a formal theory of moral values and beliefs based on the mathematical model of game theory. 31. See Ferguson (2001, 2002, 2004), Zakaria (2003), James (2006), Murphy (2007), and Bacevich (2008, 2010) 32. The Stockholm International Peace Research Institute estimated that the US 2009 military budget was $663 billion about 4.3% of GDP. An estimate for the Department of Defense budget for fiscal year 2010 is $685 billion. This expenditure has risen since 1999 when it was about 3%. However, other defense spending on Iraq and Afghanistan has brought the total for 2010 to about $ 1 trillion, about 7.5% of GDP. See the discussion in Johnson (2004) on militarism. 33. See Jacques (2010) on the rise of China as a rival, and

44. See North et al. (2009). 45. See Schofield (2010). 46. For Britain, Maddison (2007) estimates that GDP/capita grew from $1250 in 1700 to $1750 in 1820, measured in 1990 international Geary Khanis (GK) dollars. However, estimates of the real wage of building workers in 1700 and 1820 by Clark (2007) are identical. This implies inequality increased. The estimates by O’ Rourke and Williamson (1999) suggest inequality in the US only started to increase after 1890. See also Schofield (2008). 47. These are the estimates by Maddison (2007), measured in 1990 international Geary Khanis dollars. 48. Reich (2007) notes that the richest 1% received about 20% of income in 1927 but only 10% in 1970. 49. Acemoglu and Robinson (2006) and Mokyr (2010) give interesting accounts of the first period, and there are many accounts of the second. 50. Indeed, according to Hawking and Mlodonow (2010), without a theory that builds on chaos and uncertainty, we will have no understanding of our future.




Libertarian Paternalism and Growing Up Imagine that your two-year-old nephew wants an afternoon snack, and the available choices are either crunchy raw carrots or a chocolate bar. You want your nephew to be happy, and you want him to eat something in any case. Your options are to present the carrots first and only offer the chocolate if he refuses; or vice-versa; or to present both and let him choose. Most people would offer the carrots first, but this is not as straightforward as it might at first seem. What are the pros and cons of such a decision? What implicit assumptions are being made, if any? If it is the right decision for a parent or uncle, is it also the right decision for a state entity? Is there a version of this that also holds for adults, or does it only apply to children? This is an example of a recent concept in behavioral economics known as libertarian paternalism: setting up defaults and ‘choice architectures’ in such a way as to guide choices in a certain direction, while not precluding the autonomy of the individual to choose otherwise if he or she has a strong preference. It is also known as asymmetric or soft paternalism, and many people disagree with the libertarian moniker in this context, but at this point it seems destined to remain with us. The reason for the disagreement is that of course if choices are guided, then by construction anyone who wants to pick differently (e.g. your nephew choosing chocolate) has to pay some additional cost – whether in dollars or time or effort – to do so, however small. In that sense it is not fully libertarian, and again by construction it is not neutral. Probably the most well-known example of libertarian paternalism within economics is choosing a default savings rate for retirement plans: typically this has been 0%, but why not make it 3% or 6% or the maximum eligible for em-

ployer matching? It has to be something (0 isn’t neutral either), and for purely practical reasons most people are likely to want a positive number so one of the latter options will probably reduce the total time spent switching between rates. Of course there is usually a normative component as well: the employer (government?) may believe that the optimal rate for the employee, according to his or her own preferences, is not likely to be 0. This positive default is the paternalistic aspect, while the libertarian aspect is that the employee is free to choose any rate they want at any time. Other classic examples of this concept are placing healthy foods more prominently in cafeterias, or making organ donation ‘opt-out’ (i.e. presumed consent to donate) rather than ‘opt-in’ as currently exists in the US. One can even view the London sidewalk admonition for pedestrians to “Look Right!” as a so-called nudge, since it guides a certain behavior (at a small but nonzero cost of being momentarily distracted from one’s ongoing activities) but doesn’t stop anyone from crossing the street without looking right if they so wish. There is a rapidly growing literature on this idea, ranging from the general interest book “Nudge” by Thaler and Sunstein to dueling articles in law reviews, which are all quite readable and informative. However, there are several less-explored dimensions to libertarian paternalism, a couple of which are highlighted by the example above. First, what happens with children? Most people do not have a problem with being paternalistic toward children (indeed, that is the origin of the word) or others who are ‘obviously’ unable to decide for themselves, such as those who are in a psychotic state or falling-down drunk. But what makes it obvious in any of those cases that the agents in question are

likely to make a mistake with respect to their own preferences? How does one even carefully define a mistake? Assuming that there are compelling arguments in those cases, which there seem to be, would not those arguments also occasionally apply to non-psychotic adults? For instance, someone has just received a terrible and unexpected medical diagnosis and immediately opts for a risky surgical procedure. Perhaps this highly emotional state is not the best time to be making irreversible (if not possibly fatal) decisions. There is a need to draw some sort of line for determining the presumably rare (except in the case of children) situations when autonomy should be abrogated, and the presumably less rare cases when it should be slightly compromised (as in setting defaults on savings rates and organ donation). This line is unlikely to coincide perfectly with the agent’s 18th birthday, which is the current default (!) in society. Of course this goes both ways: many 15-year-olds do not require assistance in choosing what color shirt to wear in the morning, whatever their parents may think. Choosing such a line, which in addition will be different for parents versus governments, is necessarily difficult, although one important criterion will be fleshed out below. Sometimes an arbitrary line will be optimal either because the costs of the debate would outweigh the costs of policy errors or because of concerns that the debate will be corrupted by special interests. However, this does not absolve society of acknowledging the existence of such a line even now and of making some hard decisions regarding where it should lie. Another relevant consideration about children in this context is that their preferences are not yet wellformed. This makes paternalism both more legitimate but also more fraught.

Returning to the opening example, presenting your nephew with carrots first is likely to lead to more consumption of carrots over the years, which in turn has been shown to lead to a higher utility for carrots as an adult (at least if they were not forced down the throat). That is, you are to some extent choosing the final preferences of someone else, much as we all to some extent alter our own preferences e.g. by exposing ourselves to classical music or exotic foods and learning to appreciate them. Is this fair to the child? If there is no neutral framing, then there is no way not to do this, so the question becomes: what preferences would the child want to have as an adult? Does it even make sense to talk about preferences over preferences? In practice it is probably the preferences of the parents and aunts and uncles and schoolteachers and advertisers and (yes) government bureaucrats regarding the future preferences of the child that come into play, none of which are purely altruistic or benevolent. In the present example, it is probably the case that most adults would prefer to have a higher relative utility for carrots to chocolate than they currently do (which doesn’t mean that they would prefer to like carrots more than chocolate in some absolute sense), but that may not be a rigorously empirically supportable proposition. There is at least for now an undeniable subjective and capricious element to such paternalism, even on the part of well-meaning parents, but this is an exciting area for future theoretical and experimental research. The second unusual dimension of libertarian paternalism that is worth highlighting involves intertemporal choices. Libertarians believe that people should be allowed to decide for themselves, including making mistakes, as long as it does not adversely affect other individuals. This is a fairly well-accepted policy starting point for many Americans, and with good reason. But one oftenunaddressed question is how to define “other individuals.” Obvious? Well, no. In particular, are future versions of you

different individuals than the present version of you? Recent neuroimaging evidence suggests that people use exactly the same regions of the brain (especially the temporoparietal junction) to think about themselves in the future as they do to think about and predict the behavior of other people, which is known in psychology as theory of mind. But the idea goes back to the Greek philosopher Heraclitus, who said that you cannot step into the same river twice: both it and you will have changed in the interim. It seems that we think of our future selves (say, in a year) in the same way that we think about good friends: someone we care about and can predict fairly well, but still to some extent as an ‘other’ who remains foreign or alien. This is actually a standard modeling trope in many economics papers on savings behavior, in which people play strategic games with their future selves, but it appears to be more literally true than was originally imagined. In that case, who are we to make decisions on behalf of our future selves? Introspection and experience tell us that we often regret our past choices, which is further evidence that there is the potential for conflict. Unfortunately this leaves a partial quandary: with friends and neighbors, there is a natural alternative to us deciding on their behalf because they are (barring the rare exceptions discussed earlier) ready to decide for themselves. With future selves who don’t yet exist (including the adults that children will grow into), this isn’t so easy, and indeed it isn’t always possible. Many times we are forced to make decisions that will impact our future selves more than it will impact us (think of the savings rate above, or an impulsive and ill-advised email), and it can’t be helped. This is often close to unavoidable, but it’s worth keeping in mind that it is only second-best (i.e. constrained) optimal. However, this does suggest a useful criterion when debating the not-sobright line emphasized previously: how much does a given policy leave open the door to change one’s mind in the future?

