Misbehavior in Behavioral Economics
and the
Unhappy Thing about Happiness Economics(1)

Robin Neill
Economics University of Prince Edward Island
October 1, 2010

The Two-Headed Dragon
Two relatively new branches of Economics have come to the fore in the aftermath of the economic slump of 2008: Behavioral Economics and Happiness Economics. They are quite different in the way they reach into the literature of Economics, but they reach out from a common intellectual root – rejection of a basic principle of standard economic theory. No doubt, the experience of 2008 calls for a revision of our understanding of how our economic system works, but both of these emerging contentions are mistaken in the way they would make the revision. Namely, they accuse late twentieth century neoconservative economists of going too far in rejecting mid-century Keynesian economics. The rational-markets-know-best neoconservatives, so Behavioralists say, rudely brushed aside a Keynesian government-knows-best preconception. But, on the rebound, Behavioralists and Happiness Economists have rushed from the market-knows-best back to the government-knows-best preconception. That markets and governments know best when they work together seems not to be a stopping place in this intellectual sloshing back and forth.

Behavioral Economics rejects the idea that homo economicus, humankind going about the ordinary business of life, is rational. Indeed, defining homo economicus to be objectively rational, they deny the existence of homo economicus. Behavioral Economics asserts as a principle that people are not objectively correct in their assessment of the consequences of their actions. They do not objectively get the most from their economic decisions. That is, the decisions they make in the market place do not make them as happy as they could be, and/or do not lead to the greatest possible happiness in the society to which they belong. Accordingly, when decisions are left to individuals, happiness, a near synonym for economy-wide efficiency in traditional economics, is not achieved. In the Behaviorialist view, the state should step in to more advantageously direct individuals’ choices(2).

If workers are given the choice of taking sickness insurance or a larger take-home pay, most will chose the latter. If they are given the sickness insurance as part of their pay package and then allowed to choose a larger take-home pay instead, most will simply stay with the insurance. This would seem to be evidence of irrationality, because the workers make different choices on the basis of how the choice is presented to them, and not on the basis of which choice would be best for them.

Now, most of us would judge that structuring choice to favor purchasing an insurance policy, is a good thing(3). The telling questions are, however, "What general principle of state intervention in individual choice is imported with this seemingly good thing?", and, "Are maximum happiness and a maximum annual output measured by price in a market system the same thing?"

The purveyors of Happiness Economics assert that the monetary measure of success is not the same thing as happiness. That is to say, even if humankind did rationally achieve the greatest possible material welfare (meaning the largest possible Gross National Product), it would not be as well off – as “happy” -- as it could be, either individually or as a society. In their view the state should support some non-monetary kinds of development that do maximize happiness – say, longevity, or average level of education. With this approach placing government expenditures into parks and meditation salons, rather than industrial development, could be seen as the better choice(4).

Misbehavior in Behavioral Economics
The principle of Behavioral Economics is “Homo Sapiens Is a Sap”.

More precisely, Behavioral Economics starts from the postulate that human kind is “not rational”, meaning it is swayed one way or another by preconceptions, excessive fear of loss, ignorance, preference for the immediate because it is up front, sloth, and, indeed, by an army of emotions at war with reason. As we marvel at the behavior of our fellow humans, if not of ourselves, we tend to agree. But is the problem want of reason? Is it irrationality? Immanuel Kant, forerunner of the postmodern theory of how we think, thought not. He wrote,

General logic is again either pure or applied. In the former, we abstract all the
empirical conditions under which the understanding is exercised; for example,
the influence of the senses, the play of the fantasy or imagination, the laws
of the memory, the force of habit, of inclination, etc. consequently also, the
sources of prejudice -- in a word, we abstract all causes from which particular
cognitions arise .... (Kant, 1881, p. 46).

According to Kant, then, all those unfortunate things that the Behavioralists attribute to human choice do not pertain to rationality, but to “understanding”. With the term “understanding” Kant pointed to the basis of human judgement. So, there is rationality and there is judgement; and the problems of human choice arise with respect to judgement. That humankind going about the ordinary business of life is rational, that homo economicus is rational, is simply a matter of fact, and certainly economists in the past have recognized this. Consider the following from Paul Samuelson, an early American Nobel laureate in Economics

The utility analysis rests on the fundamental assumption that the individual
confronted with given prices and confined to a given total expenditure selects
that combination of goods which is highest on his preference scale. This does
not require (a) that the individual behave rationally in any other sense; (b) that
he be deliberate and self-conscious in his purchasing; or (c) that there exist
any intensive magnitude which he feels or consults (Samuelson, 1966, pp. 97-98).

