Principles of (Behavioral) Economics

The Harvard community has made this article openly available. Please share how this access benefits you. Your story matters

Citation Laibson, David, and John A. List. 2015. “Principles of (Behavioral) Economics.” American Economic Review 105 (5) (May): 385–390. doi:10.1257/aer.p20151047.

Published Version doi:10.1257/aer.p20151047

Citable link http://nrs.harvard.edu/urn-3:HUL.InstRepos:30805504

Terms of Use This article was downloaded from Harvard University’s DASH repository, and is made available under the terms and conditions applicable to Other Posted Material, as set forth at http:// nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of- use#LAA American Economic Review: Papers & Proceedings 2015, 105(5): 385–390 http://dx.doi.org/10.1257/aer.p20151047

BEHAVIORAL ECONOMICS IN THE CLASSROOM

Principles of (Behavioral) Economics†

By David Laibson and John A. List*

There are many great ways to incorporate Our choice of content for a behavioral lecture behavioral economics in a first-year under- is motivated by three factors. First, we include graduate economics class—i.e., the course that ideas that are conceptually important. Second, is typically called “Principles of Economics.” we include material that is practically import- Our preferred approach integrates behavioral ant and personally relevant to our students—we economics throughout the course e.g., see have found that such content resonates long after Acemoglu, Laibson, and List 2015 .( With the the course ends. Third, we include content that integrated approach, behavioral content) plays relates to what has been or will be taught in the a role in many of the chapters of the principles rest of the course, and therefore( serves) as a com- of economics curriculum, including chapters on plement. We want students to see that behavioral optimization, equilibrium, game theory, inter- economics is an integrated part of economics, not temporal choice, probability and risk, social a freak show that is isolated from “the standard preferences, household finance, the labor mar- ingredients” in the rest of the economics course. ket, financial intermediation, monetary policy, This paper summarizes our approach to such economic fluctuations, and financial crises. a focused behavioral lecture. In Section I, we We prefer the integrated approach because it define behavioral economics and place it in his- enables the behavioral insights to show up where torical context. In Section II, we introduce six they are conceptually most relevant. By illustra- modular principles that can be used to teach tion, it is best to combine a discussion of down- behavioral economics. We provide PowerPoint ward nominal wage rigidity i.e., the idea that notes on our home pages, which instructors workers strongly resist nominal( wage declines should feel free to edit and use. with the overall discussion of the labor market.) Whether or not an instructor integrates behav- I. Behavioral Economics Defined ioral economics throughout the principles of economics course, it makes sense to pull cen- Behavioral economics uses variants of tradi- tral materials together and dedicate a lecture tional economic assumptions often with a psy- or more to a focused discussion of behavioral chological motivation to explain( and predict economics.( ) This note describes our approach to behavior, and to provide) policy prescriptions. such a lecture, emphasizing six key principles of behavioral economics. When we teach our students this definition of behavioral economics, we like to emphasize that behavioral economics is a series of amend- * Laibson: Department of Economics, Harvard ments to, not a rejection of, traditional econom- University, Cambridge, MA 02138 and NBER e-mail: ( ics. We illustrate the complementarities between [email protected]); List: University of Chicago, 1126 traditional and behavioral economics with an E. 59th Street, Chicago, IL 60637 and NBER e-mail: [email protected] . We thank Saurabh Bhargava, (Brigitte example: if you want to get from Chicago to the Madrian, and Ted O’Donoghue) for helpful suggestions and bleachers of Fenway Park to watch the Boston feedback. We are also grateful to Daron Acemoglu, who Red Sox, standard economics will get you to directly contributed, as our Economics coauthor, to many of Cambridge, or even Boston University which is the pedagogical approaches discussed here. ( † adjacent to Fenway , but you may need behav- Go to http://dx.doi.org/10.1257/aer.p20151047 to visit ) the article page for additional materials and author disclo- ioral economics to take the final steps and find sure statement s . your seat in the bleachers. ( ) 385 386 AEA PAPERS AND PROCEEDINGS MAY 2015