For example, parents may paternalistically disallow tattoos (which have longterm implications) but on the flipside allow free will over hairstyling decisions (which do not). In terms of policy, this suggests regulating or taxing drugs (or not) less on the basis of how dire the effects are, but rather on the basis of how addictive they are. It also suggests waiting periods for everything from handgun purchases to marriage licenses. These are definitely paternalistic in that

It seems that we think of our future selves (say, in a year) in the same way that we think about good friends: someone we care about and can predict fairly well, but still to some extent as an ‘other’ who remains foreign or alien. they impose a real cost on the current self, but they also protect the future selves from potentially costly mistakes in domains where such things have been known to occur. Needless to say, this desire for flexibility is only one possible criterion among many. The underlying point is that we all agree on being paternalistic in certain contexts but not others. To better understand which situations call for which approach, and to defend those decisions, we need to have an explicit and public discussion. That discussion should be informed by economic theory; by data from economics, neuroscience, and elsewhere; by moral and political philosophy; and by our collective informed judgment. H




Behavioral Economics and Interdisciplinary Endeavours In 2002, Daniel Kahneman became the first psychologist to win the Nobel Prize in Economic Sciences. Although his research on “prospect theory” with Amos Tversky marked a clear contribution, the committee awarded Kahneman the prize “for having integrated insights from psychological research into economic science.” By importing ideas from psychology into economics research, Kahneman, Tversky, Richard Thaler, and others, founded the subfield of behavioral economics in the 1970s and 80s. Since then, behavioral economics has developed into a mature science of human decision-making with standard experimental methods, wellreplicated results, and novel theories that both explain empirical findings and generate testable predictions. This relatively young field is now an exporter of ideas and methods and has exerted a great influence on disciplines as diverse as anthropology, evolutionary biology, developmental psychology, and primate cognition. The general finding of behavioral economics is that human behavior deviates from economic models of rational selfinterest, but it does so in predictable ways. People are pro-social – they show concern for others that often goes against their immediate self-interest. But people also expect others to follow norms for cooperation and fairness and will pay to punish those who violate their expectations. These findings are not very surprising. We have all helped others in need and felt anger towards those who behave unfairly. However, behavioral economics helped to define these behaviors systematically using novel methods—economic games, in particular. Economic games are simple by design yet measure real behavior. Players stand to gain or lose money—sometimes very large stakes—based on their decisions and those of other players. For example,

in the Dictator Game, player A receives an amount of money and is told that he or she can keep all of it or give some to a complete stranger, Player B. Giving anything in this game goes against one’s selfinterest and thus, provides a measure of individual altruism. Surprisingly, a significant percentage of people give money in the game. In the Ultimatum Game, Player B has a say – he or she can reject Player A’s offer, in which case neither player gets anything. For Player B, rejecting any positive offer goes against self-interest, providing a measure of an individual’s sense of fairness. As you might suspect, people tend to reject offers that are less than 20percent of the stake. One common criticism of such games is that they are unrealistic or devoid of context. The two players remain anonymous to each other and similar situations rarely arise in life. These are valid concerns, but miss the value of the method. In novel experimental situations, people cannot respond in a rehearsed manner—they need to think about their actions and the consequences of their decisions. Offer too little in an Ultimatum Game, and you could go home with nothing. Further, with the context well controlled, the factors that affect decisions can be tested systematically. For example, if Player A knows something about Player B in a Dictator Game or if their donation decision will be known to others, then donations increase compared to the standard version of the game. But I believe the main advantage of economic games is their simplicity. These methods have been adopted by other disciplines because they can be easily adapted for different ages, different species and different cultures. And these extensions of the methods have led to several important and surprising findings. Windfall gains are rare in adult life, particularly those that come with

hints about sharing. But for children, these situations are common. Parents often give their children food or toys that could be shared with others. In this sense, the Dictator Game (DG) offers a good test of altruism in children. The economists William Harbaugh, Kate Krause and others have performed such tests and found that children give more with age and are more generous to strangers than college students. However, the DG can also be used to test the factors that influence children’s generosity. My colleague, Dave Rand, and I used this method to test the effects of currency type on children’s giving. Using stickers instead of money, we asked children to identify their favorite and least favorite stickers from a set of four types of stickers. We then had children play two DGs: half the children played with their favorite stickers first and then their least favorite and the other half played in the opposite order. We told children that their donations would go to two different children who were coming tomorrow. (To check if playing two rounds of a DG affected children’s performance, a separate group played a single round DG with their favorite stickers only. The results of that game did not differ from favorite sticker donations in the two-round game.) Comparing children’s donations of favorite and least favorite stickers, we saw a clear effect of currency. Children gave away more of their least favorite stickers than their favorite stickers. This effect held for children at all ages tested, between 3 and 6 years of age. However, when we looked only at those children who gave at least one sticker away, donations did not increase with age. All children tended to give about 50 percent of their least favorite stickers and about 40 percent of their favorite stickers. In other words, even the youngest children seemed to recognize an equal split as a norm for

sharing, but one that they could bend in their favor when the resource was highly valued. We are currently using a similar game to test how children learn these norms for giving. Other researchers have used the Ultimatum Game (UG) to assess the developmental origins of fairness and strategic thinking. Between the ages of 3 and 5 years of age, children develop perspectivetaking skills that help them to understand the point of view of others, a so-called theory of mind. A group of Japanese researchers lead by Haruto Takagishi assessed the relationship between theory of mind development and strategic play in the UG. Using 10 candies as currency, two children played the game face to face, with one proposing a division of the candies and the other responding to the offer by either accepting or rejecting it. All children also received a standard test for theory of mind. The results showed a clear effect of perspective taking skills. Children who failed the theory of mind task offered an average of about 2.5 candies to their peer, whereas those who passed the

task offered about 4.5 candies. The low offers were rejected by peers, but fair offers were not. Finding that perspective taking plays a role in games like the UG may not be terribly surprising, but this experiment and others help us to identify the origins of strategic thinking that underlie many economic interactions. By 4 or 5 years of age, children play the UG much like adults do, considering the reaction of the responder when making an offer. This simple cognitive skill appears to be unique to humans. In a clever adaptation of the UG, Keith Jensen and colleagues tested chimpanzees playing side by side. Each pair of animals faced a choice of two possible divisions of food, each on a separate tray. For example, one tray had 5 pieces of food for each player and the other had 8 pieces for the proposer and 2 for the responder. The proposer selected which tray he or she wanted to use by pulling it closer. However, in order for both animals to get their respective allocations of food, the responder had to pull the tray the rest of the way. With this set up, responders could reject the â&#x20AC;&#x153;offerâ&#x20AC;?