Often nothing more is stated than the conclusion that people behave as they
behave, a theorem which has no empirical implications ... (Samuelson, 1966, pp 91-92).

Where then does this leave Behavioral Economics? It has confused bad judgement, and, indeed, good judgement based on values, with irrationality. If it were honest with itself it would assert as its principle, “Most individuals make bad judgments, and other individuals, those with commanding positions in the state bureaucracy, make good judgments”.

Most of us are not in favor of handing over our judgment to the politicians and civil servants of Prince Edward Island, or of Canada. Why then would we approve of structuring workers choices to induce them to chose health insurance? Not because we are concerned about their bad judgment with respect to their happiness, but because we are aware that their ill health will have effects on the rest of society, perhaps through contagion. The ill-health of some will impinge upon others; so we think all with ill-health should be treated, and we think that treatment should be paid for by all who benefit;and this is the case with compulsory health insurance. In short, what seems to be a good outcome from Behavioralist transference of choice from one set of putatively irrational individuals to another set of putatively rational individuals, is not that at all. Rather it is a matter of preventing the effects of the rational though judgmentally flawed choices of one set of individuals from encroaching on the rational though judgmentally flawed choices of others. This is a different principle. In no way does it state or imply that homo economicus is irrational. One does not have to be a Behavioralist to favor regulation of economic activity; because it is one thing to tell people what they should do, and another to tell them that they should not encroach on what others choose to do.

Crudely put, the Behavioralist position is, “You have bad judgement and we have good judgement, so you should do as we say.” I don’t think so.

“The Unhappy Thing about Happiness Economics”
In its recent construction, Happiness Economics begins with the Easterlin Paradox. In the 1970s Richard Easterlin, an economist, asked, “Does economic growth improve the human lot?” (Easterlin, 1974). More generically, he began with the question, “Does the level of price-measured material goods produced annually in an economy determine the happiness of individuals in that economy? According to the purveyors of the Happiness Economics, he answered, “No”. To assess this assertion we need to know what Easterlin meant by the term “happiness”, and we need to know how, in his view, it could be measured.

Following a long discussion of the nature of happiness, Easterlin concluded that it did not need to be defined. The term, even when translated into other languages and cultures, rested on virtually the same supports – about half on economic matters, the rest on health, family tranquility, personal values realization, and the like. So, he concluded, because it was virtually the same everywhere, for purposes of measurement the definition could be left to subjective judgement. Happiness was, well, happiness. Recapitulating a number of intra- and inter-national point-of-time cross-sectional studies, he concluded that size of annual material production (GDP) was positively correlated with happiness. He then turned to data more or less supporting an answer to the specific question, “Does an intra-national increase in average income over time increase the happiness of the average citizen?” To answer the question, following standard practice in inter-temporal analysis, he examined answers to the more pointed question, “On a scale of from less to more happy, how have Americans ranked themselves at different points in time?" With some short period deviance, the patently subjective survey results showed that respondents who placed themselves in a particular rank in early years when GDP was relatively low placed themselves in the same rank in later years when GDP had increased.

This did not lead Easterlin to the conclusion that economic growth did not correlate with increased happiness. Rather, admitting the weakness of data that had not been gathered for the purpose to which he put it, he concluded that the operative supports on which happiness was based changed as GDP rose. That is to say, as GDP rose the standard of happiness rose, so that the relationships between GDP and happiness in different years were incommensurable. Clearly, the Easterlin Paradox was not a conclusion of Easterlin’s paper. Indeed, in all of his comments on cross-sectional data he concluded that there was a correlation, perhaps sometimes weak, between GDP and happiness. The Easterlin Paradox rested on some United States data to which Easterlin, himself, gave an interpretation incompatible with what has come to be called the Easterlin Paradox(5).

All of this having been said, it remains true that GDP per capita is not intended to be a measure of happiness, whatever that may be. It is a measure of effort, of economic power. GDP is higher in Canada because we have to, and are able to, clear away snow in the winter. Accordingly, on grounds other than those of the proponents of Behavioral Economics, alternatives to the quantity of annual marketed output have been suggested. Amarta Sen, for example, has proposed a “Capabilities Approach” (Sen, 1985, 1999), for which he was awarded a Nobel Prize. Happiness, he said, depends on such things as political freedom, social opportunities, and personal security with respect to property and life. As a measure of happiness rather than economic power, this would seem to be an improvement over the quantity of material goods. As Easterlin pointed out (Easterlin, 1974, passim.), however, such a measure does not escape problems of subjectivity and incomparability over time. It would be an ordinal, not cardinal measure that would take no account of a shift in the scale of ranking. It would still involve asking people how they feel and that would not produce an objective measure, especially if, as the Behavioralists say, people have bad judgement. Further again, it raises a question, “How is a quantity measure to be given to (say) a level of freedom?