In this way, behavioral economics augments PRINCIPLE 1: People try to choose the best fea- standard economic analysis. Behavioral eco- sible option, but they sometimes don’t succeed. nomics adopts and refines the three core prin- In other words, people try to make the optimal ciples of economics: optimization, equilibrium, choice—they are optimizers—but they some- and empiricism Acemoglu, Laibson, and List times make mistakes. It’s important to empha- 2015 . Both traditional( and behavioral econo- size that these mistakes are partially predictable. mists) believe that i people try to choose their One of the key explanatory factors is experi- best feasible option( ) optimization ; ii people ence and training: experienced decision-makers try to choose their best( feasible )option( ) when tend to make better choices than inexperienced interacting with others equilibrium ; and iii decision-makers. models need to be tested (with data empiricism) ( ). To illustrate these ideas, we use a range of In the next section we provide some( examples) examples. If students play the p-beauty contest of how behavioral economics refines economic game twice, they will see behavior converging analysis. toward the Nash equilibrium. The game is sim- From a historical perspective, the big bang ple enough to be played in class or on the web for behavioral economics was a paper on pref- before class . But even if you don’t( actually play erences over gambles written by two psycholo- the game in )class, you’ll be able to show the stu- gists, Daniel Kahneman and Amos Tversky, in dents easy to understand data e.g., Nagel 1995 1979. So modern behavioral economics is a lot that illustrates this convergence.( ) younger than the rest of the field of economics. If you prefer to teach the first principle using However, behavioral concepts have always field data, you could explain that credit card played a part in economic analysis though users pay fewer and fewer fees—for instance, they didn’t always have that headline (name . late payment fees—the more experience they As Ashraf, Camerer, and Loewenstein 2005) have with their card Agarwal et al. 2013 . point out, Adam Smith frequently wrote( about) Likewise, consumers switch( telephone plans,) the psychology of decision-making, including moving toward the best one, as they gain experi- the tension between a person’s “passions” and ence Miravete 2003 . their rational deliberations, which Smith refers These( examples all) illustrate that “everyone to as the “impartial spectator.” The impar- choosing optimally” is a better prediction for tial spectator is the source of “self-denial, of experienced decision-makers than for inexperi- self-government, of that command of the pas- enced decision-makers. sions which subjects all the movements of our The first principle should also be used to nature to what our own dignity and honour, explain why learning economics is so useful and the propriety of our own conduct require” to students. Economics courses have the tan- Smith 1759 1984 , I, i, v, 23 . Psychological gible benefit of increasing the optimality of assumptions( are[ as ]old as economics) itself. the students’ own decisions. We tell our stu- dents that “learning economics turns you into a ­decision-maker who is more likely to choose the II. Six Principles of Behavioral Economics best feasible alternative. By taking economics, you become a more skilled optimizer.” These principles are modular, so instructors can pick whatever subset matches their interests PRINCIPLE 2: People care in part about how and their time budget. In our experience, all six their circumstances compare( to reference) points. principles can be covered in a 1.5 hour lecture, For example, a reference point could be the but that is not what we recommend. If you wish amount of money a person expected to earn to cover all six principles, we suggest allotting during summer break, or the amount of money two lectures. that she started with when she entered a casino, or After each principle we present a series of the price she paid for 100 shares of Apple stock, or examples that illustrate and explain the princi- the price she paid for her home. It matters whether ple and engage first-year economics students. a person is losing or gaining relative to their refer- We’ve included more examples than you will ence point. Losses get far more weight than gains, probably be able to use, so we encourage you to which is called loss aversion Kahneman and pick among them. Tversky 1979 . In practice, ­people( suffer from a ) VOL. 105 NO. 5 PRINCIPLES OF (BEHAVIORAL) ECONOMICS 387 loss about twice as much as they benefit from a Fun evidence-based examples include post- gain of equal absolute magnitude. poning planned work tasks Augenblick, These phenomena have implications for Niederle, and Sprenger 2014 , placing( savings market transactions. Loss aversion discourages in a lockbox Ashraf, Karlan,) and Yin 2006; trade, since each trade generates two losses and Beshears et al.( 2013 , workplace productivity two gains the buyer has a loss and a gain and commitments Kaur, Kremer,) and Mullainathan the seller has( a loss and a gain , and the losses forthcoming , (and committing to not smoke are weighted more than the gains.) Accordingly, cigarettes or) drink alcohol Giné, Karlan, and people are prone to hold on to their endowments Zinman 2010; Schilbach 2015( . Controlling for Thaler 1980 . time of day, Read and van Leeuwen) 1998 show ( There are many) ways to illustrate this endow- that snacks chosen in advance are (overwhelm) - ment effect. For example, give half of your stu- ingly healthy, but snacks chosen for immediate dents a mug and half of your students a big consumption are not. chocolate bar, randomizing this endowment( by) switching every other seat in the classroom. Let PRINCIPLE 4: Although we mostly care about the students examine their own and their neigh- our own material payoffs, we also care about the bors’ endowments, and then ask the class who actions, intentions, and payoffs of others, even