The Empirical Evidence Against Utility Maximization 
 

Mehmet Karacuka, Ph.D.

Ege University, Department of Economics

35040, İzmir, Turkey

Phone: 90-232-3434000/5287, Fax : 90-232-3734194

E-mail: mehmet.karacuka@ege.edu.tr 

Asad Zaman, Ph.D.

International Institute of Islamic Economics

New Campus Sector H-10 
Islamabad, 44000 
Phone: 92-51-9257939, Fax: 92-51-9258019

E-mail: asadzaman@alum.mit.edu 

Mehmet Karacuka graduated Dokuz Eylul University, Izmir, Turkey and works as an Assist. Professor in the Department of Economics at Ege University. He worked as a visiting researcher at Dusseldorf Institute for Competition Economics, Heinrich Heine University;  Econometric Institute , School of Economics, Erasmus University Rotterdam; and was a guest PhD Student, in the Institute of Law and Economics, Hamburg University (2004-2005). He has publications in the Journal of Telecommunications Policy (2011), Journal of Developing Areas (forthcoming in 2012), Applied Economics Letters (forthcoming in 2012) and Journal of Business Economics and Management (forthcoming); and has been a reviewer in various journals including Telecommunications Policy, Applied Economics, Journal of Developing Areas, The Service Industries Journal. 
 

Asad Zaman graduated with BS in Mathematics at Massachusetts Institute of Technology, and got his MS degree in Statistics, and Ph.D. degree in economics at Stanford University. He has taught at the Economics Department in the University of Pennsylvania, Columbia University, Johns Hopkins University, as well as Bilkent University. He has also been a visiting professor at University Catholique de Louvain, Belgium and California Institute of Technology. Currently he is Director General at International Islamic University, Islamabad. He has many publications in leading journals such as Annals of Statistics, Journal of Econometrics, etc.  
 
 

The Empirical Evidence Against Utility Maximization 
 

ABSTRACT 

Current Economics Textbooks and Economists justify a theory of consumer behavior based on utility maximization on a priori grounds. This methodology follows Lionel Robbins�� idea that economic theory is based on logical deduction from postulates which are ��simple and indisputable facts of experience.�� Strong evidence has emerged from many different lines of research that these ��simple and indisputable facts of experience�� are contradicted by human behavior. In this article, we summarize some of main contradictions between predictions of utility theory and actual human behavior. Efforts to resolve these contradictions continue to be made within orthodox frameworks, but it appears likely that a paradigm shift is required.  

JEL Classification: D010, D100

Keywords: Behavioral Economics, Utility Theory 

INTRODUCTION 
 

Despite the claims to objectivity and factuality set out clearly in Microeconomic textbooks, the model of consumer behavior is introduced and justified on the ground of ��rationality�� – clearly a normative concept. If this microeconomic theory is positive as claimed, then it should be backed up by observations of consumer behavior which confirm the factuality of these theories. No such evidence is presented, because none exists. Indeed, as Sen (1977) has argued in his article ��Rational Fools,�� behavior which is rational according to economic axioms may actually be quite foolish. 

Over the past few decades, empirical evidence against the descriptive accuracy of economic theories of human behavior has mounted. Researchers in theories of decision making, psychology, and more recently behavioral economists who actually study human behaviour, observed the predictive failure of economic theories, and this led to substantial reconsideration of the status of such theories. Nowadays the orthodox position in many of these domains is that theories of rational behavior used in economics are normative, and not descriptively accurate; for example, see Baron (2004). This position arrived at after careful, detailed and extensive experimentation on human decision making behavior, is exactly the opposite of assertions that economic theory is positive and consists of indisputable facts of experience. One of the most spectacular pieces of evidence is the global financial crisis of 2008. As many leading economists have stated, the stock market crash is not compatible with theories of rational behavior. On the other hand, based on behavioral theories, Shiller (2005) was able to predict the crash. Because of this dramatic conflict between dominant economic theories and empirical evidence, many Nobel calibre economists have called for a fundamental re-thinking of our basic economic models. 

Margolis (1982) notes that a paradigm starts to become an intellectual handicap, when things that are obvious and obviously important can be seen more easily by a naive observer than by specialists. In this paper, we summarize the many areas of conflict between the theory of utility maximization and consumer behavior. In fact, the areas of conflict are so numerous that it is impossible to summarize them in this article. We will focus on areas where economic theory acts as an obstacle to understanding reality: the naïve observer is better able to understand observed behavior than those armed with economic theory.

Section 1:  Diverse Human Motives

The achievement of Newton in explaining a diverse set of phenomenon with a single principle was universally admired. Scientists in all fields sought to emulate this achievement by using a minimal set of explanatory principles. In particular, as Mirowski (1991) has documented, the prestige of Physics led economists to self-consciously imitate the methodology of Physics:  

��the progenitors of neoclassical economic theory boldly copied reigning physical theories in the 1870��s. The further one digs, the greater the realization that those neoclassicals did not imitate physics in a desultory or superficial manner; no, they copied their models mostly term for term and symbol for symbol, and said so.�� 

One of the aspects of this emulation was the simplification of human motives to the single one of greed. While this simplification makes mathematical modelling on the style of physics possible, it also makes the models of human behavior highly unrealistic, as we will show in this section. 

 

1.1 Concern for Fairness

 

One of the most fundamental principles of economic rationality is that people prefer more money to less. The ultimatum game strikes at the heart of microeconomic theory by showing that this is not true.  In this two player game, 10 single dollar bills are placed on the table. The first player (P) proposes a sharing rule – for example, he will keep 6 dollars and give 4 dollars to the second player. The second player (R) can either accept or reject. If he accepts, then both players get the proposed amounts. If he rejects then both players get nothing.  

Economics is the queen of the social sciences because unlike any of the others, it offers a sharp and unambiguous analysis of this game. The proposer seeks to maximize his own monetary payoff. He will offer the responder $1, and keep $9 for himself. He is secure in the knowledge that the rational responder will not reject his offer – a rejection will give both $0, and the responder will prefer to get the $1 he has been offered. This is the unique utility maximizing offer for the proposer. If he tries to keep all ten dollars and offers zero, he runs a risk of rejection, while there is no reason to go for less. This then is a Nash equilibrium of the game: The proposer offers the split ($9,$1), and the responder accepts all positive offers. 

Unfortunately, this prediction based on economic rationality bears no resemblance to the actual outcomes seen when human beings play this game. A substantial number of proposers (though still a minority) offer a fifty-fifty split ($5,$5) based on considerations of equity and fair play, even though this is not their best move in the economic sense. A majority of proposers offer 30% or more to the responders. They are aware that there is a risk of rejection if they offer less. Proposers who behave according to economic theory and offer only a small amount (less than 30%) are frequently rejected. This means that the responders act against their own best interests and refuse low percentage shares of money to punish the proposer for making an unfair offer – even when this ��unjust�� percentage is large in absolute terms.  

In fields which have more contact with actual human behavior, these ultimatum results cause no surprise. Ordinary human beings have trouble understanding what the fuss is all about; these results seem perfectly clear and natural. Only economists, conditioned to look at reality using wrong theories of human behavior, are shocked and surprised by these results. In fact, these results run so strongly against the grain of economic thinking that no major economic journal would publish these results for a long time. There were standard objections:

• results from a lab experiment would not generalize to the field of actual decisions in real economic environments.
• Monetary stakes in experiments were too low to induce people to think carefully about their decisions.
• New and unfamiliar game conditions led people to make mistakes.

 

Camerer (2003, p.60) describes the skepticism of economists about ultimatum results as follows: ��If I had a dollar for every time an economist claimed that raising stakes would drive the ultimatum behavior towards self-interest, I��d have a private jet on standby all day.�� Replication of these results in a wide variety of different context, with high stakes (equivalent to a month��s salary or more), with experienced players in realistic field experiments has led to the conclusion that these results are robust. Economists have reluctantly accepted the validity of these results. Nonetheless, there has been no progress towards integrating these results into the body of economic theory; the flat contradiction between the assumptions of micro and these observations of human behavior make it unlikely that a resolution can be found which fits into the framework of orthodox theory. 

The assumption of economists that people only care about their own share and not about fair and equitable distributions, leads economists to propose drastically wrong policies in the domains of taxation and income re-distribution. Norton and Ariely (2011) document substantial consensus in the US population about the desire for a more equitable distribution of wealth. People think the distribution of wealth is more equal than it actually is; and they think it should be much more equal than their already unrealistically-equal notion of its current state. For example, the top 20% of the US wealth distribution actually controls nearly 85% of total wealth; people think the top 20% controls under 60%; and they think it should control just over 30%. Since concern for fairness is not currently part of economic theories, these issues get swept under the rug, to the detriment of all.

 

2. Resentment, Revenge, Gratitude, Carelessness

 

P1: Adam

 
   Left                                Right

Adam: $7

Ben: $B

P2:Ben

      

Adam: $10

Ben: $5

Adam: $0

Ben: $3+x

 

                  L  R 
 
 
 
 

Empirical evidence cited below about this two player game is discussed in Goeree and Holt (2001). Initially, assume that x<2, so that $(3+x) < $5. At the first move, Player I (P1), Adam has the option of ending the game and getting the safe option of $7, by playing RIGHT. Alternatively, he can choose to play LEFT, which puts Ben on the move. Ben gets $3+x<$5 if he moves L, and he gets the higher payoff $5 if he moves R.  The standard game theoretic analysis of this game provides us with the following insights: 

1. At Node labelled (P2:Ben), utility maximization predicts Ben will choose to play R, which gives him the higher payoff $5 > $ (3+x). This implies that Adam will receive $10.
2. Adam can rely on economic theory, which states that Ben will choose R over L, as long as $5 > $ (3+x). This means that Adam can rely on receiving $10, if he puts Ben on the move by playing LEFT.
3. A rational and selfish Adam will play LEFT, since that leads to an outcome of $10, while RIGHT leads only to $7.
4. The value of B, which Ben gets if Adam plays RIGHT, is irrelevant.

 

All four of these insights are wrong. Furthermore, ordinary untrained subjects who play this game behave in ways which show deeper understanding of human behavior. Thus game theory systematically handicaps the understanding of actual observed behavior in this game. 

Experimental evidence reveals the following patterns of behavior: 

1. When the difference between $3+x and $5 is small, Adam cannot rely on Ben making the right move. Suppose 3+x =4.75, which is only a bit smaller than $5. In one experiment, 15% of the second players choose L, which gives them a quarter less than the optimal move R. This ��mistake�� substantially reduces the payoff to Adam. Anticipating this possibility, many player I��s chose to take the safe payoff of $7 by moving RIGHT.
2. As the difference between 3+x and 5 increases, the chances of player II playing R increase. In experiments, Player I anticipates this and chooses LEFT more often. None of these phenomena is predicted by game theory, showing the theory is blind to aspects of the game which untrained observers are aware off.
3. A high value of B creates resentment in player II. For example if B=$10, the LEFT by player I reduces the payoff of player II to a choice between $3+x and $5. This could easily motivate player II to take revenge by playing L and punish player I. Anticipating this, player I takes the secure option of RIGHT very often in this situation. Again this behavior shows greater wisdom than game theory.

 

When the difference between 3+x and is small, player II might ��carelessly�� choose the lesser value. This is an instance of satisficing instead of maximizing. There are many other possible explanations for this ��carelessness��; computational costs may exceed benefits, for example. A large and negative B, like B = -$100, leads to another interesting case. Now the safe choice of Right by player I incurs a large penalty to player II. Therefore, we can expect that player II will feel gratitude if player I plays Left. Player II might then choose to move Right, maximizing the payoff to player I, even if this is less than optimal for him – for example, if 3+x=6. This scenario has not been tested in experimental games, since experimenters do not use setups which involve penalties to the players for various reasons. Nonetheless, it seems plausible on intuitive grounds, while having no basis in game theory at all. 

What is the harm of reducing the complexity of human motives to the simple one of greed? We have just seen that understanding human behavior in a very simple game requires taking into account resentment, gratitude, revenge, carelessness and reciprocity. Contrary to the reductionist economic views, a vast number of market transactions are based on motives other than greed. The widely recognized phenomenon of conspicuous consumption creates an externality and hence market failure which should be regulated – however economists fail to acknowledge the phenomenon because it requires motivations other than selfish greed. An additional problem is that highlighting a single motive both legitimizes and encourages it: witness the ��Greed is Good�� maxim of Wall street. 

 

1.3 Altruism

 

There has been a fundamental paradigm shift in the last part of the twentieth century. In the early part, it was not intellectually acceptable to argue that true generosity or altruism existed. All such behavior had to be explained with reference to some underlying selfish concern. However, it has now come to be recognized that true generosity exists, and indeed is an important characteristic of human beings; for a detailed discussion, see Batson(1992). Such an understanding is fundamentally at odds with the central ideas of utility theory, which posits selfishness as the central motivation for consumer behavior.  

In the Ultimatum Game described earlier, the vast majority of offers lie between 30% and 50%, contrary to Game theoretic predictions of very low offers bordering on zero. There was some dispute as to whether the high offers are due to a preference for fairness, or due to the fear of rejection. To differentiate between these possibilities, the Dictator Game was introduced. In this variant, the offer of the proposer is automatically enforced – the responder has no choice.  Utility maximizers will always take 100%, and leave nothing to the other party. In real life games, only a small minority behaves like homo economicus. Most players leave something for the other party, sometimes a token amount. Amounts left for others are at cost to oneself, demonstrating clearly the existence of altruism.  

Aknin, L. et. al. (2010) also provide a strong global empirical evidence for generosity, based on their analysis of survey data from 136 countries.  They find that ��prosocial�� spending is consistently associated with greater happiness. In contrast to traditional economic thought, they find that the reward experienced from helping others may be deeply ingrained in human nature, emerging in diverse cultural and economic contexts.  

In general, utility maximization predicts that given a choice between a $100 dollar meal at a high class restaurant and the alternative of donating it to feed hungry children, people would prefer the first choice. In fact substantial numbers of people prefer the second option, and this is well known to most people. This illustrates our theme that intuitive lay opinion is better informed about human behavior than economic theory, in many situations.

 

1.4 CROSS-CULTURAL STUDIES

 

Is it true that a simple model of self-interest where all actors maximize material payoffs to themselves provides a good first approximation to human behavior which is universally applicable? Many economists appear to believe this, and the utility maximization model is the bedrock and foundation of economic theory. Nonetheless, there is overwhelming empirical evidence that this is not the case. Henrich et. al. (2001) searched for Homo Economicus in 15 societies but could not find him anywhere. Their experiments confirm findings which have extensive empirical support from a large number of studies:

1. Generosity, in conflict with economic theory, is widespread, though the levels vary with many factors. This means that simple utility maximization models fail to provide even a rough guide to human behavior in a wide variety of situations.
2. Preferences are not exogenous, but responsive to social and structural circumstances surrounding the choices to be made. This means that human welfare cannot be measured solely by the final consumption bundle. Both the process by which it was obtained, and the consequences to others, matter.
3. Institutional details and cultural norms impact greatly on human behavior. Thus human behavior is far more flexible than assumed by economic models, and also subject to shaping by social structures.