by simply refusing to complete the pull. The results showed that proposers almost always chose the tray with more food for themselves, and responders accepted the offer as long as they received something. However, proposers were not strategic in the game. When faced with a 5-5 versus 10-0 option, they would chose 10-0 and responders would refuse to complete the pull. The results of the UG and other tasks with chimpanzees suggest that our most recent primate ancestors did not possess the cognitive capacities required for strategic economic interactions. In fact, chimpanzees do not show concern for their peers when allocating resources even when there is no cost to themselves. When faced with a choice of one piece of food for each animal and one for me and none for you, chimpanzees choose randomly. What evolutionary change occurred to enable human strategic interactions? One possibility is that humans developed the capacity for culture in general, and cultural norms of fairness in particular. Fairness matters to us and drives re-

jections in the UG, but what we consider fair also varies by culture. In a large cross cultural study, Joseph Henrich and many anthropologists ran UGs in 15 small-scale societies, including hunter-gatherers, agriculturalists and nomadic herders. The results showed striking variation both in the offers made and responses to the offers. For example, in South America, offers of less than 20percent were common among the Machiguenga and Mapuche, but these offers were also frequently accepted. In Papua New Guinea, the Au and Gnau frequently made offers above 50percent, but these were frequently rejected. Much of the variation between cultures could be explained by the extent to which each society engaged in economic interactions outside the family. But some of the variation could only be explained by particular cultural practices. For example, the Au and the Gnau have gift-giving cultures where large offers could be seen as statusseeking behavior and accepting a large gift implies an obligation to reciprocate. Fairness is clearly an important factor in real life economic transactions, and norms of fairness vary across cultures. People are even willing to pay to enforce fairness when they will not directly benefit. In a multiple player game called the Public Goods Game (PGG), four players are each given a starting stake such as $20. Each player then decides how much to contribute to a common pool which will be multiplied and then distributed equally to all players. Thus, if everyone gives all $20, the pool of $80 is multiplied by 1.6 to $128 resulting in a payout of $32 per player. Clearly, it is best for everyone if all of the players contribute their entire stake, but there is a temptation to behave selfishly. If I withhold my stake while everyone else contributes theirs, we all receive a payout of $24, but I end up with much more than my peers ($20 + $24 = $44). This behavior—known as free-riding—begins to occur when multiple rounds of the PGG are played and quickly becomes the favored strategy. In an original variation, Ernest Fehr and Simon Gächter created a PGG with multiple rounds where each round was conducted with a new group of players. At the end of each round, players could see what the others in their group had contributed and could then pay to punish other players. Punishment was implemented as a 1:3 ratio: one could pay $1 to punish another player by subtracting $3 from their stake. With this option

in place, groups quickly converged on a cooperative norm that raised payouts for everyone. Intriguingly, players were willing to punish free-riders even though they would not encounter those players again. Punishment can help to enforce norms of fairness, but as noted above, norms of fairness vary across cultures. Benedikt Herrmann, Christian Thöni and Simon Gächter found differences in punishment by bringing the PGG described above to 16 different countries. The results were unexpected. In some countries, players punished those who made high contributions to the common pool, so-called anti-social punishment. In several cultures, such as Greece and Oman, both high and low contributions were punished, resulting in

It seems that we think of our future selves (say, in a year) in the same way that we think about good friends: someone we care about and can predict fairly well, but still to some extent as an ‘other’ who remains foreign or alien. the same level of cooperation seen when there was no punishment allowed. These results show that punishment alone cannot enhance cooperation—social norms of cooperation are needed that encourage generosity. Research into the origins of economic behavior continues to expand to different levels of analysis, with much of the work being done at Harvard. David Laibson in the Department of Economics conducts brain-imaging studies to examine the neural circuitry that supports economic decisions. Martin Nowak and colleagues at the Program for Evolutionary Dynamics use economic games and modeling techniques to investigate the biological and cultural evolution of cooperative behavior. But while extensions of behavioral economics has lead to promising advances within disciplines, progress across

disciplines is more difficult to measure. Many disciplines investigate similar problems, such as cooperation and altruism, but use very different languages and concepts to describe and categorize behaviors. For example, economists use terms like preferences and norms to describe economic behavior, but similar terms are used by psychologists to describe a broader range of behavior. Preferences can apply to ice cream, as well as to equal distribution of resources, but does similar cognition support both? Norms can be described as moral, injunctive, conventional or subjective with each term carrying different implications and supported by different empirical results. How can we situate norms of fairness among these conceptual distinctions? Even general terms like altruism are problematic. Defining an altruistic act as one where I pay a cost to confer a benefit on someone else ignores my motivations. Can there be altruism without empathy? Does it count if I’m trying to impress a third person by being nice to others? These differences in usage are not incommensurable but highlight the need for collaboration across disciplines. Economists, psychologists, anthropologists, and evolutionary biologists all benefit by using common methods. But experts in one discipline often interpret their results through post hoc references from another discipline, ignoring the complexities within that body of literature. Collaborations avoid this issue by enabling well-designed experiments that can address questions of common interest across disciplines. When delivering Harvard’s Mind, Brain and Behavior lectures in April 2008, Kahneman joked that psychology gave economists permission to use their common sense. By drawing on insights from other disciplines, behavioral economists have created an exciting body of research, which has taken on an interdisciplinary life of its own. If we are to capitalize on the use of common methods to answer broader questions about human behavior, some common sense is again warranted: cooperate across disciplines to maximize the payouts for all. H Peter Blake is a developmental psychologist and a postdoctoral fellow at the Program for Evolutionary Dynamics at Harvard.


Effect of Financial Incentives on Status-based Discrimination in a Marriage Market Of the different forms of discrimination that exist in various domains of human society, status-based discrimination is the one wherein high-status groups discriminate against low-status groups and sometimes vice versa. A prominent example of status-based discrimination is the Indian arranged marriage markets in which this type of discrimination manifests itself strongly. The existing empirical research indicates that Indian marriage markets are prone to strong discriminatory mate selection behavior based on social identity/status of an individual, where caste identity of an individual is an easily measured indicator of social status. The origin of the caste system in India can be traced back to occupation-based classification of the Hindu society. The Hindu society was historically divided into four mutually exclusive, rankable, endogamous, and hereditary caste-groups or Varnas – the Brahmins (priests), Kshatriyas (warriors), Vaisyas (traders & merchants), and Sudras (those engaged in menial jobs). However, the operational category that determines the modern social code is the Jati, which is a subset of the Varna and shares the basic attributes of the Varna. Jati refers to the specific community to which an individual is supposed to be married and forms an individual’s social identity. In modern India, impressive economic growth has seriously weakened the ominous link between caste and occupation, especially in urban India. Hence it is natural to wonder whether economic growth can make headways to weaken the castebased discrimination observed in Indian marriage markets as well? The available research on Indian marriage practices indicates that the answer to this question may not be in the affirmative. Observations from related surveys on this matter extend major support to the dominant view. For example, a survey conducted in 2006 by the CNN-IBN (the Indian subsid-

iary of CNN) found that a whopping 74% of respondents think that inter-caste marriages are unacceptable. What explains the persistence of caste-based discrimination in marriage markets in modern India? In the Hindu society, a husband’s caste determines the caste-status of the wife and children. Hence for an upper-caste Hindu female, there exists a strong disincentive to marry a lower-caste male as she and her children would experience a decline in caste-status if she were to marry down the caste ladder. Even though by the same logic lower-caste females have strong incentive to marry up, they often face harsh social sanctions if married to a higher-caste male. Thus, females face serious disincentives to cross the caste line.

In contrast, an upper-caste male does not experience a loss of his caste-status if he marries someone below his own caste, however he can also face strong pecuniary and non-pecuniary pu¬nishments for marrying outside his own caste. Newspaper reports of honor-killings to prevent or oppose inter-caste marriages are not rare in some parts of the country. Thus, disincentives to cross caste lines loom large on both sides. Do these facts imply that there is no way to break away from these caste barriers in the arranged marriage markets in India? Historically, caste-status was closely tied to economic status due to the caste-based occupational segregation. Hence, for an upper-caste female marrying a lower-caste