wants to trade with you for the good that they people outside our family. didn’t receive. Fewer than a quarter of the stu- These “social preferences” come in many sys- dents will take up this offer, but traditional eco- tematic forms, especially negative reciprocity, nomic theory predicts that half of them should behindness aversion, and social pressure. Kahneman, Knetsch, and Thaler 1990; Tversky One way to teach these ideas is to play the and( Kahneman 1991 . ultimatum game Güth, Schmittberger, and If you wish to go deeper,) show your students Schwarze 1982 . An( anonymous sender and that market experience reduces the endowment an anonymous recipient) are paired. The sender effect e.g., List 2003 . Or show them how fram- divides an endowment of $10 any division is ing manipulations( that) exploit loss aversion can allowed, rounded off to the nearest( penny . The be used to incent workers to be more produc- recipient either accepts or rejects the division.) tive e.g., Hossain and List 2012 . You could In the event of rejection, both players go home also (show your class loss aversion) in gambles: empty-handed. Most senders propose a division people won’t take an even odds gamble unless in which the recipient receives at least $2.00, the upside has twice the reward as the downside because the senders correctly anticipate that half Kahneman and Tversky 1979 . of the recipients will retaliate against an offer ( ) that is less generous than this even though the PRINCIPLE 3: People have self-control retaliation hurts the recipient . ( problems. Such social preferences )respond to incen- In a traditional economic model there is no tives, just like all other economic decisions. As gap between a person’s good intentions and the stakes get large, the recipient becomes more their actions. By contrast, in the model of pres- and more willing to accept unfair offers. For ent bias, people plan to work hard or diet, or example, Andersen et al. 2011 find that when exercise, or quit smoking, or save for (retirement, the pot to be divided is nearly( )a year’s wages, or stop borrowing on their credit card, etc. and almost no recipients reject a 20 percent offer then renege at the last second Laibson 1997;) from the sender. Showing students that prices O’Donoghue and Rabin 1999 . ( matter in the domain of social preferences helps Instructors can show how )the present-biased them develop a deeper understanding of both discount function 1, ½, ½, ½, … leads to pref- social preferences and the traditional model. erence reversals if{ studying has }an immediate effort cost of 6 and a delayed benefit of 8. In this PRINCIPLE 5: Sometimes market exchange case, studying tomorrow looks good in the eyes makes psychological factors cease to matter, of the student because ½ 6 8 1 0, but many psychological factors matter even in but immediate studying ×does[− +not ] =because> markets. 6 ½ 8 2 0 . In this simple( exam- If investors with behavioral biases are a small ple,− +studying × never= − takes< ) place. part of the total stock market, their beliefs will 388 AEA PAPERS AND PROCEEDINGS MAY 2015 not drive stock prices because ­perfectly rational are often surprised because their forecasts prove traders will sell the stocks that the biased inves- to be overly optimistic in war and peace , and tors are buying, keeping stock prices near their their preferred policies( work less well) than “rational level.” However, if biased investors anticipated. compose a large portion of the total asset market In our classes, we show students examples and marginal traders , their beliefs will matter. of paternalism that are generally thought of as ( The dot-com bubble,) which peaked in 2000, successful e.g., Social Security and Medicare illustrates this point. Dot-com fever swept the as well as paternalism( that has been unpopular) stock market and investors scooped up shares in e.g., soda bans, soda taxes, taxes on fatty foods companies that had anything to do with technol- or( disastrous e.g., alcohol prohibition . ) ogy and especially the Internet. Near the peak We also challenge( students with a policy) ques- of the bubble, some subsidiaries with a tech- tion, such as the socially optimal level of ciga- nology focus had market capitalizations that rette taxes. State taxes for cigarettes range from greatly exceeded the market capitalizations of a low of $0.17 pack Missouri to a high of their parent companies, a violation of basic arbi- $4.35 pack New/ York (, reflecting) widely dispa- trage Lamont and Thaler 2003 . For example, rate public/ views( on their) merits. Traditional eco- in early( 2000, Palm, a manufacturer) of personal nomic theory implies that cigarette smoking be organizers, was 95 percent owned by 3Com, but lightly taxed or even, subsidized, since early mor- Palm was worth much more than 3Com based tality leads to some cost savings for the govern- on the stock prices of the two companies. ment i.e., smokers tend to die at the end of their The US housing bubble, which peaked in working( lives and miss a long, socially expensive 2006, is another example of a behavioral phe- retirement, which partially offsets other negative nomenon that had a profound impact on mar- externalities . If cigarettes generate only modest kets. When this housing bubble burst, the world negative net) externalities, why are they taxed so economy sustained a long and deep recession heavily? In New York City, the combination of and many of the world’s biggest banks failed. federal, state, and local cigarette taxes sum to $6.86 per pack. Behavioral economists explain PRINCIPLE 6: In theory, limiting people’s these taxes as a way of helping people quit smok- choices could partially protect them from their ing Gruber and Ko˝szegi 2001 . Critics say that behavioral biases, but in practice, heavy-handed these( taxes are regressive and) unfair. Who is paternalism has a mixed track record and is right? Students love to debate this issue. often unpopular. In the last decade, behavioral policy recom- Behavioral insights imply that if the