 

Taking these findings on board would require radical reshaping of economic theories.

 

Section 2: Social Norms

 

Economists generally subscribe to the philosophy of ��methodological individualism.�� This means that group behavior can be reduced to individual behavior. Every group is, after all, just a collection of individuals. If we understand how individuals behave, we can aggregate over this behavior to get an understanding of group behavior. This is the intuition behind the widespread use of Dynamic Stochastic General Equilibrium (DSGE) models, which reduce the entire society to a single (aggregated) individual who optimizes consumption over various complex and uncertain scenarios. Interestingly, DSGE models have been blamed for the recent global financial crisis of 2008. Testifying before a Congressional Committee investigating the reasons for the crisis, Solow (2010) blamed the unrealistic DSGE models that currently dominate macroeconomic theory taught at the elite universities and used by leading policy makers at central banks and economic ministries. He argued that these models have nothing to say about the problem, and are systematically blind to the issues which created the crisis. 

More generally, there is substantial evidence that ��methodological individualism�� is wrong. In chemistry, the behavior of organic compounds cannot be understood in terms of the properties of atoms and molecules. The statement that ��Britain is at war with France�� cannot be reduced to any collection of sentences about citizens of Britain and France. The idea of ��emergence,�� where the whole cannot be reduced to the sum of the parts, is now well established in many different disciplines. Clayton and Davies (2006) is a convenient reference. Communities create methods to live by common rules. Choices about the language we speak, the side of the road we drive on, the legal framework we live by and many others which govern social behavior cannot be reduced to individual choices, or studied separately from the community.  

Economists generally subscribe to the idea of the ��invisible hand��; that is, individually selfish behavior leads to optimal social outcomes. This is another reason for their neglect of ��society.�� Evidence that this does not hold is given in Zaman (2011).  Biccieri (2006) writes that ��Pro-social norms of fairness, reciprocity, cooperation, and the like exist precisely because it might not be in the individual��s immediate self-interest to behave in a socially beneficial way.�� In this section, we give some examples where social norms lead us to behave in ways which conflict with predictions of economic theory.

 

2.1 Co-operation

 

The Prisoner��s Dilemma (PD) is perhaps the most widely investigated game, generating thousands of papers in diverse fields. The payoff matrix can be described as: 

Player I 	Player II	Cooperate	Betray
Cooperate		(High, High)	(Sucker, Temptation)
Betray		(Temptation, Sucker)	(Low, Low)

 

The first coordinate gives the outcome for player I and the second for player II. To create a prisoner��s dilemma, the four payoffs must be ordered as follows for both players¹: 

Sucker < Low < High < Temptation 

The socially optimal outcome is for both players to cooperate, which gives the High payoff to both. If players are selfish and value individual benefits over cooperation and social behavior, both will be tempted by the higher Temptation payoff. If both Betray, both will get the low payoff, which is both socially and individually inferior.  

Based on utility maximization, game theory offers a clear and unambiguous solution for this game. The Betrayal strategy is the dominant strategy: regardless of what the second player does, it is best for the first one to betray. If second player cooperates, one get the higher temptation payoff. If second player betrays, one is saved from the poor sucker payoff. If both players play their dominant strategy, then both will get the Low payoff.  

While this is what utility maximization predicts for rational and selfish players, it does not make intuitive sense. It seems likely that both players can figure out that both will be made better off by cooperating, and achieve the High payoff. Actual behavior corresponds to this intuition. Cooperation is much more widely observed than predicted by economic theory. A summary of results and further references are provides in Camerer (2003, p.45-46) and Kagel and Roth (1995). A few particularly strong conflicts between intuition and game theory are listed below. 

1. Economic theory predicts that ��cheap talk�� – communication between players without substantive information or binding commitments – will have no effect on the outcomes. Intuitively, we all know this is not so, and communication will lead to greater cooperation and fewer betrayals. There is strong empirical evidence that communication creates cooperation.
2. According to Game Theory, as long as the inequalities specified hold, the size of the temptation and the sucker payoff do not matter. In fact, changes in these payoffs have predictable effects – increasing temptation payoff increases betrayals, while increasing sucker payoff increases cooperation. Furthermore, intuitive players can predict this behavior and respond appropriately, while game theorists cannot.
3. With fixed finite repetitions of the PD, game theory makes the startling prediction that betrayal in all rounds is the unique dominant Nash equilibrium. This follows from a straightforward backward induction. Human beings routinely use the opportunity offered by multiple games to signal friendliness, create cooperation, and achieve far higher payoffs than homo economicus.

 

In general, untrained intuition provides a much better guide to understanding the vast literature on experimental evidence regarding the PD, while game theory serves as an obstacle to understanding. Economists often perform poorly in PD games, not achieving same levels of cooperation as those who have not been trained to think of humans as selfish and uncooperative. 

Important practical implications arise in the case of externalities. These are situations where individuals can make profits for themselves at public expense (by polluting the environment for example). Solutions proposed by economists systematically ignore the most important factors relevant to solving such problems, which relate to social norms of cooperation. This is because economic models predict that these don��t exist, or else they don��t matter. 

 

2.2 Public Goods & Free Riding

 

Public Goods have been an area of concern to economists for a long time, since economic theories predict that everyone will free-ride and inadequate quantities will be supplied. However, many experiments show that the tendency to free ride is much less than predicted by economic theories. A summary by Dawes and Thaler (1988) states that "... the analytically uncomfortable (though humanly gratifying) fact remains: from the most primitive to the most advanced societies, a higher degree of cooperation takes place than can be explained as a merely pragmatic strategy for egoistic man." 

The issues to be discussed can best be illustrated within the context of a simple public goods experiment. There are i=1,2,��,N players. Each one receives an endowment E(i). Each can choose to contribute C(i) between 0 and E(i) to the public good. The total contribution is C= C(1) + �� + C(N). The benefits from the public good are represented by a multiplier M>1. The total benefit is MC and this is distributed to equally to all players, regardless of whether or not they contributed. Thus the i-th player gets [(MC)/N] from the public good, plus E(i) – C(i) the amount he did not contribute.  

To see the conflict between public and private interest more clearly, consider a case where N=4, M=2, and E(i) = $10. If each of the four players contributes the full $10 to the public good, the total contribution of $40 is doubled and redistributed, so each player gets $20, which is the social optimum. However, the social marginal benefit from contributions differs from the private marginal benefit. Each dollar I contribute is doubled, but then it is shared by four, so that I get back only 50 cents. If I don��t contribute anything, but all others do, then $30 will be doubled to $60, and redistributed four ways, so I will get $10+$15=$25. My gains of $5 will also cause a loss of $5 to each of the other three, for a net social loss of $10. 

Standard economic theory predicts that no one will contribute anything; every one will free ride. This prediction is strongly contradicted by actual behavior over a very wide class of environments and a variety of experiments in the field and in the lab. Sally (1995) summarized the conflicts between economic theory and observed behavior in over 35 years of experimental studies. Below we list three specific conflicts which have received strong empirical support, as discussed in Kagel and Roth (1995). 
 

1. Whenever M/N < 1, the marginal private benefit is less than 1, and economic theory predicts zero contribution to public good. In fact, contributions are never zero, and respond systematically to M/N, increasing as marginal social benefit increases. Economic theory provides no explanation of this. There are competing explanations based on psychological and social theories of reasons for cooperative behavior which might justify this response.
2. In multiple rounds of play, cooperation starts out at high levels, and then declines. Numerous explanations based on economic theory have been assessed and rejected. The most plausible hypotheses to explain this empirical regularity rely on social norms, outside the scope of economic theory.
3. It turns out to be easy to manipulate the environment so as to achieve nearly full cooperation. Other manipulations achieve the economic theory outcome of nearly zero contributions. People respond to expectations of others, and these expectations can be manipulated by experimenters. In environments where high levels of cooperation are expected, high levels of cooperation are achieved. In environments of distrust, very low levels of cooperation (but rarely zero) are observed.

 

These findings have very important policy implications which have not been absorbed by economists. The vast majority of research on public goods focuses on mechanism design, incentive compatibility, and other problems created by the supposed ��free rider�� problem. In fact, human beings are trained to be social from childhood, and there exist mechanisms for inducing cooperative behavior which lie outside the bounds of current economic theory. The behavioral research reveals that solutions to public goods problems can only be found if we drop homo economicus and consider observed social behavior of human beings. 
 
 

 

2.3 Moral Values and Cheating

 

Dawes & Thaler (1988) write that economic theory assumes that people ��care nothing for outcomes to other players �� and have no qualms about their failure to do "the right thing."�� According to the standard economic model of rational and selfish human behavior, cheating is a result of deliberate calculations of the expected amount of gains from cheating, the probability of being caught in the act, and the magnitude of punishment if caught (Allingham and Sandmo 1972; Becker 1968), which can be regarded as an external cost-benefit analysis. However, psychologists show that internal values system can play an important role in decisions due to internalization of the norms and values of society (Campbell 1964; Henrich et. al. 2001).  

A series of carefully designed experiments by Mazar, Amir and Ariely (2008) bring out the conflict between the economists views of morality, and the idea of morality as a set of internalized values. Students were paid for correct answers to tests under different kinds of conditions which allowed for cheating. Some of the main results were as follows: 

1. Even when students self-report their own scores, and there are no cross checks, only about 10% to 15% of the opportunity for cheating is utilized. The economics model predicts 100% cheating in this case.
2. Increasing the payoff from cheating actually decreases the amount of cheating, opposite of the economic theory predictions. The authors explain that small payoffs make cheating a small violation of internal morals which is tolerable. High payoffs make it clear that cheating is a gross violation of internalized morals, which is done only by a few.
3. Reminders of an honor code, or of religious values, prior to testing, reduce cheating to zero. According to economic theory, such reminders should have no effect on behavior.
4. Students are not aware of their own internalized moral values. When asked to predict outcomes of these experiments on cheating, they utilize the economic model, which yields wrong predictions. This means that students are not ��rational�� in the economic sense: they cannot predict their own behavior or that of their fellow students in moral dilemmas.

 

Because economic theory neglects morality, Nobel Laureates Arrow and Solow both went astray in their analysis of an anomaly pointed out by Titmuss (1970). Titmuss argued that monetary incentives would actually reduce blood donations, since it would undermine the sense of civic duty which leads donors to donate. Both Arrow and Solow thought otherwise; they argued that the two incentives would supplement each other. Frey and Oberholfer-Gee (1997) show that crowding-out holds; monetary incentive interfere with the sense of civic duty, as argued by Titmuss.

 

Hausman and McPherson (1996) have given an excellent exposition of the problems which arise from ignoring the moral dimensions of economics. Many policy recommendations made by economists on apparently positive grounds incorporate hidden value judgments. These authors argue that it is impossible to do economics on purely positive and factual basis. Rather than hide moral judgments, it is preferable to make them explicit in the analysis.

 

2.4 Social Capital

 

The global financial crisis of 2007-8 has led many prominent economists to the realization that unrestrained greed does not lead to socially optimal outcomes. While unregulated markets cannot be trusted, governments have repeatedly failed to live up to the role of being wise, responsible, just and impartial regulators of the economy. Bowles and Gintis (2002) have emphasized the role of the community as a suitable intermediate level between the microcosm of the individual and the macrocosm of the government. Social capital refers to trust, concern, and a sense of responsibility towards the community, as well as a willingness to live by the norms of one��s community and to punish those who do not. Communities can solve certain problems in ways that neither governments nor markets can, because of two factors not taken into account in conventional economic theories: 

1. Communities have access to inside information about benefits and costs, ability and willingness to pay, skills, talents and capabilities of the members. Both markets and governments lack access to this localized information.
2. Communities can create and enforce social norms using education in formal and informal ways, as well as approbation, ostracism, and other mechanisms not available to markets or to governments.

 

The Grameen Bank succeeds in getting high repayment rates and returns in poor communities where transaction costs in terms of gathering information on creditworthiness and enforcing repayments would be too high for a commercial operation. Inside information and social pressure based on community is crucial to its success (Stiglitz, 1990). The Orangi Pilot Project succeeded in laying down sewer lines in a poor neighbourhood at minimal cost because of community involvement; see Khan (1998) for details. The community knew which members could afford to pay, and could enforce an equitable distribution of the burden. It could also exploit knowledge of relevant engineering skills available with members of the community.  

Frey and Oberholzer-Gee (1997) provide a startling example of how communities violate predictions of economic theory. Based on survey conducted in a Swiss community chosen as the site for depositing nuclear wastes, they found that about 50% of the citizens agreed to accept the site out of a sense of civic duty. With full awareness of the risks and hazards of these sites, they were prepared to sacrifice personal benefits for the sake of the good of the larger community. Offering large payments as compensation for placement of the site reduced the willingness to accept the site to 25%. The monetary incentive ��crowds-out�� the social incentive. These finding are exactly the opposite of the predictions of economic theory. 

Bowles and Gintis (2006) provide many more examples of successful operation of community based initiatives and firms in situations where conventional theories predict failure based on incentive and informational problems. They point out that communities are fragile, and government policies can make or break communities. Since communities are invisible in the conventional economic theory framework, economic policy generally ignores them, resulting in destruction of tremendous amounts of valuable social capital.  

 

Section 3: Limitations on Computational Ability

 

In the previous section, we showed that people are often concerned with payoffs to others, justice, equity, and also reciprocate both good and bad done to them. They have broader concerns than just personal selfish utility. In this section, we look at the ��maximization�� part of the utility maximization model. Herbert Simon won the Nobel Prize for showing that people often do not maximize; instead they satisfice. In some cases, maximization may fail because people lack the computational abilities to maximize. Instead they use rough and ready heuristics, which have systematic and predictable biases. Gigerenzer and Todd (1999) is a convenient reference for heuristics used by people to make decisions. These heuristics have their advantages, but they do not resemble the optimization strategies we routinely assume in economics. We give several examples of situations where people systematically fail to compute optimal strategies required to maximize utility. 

 

3.1 The Beauty Contest.

 
 

The object of this section is to present an example which makes the following points:

• Economic theory assumes (unbounded) rationality
• People only make a small (bounded) number of calculations.
• People (not Economists) know that other people only make a small number of calculations, and make accurate predictions on this basis.
• Therefore, intuition outperforms economic theory, in this context.

 

One example to demonstrate these phenomena is ��The Beauty Contest.�� The discussion and empirical evidence given below is based on Camerer (2003). The game is played with N players. Everyone writes down a number between 1 and 100. The one closest to 70% of the average wins. We now consider intuitively how we might play this game. Let us suppose all people write down numbers at random within the range specified. Then the average will be around 50, and 70% of this is 35. So if people make no strategic calculations at all, and just write down a number at random, a good choice to win the game would be 35. 