male, she would not only encounter a decline in her caste-status but would also experience a loss in economic status. However, as the caste-occupation link weakens, so does the link between caste-status and economic status. Hence in the present days, for an upper-caste female marrying a lower-caste male, a loss in her caste-status may not necessarily be associated with a loss in economic status. Instead in some cases upper-caste females may be compensated by a gain in economic-status. Since marriage is one of the most important factors that shape the economic well being of a woman in India and everywhere else, it is pertinent to inquire whether large financial incentives (for example, potential groom’s income) can enhance the willingness of higher-caste women to marry lower-caste men. This relates to a broader question of whether and what magnitude of financial incentives can diminish the incidence of status-based discrimination in real markets. To shed light on this question, we conducted a controlled field experiment in a reputable well-functioning arranged marriage market that is conducted via matrimonial advertisements in a widely circulated and well-reputed Bengali language newspaper. Bengali is the language that is primarily spoken in the state of West Bengal. West Bengal is one of the few states in India that has an impressive progressive socio-political history and has witnessed social and working class movements with strong anti-caste emphasis. According to a study by Ashwini Deshpande, West Bengal is in the middle of the spectrum amongst all the Indian states in terms of the absolute economic deprivation. However, in terms of the inter-caste economic disparity, it is one of the states with the least disparity. Against this backdrop, West Bengal appears to be one of the most favorable states to find evidence in support of our conjecture that there may exist a casteincome tradeoff. Failure to find evidence in favor of our conjecture in a socially progressive state like West Bengal would imply that there exists little or no hope about the ability of economic progress to overcome the caste-based discrimination in the sphere of arranged marriages in India. To investigate our conjecture, we posted matrimonial advertisements of grooms in the ‘brides wanted’ category of the Bengali newspaper. We systematically varied either the groom-caste (holding groom-monthly income constant)

or the groom-monthly income (holding groom-caste constant) in these advertisements, while we held constant all other groom attributes. In other words, two grooms belonging to two different caste groups (monthly income levels) appear observationally identical in every aspect to the readers except their caste affiliation (monthly income category). Marriageable males participating in this market typically advertise their caste and income information, among other attributes. Furthermore, after these advertisements appear in an edition of the newspaper, interested parties (mostly the brides’ families) contact the advertiser (mainly by letter or email) expressing their interest in that advertisement. This provides us with a confidential and reliable measure of “first intent” of the interested party. Therefore, the conjunction of an exceptionally strong status-based discriminatory mate selection behavior, a reputed market for arranged marriages where males typically advertise their income, and an exclusive correspondence process furnish us with a unique opportunity to investigate our conjecture. The responses to these advertisements would provide us information about how the caste-status and income attributes of potential grooms affect the initial response behavior of potential brides’ families. Note that it is extremely difficult to obtain caste information from official databases in India and almost implausible to obtain corresponding income information as well as other relevant marital attributes that one needs to control for to get to the bottom of the issue in question. Hence it necessitates a field experiment that systematically controls for other marital attributes and clearly focuses on the caste-income relationship. Status-based discrimination is said to arise in our context if responders from a given high-caste group send significantly lower number of responses to a lower-caste groom relative to their own-caste groom despite the fact that the own-caste groom is otherwise observationally identical to the lower-caste groom. If the number of responses from a higher-caste responder group for grooms from a given lower-caste group increases considerably with an increase in the advertised monthly income of those grooms, then we contend that the discrimination based on caste-status can be reduced if the discriminated group provides sufficient income incentive to the higher-status group.

The response data reveal that highercaste responders discriminate against the lower-caste grooms. That is, they respond significantly more to their owncaste grooms than the lower-caste grooms despite the two grooms being otherwise observationally identical. This finding is in perfect agreement with the dominant view about the Indian marriage markets. However, in spite of their discriminatory behavior, the responses from higher-caste families increase significantly with an increase in the advertised monthly income of lower-caste grooms who are otherwise observationally identical to each other. This finding indicates that higher-caste females are willing to accept a loss in the caste-status by marrying a lower-caste groom only when the prospect of economic well-being is high enough. In other words, there may exist some substitutability between loss in caste-status and gain in terms of the groom-income on the part of higher-caste females in the market we studied. The magnitudes of additional monthly income that a lower-caste groom needs to offer to higher-caste responders to completely offset their discriminatory response behavior, i.e. to be treated at par with the higher-caste grooms, are substantial. According to our data the estimated compensation was as high as Rs. 49,000, which is over three times the median monthly income of the grooms in our experimental design. This result may have important policy implications for the governments that aim to tackle the issue of caste-based discrimination in India. A case in point is the recent efforts by the state government of Madhya Pradesh (an Indian state), which has announced a host of financial incentives to reduce socio-economic inequalities. One such incentive announces to provide a fixed sum of Rs. 50,000 (over $1000) to the couples engaging in an inter-caste marriage in that state. Given that economic incentive is a powerful motivator of human behavior, one can conjecture that if the phenomenal economic growth in India succeeds in reducing inter-caste economic disparity, caste-based discrimination in the marriage market may start eroding. H - Assistant Professor, Department of Economics, University of Utah, 260 S. Central Campus Drive, Orson Spencer Hall, Rm 343, Salt Lake City, UT, 84112. Email: haimanti., Phone: 1-801-581-7481, Fax: 1-801-585-5649.



Interview with Greg Mankiw Professor N. Gregory Mankiw has been an integral part of the economics departent at Harvard University for over 25 years. He is not only a prolific author, but also a dedicated professor of Harvard’s introductory economics course Ec10 and a leading macroeconomist, who has served as Chairman of the Council of Economic Advisors. In the following interview with HCER, Professor Mankiw shares his thoughts on the financial crisis, policy responses, learning and teaching, and life in college. HCER: Two years after its breakout, the financial crisis is still puzzling in many dimensions. Has the crisis changed your views on any economic issues? Mankiw: I think one of the clear ways that the crisis has changed the study of macroeconomics is that people are more focused on the role of financial institutions. In many macro models—both those taught at the undergrad level and those taught at the graduate level— financial institutions were ignored for the purposes of policy analysis and the analysis of economic fluctuations. The current set of events has taught us that you can’t do that. Financial institutions can be at the center of economic downturns, and we need to figure out a way to incorporate that very clear fact into our models. So, in the process of revising the textbook [Principles of Economics] we use in Ec10, it got pretty clear to me that I needed to talk more about financial institutions and the basic dynamics of the financial crises. One of the things I hadn’t talked about before in Ec10 or in the textbook is the role of leverage. Economists have known for a long time how leverage works, and without understanding leverage, you can’t really understand this particular crisis. I realized that when I talked to a student in Ec10 last year. I said to the student that, in some sense, what happened was that these financial institutions invested in the housing market via mortgage backed securities; when the, price of housing fell by 30% in the crisis, these institutions found themselves in bankruptcy. Then the student said, well, if the price had only fallen by 30%, why did they lose all their money? That was a very good question, and the answer, of

course, is leverage. But I realized we didn’t really talk about the economic impact of leverage, how leverage can amplify returns, and how the financial institutions can end up being bankrupt as a result. So the next edition of the textbook will have a few pages on leverage. HCER: Now it seems that the financial sector has recovered relatively quickly whereas the real economy is still in trouble. What do you think about that? Mankiw: I don’t think that’s a surprise. The financial markets are often a leading indicator of what’s going on in the real economy. The real economy takes longer to repair itself. My colleague Ken Rogoff has done more studies about the aftermath of financial crises than anyone I know. His consistent finding is that the aftermath of a financial crisis is a very slow recovery. It seems what we are experiencing right now is pretty much in line with this historical pattern. HCER: Looking ahead, do you worry about a double-dip recession? Mankiw: Sure, I do worry a lot, for a couple of reasons. One is that the recovery so far is very meager. The speed of GDP growth is perhaps in the 2% range. 2% is below the historical average of 3%, which is what you need in order to keep unemployment stabilized, and you’ll need GDP growth rate to be higher than 3% to reduce unemployment. That is, it’s looking like we don’t have enough growth to get the unemployment down. So we are starting with the situation where the recovery is meager, and on top of that, a variety of shocks that might occur—like a crisis of confidence—can easily push us back into recession. If that happens, the Federal Reserve has very few tools left to fight it. The risk of a double-dip is larger than it is normally because we are in a situation where there’s a zero interest rate and the Fed is running out of its ammunition. HCER: Last year, you wrote in a New York Times Op-ed advocating for negative real interest rate. Now the real interest rate has been zero for a while, but recovery is still sluggish. Has your view on the ef-