gov- mendations have tilted toward nudges, which ernment is well intentioned and sophisticated, recommend or facilitate certain behaviors with- paternalistic policies might be helpful. However, out removing options or the freedom to choose heavy-handed paternalism raises new problems. Thaler and Sunstein 2008 . The leading exam- First, some government actors are self-serving, ple( is automatic enrollment) in 401 k savings so giving them expanded powers of paternal- plans Madrian and Shea 2001 . (Behavioral) ism may not make life better for the rest of us. economists( like such interventions) because they Second, government actors are prone to the same are scalable, inexpensive, highly successful in kinds of mistakes that everyone else makes—for changing behavior, and also freedom-preserving. example, overconfidence. With considerations Ask your students to consider other policy like these in mind, behavioral economists are questions. For example, is obesity a problem interested in carefully expanding the scope of that the government should try to “solve” with paternalistic policies, but skeptical about open- nudges or other types of paternalism like sugar ing the floodgates. taxes , or is obesity a reflection of personal( pref- To illustrate the tendency for governments erences) over diet and exercise with little or no to make mistakes, consider the extremely opti- role for government intervention? mistic forecasts held by both the Allies and the Central Powers at the beginning of WWI III. Conclusions Johnson 2004 . Both sides confidently believed that( they would) win in a few months, but the war Since the publication of “Prospect Theory” actually took more than four years. Governments in 1979, behavioral economics has become an VOL. 105 NO. 5 PRINCIPLES OF (BEHAVIORAL) ECONOMICS 389 important and integrated component of mod- Augenblick, Ned, Muriel Niederle, and Charles ern economic thought. In our view, behavioral Sprenger. 2014. “Working Over Time: ideas are not a fifth column, but rather a key Dynamic Inconsistency in Real Effort Tasks.” contributor to the arsenal of modern economics. http://web.stanford.edu/~niederle/ANS.Paper. Behavioral economists embrace the core prin- pdf accessed October 27, 2014 . ciples of modern economics—optimization and Beshears,( John, James J. Choi, Christopher) Har- equilibrium—and wish to develop and refine ris, David Laibson, Brigitte C. Madrian, and those ideas to make them more empirically Jung Sakong. 2013. “Self-Control and Liquid- accurate. Behavioral economists study how peo- ity: How to Design a Commitment Contract.” ple try to pick the best feasible option, includ- http://faculty.som.yale.edu/jameschoi/ ing the cases in which people, despite their best commitment.pdf accessed November 19, efforts, make mistakes. We believe that behav- 2014 . ( ioral ideas should be integrated throughout the Giné, Xavier,) Dean Karlan, and Jonathan Zin- first-year undergraduate sequence. man. 2010. “Put Your Money Where Your Butt It also makes sense to pull some key materi- Is: A Commitment Contract for Smoking Ces- als together and commit a lecture or more to a sation.” American Economic Journal: Applied focused discussion of behavioral concepts.( ) This Economics 2 4 : 213–35. note explains how we would give this lecture, Gruber, Jonathan,( )and Botond Ko˝szegi. 2001. “Is emphasizing six key modular principles. Addiction ‘Rational’? Theory and Evidence.” If you want to boil behavioral economics Quarterly Journal of Economics 116 4 : down for a classroom summary you might say 1261–1303. ( ) that most people are located somewhere between Güth, Werner, Rolf Schmittberger, and Bernd Mr. Spock and Mr. Simpson aka Homer . Like Schwarze. 1982. “An Experimental Analysis of Mr. Spock, Mr. Simpson is also( an optimizer—) Ultimatum Bargaining.” Journal of Economic he tries to choose the best feasible option. He’s Behavior and Organization 3 4 : 367–88. just not good at it. We need to study and model Hossain, Tanjim, and John A. List.( ) 2012. “The all optimizers: the good, the bad, and those in Behavioralist Visits the Factory: Increasing between. Productivity Using Simple Framing Manip- ulations.” Management Science 58 12 : 2151–67. ( ) REFERENCES Johnson, Dominic D. P. 2004. Overconfidence and War: The Havoc and Glory of Positive Illu- Acemoglu, Daron, David Laibson, and John A. sions. Cambridge: Harvard University Press. List. 2015. Economics. Upper Saddle River, Kahneman, Daniel, Jack L. Knetsch, and Richard NJ: Prentice Hall. H. Thaler. 1990. “Experimental Tests of the Agarwal, Sumit, John C. Driscoll, Xavier Gab- Endowment Effect and the Coase Theorem.” aix, and David Laibson. 2013. “Learning in the Journal of Political Economy 98 6 : 1325–48. Credit Card Market.” http://scholar.harvard. Kahneman, Daniel, and Amos Tversky.( ) 1979. edu/files/laibson/files/learning_credit_042413. “Prospect Theory: An Analysis of Decision pdf accessed November 5, 2014 . under Risk.” Econometrica 47 2 : 263–91. Andersen,( Steffen, Seda Ertaç, Uri Gneezy,) Moshe Kaur, Supreet, Michael Kremer,( )and Sendhil Hoffman, and John A. List. 2011. “Stakes Mat- Mullainathan. Forthcoming. “Self-Control at ter in Ultimatum Games.” American Economic Work.” Journal of Political Economy. Review 101 7 : 3427–39. Laibson, David. 1997. “Golden Eggs and Hyper- Ashraf, Nava, Colin( ) F. Camerer, and George Loe- bolic Discounting.” Quarterly Journal of Eco- wenstein. 2005. “Adam Smith, Behavioral nomics 112 2 : 443–77. Economist.” Journal of Economic Perspectives Lamont, Owen,( )and Richard H. Thaler. 2003. 19 3 : 131–45. “Can the Market Add and Subtract? Mispric- Ashraf,( Nava,) Dean Karlan, and Wesley Yin. 2006. ing in Tech Stock Carve-outs.” Journal of “Tying Odysseus to the Mast: Evidence from Political Economy 111 2 : 227–68. a Commitment Savings Product in the Philip- List, John A. 2003. “Does( ) Market Experience pines.” Quarterly Journal of Economics 121 Eliminate Market Anomalies?” Quarterly 2 : 635–72. Journal of Economics 118 1 : 41–71. ( ) ( ) 390 AEA PAPERS AND PROCEEDINGS MAY 2015