Now suppose all people make this calculation – this is first order strategic thinking. Then the average number written down would be 35. In this case, I can win by putting down 70% of 35, which is 24.5. Next suppose all people carry out TWO rounds of strategic thinking and assume others also do so. Then they would realize that the average number written down would be around 24.5. In this case the winning strategy would be to write down 70% of 24.5 which is about 17.5. Economic theory predicts that everyone would do infinite rounds of reasoning, which converges to zero. So all people should write down ��0��.  

What actually happens when this game is played? On the average people do about two rounds of iteration. Smart players anticipate this, and calculate that others will do around two rounds of iteration on the average, and construct winning strategies based on this assumption. In some pools, a few ��economists�� can be found who iterate to infinity and therefore predict ��zero,�� but they are a very small percentage and never win. 

What are the practical consequences of this failure of perfect rationality to economic theory? Mankiw (1985) shows that small menu costs – changing prices at a restaurant requires reprinting the menu – can lead to large business cycles. Similarly, Akerlof and Yellen (1985) show that ��near rationality�� -- or approximate maximization – can have large effects on markets. More than 800 later publications which cite these papers show that approximate instead of exact maximization could be responsible for a wide range of phenomenon, such as sticky wages, the Phillips curve, non-neutrality of money, efficiency wages, and many others. If objective functions to be maximized are nearly flat in a neighbourhood of a unique global maximum, then approximate maxima can range over a very wide set, leaving the approximate equilibria very indeterminate. Virtually no one would seriously argue that people always exactly maximize utilities, but it is widely believed that approximate maximization would lead to approximately the same results as exact maximization. The literature cited above shows that this is not true; bounds on computational ability have serious consequences for economic theory. 

 

3.2 The Winner��s Curse.

 
 

It is more or less self-evident that people are not good at solving maximization problems. We do not expect the average consumer to be able to maximize even the simplest utility functions subject to budget constraints (most people do not know calculus). This is similar to an argument that was made against rational expectations – the most informed agents are the econometricians who are armed with all the data and apply the most sophisticated modelling techniques. When their forecasts can be systematically wrong, how can ordinary people do better? The goal of the previous section was to show that people systematically make a limited number of computations. The goal of this section is to show that people can systematically be wrong, and can fail to learn and incur large losses due to errors in computations even in the long run. 
 

Winner��s Curse refers to a situation where the winning bid in an auction is systematically too high, so that the winner systematically makes losses. Such a phenomenon was reported in a number of real life auctions such as bidding for drilling rights to oil, publishing rights to books, professional baseball free agent contracts, and corporate takeover battles. See Kagel and Roth (1995, p. 537 for details). In many of these situations, bidders are experienced and the amounts of money involved are very large, so economists have strongly contested claims of failure of rationality implied by a winner��s curse. If rationality fails to hold in these ideal situations, it can hardly be relied on anywhere. Below we examine one simple case where winner��s curse can clearly be demonstrated, and is clearly due to failure of the (infinite) rationality assumption. 

We consider a Vickery second price auction. There are j=1,2,��,N bidders for an object which has value V(i) to the i-th bidder. Each bidder submits a sealed bid B(i) to the auctioneer. The highest bidder wins the object, but he must pay the second highest bid (SHB).  The profit to the winning bidder is P = V(i) – SHB. What is the optimal bidding strategy in this situation? It can easily be proven that the optimal strategy is B(i) = V(i) – the bid should equal the value of the object to the i-th bidder. Experimental evidence shows that most players cannot calculate this simple optimal strategy. Figure 7.1 of Kagel and Roth (1995, p. 509) shows that bids remain significantly above this optimal level, even after thirty rounds of play. Any bid higher than V(i) runs the risk of loss from winning in situations where  B(i) > SHB > V(i) – thus experimental results confirm the existence of a winners curse. Kagel and Roth (1995) show a wide range of alternative configurations for auctions, many of which lead to winner��s curse. 

After ruling out a number of explanations more compatible with the economist��s world view of rationality and maximization, Kagel and Roth (1995) find the plausible explanation to be a mistaken calculation by the Bidders. Bidders reason (wrongly) as follows. I don��t have to pay what I bid. If I increase my bid B(i) over my private valuation V(i), my chances of winning the auction increase a little, but my cost (the second best bid) remains the same. The correct reasoning is as follows. 

There are two cases: the second highest bid is below V(i): SHB<V(i). In that case B(i)=V(i) will win, and there is no gain from B(i)>V(i).In the second case, if SHB>V(i), then and only then will a bid B(i)>SHB>V(i) win the auction. However, in such a case winning will lead to loss since I will have to pay SHB>V(i) to win an object of value V(i). So the only case in which my probability of winning increases is the case in which winning the auction leads to a loss (SHB>V(i) case). Therefore I should never bid B(i)>V(i).In general, people have a lot of difficulty with understanding and implementing the idea that the bid B(i) should be made conditional on the assumption that it will be the winning bid. The value of the bid matters only if it is the winning bid. This conditioning can change the relevant distributions and calculations. Failure to condition in this way leads to the winners curse in many different situations, as described by Kagel and Roth (1995).   

It is empirically shown that people fail to compute the optimal strategy, and do not learn it from experience, even in very simple situations, contrary to the assumptions of utility maximization.  Almost all over the place in economic theory, we start with the assumption of rational agents figuring out the maximizing strategies in very complex environments. These are calculated to explain behavior, make predictions and derive policy implications. Empirical evidence shows that this assumption is wrong, and leads to wrong policy prescriptions by economists in many different domains of action. One striking illustration is given by Gigerenzer and Edwards (2003). Doctors and patients make decisions on diagnosis, medical tests, and prescriptions on the basis of statistical information. Professionals with an average of 14 years of experience routinely made wrong inferences and decisions when given data about prevalence of disease, sensitivity and specificity of tests in their own speciality. When the same data is presented in an easier to use and more directly accessible format, these errors are dramatically reduced. Contrary to assumptions of economists, experience and high stakes do not serve to induce rationality. 

 

3.3 Forward Looking Behavior

 

Many economic theories are based on the assumption that people look forward and maximize utilities over their lifetimes. Short-sighted behavior which may prove harmful in the long run is ruled out: wisdom prevails over temptations. Does consumer behavior correspond to this assumption, which is the basis of many economic theories such as the Life Cycle Hypothesis? There is strong empirical evidence against it, some of which we document in this section.

 

3.3.1 The Small Worlds of Savage

 

In setting up the framework of decision theory, Savage (1954) recognized that all decisions are interlinked, but to consider these linkages would make the complexity of the problem unmanageable. He therefore made the ��small world�� assumption which isolates a particular decision problem and assumes that it can be studied without reference to forward and backward linkages. Barberis and Thaler (2003) present empirical evidence favouring ��narrow framing,�� or solving problems in isolation, and how it can impact adversely on real world consumer decisions.  

Camerer (2003) provides striking evidence against multiperiod maximization hypothesis in the context of a two stage ultimatum game. In the first stage, Proposer offers Responder a share of a $20 pie. If the offer is accepted, the game ends. If it is rejected, neither party gets anything, and a second game is played where the role of the Proposer and Responder is reversed, and the size of the pie shrinks to $8.  In multi-stage games of this type, the phenomenon of ��disadvantageous counter-offers�� is frequently observed. This means that an initial offer of ($15, $5) is rejected by the responder. In the second stage, responder offers an equal share ($4, $4). Instead of offering $4, which gives him only $4, would he not be better off accepting $5 which was offered to him on the first round? Many explanations of this phenomenon which are compatible with multi-stage maximization are examined and rejected. Examining eye-movements and mouse cursor moves of the players shows that many people do not examine the payoffs in the second stage of the game and those who do, do not pay much attention to them. Optimal strategies work by backwards induction and require taking careful account of each stage, starting from the last stage. In fact players look carefully only at the current game and not too much at later stages. Each decision appears to be made in isolation, without taking into account the multi-stage character of the game.  

Many economic theories invoke the Life Cycle hypothesis, according to which humans plan their current decision taking into consideration their lifetime prospects of earnings and consumption patterns. If the forward looking hypothesis can be rejected in simple environments like this, how can we rely on it for complex lifetime utility maximization calculations?  Shefrin and Thaler (1988) show the differences which result from taking into account the quirks of human behavior, which deviates from the globally rational maximizing model.

 

3.3.2 Incorrect Probability Forecasts

 

The future is always uncertain. Forward looking rational behavior requires that people should make roughly accurate assessments of the probabilities of different types of events which might take place. There is substantial evidence that people make systematically distorted judgments about probabilities, and that these impact on economic affairs in a serious ways.  

Camerer and Kunreuther (1989) list several cases with important policy implications in situations where low probability events are involved. People suffer from a systematic ��it can��t happen to me�� bias.  They smoke cigarettes, don��t wear seatbelts, under-insure for flood risk, and undertake many actions which are not rational on the basis of correct risk calculations about the probability of future event. In the wake of 9/11, there was a massive switch to driving instead of airplanes, when the risks of accidents from driving were substantially higher than those from flying. In such situations, where people systematically make mistakes in computation of probabilities, there are several non-market mechanisms which can be used to improve welfare.  

Barberis and Thaler (2003) list several areas where probability judgements are systematically wrong, and the effects of these on stock market behavior.

For example, there is a systematic ��overconfidence�� bias. Events that people consider as certain occur with about 80% probability, while events that people consider ��impossible�� occur with 20% probability. As a typical real world illustration, Jean Drapeau, mayor of Montr��al, said that ��The Games could no more have a deficit than a man could have a baby!�� shortly before the 1967 Montreal Summer Olympics ran a $2 billion deficit (Griffin and Brenner, 2004). 

People judge probabilities by their recent experience or by salient events, which are often not representative. They do not differentiate between large samples which give reliable evidence and small samples which do not. Irrelevant details can lead people to judge the probability of an intersection of events AB to be more likely than A, which is of course impossible. Schoumaker (2004, p. 277) writes that people often do not look at relevant frequency information, and fail to differentiate between probabilities of 1 in 1000 and 1 in 10000 even though one is ten times the other.  
 

The image of a rational planner accurately calculating the probabilities and consequences of all possible outcomes cannot be maintained in face of overwhelming empirical evidence to the contrary. Dropping this model would facilitate welfare-improving policy interventions which provide training to people to make correct decisions under risk.

3.3.3 Temporal Inconsistency

 
 

There are many situations where a sequence of decisions made by a single consumer is not consistent with the idea that all of the decisions maximize a common objective function, as assumed by economic theories. There are many cases of conflicts – if one decision is optimal, then the others cannot be and vice versa. This again shows that the standard multiperiod model of consumer behavior which maximizes a sum of utilities is wrong. We give two different examples of such conflicts; many more are documented in the literature.  

Attitudes to Risk Depend on Outcomes: In a sequence of optimizing decisions, outcomes of past decisions are irrelevant In sequences of risky choices, Thaler and Johnson (1990) observe the following empirical regularities:

(1) risk aversion after prior losses;

(2) risk seeking after prior gains;

(3) changes in risk-taking behavior when one outcome (certain or risky) can allow decision makers to "break even."

This behavior, where decision making changes according to outcomes, is not consistent with the standard multi-stage maximation model. 
 
 

Figure 1. Hyperbolic discount function (Read, 2004) 

Read (2004) provides empirical evidence that people do not use exponential discounting to evaluate future prospects. The best fit to survey data appears to be some form of hyperbolic discounting. Unlike exponential discounting, hyperbolic discounting has the unpleasant property of time inconsistency, demonstrated in the Figure 1. Viewed at time 0, the larger-later reward (x₂) is preferable. However later on, for a brief period of time, the smaller-sooner reward (x₁) will be preferred. This comes under the heading of self-control problems, which have been extensively studied in behavioral economics. Anticipating this problem, the consumer might take steps to ensure that he will not be able to choose x₁ so that the temptation might be bypassed. The classic case is that of Ulysses, who has himself bound to a mast to avoid being tempted by the sirens. There is a conflict of interest between self-now and self-later which does not arise in economic models. More discussion on this topic is given in section 5.4 about savings behavior. 

The DSGE model is one example, where lifetime utility is maximized by a representative agent in a very complex, dynamic environment with uncertainty.  Testifying before a congressional committee investigating the failure of economic theories to predict the global financial crisis, Nobel Laureate Robert Solow (2010) put the blame on the DSGE models as previously mentioned above. He further states that ��a thoughtful person, faced with the thought that economic policy was being pursued on this basis (DSGE models), might reasonably wonder what planet he or she is on.�� Solow, and others, have argued that use of models where agents make rational forecasts is a major reason why economists failed to foresee the global financial crisis. Had they foreseen it, steps to prevent it could have been taken. Thus defective theories about human rationality and foresight can be held responsible for major economic crises. 
 
 

 

Section 4: Evidence from Cognitive Psychology

 

The foundations of the current methodology in economics can be traced to Robbins (1932); see Cooter and Rappoport (1984) for a description of the change in methodology that took place. Some of the errors described above can be traced to the influence of logical positivism, which was tremendously popular in that era. Although this philosophy has now been rejected, methodological foundations of economics have not been revisited. Logical positivism led to the neglect of unobservable motivations, and an over emphasis on the observables. In the field of psychology, behavioral psychology became the dominant school of thought.  Cognitive psychology emerged in reaction to this, and re-emphasized the role of internal mental states and unobservable structures of thoughts and feelings. Better models of human behavior which violate the methodology of logical positivism can also be the basis of more empirically accurate economic theories. Inter-disciplinary research by economists and psychologists incorporating insights from cognitive psychology have led to the discovery of many conflicts between actual human behavior and models of this behavior based on economic theory. Some of these conflicts are summarized here.

 

4.1 The Endowment Effect

 

Consumer decisions are influenced by status-quo biases. Status-quo bias arises when an individual��s decision is affected by the status-quo itself.  One indicator of such a bias is the willingness to accept-willingness to pay (WTA-WTP) disparity, which is one of the most controversial anomalies in choice theory. There are many goods for which the price people will accept to sell the good in their possession is very high, but the price they will pay to purchase the good when they do not own it is very low. The existence of even a single large WTA-WTP is regarded as pathology (Arrow, 1993; Diamond, 1996), since that would lead to unreasonably large income effects in the study of utility maximization (Mandler, 2004). Thus, the traditional analysis of indifference curves has no reference to current endowments. However, in many experiments, subjects require a relatively large amount of money to move from status-quo, contrary to utility theory.  

Kahneman, Knetsch and Thaler (1990) test WTA-WTP disparity and if this disparity vanishes when subjects learn in market settings. For this purpose, consumption objects (e.g., coffee mugs) are randomly given to half the subjects in an experiment and then markets for the mugs are then created. According to the conventional theory it is expected that supply and demand curves should be intersecting at their common median and half of the goods provided should change hands. However, people with mugs demand about twice as much to give up their mugs as others are willing to pay to get one. The possession of the mug creates an ��endowment effect�� and owners are unwilling to part with them.  

