NOTE ON BEHAVIORAL ECONOMICS Nick Schandler Much recent work in the area of behavioral economics has concentrated on so-called ‘‘behavioral anomalies.’’ These behavioral anomalies are seeming departures from the fundamental economic assumption of rational behavior. 1 They suggest that economics actors are not maximizing utility subject to the relevant constraints as standard theory predicts. Rather, ac- tors seem to be ‘‘leaving money on the table’’ when a costless alteration of behavior could result in superior outcomes. Some researchers have con- tented themselves with explaining such anomalies and declaring such actions ‘‘irrational.’’ Other researchers see opportunity: each anomaly presents us with the problem of asking why actors choose to behave in ways that seem to contradict standard behavioral assumptions. The former indict the actors; the latter his assumptions. Among the second group of actors are the many economists who have attempted to explain such anomalies through the concept of bounded rationality. Work on bounded rationality has focused on the actor’s limited cognitive abilities and knowledge to explain otherwise irrational behavior. Since there is a ‘‘cost’’ to thinking, actors may choose to act according to habits or ‘‘rules of thumb,’’ even though such rules may frequently result in sub-optimal outcomes. An attempt to process every bit of information available before acting would leave one frozen, unable to make even the most routine of decisions. Rather, individuals’ rationality is bounded by its Cognition and Economics Advances in Austrian Economics, Volume 9, 275–284 Copyright r 2007 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1529-2134/doi:10.1016/S1529-2134(06)09013-2 275 limited information processing capacity. We develop heuristics and habits that enable us to perform basic, everyday actions without attempting to fit these into some utility-maximizing framework. This idea of bounded rationality is often invoked to explain many of the so-called behavioral anomalies. According to this approach, we should not expect individuals to follow some utility-maximizing path through time, constantly re-adjusting to equate costs and benefits at the margin with every changing circumstance. Such behavior is simply too costly, too taxing on our cognitive resources. Vernon Smith (2003) has touched on a slightly different approach. In his 2002 Bank of Sweden Prize in Economic Sciences in Memory of Alfred Nobel, he drew a distinction between constructivist rationality and ecological rationality in economics. The constructivist approach views all social insti- tutions as the product of conscious, deductive reasoning starting from self- evident premises. By contrast, ecological rationality views rationality as a phenomenon that emerges out of cultural and biological evolutionary proc- esses. Human behavior comprises elements of both types of rationality. While many of our behaviors can be seen as constructively rational, such as when we manage our investment portfolio to achieve a specific risk–reward balance, the majority of our behavior is not so deliberately calculated. One is simply unable to calculate the ‘‘optimal’’ basket of groceries in a store carrying 30,000 items. These differing views of rationality have found relevance in today’s lit- erature on behavioral anomalies. Such anomalies have been analyzed largely through the lens of constructivism, and are, from this standard, both de- ficient and anomalous. But from an ecological perspective, such anomalies may be seen as fully rational responses to our limited knowledge concerning the relevant knowledge and effects of our actions. THE WINNER’S CURSE The ‘‘winner’s curse’’ (or, more precisely, failure to account for the winner’s curse) was one of the first behavioral ‘‘anomalies’’ to be discussed in the literature. The idea dates back to 1971, and was first applied to the bidding for oil drilling rights (See Capen, Clapp, & Campbell, 1971). The winner’s curse is the phenomenon of systematically upward-biased winning bids in an auction market. That is, the winning bid in an auction tends to be much higher than some objectively defined value of the good. 2 The basis of the anomaly is relatively simple. In an auction with a large number of buyers, NICK SCHANDLER276 each possessing imperfect information concerning the value of the auctioned good, there will be a spread of estimated values. If buyers possess rational expectations, we will expect roughly half (assuming a symmetric distribution of estimates) of the bidders to overestimate the value of the good, and roughly half to underestimate its true value. If buyers naively bid their estimated value of the go od, the winning bid will equal the most extremely over-valued estimate. Thus, the winning bid will not only be an overestimate of the good’s true value, but it will be the most extreme overestimate made by any bidder. Hence, while on average an individual’s bid may equal the actual value of the auctioned good, the winning bid will most likely be a severe overestimate of the good’s value. For this reason, bidders who naively bid their estimated value at an auction will tend to regret winning. The rational bidder will consciously take this into account, and adjust his bid downward to reflect the presence of the winner’s curse. For example, suppose that, given the number of buyers, the distribution of estimates, etc. the winning bid is expected to be made by a bidder who overestimates the true value of the good by three standard deviations. A rational bidder, then, will subtract three standard deviations from his estimate to arrive at his bid. His bid, conditional on it being accepted, is now equal to the expectation of the good’s true value. 