Philip Booth's (ed.) ... and the Pursuit of Happiness: Wellbeing and the Role of Government.
The excellent anthology under review provides a useful survey of recent research by economists on well-being and government. (1) It contains original essays by some of the most eminent researchers in the field of happiness economics. The volume is edited by professor of business Philip Booth. It is a good resource for moral and political philosophers who want a solid understanding of the results of this research (as well as its limitations).
After a forward and an introductory chapter, the essays in the anthology are gathered into three parts. The first part consists of essays on whether economists should focus on the traditional economists' measure of well-being, the Gross Domestic Product (GDP), or some new measure based on surveys of national happiness--perhaps some kind of General Well-Being (GWB). The second set of essays deals with the issue of happiness and governmental size and policy. The third set of essays explores the question of whether national happiness is delivered more reliably by governmental activity or the free market.
The foreword is by economist Mark Littlefield, who sketches out the importance of the concept of happiness for both philosophy and economics. It is obvious that utilitarianism--which equates moral rightness with the maximization of good results for everyone involved--takes the concept of happiness seriously. However, Littlefield doesn't note that the notion of maximizing happiness is of equal importance in ethical egoism--the other main variety of consequentialism--as well as in Aristotelian virtue ethics. Moreover, contemporary deontologists typically allow that maximizing good results (happiness, generally) is relevant to questions about right action, even if it is not the whole (or even the main) story. Littlefield observes that some recent economists, public policy experts, and politicians have used happiness research to argue that the traditional economic focus on GDP is misplaced; instead, the field should focus on some national measure of happiness.
Editor Philip Booth takes up these issues further in his introductory chapter, noting that British Prime Minister David Cameron urged in a 2006 speech that "[i]t's time we focused not just on GDP, but on GWB--general well-being" (p. 25). Booth notes that this led Paul Omerod and Helen Johns to produce an earlier anthology (2) critically analyzing the happiness literature. Booth notes that Cameron cited that earlier book in later speeches (in which he seemed to be more skeptical of the notion of GWB), but in 2011 started expressing attraction to the idea again, stating that "[measures of well-being] could ... lead to government policy that is more focused not just on the bottom line, but on those things that make life worthwhile" (p. 5). Booth adds that this confusion in politicians grows right out of the confusion in the economic literature on happiness, and that the papers in this anthology aim to clarify these issues.
After briefly reviewing the contributions, Booth concludes with a Bishop Butler sort of observation: a government will have the best chance of promoting well-being if it doesn't make that it objective, just as it will have the best chance of maximizing GDP if it doesn't make that its explicit goal. Instead, a government will maximize both if it focuses on maximizing the freedom of its citizens and economy.
The second chapter is by economist Paul Omerod. Omerod's focus is on disputing the notion that government policy should aim at maximizing happiness (or "well-being," which he argues is used synonymously with happiness). He notes that the U.K.'s Office for National Statistics began to collect data on self-reported happiness in 2011, pursuant to a push by Prime Minister Cameron to do so the previous year.
Omerod argues that the driving force behind this push to collect happiness data is the thought by some social scientists and policy advocates that governments have until now focused too narrowly on maximizing GDP growth, which is (they hold) too narrow an indicator of national well-being. He quotes Derek Bok as a typical case, who wrote, "People are surprisingly bad judges of what makes them happy" (p. 40). This, Omerod suggests, is a serious mistake.
First, Omerod reviews the recent elections in both the U.K. and U.S., and notes that while economic growth and prosperity matter to voters, so do a variety of other things, such as the levels of immigration and crime. And he concedes the shortcomings of GDP as a measure of social well-being, which were recognized even by Simon Kuznets, its originator. Indeed, as far back as the 1970s, economists like Bill Nordhous and James Tobin suggested incorporating measures of environmental health into GDP estimates.
However, Omerod points out that the well-being movement doesn't just want to modify or clarify GDP, but instead, replace it by measures of the happiness of a country's citizens. He adds that one motivation for this demand to replace GDP as the primary measure of national success is that surveys of individual happiness over long periods of time seem to show that the average level of happiness tends to remain flat while GDP rises significantly.