Carmon and Ariely (2000) find similar disparities on their experiments based on four studies examining buying- and selling-price estimates of tickets for college basketball games. Ariely (2008) reports that the disparity between WTP of a student who wants a ticket, and WTA of a student who holds a ticket was up to fifteen times. This implies that if a good is evaluated as a loss when it is given up, and as a gain when it is acquired, loss aversion and endowment effects induce higher values for owners than for potential buyers. These phenomena cannot be explained by standard utility maximization models. 

Another striking example of the WTA-WTP disparity arises in context of risk of death (or value of life) calculation. In order to accept even small increases in the risk of death, people ask for huge amounts of money. However, they are willing to pay very little in order to decrease risks already present in their existing circumstances. This disparity, documented in many sources such as Mart��n-Fern��ndez et al. (2010), has significant implication for economic policy involving Value of Life calculations. 

These results clearly invalidate the Coase theorem, which argues that the allocation of resources is independent of the assignment of property rights when costless trades are possible. The experimental results remain the same when there are costless trade opportunities, meaning Coase theorem predicts more trade than actually occurs. An implication of this asymmetry is that there is much less trading than economic theory predicts, since status quo is preferred by all. The Coase theorem implicitly assumes no disparity between WTA and WTP, and is the basis for many types of economic policy discussions which are based on erroneous assumptions about human behavior and psychology.

4.2 Value Ambiguity

 

People value objects along dimensions which cannot be priced in markets. Gifts can carry sentimental value which makes them more precious than the identical good in the market; this conflicts with utility theory which would price identical goods equally. Furthermore, this non-marketable value can be instantly acquired, as the experiment with mugs described in the previous section shows. Thaler (1994, p.183) states that ��Preferences �� may exhibit value ambiguity, that is, consumers may not be able to assign a unique monetary value to an object.�� We discuss several cases of value ambiguity and how it impacts on economic theory.

 

4.2.1 Some Things are Priceless

 
 

People who struggle to achieve the potential for excellence within the hearts of all human beings learn that life is infinitely precious. They would refuse to play Russian roulette² for all the gold in the world. They are surprised at the attempts to measure the value of life, since this is not commensurable with worldly goods and wealth. They would refuse to sell their wives and children at any price. They would be baffled at calculations of costs and benefits of raising children, since these relationships provide meaning to and shape the purpose of life, while gold is meaningless in terms of these larger dimensions. Some economists, on the other hand, have made all of these calculations, publishing papers on many topics normally considered outside the bounds of market. The fundamental economic paradigm reduces all things to a common scale. The pleasure derived from children is a utility like all other consumption utilities and can therefore be measured against other utilities like the pleasure derived from driving a car. Once this commensurability (which is one of axioms of utility theory) is assumed, then such horrifying calculations become possible. It then becomes possible to assign a dollar value to a million lives destroyed, and consider the tradeoff between this loss and the gain in terms of profits from oil.  

The idea that human lives cannot be purchased for money seems to be in direct conflict with the existence of a labor market, where such transactions are routine. Actual human behavior keeps breaking out of the bounds to which economic theory confines it. Human beings routinely refuse to behave as commodities which can be bought and sold on the market like any other. This refusal has a major impact on real world economics. We illustrate that ��Man does not live by bread alone,�� in a separate section which develops this critique; see also section 5.3 below. 

 

4.2.2 Values are Generated by Arbitrary Choices

 

Economic theory assumes that we have built in preferences which allow us to make choices among different types of consumer goods. There is strong evidence to show that this is not true. When faced with a choice, people make decisions using arbitrary clues to guide them. Once the choice is made, that creates a preference which we stick to for the sake of consistency. This phenomenon has been labeled ��coherent arbitrariness�� by Ariely et. al. (2003). We provide three types of evidence to show that people create preference when confronted with choices, rather than the opposite, which is assumed by economic theory. This is also in conformity the ��process�� aspect of consumer decisions, discussed later in Section 6, which describes Business School models of consumer behavior. 

The phenomenon of anchoring shows how values attached to objects are arbitrary, instead of being related to built-in preferences which are exogenous. Ariely (2008 p.28-29) describes an experiment in which students are asked to write down the last two digits of their social security numbers prior to putting down prices at which they would be willing to buy a list of items. Remarkably, there was a strong correlation between the numbers they write and the prices they offer. The student with high last digits in the range 80-99 offered between 200 to 300% higher prices than the students with low last digits in the range 00-19. Writing down these arbitrary numbers served as an anchor which unconsciously guided the decisions made regarding the value of the objects. Furthermore, the students were not aware of this process of anchoring, and did not feel that their writing down these numbers had any influence on the prices they chose. 

Savvy marketing experts are aware of the phenomenon of ��anchoring�� and use it to generate favorable prices for new products. Ariely (2008) lists several striking examples: 

• After failure of initial efforts to sell black pearls, marketers displayed them on Fifth Avenue at an outrageously high price, in the company of similarly expensive jewelry. This had the desired effect of creating a high price anchor for the product, and many sales were made at these prices.
• In a classroom exercise, Ariely offered a session of poetry reading to his students. Half the class was asked how much they would pay to attend the session, while the other half was asked how much they would require as payment in order to attend. All members of the first half offered to pay a positive amount, while all members of the second half asked for some positive payment to attend. Based entirely on the arbitrary anchor provided by the question, students evaluated the poetry reading as a ��good�� or as a ��bad�� (experience).

 

All of these experiments demonstrate that the economists�� model of price setting via supply and demand is seriously in error. Goods do not come with demand curves attached, and demands are generated by contextual references and subject to a very high degree of arbitrariness.

 

4.2.3. Comparative Evaluations

 

Instead of coming to an evaluation of objects with built in preferences, people evaluate choices they are presented with and create a preference on the basis of contextual clues. One important contextual clue is comparison. If Y clearly dominates Z then Y will be preferred to Z. Ariely (2008) documents many experiments of where three choices are offered to prospective consumers: X, Y, and Z. Choice Z is dominated by Y. For example Y is a one year subscription to both print and online versions of the Economist for $125, and Z is a one year subscription to only the print version at the same price of $125. In such situations, the presence of Z should have no effect on the option chosen – the customer should eliminate it from consideration and make the choice as if only X and Y are present. Experiments and empirical evidence presented by Ariely shows that this is not the case. Because consumers do not have well defined preferences, they focus on the fact that Y is superior to Z. No similar information is available about X. When Z is present, Y is chosen significantly more often than when only X and Y are present.   

Another confirmation and explanation of this phenomenon is given by Schoemaker (2004, p.352-353). X is a lottery which pays $15 with probability 65%. Y is a lottery which pays $35 with probability 30%. X and Y are not easily comparable. Z is a lottery which pays $14 with probability 65%, which is clearly dominated by X. Subjects were divided randomly into two pools, labeled ��conflict�� and ��dominance��. Those in the conflict condition were offered a choice between X and Y. Those in the dominance condition were offered the (easy) choice between X and Z. Both groups were given the option of purchasing a third option, different from X,Y,Z, as an additional choice. Since the value of X is the same in both conditions, standard economic theory predicts that  the conflict group will be LESS (��) interested in purchasing an outside option – the value of the third option must be superior to both X and Y to be worth purchasing. The dominance group should be MORE (��) interested since the value of the third option need only be superior to X. Empirical data shows that the opposite is true. The dominance group buys options significantly less often, because they are satisfied that they have made the right choice in X which dominates. The conflict group cannot decide which of X and Y is better and searches for a superior alternative, contrary to predictions of economic theory.

 

4.3 The Just Price

 

It is commonly believed that the ��Just Price�� is a medieval doctrine with no relevance to contemporary economics. Modern economists believe and teach that any price to which both parties agree is automatically ��just�� and hence any theory of the just price is vacuous. Kahneman et. al. (1986a, 1986b) summarize the large amount of empirical evidence against this idea. The evidence is conveniently broken into three parts: 

1. Substantial consensus exists about what are fair and unfair prices. This means that there are well defined social norms which define the just price.
2. These social norms are internalized. That is, people care about being fair to others and being treated fairly.
3. People act to punish unjust actions, and approve of and reward fair actions, even at cost to themselves.

 

Taken together, these principles mean that the concept of the just price will influence transactions in the economic domain. There is strong evidence that this is so – in many cases, firms refrain from charging profit maximizing prices because it would be unjust. The first principle has been established by means of surveys demonstrating consensus about various principles of fair pricing. We will list some examples illustrating the second and third below. 

Most people would pay three to four times for a cold drink at a restaurant, but would not pay the same price at a simple shack. This is true even if they are taking the drinks to a picnic elsewhere, so that the origin does not matter. The notion of ��fair�� price has a strong effect on consumer decisions. Similarly, at popular games and matches, it is known in advance that tickets will sell out, and will be resold at high prices by scalpers. Why don��t stadium owners price the tickets higher to prevent this from happening and capture the excess profits? Because they do not want to be seen as being unfair. 

In opposition to the dominant trend several economists have invoked a notion of fairness in their interpretations of regulation (Zajac 1978, 1985) and of the market phenomena of price and wage stickiness (Hirschman, 1970; Arrow, 1973; Akerlof 1979, 1982; Solow, 1980). Okun (1981) offered a notably detailed account of the demands of customers and employees for fair treatment and of the role of perceived unfairness in triggering a search for alternative suppliers. Okun made a strong case that many customer markets resemble labor markets more than they do pure auction models. Like labor markets, customer markets sometimes fail to clear, an observation that Okun explained by the hostility of customers to price increases that are not justified by increased costs. The opposition to price rationing as a response to a shortage is easily documented. 

A number of economic phenomena can be predicted on the assumption that the rules of fairness have some influence on the behavior of firms (Kahneman et al., 1986a). The rules of fairness tend to induce stickiness in wages and asymmetric price rigidities. They also favor a much greater use of temporary discounts than of temporary surcharges in price adjustments. Where costs for a category of goods are similar, opposition to price rationing may lead to sellouts for the most desirable items (e.g., the main game on the football calendar or the Christmas week in a ski resort). The sellers are aware that consumers will think it is unfair to charge high market clearing prices and will take revenge. Standard economic models cannot explain these phenomena. 

 

4.4 Mental Accounting

 

People separate categories of income and treat them differently. Money received as a gift, or as an unexpected bonus, might be spent on a luxury item. The same amount of money received from working overtime would not be treated in the same way. Policy examples are easy to find. Thaler (1980, 1985) records several examples of consumer behavior which are irrational for a utility maximizer, but make sense according to cognitive model which he has developed to explain them. Our first example is from Thaler (1984, p. 25) 

Mr. and Mrs. L and Mr. and Mrs. H went on a fishing trip in the Northwest and caught some salmon. They packed the fish and sent them home on an airline, but the fish were lost in transit. They received $300 from the airline. The couples took the money, went out to dinner and spent $225. They had never spent that much at a restaurant before. 

Several similar anecdotes illustrate that people mentally separate categories of income and use them for different purposes. This is not a quirk but an established regularity of human behavior, which violates the single budget constraint and utility maximization model³.  

Rebates are widely used as an incentive to sell consumer goods. Thaler analyses a case where significant rebates were offered on automobile purchases. Both the giving and the taking of rebates incur significant transaction costs which could be avoided by simply reducing the price. There is no way to explain the use and popularity of rebates other than by positing that consumers keep the purchase cost and rebate gain in separate mental accounts. Thaler (1984, p 38) gives an in-depth discussion of why conventional accounts fail to explain the observations in this context.

 

Section 5: Savings Behavior

 
 

The basic consumer decision is to allocate some portion of his income to consumption and the rest to savings. Utility maximization will force savings to zero over a suitable time frame, which is why savings behavior has not received sufficient attention within consumer theory. There are several assumptions which justify this neglect: 

1. Competitive markets equalize the average rate of return to savings in all forms – that is, for savings accounts, shares, bonds, etc.
2. Random fluctuations in savings can be smoothed out through nearly frictionless intertemporal borrowing and lending at market rates.
3. Maximization means setting savings to zero in the long run. Alternatively, there is a rationally planned bequest, based on utility attached to the coming generations .

 

If these assumptions are approximately valid, then many important real world decisions related to savings do not matter. In fact, as we will show, all of these assumptions are false, and all three impact heavily on consumption decisions. 

 

5.1 The Equity Premium Puzzle

 

There are many reasons why financial markets offer the most favourable environment for testing theories of rational behavior. The commodities being traded (stocks) have clear, precisely defined and observable market values. They are traded in a market with large numbers of buyers and sellers, and with substantial amount of information available to all. Transaction costs are minimal, and there are no externalities. Suggestions that stock prices could be systematically wrong due to irrational investor behavior were ignored by mainstream economists until the global financial crisis of 2007-8 made it apparent to all.  

Many aspects of stock market behavior are very difficult to explain under the assumption that all investors are rational. One of the most prominent of these anomalies is called the Equity Premium Puzzle. Over the entire twentieth century – from 1900 to 2000 – in every twenty year period, stocks earned a lot more than bonds. On the average, $1000 invested in bonds would be worth $1350, while stocks would more than triple to $3370 over a 20 year period. Despite this huge and virtually risk free differential in twenty year returns, long term savings of consumers were mostly held in low earning and safe instruments. Why? That is the equity premium puzzle. Note that since returns to savings differ so drastically, consumption patterns will also be heavily dependent on the choice of savings instruments, an issue which is completely ignored by conventional economic theory accounts of the consumption function. 

Barberis and Thaler (2003) offer two behavioral explanations for this puzzle. One is that consumers do ��narrow framing�� – they take their investment decisions one year at a time. For one year intervals, stocks are a lot more risky than bonds – it is only in the long run that these risks are reduced by averaging the good and bad years. The second explanation is  ��ambiguity aversion�� – stock market returns are not just risky, they are uncertain. The difference between risk and uncertainty can be summarized as the difference between roulette, where all outcomes have known probabilities, and a horse race, where probabilities of which horse will win cannot be calculated. Experimental results show that consumers stay away from uncertainty while they tolerate risk. A third possibility is that memories of the Great Depression led people to stay away from stock markets more than was objectively justifiable. This would represent the operation of the ��availability�� heuristic, which systematically distorts probability judgments; see for example Keren and Teigen (2004). 

There is a huge number of alternatives explanations, and no clear resolution of the puzzle is currently available. The important point to note here is that according to conventional economic methodology an explanation is acceptable if a model with agents acting rationally replicated observed phenomena. Conformity of actual behavior with theoretical model is neither expected nor asked for; see Friedman��s (1953; 1957) famous justification for the ��as-if�� methodology. Thus behavioral explanations which conflict with utility maximization are not considered acceptable. This makes it impossible to find the correct explanations of many observable events, as is now widely realized following the global financial crisis of 2007-8.

 

5.2 Intertemporal Choice.

 

A large number of theories of development focus on the choice between consumption and savings – consumption today versus consumption tomorrow. If savings can be increased, the rate of investment would increase and that would lead to growth and higher levels of consumption tomorrow. So how people choose to save for the future is of crucial importance to many standard models of growth like Harrod-Domar and Solow. Foreign aid is justified by arguing that savings rates are too low for rapid growth, and so the gap between desired investment levels and existing savings can be filled by foreign aid. The gap between the economic theory of how people choose to save, and the empirical data on this same choice does not receive any attention in conventional growth theory textbooks. 