3 The anomaly, then, is why this does not tend to happen. Field data has suggested that firms tend to systematically overpay for a wid e variety of assets, including other firms. 4 Experimental evidence has also fairly con- sistently replicated these results in a wide variety of contexts, even among experienced subjects who are given learning opportunities. The winner’s curse seems not to be accounted for by buyers, even by those who are given learning opportunities and have a monetary incentive to factor in the phenomenon. This is the heart of the anomaly. THE ENDOWMENT EFFECT, LOSS-AVERSION, AND STATUS QUO BIAS The endowment effect, loss-aversion, and the status quo bias are related anomalies, often hard to distinguish conceptually or in practice. The en- dowment effect can be defined as the observed pattern that people often demand much more to give up an object than they would be willing to pay to acquire it. Manifestations of the endowment effect can often also be seen as examples of the status quo bias, a preference for the current state of affairs, per se. Loss-aversion, or the practice of assigning a greater weight to Note on Behavioral Economics 277 the loss of an object than you ascribe to the acquisition of the same object, can often also be framed as an occurrence of either the endowment effect or status quo bias. Rather than three separate anomalies, the endowment effect, loss-aversion, and status quo bias can be seen as three manifestations of the same phenomenon. The experimental evidence documenting this behavior is well-established. Knetsch and Sinden (1984) provided one of the earliest laboratory demon- strations of the endowment effect. In this study, participants were endowed with either $2.00 or a lottery ticket. Each subject was offered to trade the lottery ticket for the money, or vice versa. Standard economic theory would predict that approximately 50% of participants would choose to switch their endowed good for the alternative. However, very few subjects chose to switch. Those who were given lottery tickets tended to prefer lottery tickets while those who were given cash tended to prefer the cash, even though the cash and the lottery tickets were assigned to participants arbitrarily. Kahneman, Knetsch, and Thaler (1990) found that the endowment effect survives when subjects face market discipline and have a chance to learn, though Coursey, Hovis, and Shulze (1987) had previously found that a market setting does diminish (but not eliminate) the extent of the bias. Some critics of these experiments have pointed out the existence of an income effect that could potentially account for these results. That is, participants who received a unit of good X will be richer, on average, than participants who did not receive good X. On the basis of this alone, we would expect those who received good X to have a higher reser vation price for this good than those who did not receive good X would be willing to pay, and thus we should not see a full 50% of units exchange hands. However, the value of the endowments are typically small ($10 or less), making it highly unlikely that any income effect could account for such a large deviance from predicted results. Further, experiments have been de- signed specifically to counter this criticism, with little or no change in ob- served behavior. 5 It appears that the income effect cannot account for such behavior. The status quo bias was picked up first by Samuelson and Zeckhauser (1988). In this experiment, participants were given a choice between differ- ent investment options. Researchers found that participants were more likely to choose a particu lar option if it was designated as the status quo. Experimenters divided participants into two groups. The first group was given the hypothetical option to invest in either a moderate-risk company, a high-risk company, treasury bills, or municipal bonds. The second group was given the same options. However, one choice was designated as the NICK SCHANDLER278 status quo. Specifically, the following phrase was added the passage: ‘‘A significant portion of this portfolio is invested in a moderate-risk com- pany y (The tax and broker commission consequences of any change are insignificant.).’’ Samuelson and Zeckhauser found that an option became significantly more popular if it was designated as the status quo. The same basic experimental design was replicated with different ques- tions, with no change in basic results. It should also be noted that the advantage of being designated the status quo increases with the number of alternatives. Critics have sometimes pointed out the lack of a monetary incentive for respondents to answer accurately. Such criticisms should be taken seriously, though they are beyond the scope of this paper. There are many examples, however, of behaviors that are most naturally explained using the status quo bias. The third leg of this tripartite anomaly is loss-aversion. A person expe- riences loss-aversion if losses loom larger than improvements or gains in a decision maker’s internal calculus. That is, ‘‘significant carriers of utility are not states of wealth or welfare, but changes relative to a neutral reference point’’ (Thaler, 1992, p. 70). This implies an abrupt change of slope, or kink, at the origin, or reference point. In Conflict and Cooperat ion: Institutional and Behavioral Economics (Blackwell Publishing, 2004), A. Allan Schmid attempts to integrate the study of institutional economics with recent work in behavioral economics. As Professor Schmid argues, ‘‘institutional economics is firmly rooted in the behavioral sciences and its theory is built on our best understanding of how the brain works’’ (Schmid, 2004, p. 19). That is, ‘‘to understand the impact of alternative institutions, it is necessary to g ather data reflecting the actual decision heuristics that people use’’ (p. 20). This is undoubtedly true. Ob- viously one cannot understand how institutions affect economic perfor- mance until we know the effect that a different environment has on an individual’s cognitive processes. Designating an option as the status quo may have little or no effect in a standard model, but in a model where the status quo bias is prev alent, it may radically alter the outcome. Further, these behavioral anomalies will alter the feedback that others receive, thereby altering the evolutionary path of the institutional environment itself. Clearly, institutional analysis cannot ignore the behavioral literature. Schmid goes on to explain the role of institutions in helping form heu- ristics, and in determining which heuristics are applied in any given situ- ation. Institutions determine when we employ a maximization calculus and when we apply rules of thumb, when we buck trends and when we go with the flow, when we stick to our diet an d when we have a second helping of Note on Behavioral Economics 279 dessert. Institutional analysis and behavioral economics are inextricably linked, existing side by side and simultaneously determining resul ts. It should come as no surprise, then, that psychological research has shown that context can be a powerful determinant of decision-making behavior. 6 Recent experimental studies have explored some of these context effects. 7 Schmid is surely correct to link institutional economics with the behavioral literature. Neither discipline has a monopoly on the truth, and we should take Professor Schmid’s insistence on the intricate relationship between the two disciplines seriously. While traditional behavioral literature has focused on the limited processing capacity of our brain as the reason for developing heuristics, we should also acknowledge Vernon Smith’s ecological rationality as a root cause of such heuristics. It is true that our brain is unable to comprehend all of the information we have available at our disposal, but it is also true that we do not have access to much of the information necessary to evaluate the consequences of a particular behavior. It is not simply that we engage rules to conserve on scarce brain-power; it is also that we cannot know all of the relevant information, or even what the relevant information is. Even if our brain were able to instantaneously calculate utility levels and spit out the ‘‘optimal’’ basket of 30,000 different goods in a grocery store, man would still rely on norms and rules of behavior to get by. Bounded rationality points out our limited processing skills. But the idea of ecological rationality goes further to stress the limited scope of our knowledge. Consider the winner’s curse described above. Though studies have shown that individuals fail to consciously account for the winner’s curse, buyer’s remorse does not seem widespread. EBay remains a popular medium for purchasing goods, with most of its customers being repeat purchasers. Returns and exchanges seem to be the exception rather than the rule. Consumers by and large remain happy with their purchases. How do buyers seem to avoid the winner’s curse even when the vast majority of individuals have no idea of even what the winner’s curse is, much less take conscious steps to correct for it? One way to view many of the behavioral rules that individuals follow is as ecologically rational responses to such problems. Consider the status quo bias, for example. This is typically described as an irrational desire to remain with the current state of affairs. The status quo is preferred for its own sake. While