OCTOBER 5, 2012
Seduction by Contract: Law, Economics, and Psychology in Consumer Markets
By Oren Bar-Gill
(Oxford University Press, 280 pp., $40)
WHAT DO people notice? What do they miss? In the late 1990s, the social scientists Christopher Chabris and Daniel Simons tried to make some progress on these questions by asking people to watch a two-minute movie, in which six ordinary people pass a basketball to one another. The simple task? To count the total number of passes.
After the little movie is shown, the experimenter asks people how many passes they were able to count. Then the experimenter asks: And did you see the gorilla? A lot of people laugh at the question. What gorilla? Then the movie is replayed. Now that you are not counting passes, you see a gorilla enter the scene, plain as day, and then pound its chest, and then leave. The gorilla (actually a person dressed up in a gorilla suit) is not at all hard to see. In fact you can’t miss it. But when counting passes, many people (typically about half) do miss it. I confess that when I saw the movie, I was utterly incredulous when asked whether I saw the gorilla. There wasn’t any gorilla! But there was.
Behavioral economists have been much interested in the gorilla experiment, because it shows that people are able to pay attention to only a limited number of things, and that when some of those things are not salient we ignore them, sometimes to our detriment. (Magicians and used-car dealers try to hide gorillas.) The field of behavioral economics is no longer an invisible gorilla, limited to obscure academic journals. On the contrary, it is highly salient. It has produced a number of best-sellers, including Daniel Kahneman’s Thinking, Fast and Slow and Dan Ariely’s Predictably Irrational. There is even a Behavioral Economics for Dummies. In the media and in the business community, behavioral economics seems to be everywhere. YouTube has over 1,600 videos on the topic. It also has the invisible gorilla experiment—but now that you know the gorilla is there, you will see it.
Policymakers are certainly paying attention. In the United Kingdom, Prime Minister Cameron has gone so far as to create a Behavioural Insights Team, located in his Cabinet Office. The official website states that its “work draws on insights from the growing body of academic research in the fields of behavioural economics and psychology which show how often subtle changes to the way in which decisions are framed can have big impacts.” The Team has used these insights to spur new initiatives in numerous areas, including smoking cessation, energy efficiency, organ donation, consumer protection, and compliance strategies in general.
In the United States, behavioral economics has also been noticed. In explaining the ambitious new fuel-economy regulations issued in August 2012, the Department of Transportation explicitly referred to “phenomena observed in the field of behavioral economics,” including “a lack of salience of relevant benefits (such as fuel savings, or time savings associated with refueling) to consumers at the time they make purchasing decisions.” Behavioral findings have informed new disclosure requirements and policies, including the Department of Agriculture’s widely praised Food Plate, which replaced the widely criticized Food Pyramid. Behavioral economics has also informed initiatives to simplify regulatory mandates and to promote automatic enrollment in savings plans. From September 2009 until last August, I was the administrator of the White House Office of Information and Regulatory Affairs in the Obama administration. Aware of behavioral findings, we did a great deal of work to reduce and simplify requirements and to ensure that when information is disclosed to the public, it is both meaningful and salient. We also directed agencies to test lengthy or complex forms in advance, to make sure that people actually understand them.
NOTWITHSTANDING ALL the attention and activity, there remains a lot of confusion about what behavioral economics is. In brief, behavioral economists make a series of empirical claims. They question the standard economic assumption that human beings are fully rational, and they contend that people’s departures from economic rationality are predictable. Now it is not exactly big news to say that homo sapiens is not homo economicus. Poets, novelists, philosophers, folk singers, priests, sellers of life insurance, and parents of teenagers have long been aware of that. The more striking behavioral claims, and the novel ones, involve specifying, in a systematic way and on the basis of evidence, what those departures exactly are.
From the invisible gorilla experiment (and countless others), we know that salience greatly matters, and that if information does not stand out from the background, people might simply ignore it, even if it is really important. We also know that people often use heuristics, or mental shortcuts, in thinking about risks. For many of us, calculation can be time-consuming and hard, and so in assessing risks (of crime, illness, natural disaster, and more) we often ask whether we can easily recall cases in which those risks actually came to fruition (the “availability heuristic”). This approach can lead to big mistakes. For example, people have been found to be willing to pay more to purchase flight insurance for risks from terrorism than they are willing to pay to purchase flight insurance for risks of all kinds.