fectiveness of negative real interest rate changed? Mankiw: Well, the real interest rate—the interest rate on the Treasury inflation indexed bonds—has only been negative recently (for a couple of months), but that’s a relatively short time . The standard way in which we get the economy going in recessions is that central banks cut interest rates. The Federal Reserve has cut rates, but the question is how much they need to cut interest rates given where we are. I have a version of the Taylor Rule, which is a simplified rule that gives a benchmark for where the interest should go, and my version of the Tailor Rule says the federal funds rate should be cut to -3.7% . Obviously they can’t cut it to -3.7%; zero is the lower bound. But that does suggest that if they could, they would try to stimulate the economy a lot more given how low inflation is and how high unemployment is. So I think the negative interest rate we’ve seen is only modestly negative. If the central banks could figure out a way to cut the interest rate even more, they would, but there are no easy tools at hand to do that. HCER: In a recent article, Professor Martin Feldstein wrote: “because the downturn was not caused by high interest rates, lowering them could not lift the economy out of recession.” Do you agree? Mankiw: I do agree with the first part of the statement that this wasn’t caused by the Fed. Rudiger Dornbusch once said recoveries don’t die of natural causes; they are murdered by the Fed. That may be true for the previous expansions but it’s not true for the expansion previous to this recession. This recession is not really “murdered by the Fed.” This recession was caused by the crisis in the financial system. So I agree with that part of the statement. I’m not sure what necessarily follows from that is cutting interest rates won’t help. Cuts in the interest rates do reduce the cost of capital and encourage businesses and households to spend more. We are not able to cut interest rates enough to get the economy to grow robustly right now, but I think it would be wrong to jump from that fact to the conclusion that cutting interest rates is irrelevant. I


don’t see any evidence for that. HCER: One of the causes of the sub-prime crisis is the low interest rates in the 2000s, which were aimed at reviving the economy after the burst of the dot-com bubble, which led to the housing bubble. Now, to pull the economy out of the current recession, we are setting ultra-low interest rates again. Do you think the low interest rates will lead to future problems? Mankiw: I’m less concerned about that than I am about other things. I am more worried about the fact the economy is not recovering rapidly. If we look back, there were lots of things that contributed to the financial crisis. I don’t think the low interest rates that occurred in the 2003 to 2006 period were the most important. I think some of the major problems were the failures of the financial institutions, the rating agencies, the regulators, and to a large extent, the policies pushing low-income home ownership, which both Republicans and Democrats supported. I think there were a lot of failures that contributed to the financial crisis, and I would give the low interest rates that prevailed during that period a small part of the blame. So I’m not that worried right now that low interest rates will cause some problems going forward. HCER: Now the developed world is seeing historically low interest rates while China, Brazil and some other emerging markets are raising interest rates. Do you think it is possible that capital will flow out of the developed world, and that we will have the developed world in recessions while the emerging markets are overheating? Will that lead to future problems? Mankiw: Well I’m not sure that capital flowing out of the developed world and into the developing world will necessarily be bad. Capital flowing out of the United States also means that the dollar would depreciate, and that would stimulate net exports. Some capital outflow done in an orderly way is not necessarily a bad thing. Capital flows in order to get higher returns; sometimes it’s into the United States and sometimes not. I don’t think capital outflows in themselves are things to be feared. It’s certainly true what’s been going on recently is that the emerging markets and


the developed markets are behaving a little differently, and that means the real exchange rate has to move to deal with that fact. So I’m not concerned per se that our interest rates are not in line with those of some other countries and capital will move in response to the interest rate differentials. That’s how a well-functioning financial system works. I would hope it doesn’t turn into a stampede or a case of destabilizing capital flight. As long as things are done orderly and people are rational about incentives in different countries, that’s okay. HCER: It has been long talked about that the Americans should save more and the Chinese should save less to rebalance the global economy. However, the U.S. also needs stronger domestic demand to come out of the recession. How can saving rates and domestic demand increase simultaneously? Mankiw: In order to get aggregate demand up and in order to get people back to work, demand doesn’t necessarily have to come from consumers. It could just as well come from businesses. My first choice would be to figure out mechanisms to increase business investment and replace the consumer demand. I think households did overspend. They overspent because credit was too easy, and some de-leveraging of the households is a rational response. I f we again encourage borrowing it would be another mistake. We have to think of some ways for private businesses to pick up the demand. The U.S. has tended to consume a higher share of the GDP than any other nations. There’s no reason that high consumption/GDP ratio has to continue. There’s no reason why we can’t go to a lower consumption/GDP ratio as other countries do. That means the employment is going to come from other sectors: some from business investment, and some from exports. If it comes from exports, it’s probably going to come hand-inhand with a cheaper dollar.

HCER: In recent years, you’ve been doing much research on optimal taxation, and your recent Op-Ed on income tax has also generated a heated debate. In 2006, when talking about tax reforms in an interview with the HCER, Professor Summers said, “the imperatives of reducing the distortions in capital taxation seem to me to be much smaller than the imperatives of not further exacerbating the trends that have worked to the detriment of middleincome families relative to the highest income families.” What’s your view on Professor Summers’ remarks? Mankiw: I think we have a very difficult problem ahead of us. We have a very large budget deficit— not just today, but looming over the horizon even as the economy recovers. President Obama’s budget last January shows rising debt to GDP ratios as far as the eye can see. Even after the economy fully recovers, the debt will still keep growing faster than the economy. That’s clearly not sustainable. So something has to happen, and the question is what? The first question is whether fiscal reform should be focused on the revenue side or the spending side, and to what extent? I believe

that taxes are fairly distortionary, and I would much rather see the budget gap be closed by reducing the spending through such mechanisms as raising the retirement age than raising taxes. For sure that’s going to be controversial, and none of the spending cuts will be easy. There’s no category in the budget that says “waste and abuse” that could be cut painlessly. But raising taxes is going to be difficult too. President Obama said he was going to raise taxes on the top two percent, but the budget gap is far too large to be closed by that alone. If we are going to raise taxes, we need to look very broadly at tax increases that will affect the majority of Americans. I don’t think any politicians will step out and advocate that at this point, even though on the revenue side that’s what we might have to do. If I could design the tax reform, I would do it in a variety of ways that would make the tax code be less distortionary. I have been writing for years how higher gasoline taxes are a good idea because they can reduce externalities. It not only raises revenues but also corrects market failures associated with driving and gasoline consumption. HCER: Given that you are against raising income tax for high-income earners, do you have better suggestions for addressing this social inequality problem? Mankiw: I think there’s no question that economic inequality has increased substantially in recent years. This trend started in the 1970s as far as economists can discern. The best analysis I’ve seen of rising inequality is a book written by two of my colleagues here at Harvard—Claudia Goldin and Larry Katz. Their book is called The Race between Education and Technology. The basic thesis of the book is that technological progress tends to raise the demand for skilled workers over unskilled workers. In the post-war era till the 1970s, we managed to counteract that trend by increasing the supply of the skilled workers even faster, and educational attainment was rising very rapidly. Then, around 1970, educational attainment plateaued more or less, so we stopped increasing the supply of skilled workers and technological progress kept increasing the demand for them. As a result, technology seems to be winning the race between education and technology, and the wage gap between skilled and unskilled workers has risen. If you believe that explanation, then history points pretty clearly that thinking about the educational system is probably the best route to deal with inequality. You could say we just take the inequality as given—we can’t do

anything about it—and simply redistribute through the tax system. I think a better choice is to go to the root of the problem and try to fix the educational system. That’s not something done quickly or easily. I’m not an expert on educational reform, but I think the heart of the matter of inequality is to think about education, how we can get better education systems for more Americans and allow more Americans to go to college and finish college.

crements of the knowledge from new research that’s coming out. And the truth is the increment of knowledge is smaller than the body of knowledge that you first get exposed to as a freshman. The movement of the research frontier is painstakingly slow. It’s exciting for me to see all these freshmen learning economics for the first time and to experience the newcomer’s enthusiasm about the field through their eyes.

HCER: Past crises have inspired many influential theories in economics, most notably Keynesianism. Do you anticipate any theoretical advances to come out of the present crisis?

HCER: In recent years, financial services has been one of the most popular career choices and every year a large proportion of Harvard graduates takes up jobs in the financial sector, especially in investment banking. What are your thoughts on that?

