Behavioral Economics Then and Now

“Although it has never been clear whether the consumer needs to be protected from his own folly or from the rapaciousness of those who feed on him, consumer protection is a topic of intense current interest in the courts, in the legislatures, and in the law schools." So write James J. White and Frank W. Munger Jr. in a 1971 article from the Michigan Law Review.

Today, it is not uncommon for behavioral economists to weigh in on financial regulatory policy and consumer protection. White and Munger, not economists but lawyers, played the role of behavioral economists before the phrase was even coined. They managed to anticipate many of the hypotheses and themes that would later dominate behavioral economics-- but with more informal and colorful language. A number of new legislative and judicial acts in the late 1960s provided the impetus for their study: 

"Congress has passed the Truth-in-Lending Act; the National Conference of Commissioners on Uniform State Laws has proposed the Uniform Consumer Credit Code; and many states have enacted retail installment sales acts to update and supplement their long-standing usury laws. These legislative and judicial acts have always relied, at best, on anecdotal knowledge of consumer behavior. In this Article we offer the results of an empirical study of a small slice of consumer behavior in the use of installment credit. 

In their recent efforts, the legislatures, by imposing new interest rate disclosure requirements on installment lenders,  have sought to protect the consumer against pressures to borrow money at a higher rate of interest than he can afford or need pay. The hope, if not the expectation, of the drafters of such disclosure legislation is that the consumer who is made aware of interest rates will seek the lowest-priced lender or will decide not to borrow. This migration of the consumers to the lowest-priced lender will, so the argument goes, require the higher-priced lender to reduce his rate in order to retain his business. These hopes and expectations are founded on the proposition that the consumer is largely ignorant of the interest rate that he pays; this ignorance presumably keeps him from going to a lender with cheaper rates. Knowledge of interest rates, it is believed, will rectify this defect…”

Here comes their "squatting baboon" metaphor:

“Presumably, consumers in a perfect market will behave like water in a pond, which gravitates to the lowest point-i.e., consumer borrowers should all tum to the lender that gives the cheapest loan. We began this project with a strong suspicion-based on the observations of others-that the consumer credit market is far from perfect and that water governed by the force of gravity is a poor metaphor with which to describe the behavior of consumer debtors. The consumer debtor's choice of creditor clearly involves consideration of many factors besides interest rate. Therefore, a metaphor that better describes our suspicions about the borrower's behavior in a market in which rate differences appear involves a group of monkeys in a cage with a new baboon of unknown temperament. The baboon squats in one comer of the cage near some choice, ripe bananas. In the far comer of the cage is a supply of wilted greens and spoiled bananas, the monkeys' usual fare. Some of the monkeys continue eating their usual fare because they are unaware of the new bananas and the visitor. Other monkeys observe the new bananas but do not approach them. Still others, more daring or intelligent than the rest, seek ways of snatching an occasional banana from the baboon's stock. The baboon strikes at all the brown monkeys but he permits black monkeys to eat without interference. Yet many of the black monkeys make no attempt to eat. One suspects that a social scientist who interviewed the members of the monkey tribe about their experience would find that many of those who saw and appreciated the choice bananas would be unable to articulate the reasons for their failure to eat any of them. The social scientist might also discover that a few who looked at the baboon in obvious fright would nevertheless deny that they were afraid. In addition, he might find that some were so busy picking fleas or nursing that they did not observe the choice bananas at all. We suspected that consumer borrowers had similarly diverse reasons for their behavior.

We presumed that some paid high interest rates only because of ignorance of lower rates and that others correctly concluded that they could not qualify for a cheaper loan than they received. Others, we suspected, were merely too lazy or too fearful of bankers to seek lower rates.”

Suggesting that people are just lazy, or comparing them to monkeys, has (understandably) fallen out of fashion. But pointing out that consumers are not Homo economicus has not. The authors interview people in Washtenaw County, Michigan who had purchased a new car in 1967. Most of the lenders in the area loaned money at the legal maximum add-on, while Huron Valley National Bank (HVNB) loaned at a significantly lower rate. They interview an HVNB loan officer to determine whether different borrowers would have received a loan from HVNB in 1967 and at what terms. They find that most of the consumers in their sample could have borrowed at a lower rate.

A majority of the sample did not know the rate at which they had borrowed. Most had allowed the auto dealer to arrange the loan rather than shopping for the lowest rate. Even if they knew that lower rates were available elsewhere, they declined to shop around. The authors find differences in financial sophistication, education, and job characteristics between consumers who shopped around for lower rates and those who did not. They conclude:

“The results of our study suggest that, at least with regard to auto loans, the disclosure provisions of the Truth-in-Lending Act will be largely ineffective in changing consumer behavior patterns. Certainly the Act will not improve the status of those who already know that lower rates are available elsewhere. And we discovered no evidence that knowledge of the interest rate-which, even under the Act will usually come after a tentative agreement to purchase a specified car has been reached-will stimulate a substantial percentage of consumers to shop for a lower rate elsewhere.”