Contrary to the prediction of standard economic theory, behavioral economists have found that outcomes are much affected by whether people are automatically enrolled in a program (say, for savings, or to protect their privacy online) and asked whether they want to opt out, or instead not automatically enrolled and asked whether they want to opt in. If people are automatically enrolled in a program, many of them tend to stay in; if they are not automatically enrolled, many of them tend to stay out. For this reason, default rules, specifying what happens if people do nothing, have a big impact. Inertia is a powerful force.
We also know that complex disclosure requirements, or lengthy forms, can create unexpectedly serious problems, because it is too much work for people to try to understand them. The take-up of many public and private programs is smaller than it should be, simply because it is such a headache to fill out the necessary forms. Simplification can make a major difference. We know as well that most human beings show a tendency toward unrealistic optimism, at least about their personal lives. Most people believe that they are safer than the average driver and less likely to be involved in a serious accident. As the neuroscientist Tali Sharot has demonstrated, unrealistic optimism is rooted in the workings of the human brain, which is much better at incorporating good news than bad news.
One of the central missions of the new Consumer Financial Protection Bureau is to ensure that people “know before they owe.” Showing an appreciation of behavioral economics, the Bureau has been interested in two different approaches. First, it is promoting clear and simple disclosure, so that people can learn, more or less at a glance, about the agreements that they are about to enter into. Second, it has referred to “smart disclosure,” which occurs when public or private institutions disclose detailed information in standard, machine-readable formats. This information can be used by intermediaries, often through apps that display that information in new and creative ways. It is easy to imagine consumer-friendly disclosure of the central features of a variety of credit card and mortgage agreements, allowing people to make informed comparisons and to choose the plan that is best for them.
INITIATIVES OF this kind raise some obvious questions (now being pressed by many Republicans in Congress). Do we really need a Consumer Financial Protection Bureau? Why can’t we rely on free markets? After all, there is a fierce competition in consumer markets, whether the products are mortgages, credit cards, or cell phones. In the presence of such competition, it is not so easy to identify a standard market failure that would justify regulation. If the goal really is to ensure that people know before they owe, we might think that the best solution is for government to recede and to allow the market to do its work.
Oren Bar-Gill has a straightforward answer to the critics. He believes that government regulation can be justified by “behavioral market failures,” in the form of biases and misperceptions that have been carefully studied in psychology and behavioral economics. Bar-Gill does not refer to the gorilla experiment, but he places a lot of emphasis on salience, and he contends that because consumers are imperfectly rational, they are likely to ignore important information and hence to make big mistakes. To be sure, he acknowledges that, in principle, competition could correct the problem. But he insists that, in practice, competitive forces are often the problem, not the solution. The reason is that sellers must do what the market rewards. If sellers offer people the objectively best cell phone contracts, they will end up losing out to their competitors, who are offering contracts that are less good but subjectively more appealing.
To be clear, Bar-Gill does not contend that there is literal fraud here, but urges instead that as a result of competitive pressures, sellers are forced “to exploit the biases and misperceptions of their customers.” In his view, the consequences for consumers can be extremely bad. Indeed, “seductive” contract design helped fuel the demand for subprime mortgages, thus contributing to the subprime meltdown of 2008. But how, exactly, do companies exploit these biases and misperceptions? Bar-Gill emphasizes two strategies. The first involves cost deferral. The second involves complexity.
It is true, of course, that deferral of costs is part of the very nature of loan agreements. What Bar-Gill identifies is an extreme form—cost deferral on steroids. In consumer contracts, you can often find agreements whose terms look tantalizingly favorable in the short-term, but a lot less favorable in the long-term. For credit cards, you might get short-term teaser interest rates that would cost you very little, but after a while, you might end up with long-term fees and rates that would cost you a lot. (A recent illustration from a website for a highly reputable issuer: “0% Intro APR on balance transfers for the first 15 months after account opening. After that a variable APR currently 15.99% or 24.99% depending on your creditworthiness.”) For cell phone contracts, you might get a free phone, which to some people sounds pretty appealing, even amazing—but you get the phone only if you sign a two-year lock-in contract, which contains an assortment of fees and penalties.
Why is such cost deferral so common? Bar-Gill answers by invoking two features of homo sapiens, much emphasized by behavioral economists: myopia and optimism. Consumers tend to be myopic in the sense that they focus on the short term. For many of us, the future is a bit like a foreign country, and we may well neglect it, or at least give it little weight. For this reason, we might be unduly attracted to an agreement that is terrific for the next two months but not so good for the next decade. In credit markets, the problem of myopia is compounded by unrealistic optimism. In general, people underestimate the likelihood that things will go wrong for them personally, even if they know the statistical realities. (A lot of people know that the divorce rate is close to 50 percent, but at the time of marriage, many people estimate their own divorce risk at 0 percent. Even smokers, aware of the statistical risks, have been found to think that they are less likely to get heart disease or lung cancer than the average nonsmoker.) If people are unrealistically optimistic, they will accept contract terms including high overuse fees and late fees—dismissing the risk that those fees will come back to haunt them.
Bar-Gill notes that with respect to credit cards, mortgages, and cell phones, the underlying agreements are staggeringly complex. Why are there so many confusing and disparate terms? Bar-Gill contends that from the standpoint of sellers, complexity has a big virtue, which is that it prevents consumers from easily figuring out the total cost of the relevant products. As a result, consumers must rely on what is salient. And if some terms are salient while others are not, sellers will have real opportunities to make money, and consumers will make big mistakes. Short-term prices (such as teaser rates) grab people’s attention, while high back-end fees and rates tend to seem like the invisible gorilla, or background noise.
In Bar-Gill’s view, cost deferral and complexity work together to produce behavioral market failures. As a result, many people end up with arrangements that are poorly suited to their situations, and thus lose a lot of money. Those who are in difficult economic circumstances—and who end up paying a wide range of fees and penalties—are often most vulnerable to exploitation and eventual hardship. And as the subprime crisis of 2008 suggests, the difficulties faced by individuals can be severe and eventually have systemic effects.
WHAT CAN BE DONE? Bar-Gill believes that people’s choices are not adequately informed and hence his answer is to require more effective disclosure. He strongly supports the attention now being given by federal agencies to disclosure of product attributes—of what, precisely, are the key features of agreements involving credit cards, mortgages, and cell phones. But he adds that we need to focus far more on disclosure of product use. Your cell phone plan has a set of features, and you should know what they are. But the cost of your plan does not depend only on those features. It depends also on how you use the phone, and it is important to know that, too. You ought to know, for example, the per-minute charges for minutes not included in your plan, but you also should know how likely it is that you will exceed the limit, and by how much. Many cell phone users have only a vague idea of their own usage; they need to know more.
What would product-use disclosure entail? Bar-Gill considers two possibilities. The first is average-use information, by which issuers reveal the average pattern shown by the entire population or relevant subgroups. The second is individual-use information, derived from the individual consumer’s actual behavior. Bar-Gill urges that individual-use information is better, because it is more accurate, and so it should be disclosed if it is available. Bar-Gill quotes Duncan McDonald, formerly general counsel of Citigroup’s Europe and North America card business: “No other industry in the world knows consumers and their transactions behavior better than the bank card industry.” Bar-Gill contends that because issuers know so much about use patterns—indeed, more than consumers themselves do—they should share that information with consumers.
Building on recent practices within the U.S. government, Bar-Gill argues in favor of two approaches. The first is simple, clear disclosure of total cost to consumers. Cell phone companies should be required to disclose the aggregate annual cost of owning a phone (combining standard rate information with individual use-patterns). In Bar-Gill’s view, the problems of complexity and unrealistic optimism are much diminished if consumers are able to obtain a clear sense of aggregate costs. The second approach is to require companies to disclose more comprehensive information, which would go in turn to sophisticated intermediaries. Those intermediaries would be able to assemble information about the many plans offered by cell phone companies. The intermediaries could combine that information with a particular consumer’s product-use information, thus providing the consumer with helpful comparative information. This idea, a form of “smart disclosure,” might seem a bit technical and wonkish, but it has a lot of potential to help people, in areas ranging from financial products to health care to education to energy use, to save money and to promote competition.
Much of Bar-Gill’s book consists of a detailed explanation of these general points, exploring credit cards, mortgages, and cell phones. With respect to credit cards, the level of fee complexity is truly extraordinary, and it is easy for consumers to fail to notice some of what matters. You can find “annual fees, cash-advance fees, balance-transfer fees, foreign currency-conversion fees, expedited-payment fees, no-activity fees, late fees, over-limit fees, and returned-check fees.” Owing to teaser rates and other seductive features, many consumers respond to what is salient rather than total cost, which leads to too much borrowing and an inflated demand for cards.
A major problem is that some consumers underestimate the total costs of piecemeal borrowing. Apparently people who would never take out a big loan are willing to take out a number of small loans that are big in the aggregate. One survey found that small purchases of non-essential goods (including movies and DVDs) are a major contributor to credit card debt. Financial distress, including consumer bankruptcies, is a possible consequence. Bar-Gill applauds recent legislative enactments, including the card Act and Dodd-Frank, insofar as they require significant steps toward improved disclosure, including disclosure of important product-use information. He wants regulators to be more ambitious in requiring disclosure of that information.
IN LIGHT OF the harmful effects of the subprime crisis on the economy as a whole, mortgages are of course of particular importance. Bar-Gill focuses on subprime mortgage contracts, which include a dazzling array of fees. All mortgage contracts defer costs, but for subprime agreements that feature is greatly exaggerated. Many of them have an increasing payment schedule, with seductively low interest rates for an initial period (frequently two years) and then much higher payments. For some agreements, borrowers are subject to “payment shock,” sometimes in the form of a 100 percent increase in the monthly payment. As a remedy, Bar-Gill focuses on the yet-to-be-realized potential of annual percentage rate (APR) disclosure, which can be transformed into a total-cost-of-credit measure. He urges that regulators should ensure (among other things) that APR is disclosed at a sufficiently early stage to allow comparison-shopping and that no price dimensions are excluded from APR, so that it actually works as it should. Recent statutory and regulatory changes have helped, but Bar-Gill believes that more should be done to enable consumers to make informed comparisons.
Bar-Gill’s most interesting discussion involves cell phones, which come with lock-in contracts (including early termination fees) and extraordinarily complex terms. Bar-Gill is particularly concerned with the standard three-part cell phone tariff, including a monthly charge, a specified number of minutes that the monthly charge buys you, and a per-minute price beyond the monthly limit. In Bar-Gill’s account, a big problem with the three-part tariff is that many consumers do not have an accurate sense of how many minutes they will use, leading to both underestimates and overestimates. In the words of one carrier’s usage expert, “People absolutely think they know how much they will use and it’s pretty surprising how wrong they are.”
Using a large data set, Bar-Gill finds that over 65 percent of users have the wrong plan for their needs. Extrapolating from his data, Bar-Gill estimates a total annual loss, to American consumers, of $13.35 billion. The lock-in compounds the problems insofar as it prevents consumers from switching to different carriers whose plans are a better fit. Bar-Gill acknowledges that some carriers are improving disclosure, among other things by providing product-use information, but he urges that current disclosures are both variable and insufficient, and hence “a regulatory nudge is probably required.”
TO SEE WHAT IS novel and important in Bar-Gill’s book, we need to step back a bit. A lot of the early work in behavioral economics identified potential departures from rationality and then used laboratory experiments, often with college students, to demonstrate that the departures actually occur. Some people were skeptical about this approach. They responded, not unreasonably, that we need to know whether the departures matter not in the laboratory under artificial conditions, but in the real world. On the skeptics’ view, people miss invisible gorillas in two-minute movies, but when real money is on the line, they’ll see that gorilla. In competitive markets, those who try to hide gorillas will eventually be punished, and over time those gorillas will be plain as day. More recent work in behavioral economics is exploring real-world behavior. What makes Bar-Gill’s argument both fresh and impressive is its detailed exploration of actual consumer markets and its careful argument that myopia, unrealistic optimism, and a lack of salience are playing a significant role in those markets.
But it is important to sort out the relationships among the various behavioral market failures, and Bar-Gill could have done more on that count. Standing all by itself, myopia might not justify a disclosure mandate. Imagine a special mortgage contract, perhaps called Bliss For Decades, or Sell Your Soul. Under this contract, you owe absolutely nothing for twenty years, at which time you have to pay the principal plus a ton of interest. This is a simple, fully transparent agreement, and it does not appear to exploit unrealistic optimism. The contracts that most trouble Bar-Gill, and that justify some kind of disclosure requirement, are different because they combine cost deferral with complexity and key features that are not salient.
Bar-Gill is aware that his claims will not exactly render rational choice theorists speechless. A lot of complexity may emerge in consumer markets simply because of diverse situations and tastes (not to mention the need to plan for contingencies). Reasonable people need loans of many different kinds. For some of us, it makes sense to defer costs for a long time, if only because we expect, perhaps optimistically but not unreasonably, to have more money in the future than we have now. For credit cards, mortgages, and cell phones, Bar-Gill explores potential rational-choice explanations with care, concluding that such theories are not entirely unhelpful but that they leave real gaps. Consider, for example, people’s failure to accurately anticipate their cell phone usage—and the fact that until fairly recently providers have made it difficult for consumers to monitor their usage. Or consider the fact that credit card companies have not always made it easy for people to pay their bills automatically online—and that when they do so, they often make the minimum the default payment, rather than the full amount, and thus subject you to high interest rate charges. (A little advice: Consider signing up for automatic payment today, and consider clicking that full payment box.)
Bar-Gill is therefore convincing in arguing that myopia, optimism, and a lack of salience provide at least part of the explanation for both cost deferral and complexity. So what is the best remedy? To his credit, he focuses on disclosure requirements rather than on more aggressive restrictions. As he notes, disclosure can have the important advantage of helping unsophisticated borrowers without restricting the choices available to sophisticated ones. Another advantage of disclosure requirements is that they tend to be inexpensive and will not do a lot of harm unless they are poorly designed. We might even venture, admittedly with some tentativeness, a general principle: in the face of behavioral market failures, the best response is usually disclosure and no more. Interestingly, however, Bar-Gill does not explore the range of disclosure options and restricts his discussion to purely factual information. He does not explore the possibility of warnings, which might be vivid and highly salient and thus stand out from the background.
THERE REMAINS a question to which Bar-Gill devotes too little attention. Would disclosure really help? Or might the very problems he describes—myopia, optimism, lack of salience—make it insufficiently effective? Sure, there is much to be said for disclosing total cost of ownership, and more comprehensive disclosure, used by intermediaries, has real potential to promote comparison shopping. At the same time, we could know a lot more about the actual effects of the disclosures that Bar-Gill recommends. Suppose, for example, that people are given a simple account of the total annual cost of their cell phone plan. Would that solve the problems Bar-Gill identifies? For some people, perhaps it would be a kind of gorilla, rendered invisible by appealing offers of teaser rates and free phones. In the face of those offers, the simple account of total annual cost might not much help low-income consumers, who are especially vulnerable. Maybe unrealistically optimistic consumers would say, “I’ll be just fine” (at least if they are not given information about their own usage, which might not be available at the key initial stage). It is one thing to speculate, however reasonably; it is quite another to know.
With respect to the actual effects of policies, randomized controlled trials are the gold standard, and they should be used far more often, both generally and to test disclosure strategies. To test the efficacy of drugs, the best approach is to compare a control group with a similarly situated treatment group and to see what, exactly, the effects of the drug turn out to be. Something of this sort should be done, where feasible, to test disclosure requirements. Perhaps different groups could see different disclosures relating to cell phone costs in New York, California, and Illinois. If so, we could actually test their impact on people’s choices, and see whether Bar-Gill’s remedy would help with the problem.
There is also a lurking question. If we take Bar-Gill’s diagnosis seriously, should we consider more aggressive reforms? He does not really say, beyond noting (in just a sentence) the possibility that bans on abusive practices (which he does not define) might be justified. But a sympathetic reader might wonder whether regulators should be willing to explore stronger steps to discourage exploitation and to promote simplicity. With respect to credit cards, for example, Bar-Gill can be read to favor a more transparent market in which issuers are charging visible, higher annual fees while reducing penalties and nonsalient fees. One problem with moving the market toward this approach is that one size does not fit all. Another problem is that more aggressive regulatory approaches could end up imposing real costs on careful or sophisticated users, who avoid penalties and fees. Even in the face of behavioral market failures, regulators should be alert to the risk of unintended bad consequences, particularly if they impose a mandate or a ban. If, for example, regulators prohibited credit card companies from charging high interest rates, some people might be priced out of the credit card market, and they might end up going to payday lenders and others whose interest rates are higher still. With respect to mandates and bans, the risk of unintended consequences needs especially careful attention.
Along with many other nations, the United States is in the midst of fundamental thinking about consumer markets, and in particular about the best approach to consumer protection amidst fierce competition, constantly evolving technology, consumer-friendly apps, a dazzling array of options, impossibly seductive offers, and widespread consumer confusion alongside endless opportunities for learning. Bar-Gill’s word is far from the last. But he is right to emphasize the need to bring financial gorillas out of the background, so that people can plainly see them.
Cass Sunstein is a contributing editor at The New Republic. This article appeared in the October 25, 2012 issue of the magazine under the headline “Show Me the Money”