Mankiw: Yes, but I don’t know what they are yet. The thing about research is that it’s very hard to predict. If I knew where the next great discovery was to come, I would have discovered it already. So I think the research process by itself is pretty unpredictable. On the other hand, I’m very heartened by the fact that many very good students and faculty members are spending lots of time thinking about the economics of our crisis, and no doubt there will be lots of fascinating research coming from those efforts. Looking back to the history of macroeconomics, many of the great macroeconomists of the 20th century, like Milton Friedman, Robert Solow and James Tobin, came to macroeconomics because they were fascinated by the Great Depression; they wanted to find out what went wrong and how to prevent society from experiencing that horrific event again. I have no doubt that the next generation of great macroeconomists will in some sense be motivated by this financial crisis, and that will be the silver lining of this grey cloud. HCER: You’ve been teaching Ec10 and revising the Principles of Economics textbook for many years. The materials covered are the very basics of economics, and it must all be very easy for you. What do you get out of these simple tasks? Mankiw: I love teaching introductory classes, and I love teaching freshmen because they are seeing economics for the first time. To be honest, in my own life, the most exciting time for me was introductory economics. For me then, the field was brand new, and there’s so much to learn, and that’s very, very exciting. One of the frustrating things once you become a professional economist is that you don’t have this great body of knowledge to consume, to learn, to absorb. You are only absorbing the in-

Mankiw: Harvard has the best undergraduates in the world. When I see a very high fraction of the best students going to finance, I do worry that it does not seem to be an optimal allocation of resources for society. It’s not I’m thinking about intervening and telling you guys where to go, but something might be wrong that a very high fraction of the best talent is going to one sector. I think this crisis may change that. My colleague Ken Rogoff has argued that the financial sector is likely to shrink in the coming years, If that’s the case, my guess is that the fraction of Harvard students going to finance should decrease with it. That’s how market forces works. The sizes of the industries rise and fall, and I’m sure where Harvard students will go to take their first jobs will reflect those broader macroeconomic trends. My suggestion is not thinking too much about finance because I think college is a time for broader reflection; it shouldn’t be just for preprofessionalism. It should be a time to sit back and think about your life, the role of society, how the society works. Harvard students are all talented and they will all get great jobs—I have no doubt about that. Don’t worry. The economy is going to treat you guys well. People as talented as Harvard students will do fine. Spend four years being more reflective and enjoy college for itself. HCER: Your pet dogs Keynes and Tobin are both named after great economists. Would you mind if your students were to name their pets after you? Mankiw: I would be honored. Someone once sent me a picture of a goldfish named Mankiw. If I could be promoted to a dog some day, I would be absolutely delighted. H



Interview with Yasheng Huang ’85 Yasheng Huang (Harvard ’85) is professor international management at Sloan School of Management, Massachusetts Institute of Technology where he also holds the International Program Professorship in Chinese Economy and Business. Professor Huang has dedicated decades of meticulous research to understanding different aspects of China’s economic reforms. Some of his major works in the recent years focused on the role of Foreign Direct Investment in China (summarized in his 2003 book Selling China) and the contributions of rural private enterprises in the 1980s (summarized in his 2008 book Capitalism with Chinese Characteristics). Based on detailed archival and quantitative evidence from late 1970s to early 2000s, his research shows that private entrepreneurship, facilitated by financial liberalization and microeconomic flexibility, played a central role in China’s economic miracle. The book Capitalism with Chinese Characteristics was also selected as the best book of the year by The Economist. In the following interview with HCER, professor Huang shares his views on China’s economic reforms in the past and at present.

Huang: I’m not even sure if we know the answers now. It is still a continuous process of asking questions and exploring the answers. These are very complicated questions about economics, politics and society--not narrow technical questions-and some are even very deep, philosophical questions. But I think by and large, though different people have different views, my own view is that the market economy is really the best system. That’s what we have come to know, even though where the appropriate boundary should be

is a bit tricky. The political questions are a bit more controversial. Even though still some people think the one-party system is the best system, my own view is democracy is the best system, and making a transition to democracy won’t necessarily inhibit economic growth. In my latest book [Capitalism with Chinese Characteristics] I mainly talked about the 1980s, which was a time of a lot of meaningful reforms and the best decade of economic performance as far as personal income is concerned. The 1980s was a superior period both in terms

HCER: You came to the US for college in 1981, a little while after China’s reform started in 1978. Did you have any opinion about the Chinese economy and economic reform at that time? Huang: There were a lot of people coming [from China] to the US in the 1980s and we didn’t have lots of opinions but we did have a lot of questions. We had many questions about why the Chinese economy was not performing well especially after we learned about Japan, Taiwan and South Korea’s miracles, and why our politics had so many problems and so many people died in the Cultural Revolution. We might be a bit naïve in thinking the answers were ready and could be taken off from the book shelves, but we really wanted to explore the answers to these questions. HCER: Did your questions get answered in the following years?

Jindal School of International Affairs



of growth and in terms of distribution. In the later years, the GDP growth was very fast but the distribution was terrible, and that actually coincided with tighter political control. A lot of people are talking about political reform now in China, and the view I have had for many years is you need political reforms [to help the economy grow], whereas many Chinese academics believed it’s OK as long as you have GDP growth. I don’t agree. I think GDP growth is less important than personal income growth, and for a period of time, the Chinese personal income growth was not actually as impressive as many people would think. I guess these are the answers if you put it that way. In fact, economics and politics are connected, and you can’t have true market economy in the absence of constraints on the government. I’m not saying that China should be exactly the same as the US, just as people say you can’t have one size of the shoe that fits every foot, which I agree. But before we talk about the size of the shoe, there is a prior question which is: do you wear a shoe or not? Similarly, you have to decide whether you want to become a democracy or you want to have a one-party system--that’s equivalent to answering whether or not you want to wear a shoe. If you decide to become a democracy, then we can talk about what kind of democracy you want to have—that’s a separate and entirely legitimate question, but first of all, I believe people should first decide whether to wear shoes and then debate about the size of the shoes. What kind of shoes, what color, what size—we can debate about that, do research and come up with an answer, but I believe people should wear shoes and China should have a market economy and democracy. HCER: Has your opinions about the Chinese economy changed over the years? Huang: Yes, there are some changes. For example, ten years ago maybe I didn’t think that China needed democracy. Though I was always sympathetic with democracy and I always believed that I would be happier under a democracy than under a one-party system, ten years ago if I thought: in order to make the transi-

tion from a poor country to a middle income country, would China require democracy? I guess my answer at that time was probably no. Now I have changed, and that’s because I have more data and that’s also because my view of the 1980s has changed, which came from my research. I like to think about myself as a scholar whose view should change if new data and evidence are presented and I believe we shouldn’t be stuck in ideology. Another change, as briefly mentioned before, is my view of the 1980s. Maybe six or seven years ago my view of the 1980s was very similar to many other people’s views, which is “1980s was OK, we had small and gradual reforms, but really the big changes happened in the 1990s and after China joined the WTO [in 2001]”— that’s a standard view among scholars both in China and abroad. The reason why my view changed is associated with my previous research and book about Foreign Direct Investment (FDI) in China [Selling China]. The standard FDI theory would tell you that FDI is not money, it is technology. That’s only true if you have the money; if you don’t have the money, then FDI is the money. In that book, I showed China had a lot of FDI because of tight credit constraints and suppression of the private sector and my conclusion is much of the FDI in China was the money that funded the private entrepreneurs, rather than technology. Now this view is actually accepted by a lot of people in China; my book was translated into Chinese and there were actually a number of fairly high level Chinese officials who actually read my book. In 2007, China began to rethink about FDI policies. I probably wouldn’t reach the same conclusions as many Chinese officials who are now anti-FDI; I think this is not right either. My view is that it was actually very fortunate for China to open to FDI because the domestic financial system was not financing private entrepreneurship; it was FDI that supported the private entrepreneurs. The reason why I began to rethink about the 1980s is, after I published the book on FDI, there was a puzzle in my mind: if in the 1990s it was FDI that financed growth and private entrepreneurship, then how do you explain the growth