Despite these difficulties, with encouragement from the Behavioralists, an intellectual industry has grown around the development of a possible, more happiness revealing, measure of personal and national well-being (Anand et.al, 2009). Admittedly, if the social engineers are to succeed in their quest to make us all happier, an objective measure of happiness will be required. But what if one is found? What if some mechanical device, chemical procedure, or even some statistical social barometer is invented by which individual and group happiness can be determined. A happiness meter will then be constructed by means of which social engineers can determine if people are objectively happy whether or not their subjective feelings tell them they are.

Putting aside broad consideration of a dystopia in which a public agent would be charged with telling individuals what they should do to be happy, think of the implications of social engineers coming to a definition of “objective happiness”. Without subjective input, happiness would be the same for all. But, of course, from a subjective point of view happiness producing states vary from person to person, to say nothing of from culture to culture. In a socially engineered state of objective happiness some would feel more happy than others, and some would not be happy at all. It would be a grand case of what economists call externalities. What generates happiness for one spills over into unhappiness for another. The tendency for “kindred spirit” splinter groups to break off from the whole of society would produce smaller and smaller groups until everyone was his or her own social engineer and there would be no need for an objective measure of happiness.

There is room for a term indicating “happiness”, meaning something beyond an efficient allocation of resources as determined by market price; and indeed, another term has been suggested, “ophelimity”. Ophelimity has been defined in a number of ways (Mclure, 2009), and one mathematically bent nineteenth century economist foreshadowed our would-be social engineers by suggesting the need for a “hedonimeter” to objectively measure Ophelimity(6).

Niceties in nuanced expressions of various states of the human soul are not to the main point with respect to the entailments of a putative objective measure of happiness. The point is use of an objective measure of happiness entails the existence of a totalitarian regime with a high tech enforcer of officially happiness generating behavior.

Conclusion
“In these troubled times”, when the admitted failures of the capitalist market system have captured the headlines of newspapers and become a prominent subject of discussion on the internet, the case presented by Behavioralists and Hedonimists is seductive. The capitalist market system has its failures. Market-knows-best Neoconservatives went too far in rejecting the implicit centralizing tendencies flowing from a Keynesian presumption that the expectations of economic agents are “irrational”. Granted, the expectations of economic agents do exhibit a seeming irrationality. Their seeming irrationality, however, is rooted in flawed judgment, not flawed reason, and their judgement is not completely off. Clearly, there are less than perfect market structures. There are blocks to economic mobility, legal and informational. Short run benefits for a few do lead to long run costs for many. At times people make bad choices. There are negative externalities. The solution for these failures is to forbid activities that damage third parties. That is, to regulate in such a way as to prevent harm. Any attempt to solve these problems by positively directing people to do what the Behavioralists and the Hedonimists would have them do, would kill initiative, energy, and invention, and certainly would not eliminate bad judgement on the part of the directing bureaucracy. Social chaos accompanying the inevitable failure of such a regime would entail greater unhappiness than the unhappiness that its proponents intend to eliminate.

Notes

1. The June 8, 2010, issue of the Charlottetown Guardian contained an editorial in which Gwynne Dyer alleged that research into “happiness” was a useless activity undertaken by underemployed sociologists and economists looking for something to write about; because, Dyer wrote, no one knows what “happiness” is. That may be true, but there is more to the rash of publications on happiness than appears in Dyer’s account. Herman Daly (“Towards some Operational Principles of Sustainable Development”, Ecological Economics, vol. 2, pp1-6, 1990) has suggested that economic development need not require growth of GDP. For example, growth may follow from the advent of something in the field of information that entails a decline in the need for material resources. Evidently, powerful environmental activists are on board with the purveyors of “happiness” who think GDP is not what policy should be pursuing. Further, a prominent Behavioralist, who also thinks that GDP is an inadequate measure of good economic outcomes, has achieved considerable influence in United States economic policy. Austan Goolsbee, Professor in the University of Chicago’s Booth School of Business, and a leading Behavioralist, is Chair of the President’s Council of Economic Advisors, and a member of the President’s Economic Recovery Board. Professor Goolsbee was Obama’s economic advisor during his run up to the Presidency. Indeed, when asked, the President declared himself to be a Behavioralist in Economics.