Do people save in a way which approximately corresponds to the assumptions of economic theory?  Rooij et. al. (2011) summarize evidence that the majority of workers are ignorant about the type of pension plans they have and what kind of retirement income they can expect. Furthermore, this ignorance impacts heavily on their future financial prospects; thus it is not the case that people magically make the optimal savings decisions without doing any financial planning. In fact, the literature reviewed shows that good financial planning leads to wealth building. This means that our standard growth models do not accurately reflect consumer decisions about savings for the majority of the work force.  

According to standard growth theory models currently in use in economic theory, savings behavior is a crucial determinant of growth. Actual observed savings behavior does not match what is assumed in economic theory. In particular, it is very far from maximizing utility, as surveys of large numbers of people who failed to plan adequately for retirement show. Observations provide the following insights and policy implications which are outside the bounds of conventional economic theory. 

Conflict between Current and Future Self: The economic model assumes that my interests are identical to those of my future self, and hence we can maximize a common utility function. Empirical research reveals that the opposite is true; Caplin (2003) states that ��conflicts of interest are close to universal�� and provides references documenting this finding. By saving today, I can enjoy a much better life during retirement. However, most people lack the self-control to curb or reduce today��s consumption in order to save for the future. Forced savings devices which lock money into accounts which cannot be accessed are very popular because they resolve the problem of self-control. When corporations introduce pension plans, the amount of discretionary personal saving should be roughly offset by the increased saving created by the pension plan. Contrary to this prediction of economic theory, Munell (1976) shows that personal saving is not offset. Similarly, Rooij et. al. (2011) provide evidence on the impact of financial planning on wealth accumulation. This variable is not present in economic theory because of the assumptions of identity (instead of conflict) of interest, as well as the ability to foresee and rationally plan for the future, none of which holds in real life. 

Intertemporal Borrowing: This idea which is assumed in economic theories is so far from reality that it is hardly worth discussing. If we can accurately forecast future income streams, then we can smooth consumption by costlessly borrowing from surplus periods, and spending more in low income periods. The borrowing and lending should be done at prevailing market interest rates. In reality, people cannot frictionlessly borrow and lend at market rates. Contrast these idealizations with the current realities of millions who lost their homes in the recent mortgage crisis; these consumers obviously did not forecast their income streams and expense streams accurately. The reality is that personal discount rates differ dramatically from market rates, and lead to behavior which is strange by economic theory standards. People do not have enough information and self-control to do the type of consumption smoothing that conventional economic models assume. There is a huge literature documenting these facts; convenient entry points to the literature which provide further references are Ameriks et. al. (2003) and Read (2004).

 

5.3 Intergenerational Justice.

 

The empirical evidence shows that we do not even treat our own self in the future very fairly or justly, depriving our future self of health and money so that we can enjoy our present self more. The problem is far greater when it comes to treating future generations equitably and justly.  

Older economists like Adam Smith used the concept of ��enlightened�� self-interest, and not the narrowly defined selfishness currently used to model human behavior by economists. Enlightened self-interest would look beyond current local concerns and note that being kind to the poor would create a caring society and also provide insurance for myself, should I need it. It would similarly encourage self-control in situations where acting on selfish impulses would bring temporary pleasure but cause damage to society – we all know that society can bite back. Many of the problems with conventional economic theories discussed earlier could be resolved by shifting the paradigm of consumer behavior from selfishness to enlightened self-interest. However, the problem of intergenerational justice cannot be solved in this way. Our actions will impact on the fates of the unborn generations, even to the extent of preventing them from coming into existence. Distant generations are not sufficiently concrete to be visualized and evoke sympathy. They are not present to negotiate with us, and no market exists for trading with them. Only certain conceptions of morality built into human beings can be invoked to deal with them justly and fairly.  

A recent volume of readings on Intergenerational Equity and Sustainability by Roemer and Suzumura (2007) illustrates very well the inadequacy of economic theory in dealing with real world issues. Of the 17 articles in the volume, Takayama (2007) is the only one who deals with a real world issue, namely the problem of funding retirement plans for the aging population of Japan. Among the many practical issues, we focus on three points which are made in this article: 

1. Mistakes in demographical calculations and levels of retirement benefits have led to a huge burden of taxes/transfers from the working class to the retired population.
2. There is substantial consensus about what are fair transfer payments from the young to the old, and consensus that the current burdens are not equitable. Thus there is a social problem created by older policies devised without sufficient information.
3. The problem with implementing solutions lies in political issues about how to share the burden of financing the retirement of the old among different classes of workers, with differing amounts of power.

 

The other sixteen papers deal with the mathematical calculations that an omniscient planner would make to ensure that the twin goals of equity and efficiency can be achieved within an overlapping generation model. These mathematical models start out by assuming that none of these three real world problems exist.

• The omniscient planner knows exactly the population growth patterns and the incomes to be earned from now to eternity. In the real world, it is the lack of this knowledge which has created the problem currently being faced in Japan.
• How to define intergenerational justice is one of the major issues which receives substantial importance in Roemer and Suzumara (2007). Different mathematical formulations and arguments are presented and evaluated. None of the authors considers what are the prevailing social norms for fairness  and intergenerational equity in the society under discussion. In a different context, behavioral economists Kahneman et. al. (1986a) used survey data to find out what people consider fair. These internalized rules play a very important role in proposals made for equitable retirements and political feasibility or otherwise of such proposals. Economic theory proceeds as if such social norms do not exist, and seeks to derive them mathematically from first principles and arguments about ethics, which makes it irrelevant to real world societies.
• There is no omniscient planner who is in control. So any practical plan for intergenerational justice must take into account the political environment. How is the burden of financing retirement being shared among different classes, with differing amount of political clout? Political institutions will be critical in determining which plans are and are not politically feasible. Economists have nothing to say about these issues.

 

Because economists focus on markets and marketable commodities, their discussion focus on money transfers between generations as an abstraction for all material transfers. However, many issues related to intergenerational justice cannot be measured in monetary terms. Can we evaluate the cost of species which become extinct, unique environments like coral reefs and rainforests which are destroyed and can never be re-created at any price? What about the many colourful cultures destroyed beyond the possibility of restoration by the inexorable processes of growth and development? The ruthless exploitation of Earth for current profits has led many to the realization that this cannot be sustained in the long run. What will our posterity inherit after our destruction of rain forests and the ozone layer, pollution of rivers and atmosphere, building up of greenhouse gases, depletion of coal and oil reserves, destruction of many, many species of plants and animals? The idea of ��sustainable development�� came into being to deal with these issues, which economists are singularly ill-equipped to handle.

 

Section 6: Views from the Business School

 
 

If anybody knows how consumers behave, it should be the folks in the Business schools. Their livelihoods depend on getting it right. It is interesting that there are large number of texts with ��Consumer Behavior/Theory�� in their titles which are taught in business schools. The contents of these texts have virtually no overlap with what is taught about the same subject in economics departments. Furthermore, it is often the case that students in one department know nothing about what is being taught in the other. In this section we go outside discipline boundaries to take a peek at what the folks in Business have to say about consumer behavior.  

Schiffman and Kanuk (2004, p 550) describe four models of consumer behavior, starting with ��An Economic View.�� They write that the economic model is ��often rejected as too idealistic and simplistic��; none of the three assumptions required for rational behavior is valid:

1. Consumers have knowledge of the range of options available
2. They are able to rank the alternatives.
3. They are able to identify the best one.

They further state that recent research rejects the economic assumption that customers are out to get the best value for their money which is the basis of the model of utility maximization under a budget constraint. For example, consumers haggle for prices for social reasons of ��achievement, affiliation, and dominance�� instead of obtaining good value for money. 

After flatly rejecting the economic model, Schiffman and Kanuk (2004) go on to formulate more realistic models of how consumers make decisions. A second model is the ��passive�� model, where the consumer decisions are entirely shaped by advertising and marketers. After rejecting this as well, the textbook develops a complex model of consumer behavior based on a cognitive/emotional framework, where decision making is broken down into three separate components. Taking these three dimensions of consumer decision making into account would substantially improve and enrich the microeconomic models we use in economics, as we discuss below.  

1. The input – Knowledge of the environment 

The ��input�� is the background environment in which the consumer makes choices. This complex environment replaces the idea that consumers have full knowledge of all available alternatives. The input is shaped by marketers�� efforts, as well as social and cultural influences, none of which play a role in economic theory. 

Taking this aspect into account would involve making a deeper study of the background assumptions which are used by consumers in making decisions. Advertising is a major determinant of consumer behavior. Social norms determined by what is and is not approved or admired in the relevant community is another major determinant. These interact to determine the framework and background knowledge which the consumer has in evaluating the range of products on the market and their attributes. Crucial questions which have not been sufficiently explored are: the extent to which advertising provides information, as well as the extent to which it shapes and molds tastes. If advertising works by creating desires, and making unnecessary object appear essential and attractive, then it creates disutility for large portions of the population who cannot have the object advertised. Any realistic theory of consumer behavior must take this into account. 

2. The process: Given knowledge of the alternatives and their attributes, how do consumers select a best one?

The ��process�� is what is used by consumers to arrive at a decision. This involves gathering information about what products are available (searching for knowledge), defining and recognizing ones needs (the utility function), and then evaluating and selecting a particular product (the maximization). All three parts of the process require effort on part of the consumer, and are mediated by social influences as well as complex motivations which drive consumption decisions. Experience is gained during this process which becomes the basis of future decisions. 

There is now a large literature on the heuristics that people use to make decisions. These differ dramatically from the process envisaged in microeconomic textbooks. Our homo economicus knows the entire range of alternatives available, and can clearly rank the choices and pick the best one. In the real world, information processing, evaluation of alternatives, and making decisions, all involve complexities not envisioned in economics. The heuristic literature provides guidance on human behavior in one of these dimensions. How this impacts on economic theory is discussed by Gigerenzer (2000). 

3. The Output: Evaluation of the results of decision making is done, and stored in memory for use in later decision making.  

The ��output�� is the end result of this decision making activity. Some product is selected, and then evaluated. But the evaluation, or the utility derived from the product, is not just the consumption of the product, but the entire experience involved in the purchase and consumption, contrary to economic theory. Ariely (2008) gives the example of how Starbucks sells coffee at five times the price of comparables because it markets the consumption experience and not just the coffee. 

Consumers evaluate the experience and record successes and failures. These are used as inputs in subsequent decisions. After sequences of successes, the consumers encode the methodology into routines which self-activate. This is of great benefit in reducing decision making costs, but it lead to habit formation, and a failure to respond adequately to changes in environments. There are many illustrations of how these types of issues impact in major ways on economies, instead of being inessential quirks of consumers behavior. Haucap and Heimeshoff (2011) find that consumers overestimate the savings resulting from their tariff choices in telecommunications. Barth and Graf (2011) provide an experimental evidence how consumers stick to calling plans (tariffs) of mobile phone networks that are not always cost-minimizing.  

The point here is that realistic models of consumer decision making differ in very significant ways from the economic ones. Of course, at some level, this is well known to everyone. The economists regard these as inessential complexities which distract from seeing the big picture. Stripping these away allows us to get to the heart of the matter, a simplified model which captures the essence of consumer behavior. In fact, as we have demonstrated through a wide range of examples, the simplifications of the economic models throw out the baby with the bath water.

 

Section 7: The Polanyi Critique: Land, Labor, Money

 

In his landmark study The Great Transformation, Karl Polanyi (1944) offers a deep critique of modern economic theory. This is based on the history of the emergence of economics as a discipline, and how this was affected by particular historical circumstances in England. It describes the great transformation that took place as markets, which were originally peripheral to society, became central institutions. It is not possible for us to treat these deep and complex issues in the present essay. We only take up one aspect of the Polanyi critique which impacts on consumer behavior in market societies.  

Polanyi defines a market society as one where an ��unregulated market�� is central to the functioning of the society. A good way to understand the concept of a market society is to contrast it with a polar opposite: a self-sufficient community, or a collection of nearly self-sufficient communities. Within such communities, production, acquisition, and distribution of goods could be organized in ways radically different from those of market societies. Trading across communities would be a peripheral function, not a central one, as it is in market societies.  

Polanyi makes the following three points about market societies relevant for our present purposes: 

1. Because of the global dominance of market societies, we have come to think of them as natural and inevitable ways of organizing economic affairs; it is hard for us to imagine alternatives. History teaches the opposite: other than our present civilization, none of the cultures/societies over the span of centuries have been market societies.
2. Market societies organize economic in ways which are in violent conflict with natural social tendencies of human beings. They come into being by replacing and suppressing (rather than adding to) other modes of organizing economic affairs. Markets come to dominate society rather than the other way around.
3. Market societies promote greed, indifference to poor and other characteristics universally condemned. It is the triumph of markets over society which is the answer to the question of Hirschman (1977): ��How did commercial, banking, and similar money-making pursuits become honorable at some point in the modern age after having stood condemned or despised as greed, love of lucre, and avarice for centuries past? ��

 

Polanyi says that there are three things (labor, land, money) which are not standard commodities like others traded in markets. A market society must force them to become marketable commodities like others, but these resist such a treatment. There is a tension between the requirements of market society, and certain natural tendencies of social behavior. These tensions lead to failure of economic theory to operate in these dimensions. We now document these failures.

 

7.1 The Labor Market

 

It is now a widely accepted social norm that human beings are not commodities and cannot be bought or sold in markets. Our children cannot be treated as our private property, even though there is substantial logical justification to allow this. Even more, we cannot even sell our own self voluntarily to another. The logic of the labor market is in conflict with this ethic. If I can sell 8 hours of my time, I can also sell 12 or even 24 hours of my time. This is a clear illustration of how market-based values conflict with social values as Polanyi argues. This conflict is clearly illustrated in many labor market phenomena which do not square with economic theories, as we show below.

 

7.1.1 Excess Supply of Labor Can Persist.

 

Large and persistent unemployment during the Great Depression showed to all that the forces of supply and demand fail to achieve equilibrium in the labor market. Keynesian economics was invented to show that government must intervene to bring about full employment, since market forces would not do so, contrary to classical economic theory. The problem Keynes struggled with was: why does not wage fall to eliminate unemployment, as predicted by supply and demand?  

Among competing explanations, one with strong empirical support is the following. People refuse to be treated as marketable commodities. Commodities are not insulted by price cuts and do not feel honoured if the price rises. On the other hand, wage cuts are seen as hostile actions, and people lower their efforts causing more damage to the firms than the savings in wages. A wage offer above equilibrium is considered as a gift, and workers return the favour by putting in more effort. For both of these reasons, labor markets do not work according to the standard supply and demand theory.  