the status quo bias may indeed be costly to somebody who passes up a profitable opportunity by stubbornly clinging to the status quo, it also helps individuals avoid the winner’s curse. A person who would otherwise jump at all the ‘‘deals’’ that come his way now has something tugging at his NICK SCHANDLER280 sleeve, obstinately telling him to just stick with what has worked in the past. Such behavior helps him avoid ruin. At some level, this is acknowledged by most behavioral economists. For example, Thaler states, ‘‘following the rule ‘don’t accept an offer that looks too good to be true’ protects people from disaster (at the cost of passing up an occasional really good deal)’’ (Thaler, 1996, p. 229). Schmid (2004, p. 44) then adds, ‘‘Thaler uses this observation to destroy the argument that people must be rational to survive in the long run’’. In what way can such a rule be considered irrational? Only in the constructivist sense, the only sense in which Thaler and Schmid seem to use the word ‘‘rational.’’ To view the only form of rationality as rationality in the constructivist sense, however, is to misunderstand its nature and to deny its limitations. The endowment effect and loss-aversion similarly help individuals avoid ruin. Consider the flip side of the winner’s curse, what could be called ‘‘seller’s remorse.’’ Just as those buyers who most severely overestimate the value of a good will tend to win auctions and overpay (absent any constructivist or ecological response), those sellers who most severely un- derestimate the value of a good will tend to dispose of their goods at too low a cost. But such ‘‘seller’s remorse,’’ like ‘‘buyer’s remorse,’’ seems to be the ex- ception, and not the rule. How do sellers avoid disaster? Not through any constructivist adjustment to account for the phenomenon, but rather through exhibiting loss-aversion and the endowment effect. Sellers are loath to part with a good they know little about, even if they can’t explain why. Like the status quo bias with buyers, these behavioral rules prevent indi- viduals from making costly mistakes. Experimental evidence reinforces the beneficial functions these behavioral rules. Genesove and Mayer (2001) note that investors who don’t live in their condos exhibit less loss-aversion than owners who do. A field experiment by List (2003) found that amateur sports paraphernalia collectors who do not trade very often showed an endowment effect, but professional dealers and amateurs who trade a lot did not. Further, by revisiting the same traders a year later, List showed that it was trader experience that reduced endow- ment effects, rather than self-selection. This research indicates that such behavioral anomalies are used primarily by individuals when placed in an unfamiliar environment, where the likelihood of making costly buying or selling decisions is large. As individuals begin to become more famili ar with an environment, and thus less prone to exhibit the winner’s curse or seller’s remorse, they discard such rules, and apply a more standard maximization calculus. Note on Behavioral Economics 281 CONCLUSION Professor Schmid is surely correct to link institutional analysis and behavioral economics. And his focus on the limited processing capabilities of the human mind gives us one way to analyze the role of many institutions. But this is not the whole story. Institutions not only conserve on limited processing power, but in fact make use of much knowledge that remains elusive to any single actor. It is not only that we cannot process all the knowledge that we have access to, but also that there is a great deal of knowledge that we do not have access to, that we cannot retrieve, and that we do not even know exists. This is a major function of institutions, and one we would do well to recognize. NOTES 1. The term ‘‘rational’’ is here used in the more common neoclassical sense of maximizing expected utility subject to the constraints the actor faces. The term ‘‘rational’’ in economics is also sometimes used in a more general sense to mean ‘‘purposive action,’’ e.g. see Ludwig von Mises, Human Action, pp. 19–22. However, since even so-called behavioral anomalies are rational under this definition, we will eschew such usage for now and focus instead only on the more narrow neoclassical sense. 2. We must assume that the good being bid upon has some ‘‘objective’’ value that could be reasonably agreed upon by the bidders in the market. Thus, research in this area has tended to focus on business decisions, where non-monetary factors can be assumed to be negligible. Experiments that have auctioned off goods to individuals have tended to be goods whose utility is primarily or entirely based on the income they provide, e.g. jars of pennies. In this way, the ‘‘objective’’ value of the good can be reasonably approximated. Heiner (1985) has used this argument to conjecture that the results of such experiments need not generalize to all market settings. 