The authors come down as pessimistic about the Truth-in-Lending Act, but make no new policy recommendations of their own. If they were writing today, instead of just predicting that a policy would be ineffective, they might suggest ways to design the policy to "nudge" consumers to make different decisions. The Truth-in-Lending Act has been amended numerous times over the years, and was placed under the authority of the Consumer Financial Protection Bureau (CFPB) by the Dodd-Frank Act. Behavioral economics has played a central role in the work of the CFPB. But the active application of behavioral law and economics to regulatory policy is not universally accepted. For example, Todd Zwicki writes:

"We argue that even if the findings of behavioral economics are sound and robust, and the recommendations of behavioral law and economics are coherent (two heroic assumptions, especially the latter), there still remain vexing problems of actually implementing behavioral economics in a real-world political context. In particular, the realities of the political and regulatory process suggests that the trip from the laboratory to the real-world crafting of regulations that will improve consumers’ lives is a long and rocky one."

Zwicki expands this argument in a paper coauthored with Adam Christopher Smith. While I'm not convinced that their rather negative portrayal of the CFPB is warranted, I do think the paper presents some provocative cautions about how behavioral economics is applied to policy--especially the warning against "selective modeling of behavioral bias," which I have heard even top behavioral economists caution against.

The Limited Political Implications of Behavioral Economics

A recent post on Marginal Revolution contends that progressives use findings from behavioral economics to support the economic policies they favor, while ignoring the implications that support conservative policies. The short post, originally a comment by blogger and computational biologist Luis Pedro Coelho, is perhaps intentionally controversial, arguing that loss aversion is a case against redistributive policies and social mobility:

"Taking from the higher-incomes to give it to the lower incomes may be negative utility as the higher incomes are valuing their loss at an exaggerated rate (it’s a loss), while the lower income recipients under value it... 

...if your utility function is heavily rank-based (a standard left-wing view) and you accept loss-aversion from the behavioral literature, then social mobility is suspect from an utility point-of-view."

Tyler Cowen made a similar point a few years ago, arguing that "For a given level of income, if some are moving up others are moving down... More upward — and thus downward — relative mobility probably means less aggregate happiness, due to habit formation and frame of reference effects."

I don't think loss aversion, habit formation, and the like make a strong case against (or for) redistribution or social mobility, but I do think Coelho has a point that economists need to watch out for our own confirmation bias when we go pointing out other behavioral biases to support our favorite policies. Simply appealing to behavioral economics, in general, or to loss aversion or any number of documented decision-making biases, rarely makes a strong case for or against broad policy aims or strategies. The reason is best summarized by Wolfgang Pesendorfer in "Behavioral Economics Comes of Age":

Behavioral economics argues that economists ignore important variables that affect behavior. The new variables are typically shown to affect decisions in experimental settings. For economists, the difficulty is that these new variables may be unobservable or even difficult to define in economic settings with economic data. From the perspective of an economist, the unobservable variable amounts to a free parameter in the utility function. Having too many such parameters already, the economist finds it difficult to utilize the experimental finding.

All economic models require making drastic simplifications of reality. Whether they can say anything useful depends on how well they can capture those aspects of reality that are relevant to the question at hand and leave out those that aren't. Behavioral economics has done a good job of pointing out some aspects of reality that standard models leave out, but not always of telling us exactly when these are more relevant than dozens of other aspects of reality we also leave out without second thought. For example, "default bias" seems to be a hugely important factor in retirement savings, so it should definitely be a consideration in the design of very narrow policies regarding 401(K) plan participation, but that does not mean we need to also include it in every macroeconomic model.

Soda Calories and the Ethics of Nudge

The “surprisingly simple way to get people to stop buying soda,” according to a new and highly-hyped study, is to tell them how long they will have to run in order to burn off the calories in the soda. Since so many of my friends are economists, runners, or both, articles about this study ended up plastered all over my Facebook and Twitter pages, followed by a lot of commentary and debate. Some of the issues that came up included paternalism, the identification of social goods and ills, the replicability of field experiments on a larger scale or longer time frame, and the ethics of nudge policies.