A fascinating series of carefully controlled experiments described by Fehr and Gachter (2000) show that reciprocity provides a far better explanation of observed behavior than conventional economic theory.  Experimental employers offer higher than equilibrium wages, and experimental employees respond by putting in additional effort, even though it is costly to do so, and they would incur no loss from not doing so.  The lessons are striking: humans do not obey the laws of supply and demand, and modelling them as homo economicus who does obey these laws can prevent us from understanding major economic events like the Great Depression or the global financial crisis of 2007-8.

 

7.1.2 Social Norms Govern Labor Market

 

Several studies have established the following anomaly: laborers with identical skills earn significantly different wages in different industries⁴. According to standard economic theory, wages should equalize across industries. More laborers would go to higher paying industries, reducing wages to bring about equilibrium. Strong empirical support is available for the proposition that industries which are more profitable are expected to share their profits with the laborers. Unlike marketable commodities, human beings have conceptions of fairness, and try to achieve goals in accordance with these conceptions.  

Another anomaly reported by Card and Krueger (1997) is that lower minimum wage requirement do not  create unemployment, in conflict with the dicta of labor theory. Why? One possibility is the efficiency wage explanation. Another is the supply and demand analysis does not hold; this is the explanation Card gave in an interview.  
 

 

7.1.3 Man does not live by bread alone

 

Economists believe that we can get people to work more and/or harder by paying more. Experimental evidence shows that this is not true. Over a broad range of circumstances, money is not a strong motivational factor for labor. People respond much more strongly to being appreciated, trusted, challenged, as we would expect from human beings. In many cases non-monetary incentives perform better than monetary incentives in improving performance (Ariely, 2008). 

This experimental evidence is also very damaging to the principal agent literature. Explicit contracts which monitor employee effort and punish shirking and reward additional effort turn out to be inefficient and costly relative to the reciprocity mechanism. The solutions offered by principal agent theory are radically different from those used by practical managers who argue "that workers have so many opportunities to take advantage of employers that it is not wise to depend on coercion and financial incentives alone as motivators. ... Employers believe that other motivators are necessary, which are best thought of as having to do with generosity" (Bewley, 1995, p. 252)⁵.   

As the empirical literature shows, the wisdom of economic theory in form of the principal agent theory designed to solve problems of motivation, turns out to be folly in the field of actual behavior. Carefully designed contracts which monitor employee performance reward merit, and punish slacking, fail to have the desired effect because employees do not like the implied lack of trust. The idea of trusting employees to repay generosity and concern by loyalty and effort seems foolish from the point of view of economic theory. However, wise employers use precisely this strategy, and experiments confirm its superiority (see Fehr and Gachter, 2000). Gneezy, Meier and Rey-Biel (2011) collect empirical evidences that monetary incentives may not work to find solutions to principle-agent problems for desired outcomes, but crowd-out or promote just the opposite behaviors. Their finding is extrinsic incentives (monetary) may come into conflict other intrinsic motivations such as being ��good�� or ��trustworthy�� or ��useful��. Titmus (1970) for example, argue that paying people to donate blood would result in a reduction of people who wish to donate. Mellstörm and Johanesson (2008) support this argument in a field study. The theory of economic incentives should take the conflicts between external benefits and internal values in to account, as empirical evidences show that these values can be more important for humans. An incentive to work harder may signal the agent that the principal does not believe that he has the necessary intrinsic motivation to his best.    
 
 

 

7.2 The Market for Land

 

Is land a commodity for sale and purchase like others?  The phrase ��Mother Earth�� captures the natural social relations between humans and the land and water on the surface of the planet. We respect nature and tread lightly on the planet, and it nurtures us and provides us with all the resources necessary for human life on the planet. As Polanyi argues, these natural conceptions are in violent conflict with the commodification of land required by a market society. Below we focus on two such conflicts and their implications for economics and economic theory. 

1. Natural Resources are public property. Rivers, Forests, Mountains, Wildlife, Oil, Coal, Mineral, Botanical treasures, are the common heritage of mankind. No one has the right to privatize or exclude some from this heritage, and use it for personal profits.
2. We are not ��owners�� of the land but only stewards; we receive it from those who held it before us, and we hold it in trust for the generations to come.

 

Polanyi predicted accurately that treating land as a marketable commodity would lead to disaster: ��Nature would be reduced to its elements, neighborhoods and landscapes defiled, rivers polluted, military safety jeopardized, the power to produce food and raw materials destroyed.�� It is the reduction of land to private property that has allowed large scale exploitation, resulting in incalculable damage to the environment.  The main argument given in favor of capitalism is that it allows for rapid growth of wealth. Schoumaker (2004), Douthwaite (1999) and many others have argued that this growth is an illusion. If we take into account the costs paid in terms of damage to environment, community, society, the balance is negative. Taking into account the environment within which we operate, and the true social and environmental costs of consumer activities related to land and natural resources would have impact on both prices and budget constraints and substantially change the cost-benefit calculus of conventional economic theory. 

The second conflict with traditional consumer theory comes in the assumption of private property which is built so deeply into the micro models it does not receive any scrutiny. When we consider an abstract exchange economy, we find that each agent has a certain endowment, but our model does not contain natural resources and land. One is a common endowment for all agents, while the second is not private property in the same way that the other goods are. Varying levels of ownership which are crucial to society, and to human life on the planet, and to good economic policy, are entirely out of sight of the conventional economic theory models, and hence of economists.  Kogl (2005) provides historical details of how the notion of private property emerged in England, and how it replaced the previous ��commons,�� resulting in drastic reductions of consumption streams of the poor

 

7.3 Money: Marketing Trust

 

The third artificial commodity which market societies create is money. Polanyi (1944) argues that commodity monies like gold and silver cannot satisfy the needs of a market economy. Market economies continuously produce surplus and expand, and a fixed or narrow gold base would require deflation in prices, which is intolerable in a market economy. Hence the innovation of a paper currency comes into existence, which is flexible enough to expand with needs. Market expansion puts stress on the relation between the paper and the real assets (gold) it represents.  Unless paper is issued freely in response to needs of the market, without necessarily having sufficient backing, deflation will again result. This is the inexorable logic behind the eventual emergence of floating unbacked paper currencies currently in universal use. However, an unbacked paper currency serves none of the four traditional functions of money:

1. Unit of Account: The worth of the unit (paper money) itself is indeterminate, not being linked to any real asset.
2. Medium of Exchange: This holds true by social convention, or by government fiat. An individual market transaction is a highly unequal exchange of a piece of paper for a real good. This makes sense only because society has agreed to accept the paper universally.
3. Store of Value. Again paper itself does not carry value, and cannot store it. The value is stored in an invisible place – the level of trust placed by the society in the issuer of the currency.
4. Standard of Deferred Payments. If one could rely on stable value of money, this would hold. But over a hundred monetary crises in the last century testify otherwise.

 

To understand the nature of money, we must unlearn ways of thinking imposed upon us by needs of a market society. In social exchange, when someone takes something from another, this generates an implicit or explicit claim. Trust is the basis of social exchange, and people can rely on to fulfill commitments on the basis of social norms. Paper currency is an attempt to market this trust, and make it anonymous. Thus the paper is a generalized obligation of the society as a whole (or that of the government). This is the fundamental point of difference between heterodox theories of money and conventional ones. Many such theories are surveyed in Arestis & Sawyer (2006). After reviewing the difficulties of formal economic theories in explaining money��s existence, and assigning it an essential role, Ingham (1996) suggests that these can be resolved by conceptualizing it as a ��structure of social relations.�� 

Conventional economic theory models of consumers, firms, and financial markets are defective because they do not have a realistic theory of money. Consumers decide on their purchases by maximizing utility subject to a budget constraint in theory. This budget is given in nominal terms, and the value of goods is converted to a nominal value by multiplying by prices. Thus the budget constraint states that the (money) value of all the goods purchased must be less than or equal to the value of money income or wealth in my possession. How do we learn what is the value of money? According to conventional theory, money is just a unit of account, with no intrinsic value. Its real value is settled by dividing by a price index. However the prices are determined partly by the demand functions which come out of this utility maximization. Problems created by this circular logic for basic supply and demand theory of equilibrium price determination are highlighted in Zaman and Saglam (2010).  

In general, the value of commodities is determined by consumer preferences. However, money itself does not generate utility according to conventional models, which do not take into account the utility that misers derive from the hoarding of gold. Thus money is considered purely as a unit of account which has no real effects on the economy. In fact the value of money is crucially determined by trust, a variable that has only recently started receiving attention in economics. The level of trust can vary without any change in any of the variables considered relevant in conventional economic theory. The role of trust and its relation to monetary crises has been highlighted by Eichengreen (2004), who discusses two mechanisms relevant to crises. Government guarantees of bank deposits can prevent runs on the banks, since runs are cause by loss of trust. At the same time, government guarantees can cause crises if institutions take unnecessary risks, since they are secure from the possibility of loss.  

A proper understanding of money requires moving back from the theory of markets to social theories about how trust is created and maintained in society. A currency is just what it says on the face of the note: ��a promise to pay��. Promises are not marketable commodities like others. Analysis of how much promises are worth requires studying real human beings and societies and not homo economicus. 
 

 

Section 8: Happiness and Welfare

 

Ultimately, the subject of economics is (or ought to be) about how material resources can be used for the benefit of human beings. So the question of what constitutes welfare is (or ought to be) central to the study of economics. For nearly a century, a methodology based on the flawed philosophy of logical positivism has side-stepped these questions by assuming that: 

1. People know what is good for their own welfare, and this is the only valid criteria – we have no external ways to assess welfare.
2. People make choices in accordance with this knowledge.

 

Both of these assumptions are very easy to refute. Divorces reflect errors of judgment which inflict heavy personal and social costs, showing that one of the two decisions was taken without adequate knowledge. Interviews of people near end of life in hospices show widespread consensus that careers and materials were wrongly prized over social relations. Thus people acted in ways which conflicted with their long term welfare. Some empirical evidence against the economist��s views of welfare is collected below:

 

8.1 Bounded Self-Control

 

Thaler and Sunstein (2008) give examples of how people are vulnerable to temptation in their daily lives. One person may decide to exercise in the morning instead of watching a game on afternoon. However, in many cases, when the game starts, he may decide to watch the game. Preferences can be inconsistent in a dynamic time span. People may prefer A to B initially, but later they may choose B to A.  Such examples give solid evidence against the theoretical propositions of positive economics which regard people as always making the best choices. Similar problems arise in daily decisions on health (smoking vs. non-smoking, exercising vs. watching TV, and dieting), on financial welfare (savings vs. spending) and career (working vs. leisure), that the decisions are not for the best alternatives.  

Ariely (2008) presents an experiment to understand how the performances of students are affected when they are allowed to make their decisions. In this experiment students in three different classes are asked to complete 3 projects in one term. The students in the first class were dictated three deadlines for three projects. The second class were not given deadlines, and allowed to finish at the end of semester. However, they could turn the papers in early, but there was no grade benefit to doing so. In a way, they were given complete flexibility and freedom of choice. The students in the third class were allowed to fix deadlines for each project. However, once the deadlines were set by students, they were not allowed to change, and late papers would be penalized at the rate one percent off the grade for each day. The rational behavior for a student in this class would be to set a deadline at the end of the semester, since there was no benefit of handing the projects early, but punishment for handing them out after the fixed deadline. 

This experiment shows that self-control is not easy and people do not make choices that maximize their utilities, (grades in this case). The objective evaluation of the papers shows that the first class got the best grades, The second class with complete flexibility without any deadlines were the least successful. One interesting point of this study is that although the third class had the opportunity to choose the latest deadline, end of semester, most of them did not so, and scheduled their deadlines earlier. This behavior shows that people are willing to self-impose meaningful (i.e., costly) deadlines to overcome the problems of procrastination and they understand that inconsistent preferences can occur as the time evolves. Another point is that, given the opportunity to evaluate their future behaviors, people are too optimistic or over confident about the future that leads higher risk taking behavior as seen in this example that when self-imposing deadlines are set by individuals they do not set them optimally, for maximum utility. Thaler and Sunstein (2008) show numerous examples (drug addiction, saving, spending, education) on self-control problems. Due to the psychological facts that people can be short-sighted, or may have inertia and have unrealistic optimism, combined with the heuristics explained above, the choices of people can result in sub-optimal welfare levels.

 

8.2 The Easterlin Paradox

 

In a seminal paper of Easterlin (1974) he finds that within any one country, in cross sectional studies, there was a strong correlation between income and happiness. One would easily conclude that money can buy happiness. However, looking at a cross section of countries, one comes to a different conclusion. Figure 2 is taken from the Easterlin article: 
 
 

Figure 2. Personal happiness rating and GNP per capita (Easterlin, 1974).  

For 10 of the 14 countries surveyed, the happiness ranking is about the same, even though the income per capita changes by a factor of 30 from $140 to $2000. There are a few outliers, but these also confirm this finding. Cuba and USA have similar levels of happiness even though the GNP per capita is 6 times greater in the USA. Dominican Republic is extremely unhappy relative to India despite having twice as much GNP per capita.  

The finding of strong correlation between income and happiness disappears when comparisons are made across countries. Similarly, there is no correlation between happiness and income in the long run within a single country. For example, between 1962 and 1996, the real income of Japan quadrupled, but there was no change in the level of happiness. Similarly, Easterlin (2001) cites several studies which show that, despite tremendous increases in GNP per capita, the level of happiness in European and Latin American has remained virtually constant over decades.  

The startling implication of these empirical findings is that the stress being placed on economic growth is entirely misplaced. Growth has no clear relation to happiness. The profession of economics, as well as policy makers all over the world are directly threatened by these findings, which suggest radical changes in how to organize economic affairs. There have been three counterarguments made in the literature, which we now discuss: 

1. The first and most common one is to ignore these findings. Economists have objective measures of progress. People have refrigerators, cars, washing machines, and a tremendous number of consumer goods that they did not have before. Life expectancies have increased. We cannot ignore all this objective evidence just because people say that they don��t feel happier. Perhaps the surveys are not conducted properly. Perhaps feelings vary with moods and are too unstable and unreliable as an instrument to measure welfare. Easterlin (2010) has given a detailed examination of these critiques and shown that surveys give robust and reliable measures of happiness. To ignore these is equivalent to saying that economists as outsiders are better able to measure life satisfaction using objective criteria then the individuals themselves can feel and articulate- an extremely paternalistic attitude.
2. The second objection comes from failure to understand the nature of Easterlin��s paradox, which comes under the heading of ��fallacy of composition.�� It is true, and can be easily proven using a wide variety of evidence, that individuals do become happier if they are given more money. What is true for each individual is not true for the group as a whole. If we double the wealth of every single person in the society, then happiness levels will not change in the long run, because happiness is governed by relative income, and not absolute income.
3. The third objection is based on short term studies which do find correlation between income and happiness. Just like there is a strong correlation between the two in cross section studies, so there is a good correlation in short time periods. As income increases, everybody becomes happier in the short run, because it takes a while to realize that average income has increased. Easterlin (2010) says that over a period of ten years or more, these income changes become incorporated into our benchmarks for comparison and correlations between income and happiness decline to zero. He shows how some studies came to the wrong conclusion because they used short run data.