3. In a thin market, a bidder may choose to reduce his estimated bid by more than his expected overestimate conditional on tendering the successful bid. Specifically, a bidder may attempt to maximize p(x)*u(x), where p(x) is the probability of tendering the winning bid, and u(x) is the expected utility of winning the auction. That is, he will want to increase the size of his ‘‘cushion’’ so that winning the auction is an expected utility-enhancing outcome; but this tendency will be tempered by the de- creasing probability of winning the auction as his ‘‘cushion’’ becomes larger. In a thick market, we may assume that competition among bidders forces the cushion to equal the estimated overestimate conditional on winning, eliminating this profit op- portunity and reducing the expected utility of winning the auction to 0. This assumes that bidders have perfect information concerning the size of the standard deviation of estimates, the shape of the distribution, etc. Merely assuming RE on the part of bidders here does not solve the problem, as now those bidders who tended to NICK SCHANDLER282 underestimate the standard deviation of estimates will tend to ‘‘under-cushion’’ their bids and thus still experience a meta-winner’s curse. Thus, consciously accounting for the winner’s curse is not as simple as it first appears – bidders must account for the winner’s curse in estimates, and the winner’s curse in estimates of estimates, etc. 4. Of course, the winner’s curse is not the only possible explanation of such behavior. Much work has focused on the non-monetary incentives of agents to expand their businesses beyond that size or scope which maximizes the value of the company. 5. For example, one such study is documented in Kahneman et al. (1990). 6. See Goldstein and Weber (1995); Loewenstein (2001). 7. See Cooper, Kagel, Wei, and Qilo (1999); Hoffman, McCabe, Shachat, and Smith (1994). REFERENCES Capen, E. C., Clapp, R. V., & Campbell, W. M. (1971). Competitive bidding in high-risk situations. Journal of Petroleum Technology, 23(June), 641–653. Cooper, D. J., Kagel, J. H., Wei, L., & Qilo, L. G. (1999). Gaming against managers in incentive systems: Experimental results with Chinese students and Chinese managers. American Economic Review, 89(4), 781–804. Coursey, D. L., Hovis, J. L., & Shulze, W. D. (1987). The disparity between willingness to accept and willingness to pay measures of value. The Quarterly Journal of Economics, 102, 679–690. Genesove, D., & Mayer, C. (2001). Loss aversion and seller behavior: Evidence from the housing market. Quarterly Journal of Economics, 116(4), 1233–1260. Goldstein, W. M., & Weber, E. U. (1995). Content and its discontents: The use of knowledge in decision making. In: J. R. Busemeyer, R. Hastie & D. L. Medin (Eds), Decision making from a cognitive perspective: The psychology of learning and motivation, Vol. 32. New York: Academic Press. Heiner, R. A. (1985). Experimental economics: Comment. American Economic Review, 75(March), 260–263. Hoffman, E., McCabe, K., Shachat, K., & Smith, V. L. (1994). Preferences, property rights, and anonymity in bargaining games. Games and Economic Behavior, 7, 346–380. Kahneman, D., Knetsch, J. L., & Thaler, R. H. (1990). Experimental tests of the endowment effect and the coase theorem. Journal of Political Economy, 98(December), 1325–1348. Knetsch, J. L., & Sinden, J. A. (1984). Willingness to pay and compensation demanded: Ex- perimental evidence of an unexpected disparity in measures of value. Quarterly Journal of Economics, 99, 507–521. List, J. A. (2003). Does market experience eliminate market anomalies? Quarterly Journal of Economics, 118(1), 41–71. Loewenstein, G. (2001). The creative destruction of decision research. Journal of Consumer Research, 28(3), 499–505. Samuelson, W., & Zeckhauser, R. (1988). Status quo bias in decision making. Journal of Risk and Uncertainty, 1, 7–59. Note on Behavioral Economics 283 Schmid, A. A. (2004). Conflict and cooperation: Institutional and behavioral economics. Bodmin, Cornwall: Blackwell Publishing. Smith, V. L. (2003). Constructivist and ecological rationality in economics. Central Bank of Sweden Prize in Economic Sciences in Memory of Alfred Nobel Lecture, pp. 1–94. Thaler, R. H. (1992). The winner’s curse: Paradoxes and anomalies of economic life. Princeton, NJ: Princeton University Press. Thaler, R. H. (1996). Doing economics without homo economicus. In: S. G. Medema & W. Samuels (Eds), Foundations of research in economics: How do economists do economics (pp. 227–237). Cheltenham: Edward Elgar. NICK SCHANDLER284 . decision making. Journal of Risk and Uncertainty, 1, 7–59. Note on Behavioral Economics 283 Schmid, A. A. (2004). Conflict and cooperation: Institutional and behavioral economics. Bodmin, Cornwall:. routine of decisions. Rather, individuals’ rationality is bounded by its Cognition and Economics Advances in Austrian Economics, Volume 9, 275–284 Copyright r 2007 by Elsevier Ltd. All rights. cash and the lottery tickets were assigned to participants arbitrarily. Kahneman, Knetsch, and Thaler (1990) found that the endowment effect survives when subjects face market discipline and have