Madrian, Brigitte C., and Dennis F. Shea. 2001. Schilbach, Frank. 2015. “Alcohol and Self- “The Power of Suggestion: Inertia in 401 k Control: A Field Experiment in India.” http:// Participation and Savings Behavior.” Quar( )- scholar.harvard.edu/files/schilbach/files/jmp_ terly Journal of Economics 116 4 : 1149– frank_schilbach.pdf accessed December 1, 87. ( ) 2014 . ( Miravete, Eugenio J. 2003. “Choosing the Wrong Smith, Adam.) 1759. The Theory of Moral Senti- Calling Plan? Ignorance and Learning.” Amer- ments. Edited by D. D. Raphael and A. L. Mac- ican Economic Review 93 1 : 297–310. fie. Indianapolis: Liberty Fund, 1984. Nagel, Rosemarie. 1995. “Unraveling( ) in Guessing Thaler, Richard H. 1980. “Toward a Positive The- Games: An Experimental Study.” American ory of Consumer Choice.” Journal of Economic Economic Review 85 5 : 1313–26. Behavior and Organization 1 1 : 39–60. O’Donoghue, Ted, and (Matthew) Rabin. 1999. Thaler, Richard H., and Cass R.( Sunstein.) 2008. “Doing It Now or Later.” American Economic Nudge: Improving Decisions About Health, Review 89 1 : 103–24. Wealth, and Happiness. New Haven: Yale Uni- Read, Daniel,( and) Barbara van Leeuwen. 1998. versity Press. “Predicting Hunger: The Effects of Appetite Tversky, Amos, and Daniel Kahneman. 1991. and Delay on Choice.” Organizational Behav- “Loss Aversion in Riskless Choice: A ior and Human Decision Processes 76 2 : ­Reference-Dependent Model.” Quarterly 189–205. ( ) Journal of Economics 106 4 : 1039–61. ( )