in the 1980s? I didn’t even think about that question before. It is really puzzling because if you believe the Chinese economic system and political system were suppressing private entrepreneurship in the 1980s, and there was no FDI, the ordinary logic would say there couldn’t be growth. And yet we know in the 1980s, the Chinese economy was growing very fast. So for my 2008 book [Capitalism with Chinese Characteristics], I began to dig into history, really trying to fully understand the 1980s. There are actually not a lot data about the 1980s; also, most scholars didn’t begin studying China till the 1990s, so nor did many people write about the 1980s. I had to locate my own information. Fortunately, in Hong Kong, I found a collection of bank documents from 1979 all the way to 2004—twenty volumes and tons of thousands of pages. I went through these documents and my view of the 1980s completely changed. I found the 1980s to be very liberal and progressive. There were lots of policy innovations and substantial financial reforms, and almost all of that took place in the rural part of the country, which was poorly documented because of observation bias. But remember, China in the 1980s was 80% rural in terms of population. So what happened in the rural area was far more important than what happened in the urban area. Then in the 1990s, there were policy reversals, suppressing the private sector. China did not undertake many political reforms in the last 30 years. This is true but every political reform they did undertake was undertaken in the 1980s. Now I believe the reforms in the 1980s went far beyond what many people believed to be the case. I have to say the changes are totally natural because of the discovery of new facts. HCER: You have pointed out that the beneficial growth of the marketdriven rural China is suppressed by the distortive growth of the stateled urban China, since the state is intentionally directing capital away from the rural firms to support urban growth. Therefore, the rural firms are lacking adequate access to capital. However, if there’s no government intervention, will banks actively lend to the Township and Village Enterprises


(TVE) given that State Owned Enterprises (SOE) are backed up by the government and therefore much safer debtors? Huang: Almost 100% of the non-performing loans are in the state owned sectors. Micro finance in other countries also shows that rural households are very responsible for their financial decisions and repayments; the default rates are in face very low. As for banks’ lending decisions, there are good reasons that banks would not lend if they are located in the cities. The reason is that in the countryside you need to make many small loans rather than one large loan, so that the auditing costs and loan servicing costs are high. But this was where China used to have a huge advantage. In the 1980s, the Chinese banking institutions had many branches in the countryside--the network of branches was very substantial, unlike India, for example. If an Indian bank, based in Mumbai, wants to expand, it has to build new branches. That will incur very high fixed costs. However, China already paid for the fixed costs in the 1950s and 1960s. It was very unbelievable that in the 1990s Chinese banks got rid of these branches. For example, the Agriculture Bank of China shut down their rural branches, which was absolutely amazing, because in India the most profitable business now is in the countryside. The main reason that the local banks in India are able to compete with American banks and foreign banks is they have rural operations. But Chinese banks gave it up. Now the American banks want to go to the Chinese countryside because the returns could be extremely high if you figure out the effective ways to deliver the products to your customers. HCER: Some people have noticed a somewhat worrisome phenomenon in recent years. Nowadays, many wealthy private entrepreneurs are using an overwhelming amount of their money to speculate in the real estate markets and commodity markets instead of investing in the real economy. What are your thoughts on that?

Huang: The speculative markets have a self-fulfilling prophecy. If you believe in the value in the real estate market or the stock market, it goes up. It’s very important to think about how these speculations started. I haven’t done any academic research on this, so I only have some ideas but not research findings. One is very high level of income inequality in China may have contributed to the high level of speculative activities. Another reason is, with the tight capital control, you can’t invest outside of the country. If you want to speculate, there really aren’t many things you can do other than investing in real estate. HCER: But why do the entrepreneurs choose to speculate instead of further investing in their business in the first place? Huang: That’s a very good question. I think we need to think about what are the things that are needed to motivate people to make investment in R&D and technological innovations. If you invest in the real economy, the returns will come a few years down the road. You would only want to do that if the property rights are secure and if you have appropriate supporting systems that motivate you to make those investments. In this sense, it’s much easier to speculate than to invest in technology because the latter would require more legal protection. This is not unique in China, and the issue is whether in China there is excessive speculation. Also, I think investment in technology in particular would require truly research oriented universities, academic freedom, and truth-oriented research as opposed to policy-oriented or short-term oriented research. China currently doesn’t have those. You would need those complementarities, including the legal system, protection of property rights and intellectual property rights, to make the technology investments worthwhile. You would need these complementarities to narrow the gap between speculation and longer-term investments. HCER: You have criticized the Chinese government’s macroeconomic policies in response to the financial crisis, es-


pecially the stimulus package, pointing out it has created excessive liquidity and inflation. What policy would you like to see a response to the crisis going back to 2008? Huang: Strictly speaking, these were not macroeconomic policies but micro ones—micro management and micro directions, channeling money into certain parts of the economy and to state-owned enterprises in many detailed ways. It’s an important distinction we need to make. As for macro policies like interest rates, actually they haven’t done much. I also don’t think that the government directed investment fall into the category into fiscal stimulus because if you look at the government budget, there is actually a budget surplus, and in this sense, the fiscal policy is not expansionary but contractionary. What policy response I would like to see is more reform, as opposed to more spending. I can see a way of combining policy actions with reforms, but what we have seen in reality is exactly the opposite-massively pouring credit (into the stateowned sectors) and a reversal of reforms. I think this is terrible. I would like to see more reforms, perhaps in combination with spending on infrastructure; I’m not saying you don’t need policy actions, but you should have policy actions plus reforms, or maybe reforms plus some policy actions. HCER: Many people think China is in a situation similar to Japan in the 1970s and 1980s, with problems of exchange rate and bubbles in asset markets. What will the optimal monetary policy for China now? Huang: I think they should definitely raise the interest rate. Whenever you have a low interest rate, there are more people waiting to get the money than who can actually receive the money, which means that those who are politically powerful are the ones who get the money, and they are the state-owned enterprises. Raising the interest rate is actually going to turn out to provide more money to the private sector, paradoxically. Some people would say, if you are increasing the interest rate,



aren’t you increasing the cost of borrowing for the private sector? Actually, this is not really true. You may be increasing the cost of capital, but this is better than not getting capital at all even though they can afford it. HCER: In spite of China’s many problems, it is still one the most successful examples among all the countries that have been making the uneasy transition from communism to capitalism. What do you think went wrong with Russia and some other post-communist countries in Eastern Europe or central Asia? Huang: I think we need to separate Russia from the Eastern European countries which, until the current the Great Recession, were doing very well. In terms of growth, the Eastern European countries had difficulties in the early 1990s, but what I have read indicate they have long recovered and are doing pretty well now. Russia is a totally different story. There is no question that Russia went into huge problems in the 1990s. If you look at the data, Russia began to perform badly before they began the political reforms in the early 1990s. Their productivity growth was nearly zero and their life expectancy was declining before 1991. Then after the reforms started, it went down even more. A difference between China and Russia that makes me optimistic about China is something I don’t know how to explain-Chinese are just very entrepreneurial, and you see that every day. Every Chinese is a businessman or a businesswoman— very hardworking and very energetic. Some people say the Chinese are very ho-

mogeneous and not creative. I think a lot of these criticisms are about the Chinese academics that are not producing innovative paper, but the Chinese people are extremely innovative in life, entrepreneurial and very sensitive to business opportunities. Another difference is, in spite of the Cultural Revolution and the central planning system, the vast majority of China was never completely centralized. The rural China was kind of left alone. However, if you look at Russia, for seventy years, it was completely centralized. Therefore, there was more capitalism (in China) even before China started the reforms. Also, it seems the Chinese can tolerate many more policy mistakes than in other countries. Another thing is the high flexibility and adaptability. People always come up with something on their own (in response to the policies). I once gave a speech to a group of entrepreneurs in Zhejiang Province; I talked about the policy developments and the entrepreneurs said “oh professor, everything you said is right, but I believe we can totally deal with these things on our own”. I thought it was very interesting. They were not so much optimistic about the policies as they were optimistic about themselves. Though scholars and academics tend to say if you have one policy, you will have these effects, and if you have another policy, you will have different effects, the reality may be more complicated than that. This is the reason why I am undertaking some of the companyprojects in China. I founded a China Lab and I also have an India Lab, and we send our MBA students to help the Chinese entrepreneurs. I believe that though policies are important and there’s no ques-

tion about it, let’s not have the view that policy is everything. Policy may count for about 60%, so there still are things I can do to help the private entrepreneurs, like improving the management of the firms and help them with market entry plans— these are very tangible things. I guess that’s another change I have experienced. HCER: We are proud that you are a Harvard alumnus. How did you like your time at Harvard? Huang: I loved it. I loved the broad liberal arts education which I truly valued. It helped me form the habit of thinking about things from different perspectives. It terms of opening your mind to different ideas, Harvard is probably the best. HCER: Did you have a favorite course or favorite extracurricular activity? Huang: I didn’t have many extracurricular activities because I was so poor and I worked as a security guard for the Harvard police and I also delivered mail, so basically my life was studying and working to make some money and buy my textbooks, but it was great. As for courses, I think the moral reasoning courses were the most interesting. They inspired me to think about those big questions—democracy and etc. In the 1980s we were not terribly interested in the stock market, and we didn’t even know those things very much. It was really the big questions that were very attractive. Later on, the courses became more technical, but it’s perhaps a progress because you would also need the techniques for your daily lives. H