My friend and classmate David Berger has allowed me to share his take:

Alright, this keeps coming up: public health types saying stupidly pessimistic things about the number of hours you have to burn off x amount of calories. Friends, it's easy to deceive yourself about how many calories you actually burn doing cardio. And then there's a whole bunch of people--treadmill manufacturers, for example--who want to inflate the numbers. But this trend of public health know-it-alls using the most pessimistic calculations needs to stop. It's just wrong. These people convinced teenagers that it would take 50 minutes of running to burn off one soda. They must be targeting non-runners. 

Actually, who they are targeting is baffling. They base their calculations on the activity-energy equivalents for a 110 pound fifteen year old. Nowhere do they indicate pace, although when I use the calculator on runner's world to give a 110 pound a 15 min/mile pace (a pace walkers in comfortable clothing can manage), it gives me 277 calories for 50 minutes. 

Let's do a real calculation. If you are less worried about 110 pounders drinking soda, try a 200 pounder. Let's give the same walking pace of 15 min/mile, and keep it at 50 minutes. 504 calories. Alternately, that's 25 minutes to work off a soda. 

Or, suppose you expect someone to aspire to something, and someone reading this public service message to know the difference between walking and running, and to be able to determine whether they can maintain a running pace for 50 minutes. I won't even make them a good runner, just a 12:30 min/mile, which someone starting out can manage in most cases. The same 50 minutes goes up to 605 calories. Granted 50 minutes might be much for someone starting out, but then they only need 21 minutes to manage a 250 calorie soda. 

Now, calories/hour will be lower if you weigh less than 200 pounds. But then you will probably be able to run faster than 12:30. I understand this push society is making overall: there's too much false hope in the ability of exercise to compensate for constantly immoderate caloric choices, and there's too much acceptance of empty nutrition (like soda) as a source of calories. But if the next heavy-handed social tactic is outright lying, can we please just not?

Consumption Contagion and Income Inequality

The trends of rising income inequality and the declining national savings rate since the early 1980s may be related, according to a paper by Marianne Bertrand and Adair Morse. The authors find that higher levels of visible consumption by increasingly better-off households at the top of the income distribution induces consumers in the lower parts of the income distribution to spend a higher share of their disposable income. From "Consumption Contagion: Does the Consumption of the Rich Drive the Consumption of the Less Rich?":

"Our empirical strategy exploits variation across geographic markets and over time to identify the effect of expenditures by the rich on that of the non-rich. We ask whether, everything else held constant, higher levels of consumption by the rich living in a household’s relevant market (which we define to be either a state or an MSA in a given year) predicts a higher propensity to consume out of disposable income for the non-rich household. After establishing that such vertical consumption correlations occur, we then explore possible mechanisms. Our results are most consistent with the view that visible increased consumption by the rich induces status-seeking or status-maintaining consumption by the less rich."

Bertrand and Morse define the rich households in each state as those with above the 80th percentile of income in that state. Their baseline regression shows that a 1 percent increase in consumption (excluding housing) among the rich in a particular state translates into a 0.07 percent increase in consumption among the less-rich. They find no evidence that this could be explained by the permanent income hypothesis; rising consumption by the rich in a particular state is not predictive of faster future income growth by the state's less-rich. Thus, they conclude that "Our preferred explanation for the vertical consumption spillovers we observed in our basic results is that low and middle income households witness the higher consumption levels by the rich and are tempted to also consume more."

To test this explanation further, they use data from the Consumer Expenditure Survey and use the Ori Heffetz (2011) index to rank goods into seven categories of increasing visibility. Highly visible consumption items include cars, clothing (except underwear!), shoes, and cigarettes; minimally visible consumption items include health or legal accounting services and, yes, underwear. They replicate the analysis by goods category, and find strongest effects in the most visible consumption categories, consistent with a "consumption contagion" explanation.

As another test, they replicate the analysis using Census Metropolitan Statistical Areas (MSAs) instead of states. They use a measure of community segregation, indicating how closely the rich live to the less-rich in each MSA. In MSAs where the rich and less-rich live closer together, there is more consumption contagion.

Overall, the authors estimate that the savings rate of median-income households would be one to two percentage points higher in the absence of this "consumption contagion" effect. This is non-trivial but also not huge. What is most important is the empirical support of a particular type of departure from the Permanent Income Hypothesis. Many types of departures have been hypothesized, but quantifying their relative importance and carefully tracing out their implications for macro models is an ongoing task.

Japan and the Formation of Inflation Expectations

With the Bank of Japan's adoption of a 2% inflation target making headline news, it seems like a good time to discuss some recent research on the psychology of inflation expecations by Berkeley Professor Ulrike Malmendier, who was recently awarded the 2013 Fischer Black Prize from the American Finance Association. This biennial prize honors the top finance scholar under the age of 40 years old. Malmendier works in the intersection between finance and behavioral economics and is known for her incredible creativity.

Here is the abstract of Malmendier's paper with Steven Nagel titled "Learning from Inflation Experiences":

How do individuals form expectations about future inflation? We propose that past inflation experiences are an important determinant absent from existing models. Individuals overweigh inflation rates experienced during their life-times so far, relative to other historical data on inflation. Differently from adaptive-learning models, experience-based learning implies that young individuals place more weight on recently experienced inflation than older individuals since recent experiences make up a larger part of their life-times so far. Averaged across cohorts, expectations resemble those obtained from constant-gain learning algorithms common in macroeconomics, but the speed of learning differs between cohorts.
Using 54 years of microdata on inflation expectations from the Reuters/Michigan Survey of Consumers, we show that differences in life-time experiences strongly predict differences in subjective inflation expectations. As implied by the model, young individuals place more weight on recently experienced inflation than older individuals. We find substantial disagreement between young and old individuals about future inflation rates in periods of high surprise inflation, such as the 1970s. The experience effect also helps to predict the time-series of forecast errors in the Reuters/Michigan survey and the Survey of Professional Forecasters, as well as the excess returns on nominal long-term bonds.

Malmendier and Nagel's paper over a time period covering several monetary policy regimes, which differ markedly from that of Japan. But a related paper by David Blanchflower and Conall Mac Coille focuses on the UK, which practices inflation targeting.  "The Formation of Inflation Expectations: an Empirical Analysis for the UK" (2009) includes a summary of why inflation expectations matter for monetary policy, and how this is relevant to inflation targeting: 

In the neo-Keynesian model (see, for example, Clarida et al. 2000), sticky prices result in forward looking behaviour; inflation today is a function of expected future inflation as well as the pressure of demand, captured in an output gap term. Thus, expectations are deemed to be an important link in the monetary transmission mechanism. Monetary policy can be more successful when long-term inflation expectations are well anchored. Hence, many studies have focused on the question of how to assess the response of inflation expectations to macroeconomic shocks, and whether this is likely to be lower in inflation targeting regimes. 

Blanchflower and Mac Coille also summarize three paths through which inflation expectations matter:

Wages are set on an infrequent basis, thus wage setters have to form a view on future inflation.  If inflation is expected to be persistently higher in the future, employees may seek higher nominal wages in order to maintain their purchasing power.  This in turn could lead to upward pressure on companies’ output prices, and hence higher consumer prices.  Additionally, if companies expect general inflation to be higher in the future, they may be more inclined to raise prices, believing that they can do so without suffering a drop in demand for their output.  A third path by which inflation expectations could potentially impact inflation is through their influence on consumption and investment decisions.  For a given path of nominal market interest rates, if households and companies expect higher inflation, this implies lower expected real interest rates, making spending more attractive relative to saving. 

In Japan, the third path may be most important. The higher inflation target is intended to lower real interest rates and boost consumption and investment. But there is a fourth reason, not listed by Blanchflower and Mac Coille, of particular relevance to Japan. Foreigners' expectations of future inflation affect the value of the currency. the Japanese Ministry of Finance recently revealed a 222.4 billion yen ($2.5 billion) current account deficit-- a measure of how much imports exceed exports. When people expect Japanese inflation to be higher in the future, the yen gets less valuable now, because it won't be able to buy as much stuff later; the yen weakens. But in this case, weakness is not necessarily bad. A weaker yen means that Japanese people will find it more expensive to import stuff, so they will import less. Likewise, people outside of Japan will find it cheaper to buy Japanese stuff, so Japan will export more. This helps shrink the current account deficit. And depending on the sizes of the income and substitution effects, Japanese consumers may buy more Japanese products.

Under inflation targeting in the UK, even though inflation expectations are reasonably well anchored, and median expectations are around the inflation target, there is substantial heterogeneity across agents in their inflation expectations. Malmendier and Nagel's paper provides a behavioral theory to explain part of this heterogeneity based on agents' heterogeneous past experiences of inflation. Heterogeneous inflation expectations have the potential to affect the workings of all the paths through which inflation expectations matter. We need to understand not only how Japan's inflation target will influence median inflation expectations, but also how it will affect expectations of price setters, wage setters, borrowers, savers, exporters, trade partners, etc. More than likely, these groups differ significantly in their demographics, have had different experiences, and thus form different expectations of inflation. (For reference, the graph below displays Japanese inflation, interest rates, real GDP per capita growth rate, and M2 growth rate. Japan has not seen 2% inflation since 1997.) Extensions of Malmendier's research to other countries and monetary regimes will be very useful in understanding the effects of monetary policy.