This phenomenon has a name: the Easterlin Paradox, named after economist Richard Easterlin. He concluded (after first studying Japanese data on self-reported happiness over the decades of its rapid post-World War II economic expansion) that after an initial rise, happiness levels off despite the increasing per capita income. This result is puzzling, because under standard economic theory happiness consists in having one's preferences met, and money is the prime mechanism for getting what one prefers. Easterlin and others drew the conclusion that it is not wealth that causes happiness, but equality of wealth that does. From this paradox, progressive economists (such as Richard Layard) draw out policy implications that conveniently fit their preexisting worldview. These include various schemes for wealth redistribution and steeply progressive income taxes.
Omerod questions the logic of the policy argument. He notes by analogy that during the past forty years in the U.K., while happiness levels remained flat, government spending rose dramatically (by 60%, in fact). Yet few progressive economists would argue that because of this, government spending should not be increased because it doesn't make the people happier. Similarly, during that period, inequality in the U.K. (as measured by the Gini index) rose dramatically, again while happiness remained constant, which would seem to indicate that we should not try to correct inequality of income, since it hasn't negatively correlated with happiness level.
Again, Omerod notes that U.S. data show that during the roughly past forty years, while happiness levels remained flat, life-spans increased significantly and gender pay inequality dropped sharply. But few progressive economists--or anyone else--would conclude that longevity and women's equality are irrelevant to happiness.
Omerod's conclusion is that happiness data as it is now measured is flawed. He notes that happiness is usually measured by surveying people and asking them to rank their happiness on a small, discrete scale (say, "1" for "not happy," "2" for "fairly happy," and "3" for "very happy"), and averaged over the whole population. That is empirically crude, in that it only records increases or decreases in happiness when they are relatively large, and beyond the top number, no increases are able to be recorded.
To show how crude this data measurement technique is, Omerod notes that if 1% of the population were to move a step up on this scale, the average happiness only bumps up by 0.01. Looking at the U.S., surveys that use a 3-point scale show an average happiness score of 2.2. It would take more than 20% of the population moving up a full point (very difficult to achieve, obviously) to see the average go up a modest 10%.
Moreover, while GDP has no theoretical upper limit (and in fact has risen significantly over the last two centuries), all existing happiness measures are on point scales that each have a maximum value, and so are incapable by construction of showing persistent long-term trends. This raises tricky mathematical issues of interpreting putative correlations between GDP and average happiness scores. (3)
Omerod then briefly reviews more recent (and more sophisticated) work on happiness. He notes that recent research indicates that the positive correlation between happiness and wealth has no upper limit. He doesn't name the two scholars who have done yeoman's work to refute the Easterlin Paradox--Betsey Stevenson and Justin Wolfers--but they are well represented in this book.
Omerod also sketches the recent work by Daniel Kahneman (one of only a few non-economists to win a Nobel Prize in economics) and Angus Deaton, in which they first disambiguate the concept of happiness, and then use it to explore a database of nearly a half million responses by Americans to various happiness queries. (4) Kahneman and Deaton distinguish the concept of life satisfaction, which is one's feeling about how well one's life is turning out, from emotional well-being, which is one's feeling about the happiness of one's current life (i.e., how much joy, anger, or sadness one is experiencing). They showed that while life satisfaction is clearly strongly positively correlated with income, emotional well-being isn't.
Omerod concludes by noting that these recent developments in happiness data studies haven't stopped progressive policy analysts from trying to use happiness measures to replace GDP and justify progressive policy prescriptions, even though that same data can justify conservative policy prescriptions. (The data show, for example, that being married and being religious both strongly positively correlate with happiness. These facts could be used to support policies intended to encourage marriage and religious observance, but progressive policy advocates never draw that inference.)
Nor has the past failure of the econometric uses of GDP to successfully predict--much less guide policies which forestall--recessions and rises in unemployment deterred these progressive policy analysts from attempting to use the happiness index. This leads Omerod to observe:
Of course, the fact that economics has made little or no progress in its ability to predict and control the macro-economy does not necessarily mean that the same fate awaits the happiness index and its devotees. Changes in both real GDP and happiness over time share a deep common feature, however. Namely, that they are, across the Western world as a whole, scarcely indistinguishable from purely random series. There is a small amount of pattern, of potential information, in the US GDP data, but it is small. And, more generally, these data series are dominated by random noise rather than by any consistent signal. (p. 55)
I will return to this skeptical point below.