2. Writing in the early 1950s, Milton Friedman accepted objective rationality as a useful assumption in economic research. He also denied the need for assumptions to correspond to reality in economic research. Herbert Simon then accused traditional economics of assuming that economic agents were objectively rational, which, he said, and Friedman agree, they were not. John Muth responded to Simon with the assertion that in fact economic agents were objectively rational (Donald Mackenzie, An Engine, Not a Camera, p. 39). Thereafter "rational expectations" became entrenched as objective truth in the cutting edge of policy-related economics, and the notion of rational behavior found in traditional economics came to be ignored.

3. We make this judgement even though dependence on various insurance schemes reduced the North American saving rate to near zero, and this is an important factor underlying the United States’ unsustainable reliance on a near 40% saving rate in China.

4. There are descriptions of and lists of examples of the literatures of Behavioral and Happiness Economics on Google. What you will not find on Google is a critical analysis of either. I begin with “misbehavior in Behavioral Economics”, and then take up the “unhappy thing about Happiness Economics”. The second of these phrases is borrowed from Helen Johns and Paul Omerod (2008).

5. Subsequently, in a study in which the relationship between subjective happiness and GDP over time was measured for the specific purpose of determining what that relationship was, Stevenson and Wolfers (2008) found that happiness has risen with rising GDP. I should add here that on July 14, 2010, George Loewenstein and Peter Ubel, the one at Carnegie Mellon University and the other at Duke, published their finding that behavioral based policies are much less effective than tax and subsidy policies based on standard economics.

6. For a more fundamental and technical treatment of this matter consult Collander (2007) in which a virtual debate between Francis Edgeworth and Irving Fisher is recounted and related to current Behavioralist assertions. For an even more complete, technical, and indeed tedious discussion consult Stigler (1950). The main channel of discussion of the merits of the concept of utility, ophelimity, or happiness (choice of descriptor depending on one’s point of view) begins with an assertion that it is not a basis for scientific investigation because it is not measurable. It ends in frustration with an assertion that it has to be measurable because without an objective measure social welfare cannot be objectively maximized. Perhaps the most readable and enlightening treatment of non-price-based economics can be found in No Wealth But Life edited by Roger Backhouse and Tamotsu Nishizawa.

References

Anand, Paul et al.,
2009, “The Development of Capability Indicators”, Journal of Human Development
and Capabilities, vol. 10, pp. 125-152.

Backhouse, Roger J., and Nishizawa, Tamotsu,
2010, No Wealth But Life, Cambridge University Press, Cambridge Mass.

Camerer, Colin F., Loenstein, George, and Rabin, Matthew,
2004, Advances in Behavioral Economics, Princeton University Press, Princeton.

Collander, David,
2007, “Retrospective: Edgeworth’s Hedonimeter and the Quest to Measure Utility”,
Journal of Economic Perspectives, vol. 21, pp. 215-226.

Daly, Herman E. and Kenneth N. Townsend,
1993, Valuing the Earth: Economic, Ecology, Ethics, MIT Press, Cambridge.

Easterlin, Richard A.
1974, “Does Economic Growth Improve the Human Lot? Some Empirical
Evidence” in Paul A. David and Melvin W. Reder, National Households
in Economic Growth
, Academic Press, New York, pp. 89-125.

Johns, Helen, and Omerod, Paul,
2008, “The unhappy thing about happiness economics”, Real World Economic
Economics Review
, no. 46, pp. 139-146.

Kant, Immanuel,
1881 [1871] Critique of Pure Reason, Macmillan, London.

Lowenstein, George and Peter Ubel,
2010, “Economics Behaving Badly”, New York Times, July 14, 2010.
http://www.nytimes.com/2010/07/15/opinion/15loewenstein.html
Accessed 7/22/2010, 7:59am.

Mackenzie, Donald,
2006, An Engine, Not a Camera, MIT Press, Cambridge, Mass.

Mclure, Michael,
2009, “Pareto and Pigou on Ophelimity , Utility, and Welfare: Implications for
Public Finance”, University of Western Australia, Discussion Paper 09.13.

Samuelson, P.A.,
1966, Foundations of Economic Analysis,, Harvard University Press, Cambridge Ma.

Sen, Amartya K.,
1985, Commodities and Capabilities, Oxford University Press, Oxford.
1999, Development and Freedom, Knopf, New York.

Stevenson, Betsey and Justin Wolfers,
2008, “Economic Growth and Subjective Well-Being: Reassessing the Easterlin Paradox”,
Brookings Papers on Economic Activity.

Stigler, George,
1950, “The Development of Utility Theory” (2 parts) Journal of Political Economy,
Part 1, Vol. 58, 307-327; Part 2, pp. 373-396.