An Economic Lesson from the Sports Industry It is a truism, most would say, that sports leagues have one fundamental difference from other industries. In order to produce their output, the teams must cooperate. While Coca Cola can make soft drinks without Pepsi, the Red Sox can’t play games without an opposing team. Not only must the Red Sox have an opposing team for each game, but they need a full league of opponents to conduct a championship season. Further, for the championship season to hold the fans’ interest, there must be a significant degree of uncertainty as to which team will emerge as the winner. This uncertainty is engendered by having a level of competitive balance among the teams. Herein lies a major conundrum of sports leagues. Given that some teams come from large cities with a deep corporate base and others come from smaller cities with an exiguous corporate base, and that all teams compete in the same market for players, how can these teams with very disparate economic underpinnings compete with each other in a meaningful way? Sports leagues have attempted to an-

swer this question in different ways. With varying degrees of success, U.S. leagues have used a combination of a team payroll limits (salary caps), individual player salary limits, franchise debt restrictions, luxury taxes on high payrolls, revenue sharing among the teams, and reverse order amateur drafts, inter alia. European soccer leagues generally employ only minimal revenue sharing and no salary caps; instead, they are organized within each country in a hierarchy of promotion/relegation leagues. The bottom three or so teams in the top league are relegated to the second league the next year, while the top three or so teams in the second league are promoted to the top league. The same interleague mobility applies for the second and third leagues, the third and fourth, and so on. This open system has several effects on competitive balance and fan avidity. First, fans have at least two competitions to keep them interested in the outcome of games: (1) whether their team will finish at the top of its league and (2) whether the team will finish near the bot-

tom and be relegated. In the English Premier League (EPL), the race to finish first is complemented by the race to finish in the top four (the top four clubs compete in the Pan-European Champions League the following year) and in the top seven (the fifth, sixth and seventh best EPL teams normally qualify for the UEFA Europa League competition). On the other end, there are usually six teams that are within a few points of finishing in the bottom three teams during the last several matches of the season. Thus, aside from a handful of clubs in the middle, the fans of a strong majority of the EPL’s twenty clubs have something significant to root for throughout the season. Second, the open European soccer leagues have an automatic market-balancing system. The process by which a team enters a country’s top league is not by buying its way in. It is by working its way up. That is, an entrepreneur sees an opportunity in an available market to build a soccer club. He/she enters the team in a low level league, invests in its players and a facility, and, if the team succeeds, it goes to



the next level. Thus, underdeveloped markets beckon investment. Greater London has had between five and seven clubs in the EPL during past decade. If there were three, let alone six or seven, MLB teams in greater New York City, it is unlikely that the Yankees would be the perennially dominant team they are today. What about the U.S. leagues? Of the big four leagues (MLB, NFL, NBA and NHL), only the MLB does not have a salary cap; instead it has a luxury tax on high team payrolls and debt limits. All the leagues have some form of revenue sharing and a reverse order amateur draft. The NFL has both the hardest salary cap and the most extensive revenue sharing, with roughly 75 percent of all league revenues subject to sharing. There are many ways to measure competitive balance in a league. For instance, competitive balance can be reckoned within a season or over a period of years (e.g., whether the same teams go to the postseason every year). At the risk of some simplification, here I shall focus only on intraseason balance. The most common metric used by sports economists to assess intraseason balance is the ratio of (a) the actual standard deviation of win percentages across teams in a given year, to (b) the idealized standard deviation of win percentages across teams in that year. The latter projects what the standard deviation would be if each of the teams had precisely a 50 percent chance of winning each game, i.e., if the playing talent on the teams was distributed equally. Hence, the ratio provides a measure of how far the league deviates from full parity. By this metric, the NFL has consistently had the most competitive balance, followed, in order, by MLB, the NHL and the NBA. Consider a more detailed look at MLB. Baseball’s competitive balance ratio has trended downward over the decades, as depicted in Table 1. The reason for the downward trend (i.e., toward greater balance) is secular talent compression. This has occurred as the percentage of the population (from the United States and other baseball countries) that plays in the major leagues decreased over time, since population has grown considerably faster than the number of teams. Note, however, that the downward trend of the ratio continues into the 1980s



Ratio of Actual to Idealized Standard Deviation of Win Percentages, by decade Decade MLB 1901-1909 2.92 1910-1919 2.54 1920-1929 2.35 1930-1939 2.58 1940-1949 2.40 1950-1959 2.29 1960-1969 2.10 1970-1979 1.90 1980-1989 1.72 1990-1999 1.74 2000-2009 1.86 and then experiences a slight reversal -- a reversal that takes hold paradoxically during the decades when MLB introduces its revenue sharing and luxury tax systems. The amount of revenue sharing in MLB grew rapidly from below $25 million in 1995 to above $420 million in 2009. Baseball’s revenue sharing was introduced and extended in the name of increasing competitive balance. What happened? In short, at least through 2006, MLB’s revenue sharing system contained perverse incentives. The marginal tax rate faced by the top half of teams (ranked by revenue) was roughly 39 percent, while that faced by the bottom half was around 48 percent. That is, if a top half team increased its local revenues by $10 million, it had to pay $3.9 million in additional taxes into the sharing plan. Yet, if a bottom half team grew its revenues by $10 million, it received $4.8 million less in revenue transfers. Thus, the bottom half of teams had a stronger disincentive to invest in improving team performance (such as by increasing payroll) than the top half. As a result of this, payroll distribution across the teams grew more unequal over the period. As shown in Table 2, the coefficient of variation (standard deviation divided by the mean) grew steadily since 1985. (Of course, it is at least possible that without revenue sharing the disparity may have grown more rapidly.) Taking revenue from the rich teams and giving it to the poor teams was intended to lower the payroll of the rich teams and

Coefficient of Variation of Team Payrolls in MLB, by quinquennium 1985-89 1990-94 1995-99 2000-04 2005-09

0.261 0.284 0.395 0.403 0.417

raise the payroll of the poor teams. The system’s underlying incentives, however, encouraged the opposite to occur. No one should have been surprised by the recent release of club financials by, showing that certain low-revenue clubs did not put their transfers into increasing player payroll. It is arguable that MLB’s revenue sharing system, in effect, became a wealth transfer policy, rather than a competitive balance policy. Although sports leagues don’t always seem to get it, the lesson here is apparent. Incentives matter. H Andrew Zimbalist is the Robert A. Woods Professor of Economics, Smith College, Northampton, Ma. Dr. Zimbalist has written extensively about and consulted widely in the sports industry. His latest book, Circling the Bases: Essays on the Challenges and Prospects of the Sports Industry, was published by Temple University Press in November 2010.

Announcing the John Dunlop Thesis Prize in Business and Government $500 awarded by the Mossavar-Rahmani Center for Business & Government at Harvard Kennedy School to a Harvard College graduating senior for the best thesis on a challenging public policy issue at the interface of business and government Application Deadline: May 5, 2011 To apply, please visit: