By OREN BAR-GILL
Review of HAPPINESS AND THE LAW, by John Bronsteen, Christopher Buccafusco and Jonathan S. Masur
University of Chicago Press, 2014
Happiness and the Law is an important book. Bronsteen, Buccafusco and Masur (BBM) provide a well-written, thought-provoking, rigorous introduction to hedonic psychology and its many potential applications in law and policy. Numerous lessons are already ripe for consumption by policymakers. Other ideas set the stage for a fruitful research agenda that will influence policy in years to come.
The book’s point of departure is that happiness is a major objective of legal policy. It is intimately linked to utilitarianism and to welfarism. The importance, to policy, of measuring happiness and of hedonic psychology more generally seem beyond question.
This review will focus on the first two Parts of the book – Part I: Analyzing Laws’ Effects on Well-Being, and Part II: Viewing Two Core Areas of the Law through the Lens of Hedonics. Part III: Well-Being engages the “philosophical claims about happiness” (p. 133). It will be left for another day, or another review. In any event, as BBM note at the beginning of Part III, “nothing in the previous chapters depends at all on [these philosophical issues].”
The review proceeds as follows. I begin, in Section 1, with a general discussion of hedonic psychology, its methodology and main results; tracking chapter 1 of the book. Section 2 then turns to well-being analysis – BBM’s application of hedonic psychology to the evaluation of regulation; tracking chapters 2 and 3 of the book. Section 3 considers BBM’s two main non-regulatory applications – criminal justice and punishment, and civil litigation and settlement; tracking chapters 4 and 5 of the book. Section 4 offers a brief conclusion.
In my review of the different chapters, I try to provide a summary and synthesis of the main findings – for the benefit of those readers who are not versed in this literature. The more critical comments (that inevitably follow) are not intended to detract from my great admiration for what BBM have accomplished with this Happiness and the Law.
The book begins with an introduction to hedonic psychology. How can we measure happiness and the effect of different events, and policies, on happiness? The basic answer is: self-reporting – we ask people how they feel at different moments during their day, or during their life. Different instruments have been used to measure this “experienced utility.” The “gold standard,” according to BBM, is the Experience Sampling Method (ESM): The subject’s smartphone beeps randomly throughout the day asking her to report what she is doing and how she feels about it. Another influential approach, the Day Reconstruction Method (DRM) uses daily diary entries. The oldest method is the life satisfaction survey, asking individuals: “All things considered, how satisfied with your life are you these days?” These different instruments vary with respect to the resolution – in terms of time and event – of the data they produce.
BBM acknowledge that the self-reporting methodology is not without problems but argue, persuasively, that the science of happiness has evolved to a stage where the data produced is reliable. In technical terms, the different measures of happiness, or Subjective Well Being (SWB), have been scientifically validated. In essence, this means that the measures produce logical results: people report being happier when good things happen and less happy when bad things happen. Of course, we don’t need hedonic psychology to know that. The idea is to validate the instrument with these common sense understandings of what makes people happy, and then rely on the instrument even when it produces results that might initially seem counter-intuitive. (Another use of the validated instrument is in quantifying the common sense understandings. More on this below.)
So what are the main counter-intuitive results? First, money matters much less than people think. Indeed, BBM cite evidence that money doesn’t matter at all beyond $75,000 a year (p. 16; in other places, BBM cite evidence that money can matter, even beyond an annual income of $75,000, but that the effects of additional income is minimal). Second, disability, e.g., losing an arm in an accident, doesn’t matter very much (p. 17). These results are striking. Do we believe them?
On the irrelevance of money, BBM reject the economist dogma that people spend money in ways that make them happy (pp. 53-54). I am an economist, and so I am probably a victim of this dogma. I am, however, willing to accept that people buy things that don’t really make them happy. But they also buy things that do make them happy, right? Increasing your annual income from $75,000 to $100,000, say, could mean a 2-bedroom apartment, rather than a 1-bedroom apartment (depending on where you live). It could allow you to pay for piano lessons for your child. It might enable you to dine out once a week or treat yourself, and your family, to an occasional movie or to the opera. Are these irrelevant to happiness?
And what about disability? BBM cite evidence that people with limb deficiencies, spinal cord injuries, burns and other disabilities report levels of well-being similar to those of healthy controls. (BBM are careful to note that this “irrelevance of disability” does not hold for all types of disability.) How can this be?
Hedonic psychology’s answer is adaptation. We have a powerful “psychological immune system” and it allows us to adapt even to severe disability. Similarly, we adapt to having more money. BBM’s discussion of adaptation reminds me of the power of reference points in Kahneman and Tverskey’s Prospect Theory. These shocks – to our income or to our physical health – affect our utility, or happiness. But then our reference point recalibrates: the high income or poor health becomes the new normal. As BBM write: “People come to understand themselves as rich people (if they have won the lottery) or widows, and they focus less and less on the changes that brought them to that state.” (p. 20)
Adaptation, or a new reference point, explains the data. It makes sense of self-reported SWB levels that are indifferent to income or disability. But it doesn’t necessarily mean that these data are normatively relevant or, to be more careful, it doesn’t mean that these data capture everything that is normatively relevant. If we ask a person who received a $25,000 raise one year ago (to allow for adaptation), whether she would be willing to forgo the extra income – will she say ‘sure’? If we ask a person who lost a limb one year ago, whether he would have preferred to have that lost limb – will he say ‘no, thanks”? Is the predicted refusal to forgo the raise or the preference for having all four limbs a normatively-irrelevant artifact of people’s misperceptions about what makes them happy? Or is the methodology missing something important? Could it be that the SWB self-reports provide important measurements of happiness vis-à-vis an adaptable reference point, while missing important differences between two reference points?
Kahneman and Deaton, in a study cited by BBM, distinguish between two aspects of subjective well-being: emotional well-being or “the emotional quality of an individual's everyday experience” [this is what BBM refer to as happiness] and life evaluation or “the thoughts that people have about their life when they think about it.” They find that “high income buys life satisfaction but not happiness.” Perhaps emotional well-being, or happiness, takes the reference point as given, whereas life satisfaction captures differences between reference points. Kahneman and Deaton see both as components of SWB.
To be clear, BBM convinced me that many people (and some economists) overestimate the impact of money and of disability on their SWB, and that these findings are relevant for policy. It is the apparent strength of the hedonic psychology results – money doesn’t matter (beyond $75,000) and three limbs are as good as four – that weaken their persuasive power.
2. Well-Being Analysis
Perhaps the main policy move in the book happens in chapters 2 and 3, where BBM propose to replace conventional Cost Benefit Analysis (CBA) with Well-Being Analysis (WBA). Rather than relying on monetary proxies for well-being (usually willingness-to-pay) as CBA does, WBA uses direct measures of well-being informed by hedonic psychology. WBA is based on the notion of a Well-Being Unit, or WBU (p. 34). A policy that increases the SWB of one person by one unit, as measured on a scale of 1-10, for a period of 1 year produces 1 WBU. A policy that increases the average SWB of a million people by one unit, and this benefit is expected to last for 10 years, produces 10 million WBUs.
A few “technical” details of WBA should be noted. First, regulation generally entails monetary implementation costs. WBA converts implementation costs from dollars to WBUs, by assuming that these costs are distributed across a large number of individuals (e.g., shareholders, employees, consumers) and using the data described in Section 1 above to link money and happiness at the individual level. Second, for any regulation that reduces the risk of death, WBA (like CBA) must measure the value of a life. BBM use the average WBU of a living person and multiply it by the average number of years saved by the regulation. In their discussion of these issues, on p. 42, BBM neglect to consider the anxiety that precedes death (- from cancer, in their example) and the unhappiness it causes, but later in the book argue they argue that WBA takes this anxiety into account (p. 82).
Third, many regulations generate benefits and costs that are spread over time. CBA has struggled with the proper aggregation of temporally disparate benefits and costs. The standard approach is to discount future benefits and costs into the present, but identifying the appropriate discount rate has proven contentious. BBM identify two sets of reasons for discounting: (i) those related to money – we discount future monetary benefits and costs, because the value of money changes over time; (ii) those unrelated to money – we discount future benefits and costs, because the future is uncertain or because people prefer to be happy now as compared to being happy later. BBM note, correctly, that the first set of issues is irrelevant for WBA and, therefore, suggest that discounting is less of a problem for WBA (pp. 88-91).
WBA is attractive, especially given the limitations of CBA. Indeed, BBM’s strategy is to argue that, whatever shortcoming WBA has, CBA has similar or more severe problems. The logic works, but is somewhat unsatisfying.
3. Beyond Regulation
BBM argue, convincingly, that hedonic psychology is relevant to legal policy beyond WBA and the assessment of regulation. They offer two applications: (1) criminal justice, and (2) civil litigation and settlement. However, BBM’s analysis of these applications is relatively thin, especially in comparison to their account of regulation and WBA.
3.1 Criminal Justice
BBM’s basic point, in the Criminal Justice application, seems absolutely right: Measuring criminals’ hedonic response to punishment is critical for both deterrence-based and retribution-based theories of punishment. BBM report two main findings: (1) People adapt to prison; (2) Having been in prison, even for a short period, has substantial negative post-release effects. A major implication (of both findings) is that different lengths of prison sentences achieve less differentiation, in hedonic effect, than we might otherwise believe. Since such differentiation is important – for both deterrence and retribution – this is a problem.
These findings are clearly important from a legal policy perspective. But they are not new. BBM acknowledge this, when they write: “Of course, it is not news that convicted criminals face reduced opportunities – particularly economic ones – after release from prison.” (p. 110) So what does hedonic psychology add? Does it help quantify these effects? BBM have little to say here.
BBM make two additional points: First, they emphasize that because of unanticipated adaptation, we get “deterrence at reduced cost” (pp. 104-106). In essence, potential offenders perceive incarceration as more painful than it really is. This means that a given prison sentence, which entails a given cost, has a larger deterrence effect due to potential offenders’ failure to anticipated adaptation. Put differently, and with a slightly less positive spin, adaptation reduces the deterrence effect of incarceration, but this reduction is mitigated by the limited anticipation of adaptation. The behavioral economics literature has identified other effects of misperception that are somewhat related. In particular, it has been noted that deterrence can be increased at a low cost by raising the perceived probability of detection, without taking the much more costly step of increasing the actual probability of detection. (See: Christine Jolls, “On Law Enforcement with Boundedly Rational Actors,” in Francesco Parisi and Vernon L. Smith (eds.), The Law and Economics of Irrational Behavior, p. 268 (Stanford University Press, 2005).)
Second, BBM note that adaptation reduces the deterrence effects not only of incarceration, but also of monetary sanctions. Specifically, higher fines will result in a smaller deterrence effect than we might otherwise think. Again, it is interesting to note that a similar result can be derived from a different, non-hedonic behavioral model: Prospect Theory with its convex utility function in the loss frame also implies reduced marginal deterrence from increased fines.
3.2 Civil Litigation, and Settlement
BBM’s application of hedonic psychology to pre-trial settlement negotiations produces the following main prediction: For adaptable injuries, tort plaintiffs’ reservation prices will go down over time, between injury/filing and trial, thus raising the prospect of settlement. This suggests that long delays before trial can be good for settlement. From a policy perspective, the counterintuitive implication is that investing resources in reducing backlog and minimizing delay before trial can be harmful – it can reduce the likelihood of settlement and increase the number of costly trials.
BBM make another basic, yet important point, about the relevance of hedonic psychology to civil litigation (although this point is made in Ch. 1, not in Ch. 5). They argue convincingly that when a court determines damages, e.g., in a tort case, the amount of money that would compensate the plaintiff for harm suffered should be informed by hedonic psychology.
I conclude with a final thought. There is a basic distinction between the following two categories of results from hedonic psychology: (i) Results that provide information about preferences – what makes us happy; hedonic psychology offers a richer understanding of preferences, of the utility function. And (ii) Results about perceptions of preferences, specifically the failure to anticipate hedonic adaptation. BBM are very clear about this distinction, although it may deserve some additional emphasis. Focusing on perceptions of preferences, it may also be interesting to compare underestimation of adaptation with other types of misperceived preferences, e.g., misperception of time preferences (as in the model of the naïve hyperbolic discounter).
BBM’s Happiness and the Law is a terrific book. It provides an accessible, yet thorough introduction to hedonic psychology and its wide-ranging applications in legal policymaking. The book should be of great interest to any student of law and policy. And, perhaps more importantly, it should inform policymakers – legislators, regulators and judges – as they make policy choices that inevitably influence our happiness.
OREN BAR-GILL is Professor of Law at Harvard University.