 

The implicit proposition of utility theory that the sole route to happiness is maximization of consumption contradicts with the empirical evidence: this proposition is true only in the short run. This short run validity creates a dangerous illusion of long run validity; understanding this has dramatic policy implications. If happiness is determined by relative comparisons, then one can achieve greater happiness by reducing inequalities, and also by reducing the standards of living for everyone. This will lower the benchmark and make it easier for everyone on the planet to be happy in comparison with this benchmark.

 

8.3 Other Routes to Happiness

 

If utility maximization is not the route to happiness, then what is? The seminal paper of Easterlin (1974) launched the field of ��Happiness Studies�� which has been engaged in researching this question over the past several decades. Many important insights have emerged which have an important bearing on consumer theory. We summarize some of the key findings below.

 

8.3.1 The Utility of Work

 

Standard economic theory treats work as having disutility, while leisure provides utility. As discussed earlier in the Polanyi critique, human lives are valuable, and when we invest a lot of time in an activity (like work) it acquires depth and meaning in our lives. There is a huge variation in types of work and the match between laborers and work, which means that wide ranging generalizations are not possible.

Summarizing numerous other authors who have written on the topic, Fagin and Little (1984) list seven major functions of work. Of these, we mention a few relevant to our current concerns:

(1) Work as a source of identity—the work people do classifies them in terms of class status and influence and established hierarchies and groupings from which people demand a sense of security, recognition, belonging and understanding. Work identity may be transferred to children or other family members but may be lost on retirement or unemployment.

(2) Work as a source of relationships outside the nuclear family. Work allows emotional outlet in family relationships as well as enriching the scope of interpersonal relationships which in turn has benefits on family life.

 (3) Work as an opportunity to develop skills and creativity—although work can satisfy sensual and aggressive instincts, as well as allay fears or anxieties, it can allow for the mastery, control or altering of the environment. There is considerable satisfaction in the integrating and co-ordinating of intellectual and motor functions which lead, over time, to the development of skills.

 (4) Work as a sense of purpose—at best work prevents classic signs of alienation such as feelings of powerlessness, self-estrangement, isolation and meaninglessness. While at best work ensures interdependence with others which helps in the development and achieving of life goals.

(5) Work is a source of income and control—work means putting oneself in the hands of employers during working hours so long as it provides sufficient money to assure oneself of independence and free choice of leisure and future outside the work place.  

There is strong evidence that both work and leisure provide us utility, although there is substantial variation in the levels provided.  Given that motives of work are diverse, as listed above, the neoclassical theory of the labor market fails to explain labor-leisure choice. If work can provide utility, we can increase levels of human happiness by providing better quality jobs, in ways that go outside dimensions considered in economic theory. In addition, failing to consider these aspects would lead economists to underestimate the effects of unemployment. Thus the recommendation to allow an industry to collapse because it is no longer competitive would not take into account the full cost of the transitional unemployment created.

 

8.3.2. Action, not consumption, leads to happiness

 

One of the most critical findings of several decades of research on happiness has been termed the ��Hedonic Treadmill�� (Brickman and Campbell, 1971).  One quickly gets accustomed to a given level of consumption – now called a ��set point��.  Then happiness results only from further increases beyond the setpoint. What this means is the increases in consumption only raise happiness temporarily. In the long run, such increases can be harmful because the increased setpoint means that you must work harder and consume more to achieve the same level of happiness that was previously available at a lower level.  

Since consumption only leads to short term and temporary increases in happiness, with negative long run consequences from habituation, researchers have focussed on what can raise long run happiness levels. Sheldon and Lubyamiroski (2006) document three studies showing that sustained increases in level of happiness can be brought about by ��practicing certain virtues, such as gratitude, forgiveness , and thoughtful self-reflection.�� This is opposed to changed circumstances, including increased consumption levels, which result in short term and temporary increases. In a later study, Boehm & Lubyamiroski (2010) documents several controlled experiments showing that ��committing acts of kindness, expressing gratitude or optimism, and savoring joyful life events, represent the most promising route to sustaining enhanced happiness.�� Still other studies shows what everyone knows, but economists have neglected – ��Life Goals Matter to Happiness�� is the title of one such study by Headey (2008). 

Exactly the opposite of the economist��s model, real world studies show that generosity and other virtuous actions create sustained happiness in the long run. Furthermore, the way that people behave is not genetically determined but depends a lot on the received education. Mitchell et. al. (2008) show the influence of college majors on attitudes towards careers and family. Ariely (2008) reports that in public good experiments, economics majors are prone to free-riding significantly more than others. By suggesting that selfishness is natural and rational, students are encouraged to develop these traits. 

 

8.4 Bounded Wisdom

 

Since Herbert Simon��s (1957) formulation of ��Bounded Rationality�� showed the limits of human capacity to reason, many additional failings of real human beings in comparison with homo economicus have been recorded. These can be categorized under ��Bounded Selfishness�� and ��Bounded Willpower�� – showing that people act for motives other than selfishness, and also that they cannot always implement plans which they know would be good for them. In this section, we would like to argue that people often do not know what is good for them.  

Economists assume, without discussion, that everyone automatically knows what is best for them. There is no reflection on the enormity of this assumption: how did people get to know what is best for them? It is universally agreed that children do not start out knowing what is good for them, and parents must train them to distinguish good from evil. Is it true that all parents automatically know what is right and wrong, and that their training is always effective? Since this is obviously not so, and a massive amount of evidence can be brought in to prove this, it follows immediately that some/many children never learn what is best for them. Therefore the choices they make do not reflect the choices they would make if they were wiser.  

Statistics show that about 93% of the road accidents are due to driver error, showing that people make fatal mistakes quite frequently. This is as clear a demonstration of non-maximizing choices as any. Studies show that most drivers are overconfident – more than 90% feel they are better than average. Experience actually feeds this overconfidence – a close escape is attributed to superior driving skill rather than luck.  

Thaler and Sunstein (2008) have argued for the ��Nudge Theory�� of libertarian paternalism – steering a course between the Scylla of paternalism, and the Charybdis of complete non-interference required by libertarianism. People often make choices without giving much thought to what is best. Therefore it is possible to push them in the right direction in a non-interfering way. For example, a current debate raging in the UK is about organ donations. Many people die for the want of a suitable donor. People who are dying must fill out a card indicating that they want their organs donated. Thaler and Sunstein (2008) suggest that we change the default option. It is assumed that everybody wants their organs donated unless they say otherwise. According to maximization theories, this would make no difference. People would automatically choose the option that they prefer. However, this is not the case in reality. There is strong empirical evidence that change of the default option would lead to substantially greater organs being available. Few people would go to the trouble of explicitly forbidding transplants, just as currently few people go to the trouble of explicitly allowing transplants.  

Thaler and Sunstein (2008) show that financial planning for retirement leads to substantially better outcomes. Credit Card Counselling shows that people get deep into debt without realizing it, and face severe consequences as a result. Patients near end of life in hospices show uniform regrets at choosing careers over families. 

All of this shows that planning and maximizing does not come naturally to people, as assumed by economists. Providing information about important choices in life, and especially providing information about experiences of those who have made these choices and their long term consequences can be extremely helpful to make good decisions. 

Life Goals are chosen, and then all things are evaluated with respect to these. Studies show that different types of goals are different with respect to their influence on welfare and happiness. How much consumption is done, and how it relates to life goals is crucial to building a correct theory of consumer behavior. Clearly consumer theory for the Amish community will be different from that for Manhattan. Ignoring this, as conventional microeconomics does, is to ignore the most important factors relating to consumer behavior. 

 

9. Conclusions

 

There can be no doubt that Aristotle was among the brightest men who have walked on this planet. Even though they are 2400 years old, his books continue to be studied and discussed at leading universities today. Yet he stated that ��heavier stones fall faster then lighter ones,�� and that ��women have fewer teeth then men�� without bothering to verify these theories by observations. Why, when this would have been so easy to do? The early methodology of science was axiomatic and deductive, borrowed from geometry. Starting from self-evident propositions, logic was used to deduce sound conclusions. Because they were logically valid, there was no need to check them empirically – we do not try to check whether or not two parallels to a straight line can be drawn through a given point. Observed regularities in nature were suspect because they could be accidental; so empirical observation was not a good path to knowledge. Giving primacy to observations and induction is the basis of the scientific method which revolutionized the world of knowledge. It is worth noting that the Greeks were right about induction – it cannot lead to certain knowledge, as has been forcefully stated by Popper (1959). But it turns out that this uncertainty is the price we have to pay for advances in knowledge. The axiomatic deductive framework is suitable for mathematics, but not well adapted to real world science. As Kuhn (1962) has argued in The Structure of Scientific Revolutions, spectacular scientific advances of the past few centuries can be traced to attempts to explain observations in conflict with widely accepted theories.  

The methodology of modern economics is firmly set in the mold described by Lionel Robbins (1935) ��The propositions of economic theory, like all scientific theory, are obviously deductions from a series of postulates. And the chief of these postulates are all assumptions involving in some way simple and indisputable facts of experience����  This is a precise description of the pre-scientific methodology of Greeks. Commitment to this methodology led to serious errors by very intelligent people. Today��s economists do not bother to check whether their theories are actually aligned with real world behavior. All conventional economic textbooks use the axiomatic method to derive theoretical results which are never cross checked against observations of the real world. The straitjacket of a wrong, pre-scientific, methodology can lead even the best minds to theories grossly in conflict with observations. Most of the current efforts at reform do not go far enough in challenging methodology. They seek to achieve conformity with observations while retaining existing economic methodology. We feel is that this is not sufficient. Radical methodological changes are required for progress. The world of economics awaits its Copernicus.

 

10. References

 

Akerlof, George A., and Janet L. Yellen. (1985). "A Near Rational Model of the Business Cycle with Wage and Price Inertia," Quarterly Journal of Economics, 100 (Supplement): 823-38. 

Akerlof, George A. (1979) "The Case against Conservative Macroeconomics: An Inaugural Lecture," Economica, vol. 46(183): 219-37. 

Akerlof, George A. (1982). "Labor Contracts as Partial Gift Exchange," The Quarterly Journal of Economics, vol. 97(4): 543-69. 

Aknin, Lara B.,  Christopher P. Barrington-Leigh, Elizabeth W. Dunn et al.  (2010). Prosocial Spending and Well-Being: Cross-Cultural Evidence for a Psychological Universal. NBER Working Paper 16415, downloadable from: http://www.nber.org/papers/w16415 

Allingham, Michael G. and Agnar Sandmo (1972). ��Income Tax Evasion: A Theoretical Analysis�� Journal of  Public Economics , vol.1: 323-338.     

Ameriks, John; Caplin, Andrew and John Leahy (2003). ��Wealth Accumulation and the Propensity to Plan,��  Quarterly Journal of Economics, vol. 118(3): 1007-1047. 

Arestis, Philip and Malcolm Sawyer, 2006, Handbook of Alternative Monetary Economics, Edward Elgar Publishers. 

Ariely, Dan (2008). Predictably Irrational: The Hidden Forces That Shape Our Decisions. New York: HarperCollins.  

Ariely, Dan, Loewenstein, George and Drazen Prelec (2003). ��"Coherent Arbitrariness": Stable Demand Curves without Stable Preferences��, The Quarterly Journal of Economics, vol. 118(1): 73-105  

Arrow, Kenneth (1973) ��The Theory of Discrimination��, in Discrimination in Labor Markets, ed. O.A. Ashenfelter and A. Rees, pp. 3-33. Princeton, NJ: Princeton University Press. 

Arrow, Kenneth. (1993). ��Contingent valuation of nonuse values: observations and questions��, in Contingent Valuation: a Critical Assessment, ed. Hausman, J. Amsterdam: Elsevier Science. 

Barberis, Nicholas and Richard H. Thaler (2003). "A survey of behavioral finance," in Handbook of the Economics of Finance, vol. 1. ed. G.M. Constantinides, M. Harris and R. M. Stulz, pp. 1053-1128, North Holland:  Elsevier. 

Baron, Jonathan (2004). ��Normative Models of Judgment and Decision Making,�� in Blackwell Handbook of Judgment and Decision Making, ed. Koehler, Derek J. and Nigel Harvey: p. 19-36, Boston: Blackwell Publishing. 
 
 

Barth, Anne-Kathrin and Julia Graf (2011). Irrationality Rings! Experimental Evidence on Mobile Tariff Choices. DICE Discussion Papers, No:36, D��sseldorf: Heinrich Heine University.  

Batson, C. Daniel (1992). ��Experimental Tests for the Existence of Altruism,�� PSA: Proceedings of the Biennial Meeting of the Philosophy of Science Association, vol. 2: 69-78 

Becker, Gary (1968). "Crime and Punishment: An Economic Approach". The Journal of Political Economy, vol.76(2): 169–217 . 

Bergstrom, Theodore; Blume, Laurence and Hal Varian (1986). ��On the Private Provision of Public Goods��,  Journal of Public Economics, 29: 25–59. 

Bewley, Truman F.  (1995). ��A  Depressed  Labor  Market  as  Explained  by  Participants,��  American Economic Review 85(2):250-54 

Bicchieri, Cristina (2006). The Grammar of Society: The Nature and Dynamics of Social Norms. Cambridge: Cambridge University Press. 

Boehm, J. K., & Lyubomirsky, S. (in press). ��The Promise of Sustainable Happiness,�� in Handbook of positive psychology (2^nd ed.). ed. S. J. Lopez, Oxford: Oxford University Press.

Online: http://www.faculty.ucr.edu/~sonja/papers/BLinpressa.pdf 

Bowles, Samuel and Herbert Gintis (2002). "Social Capital and Community Governance," Economic Journal, vol. 112(483): 419-436. 

Bowles, Samuel and Herbert Gintis (2006). "Evolutionary Origins of Collective Action," in Oxford Handbook of Political Economy ed.Wittman, Donald and Barry Weingast, Oxford: Oxford University Press. 

Brickman, Philip and Donald. T. Campbell (1971). ��Hedonic Relativism and Planning the Good Society,�� in Adaptation-Level Theory, ed. M. H. Appley, New York: Academic Press. 

Camerer, Colin F. (2003). Behavioral Game Theory : Experiments in Strategic Interaction,  Princeton: Princeton University Press. 

Camerer. Colin F. and Howard Kunreuther (1989) ��Decision Processes for Low Probability Events: Policy Implications,�� Journal of Policy Analysis and Management, vol. 8(4): 565-592 

Camerer, Colin F., Loewenstein, George and Matthew Rabin (2004). Advances in Behavioral Economics. Princeton: Princeton University Press. 

Campbell, Ernest Q. (1964). ��The Internalization of Moral Norms��, Sociometry, vol.27(4): 391- 412.

Caplin, Andrew (2003). ��Fear as a Policy Instrument��, in Time and Decision, ed.R. Baumeister, G. Loewenstein and D. Read, Russell Sage, New York. 

Card, David and Alan B. Krueger (1995).  Myth and Measurement: The New Economics of the Minimum Wage. Princeton: Princeton University Press.  

Carmon, Ziv and  Dan Ariely (2000). "Focusing on the forgone: How value can appear so different to buyers and sellers," Journal of Consumer Research, 27 (3): 360-370. 

Clayton, Phillip and Paul Davies (2006). The Re-Emergence of Emergence: The Emergentist Hypothesis from Science to Religion. New York: Oxford University Press. 

Cooter, Robert and Peter Rappoport (1984). ��Were the Ordinalists Wrong About Welfare Economics?�� Journal of Economic Literature, vol. 22(2): 507-530 

Dawes, Robyn M and Richard H. Thaler (1988). "Anomalies: Cooperation," Journal of Economic Perspectives, vol. 2(3): 187-97. 

Diamond, Peter (1996). ��Discussion of the conceptual underpinnings of the contingent valuation method by A. C. Fisher��, in The Contingent Valuation of Environmental Resources, Bjornstad, D. and Kahn, J. (eds.), Cheltenham: Edward Elgar. 

Douthwaite, Richard (1999). The Growth Illu$ion: How Economic Growth Has Enriched the Few, Impoverished the Many and Endangered the Planet.  Foxhole: Green Books Ltd.  

Easterlin, Richard .A. (1974). ��Does Economic Growth Improve the Human Lot? Some Empirical Evidence��, in Nations and Households in Economic Growth ed. P.A. David, W.R. Melvin, pp. 89–125 New York: Academic Press. 

Easterlin, Richard A. (2001). ��Income and Happiness: Towards a Unified Theory,�� The Economic Journal, 111(473): 465-484. 

Easterlin Richard A., McVey, Laura Angelescu and Malgorzata Switek, et al. (2010). The Happiness-Income Paradox Revisited. Proceedings of the National Academy of Sciences, vol. 107 (52), 22463-22468 

Eichengreen, Barry, J. (2004). Capital Flows and Crises. Mass: The MIT Press.  

Fagin, L. and Little, M. (1984). The forsaken families. Harmondsworth: Penguin, 1984. cited on p. 237 of ��Work and Leisure Satisfaction,�� by Adrian Furnham, Chapter 12 of Subjective Well-Being An Interdisciplinary Perspective, edited by Fritz Strack, Michael Argyle and Norbert Schwarz, Pergamon Press 1991. 

Fehr, Ernst and Simon Gächter (2000). ��Fairness and Retaliation: The Economics of Reciprocity,�� The Journal of Economic Perspectives, vol. 14(3): 159-181 

The Cost of Price Incentives: An Empirical Analysis of Motivation Crowding- Out

Bruno S. Frey and Felix Oberholzer-Gee

The American Economic Review

Vol. 87, No. 4 (Sep., 1997), pp. 746-755 

Friedman, Milton (1953).  "The Methodology of Positive Economics," in Essays in Positive Economics,  ed. Friedman, Milton, pp.  3-43. Chicago: Chicago University Press, 1953 

Friedman, Milton (1957). A Theory of the Consumption Function. Princeton: NBER. 

Gigerenzer, Gerd (2000). Adaptive thinking: Rationality in the real world. New York: Oxford University Press. 

Gigerenzer, Gerd and Peter M. Todd (1999). Simple heuristics that make us smart, New York, NY, US: Oxford University Press 

Griffin, Dale and Lyle Brenner (2004). ��Perspectives on Probability Judgment Calibration�� in Blackwell Handbook of Judgment and Decision Making, ed. Koehler, Derek J. and Nigel Harvey: 177-99, Boston: Blackwell Publishing. 

Gigerenzer, Gerd and Adrian Edwards (2003) ��Simple Tools for understanding risks: from innumeracy to insight,�� British Medical Journal v. 327, no. 7417 published 25^th Sep. 2003. 

Gneezy, Uri; Meier, Stephan and Pedro Rey-Biel (2011). ��When and Why Incentives (Don��t) Work to Modify Behavior��, Journal of Economic Perspectives, vol. 25(4): 191-210. 

Goeree, Jacob K. and Charles A. Holt (2001). "Ten Little Treasures of Game Theory and Ten Intuitive Contradictions," American Economic Review, vol. 91(5): 1402-1422. 

Haucap, Justus and Ulrich Heimeshoff (2011). "Consumer behavior towards on-net/off-net price differentiation," Telecommunications Policy, vol. 35(4): 325-332 

Hausman, Daniel M. and  Michael S. Mcpherson (1996). Economic Analysis, Moral Philosophy, and Public Policy. Cambridge Surveys of Economic Literature, Cambridge: Cambridge University Press.  

Headey, B. (2008). Life Goals Matter to Happiness: A Revision of Set-Point Theory. Social Indicators Research. Vol. 86. No. 2, pp. 213 

Henrich J, Boyd R, Bowles S, Camerer C, Gintis H, et al. (2001). ��In Search of Homo Economicus: Behavioral Experiments in 15 Small-Scale Societies�� American Economic Review, 91:73–79. 

Hirschman, Albert O. (1970). Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States. Cambridge, MA: Harvard University Press. 

Hirschman, Albert, O. (1977). The Passions and the Interests: Political Arguments for Capitalism before Its Triumph. Princeton: Princeton University Press. 

Ingham, Geffrey (1996) ��Money is a Social Relation,�� Review of Social Economy, vol 54, issue 4, p507-529 

Kahneman, Daniel; Knetsch, Jack L. and Richard H. Thaler (1990) Experimental Tests of the Endowment Effect and the Coase Theorem, The Journal of Political Economy, vol. 98(6): 1325-1348 

Kahneman, Daniel and Amos Tversky (1979). ��Prospect Theory: An Analysis of Decision under Risk,�� Econometrica vol. 47(2): 263-292  

Kahneman, D., Knetsch, Jack . L., and Richard Thaler (1986a). ��Fairness and the Assumptions of Economics�� Journal of Business, 59: 285–300.  

Kahneman, D., Knetsch, Jack . L., and Richard Thaler (1986b). ��Fairness as a Constraint on Profit Seeking Entitlements in the Market�� The American Economic Review, 76: 728–41. 

Kagel, John H. and Alvin Roth (1995). The Handbook of Experimental Economics, Princeton University Press. 

Keren, Gideon and Karl  H. Teigen (2004) ��Yet Another Look at the Heuristics and Biases Approach,�� in Blackwell Handbook of Judgment and Decision Making ed. Koehler, Derek J.  and Nigel Harvey, Malden, MA: Blackwell Publishing. 

Khan, Akhtar Hameed (1998) The Orangi Pilot Project, Reminiscences and Reflections, New York: Oxford University Press. 

Koehler, Derek J. and Nigel Harvey (2004). Blackwell Handbook of Judgment and Decision Making. Malden, MA: Blackwell Publishing. 

Kogl, Alexandra (2005). Enclosure And Exclusion: The Invention Of Private Property. Paper presented at the annual meeting of the The Midwest Political Science Association, Chicago, IL, April 7-10 2005. 

Kuhn, Thomas (1962). The Structure of Scientific Revolutions. Chicago: The University of Chicago Press. 

Mandler, Michael (2004). ��Status Quo Maintenance Reconsidered: Changing or Incomplete Preferences?�� Economic Journal, vol. 114(499): F518-F535. 

Mankiw, N. Gregory (1985) ��Small Menu Costs and Large Business Cycles: A Macroeconomic Model of Monopoly,�� The Quarterly Journal of Economics

Vol. 100, No. 2 (May, 1985), pp. 529-537 

Margolis, Howard (1982). Selfishness, Altruism, and Rationality. Cambridge: Cambridge Univ. Press 

Mart��n-Fern��ndez, Jesus; del Cura-Gonz��lez, Ma Isabel and Tom��s G��mez-Gasc��n et al. (2010). ��Differences between willingness to pay and willingness to accept for visits by a familyphysician: A contingent valuation study��BMC Public Health 2010, 10:236

http://www.biomedcentral.com/1471-2458/10/236 

Mazar, Nina, On Amir, and Dan Ariely (2008). "More Ways to Cheat - Expanding the Scope of Dishonesty",  Journal of Marketing Research, 45 (6): 651-653. 

Mellströrm, Carl, and Magnus Johaneson (2008). ��Crowding Out in Blood Donation: Was Titmuss Right?�� Journal of European Economic Association, vol.6(4): 845-63. 

Mitchell, Colter; Kimball,Miles; Thornton, Arland and Linda Young-DeMarco (2008). ��The Reciprocal Relationship between College Major and Values: Family, Careers and Society�� – download from http://www.princeton.edu/~cmitchel/ 

Mirowski, Phillips (1991). More Heat than Light: Economics as Social Physics, Physics as Nature's Economics, Cambridge: Cambridge University Press. 

Morgan, Charles J. (1985). ��Natural selection for altruism in structured populations,�� Ethology and Sociobiology, vol. 6(4): 211-18.  

Munnell, Alicia H. (1976). ��Private Pensions and Saving: New Evidence,�� Journal of Political Economy, 84(5): 1013–1032. 

Norton, Michael I. & Dan Ariely (2011). ��Building a Better America—One Wealth Quintile at a Time�� Perspectives on Psychological Science, vol. 6 no. 1: 9-12. 

Okun, Arthur (1981). Prices and Quantities: A Macroeconomic Analysis. Washington, DC: Brookings Institution. 

Oliner, Samuel. P. and Pearl M. Oliner (1988). The Altruistic Personality: Rescuers of Jews in Nazi Europe. New York: Free Press. 

Polanyi, Karl (1944). The Great Transformation. New York: Rinehart. 

Popper, Karl (1959). The Logic of Scientific Discovery. New York: Basic Books.  

Read, Daniel (2004) ��Intertemporal Choice�� Chapter 21 of Blackwell Handbook of Judgment and Decision Making, (eds) Koehler, Derek J.  and Nigel Harvey. 

Robbins, Lionel (1932). An essay on the nature and significance of economic science. London: Macmillan.  

Roberts, Russel D. (1984). A positive model of private charity and public transfers. Journal of Political Economy, vol. 92(1): 136–148. 

Roemer, John. and Kotaro Suzumura (2007). Intergerational Equity and Sustainability, Hampshire: Palgrave Macmillan. 

Rooij, Maarten van, Annamaria Lusardi, and Rob J. Alessie (2011). Financial Literacy, Retirement Planning, and Household Wealth. NBER Working Paper No. 17339, August 2011 

Sally, David (1995). ��Conversation and Cooperation in Social Dilemmas: A Meta-Analysis of

Experiments from 1958 to 1992��  Rationality and Society, 7: 58–92. 

Savage, Leonard J. (1954). The Foundations of Statistics. New York: John Wiley. 

Schiffman, Leon G. and Leslie L. Kanuk (2004). Consumer Behavior. 8th edition. India: Prentice Hall. 

Shefrin, Hersh M. and Thaler, Richard H. (1988). "The behavioral life-cycle hypothesis" (1988) Economic Inquiry, 26(4): 609-643. 

Schoemaker, Paul J. H. (2004) ��Forecasting and Scenario Planning: The Challenges of Uncertainty and Complexity,�� in Blackwell Handbook of Judgment and Decision Making Koehler, Derek J.  and Nigel Harvey (eds.), Malden, MA: Blackwell Publishing. 

Sen, Amartya K. (1977). Rational Fools: A Critique of the Behavioral Foundations of Economic Theory, Philosophy & Public Affairs, vol. 6(4): 317-344. 

Sheldon, Kennon. M., and Sonja Lyubomirsky (2006). ��Achieving sustainable gains in happiness: Change your actions, not your circumstances,�� Journal of Happiness Studies, vol. 7(1): 55-86. 

Shiller, Robert J. (2005). Irrational Exuberance. Princeton: Princeton University Press. 

Simon, Herbert A. (1957). Models of Man, Social and Rational: Mathematical Essays on Rational Human Behavior in a Social Setting. New York: John Wiley and Sons.   

Solow, Robert M. 1980. "On Theories of Unemployment," American Economic Review, vol. 70(1): 1-11.  

Solow, Robert, (July 20, 2010). ��Building a Science of Economics for the Real World�� Prepared Statement present to House Committee on Science and Technology Subcommittee on Investigations and Oversight

Download: http://science.house.gov/sites/republicans.science.house.gov/files/documents/hearings/072010_Solow.pdf 

Stiglitz, Joseph E. (1990). ��Peer Monitoring and Credit Markets�� World Bank Economic Review, vol. 4(3): 351-366.  

Takayama, Noriyoki (2007).  ��Social Security Pensions and Intergenerational Equity: The Japanese Case,�� in Intergenerational Equity and Sustainability, ed. Roemer, J. and Suzumura, K.,  pp.36-52, Hampshire: Palgrave Macmillan. 

Thaler, Richard H. (1980). "Toward a positive theory of consumer choice," Journal of Economic Behavior & Organization, vol. 1(1): 39-60.  

Thaler, Richard H. (1985). "Mental Accounting and Consumer Choice," Marketing Science, vol. 4(3): 199-214. 

Thaler, Richard H. (1994). Quasi Rational Economics. New York: Russel Sage Foundations 

Thaler, Richard H. and Eric J. Johnson (1990). "Gambling with the House Money and Trying to Break Even: The Effects of Prior Outcomes on Risky Choice," Management Science vol. 36(6): 643-660 

Thaler, Richard H. and Sunstein, Cass R. (2008). Nudge: Improving decisions about health, wealth, and happiness. New Haven:Yale University Press.   

Titmus, Richard M. (1970). The Gift Relationship. London: Allen and Unwin. 

Zajac, Edward E. (1978). Fairness or Efficiency: An introduction to public utility pricing . Cambridge, Mass: Ballinger Pub. Co. 

Zajac, Edward E. (1985) ��Perceived Economic Justice�� in: Cost Allocation Methods, Principles, Applications,  ed. Young, Peyton,  Amsterdam: Elsevier Science Publishers B.V. 

Zaman, Asad and Ismail Saglam (2010) The Conflict between General Equilibrium and the Marshallian Cross.  MPRA Paper No. 33256 Online at http://mpra.ub.uni-muenchen.de/33256/ 

Zaman, Asad and Syed Kanwar Abbas, "Efficiency Wage Hypothesis – the case of Pakistan.��  Pakistan Development Review, Vol 44  number 4, Winter 2005, 1051-1066

 

 

 

¹ We also need to assume that Temptation+Sucker<2High. If this inequality is violated, then the social optimum becomes one where two player take turns betraying each other. We do not discuss this case to avoid irrelevant distractions from the main themes.

²Russian roulette is a potentially lethal game of chance in which participants place a single round in a revolver, spin the cylinder, place the muzzle against their head and pull the trigger.

³Note that this phenomenon is not explained by differentiating between permanent and transient income, since the two types of income are used in different ways.

⁴ See Zaman and Abbas (2005) for evidence and further references.

⁵ The details of the actual experiments which confirm the broad picture painted above are too complex to describe here. The interested reader is referred to Fehr and Gachter (2000) and Camerer Loewenstein Rabin (2004)  for a guide to the literature.