The third chapter is by economists Daniel Sacks, Betsey Stevenson, and Justin Wolfers. It is an update of the classic 2008 paper (5) by Stevenson and Wolfers that so rocked the Easterlin Paradox literature that it made news in The New York Times. (6)
Sacks/Stevenson/Wolfers used data sets of wealth and reported subjective well-being that together covered 140 countries and almost all of the world's population. They note that Easterlin had argued in a number of influential papers over thirty years that while within a country, wealthier individuals report higher levels of happiness on average than do poorer ones, across countries he found no statistical link between per capita wealth and average happiness. This is the paradox. The conclusion Easterlin and others draw is that it is not absolute, but rather, relative wealth that determines an individual's well-being. So it would seem that policies aimed at increasing national wealth/GDP--even if successful--would fail to make people happier on average.
Sacks/Stevenson/Wolfers also note two other views advanced by some economists that reinforce the Easterlin findings. First, some hold that subjective well-being adjusts to circumstances over time. This would seem to suggest that as wealth in a society increases, people adapt to their higher income with just higher expectations (so experience no greater feeling of well being). (This is often called "the hedonic treadmill.") Furthermore, some argue that individuals have a point of satiation beyond which more wealth will not bring more happiness.
Sacks/Stevenson/Wolfers refute these claims with three major findings. First, universally, in all 140 countries they studied, wealthier people report higher levels of satisfaction with their lives. Indeed, universally, it appears that the relation between well-being and income is functionally simple: the rise in well-being is correlated with the rise in the logarithm of income (as opposed to merely the increase in income). So while going from $500 to $750 in income will correlate with a 50% increase in average reported well-being, it would take an increase of $250,000 to achieve the same average increase starting with a base of $500,000.
Second, the authors found that there is a statistically significant positive correlation between the average level of well-being of a country and the log of its GDP. This helps resolve the paradox: the authors argue that Easterlin's data sets were too small (and in the case of the Japanese data, failed to take into account a change in the survey questions) to show a correlation; he took absence of evidence of a correlation to be evidence of the absence of a correlation. Moreover, to the claim of satiation, Sacks/Stevenson/Wolfers argue that their data analysis shows no upper limit beyond which the correlation fails to hold.
Third, the authors offer analyses of time-series data sets that span twenty years and dozens of countries. They show that, over time, increasing economic growth correlates significantly with increasing well-being. This further refutes the paradox, as well as the notion that people's expectations adapt to rising incomes.
Finally, Sacks/Stevenson/Wolfers show that their results are robust even when other measures of subjective well-being are employed. This includes reported happiness as well as "effect-specific measures of subjective well-being," such as feelings of enjoyment or love, and absence of pain. Looking at a variety of measures is important, for as the authors note (p. 63), the Easterlin literature has tended to look only at life-satisfaction data (which is often simply characterized as "happiness" data).
The authors make another important point, one that ties in the happiness research done by economists with that done by psychologists. They note that while some economists have criticized data consisting of self-reported feelings of life satisfaction as being overly subjective, in fact, psychologists (such as Kahneman and Diener) have shown that there is a strong correlation between that "subjective data" and objective anchors such as heart rate, health, and sleep quality, as well as independently gathered reports by friends. Furthermore, this data is stable over both time and retesting.
The Sacks/Stevenson/Wolfers piece strikes me as the most important one of the anthology, both for the wealth and quality of its data and robustness of its results, as well as the fact that it takes on the Easterlin paradox directly. However, one aspect of their analysis is questionable. The authors seem to think that there are two major explanations for within-country correlation between income and life satisfaction, both hedonistic ones (p. 70). Specifically, either more income causes more satisfaction (because it allows more purchases of leisure, health care, good food, and so on), or people are more concerned with how their income compare with some fixed point (such as the average income in the country or people's previous income). Their data analyses rule out the second explanation, and so they favor the first. This leads them to suggest that "transferring a given amount of money from rich to poor countries could raise life satisfaction, because $100 is a larger percentage of income in poor countries than rich countries" (p. 3). I will return to this point.
The fourth chapter is by historian Christopher Snowden. Snowden acknowledges that some scholars have argued that Easterlin was wrong in holding that, in recent decades, rising GDP has not been correlated with a rise in happiness. However, he takes that lack of correlation as a fact, and examines the question about how much comfort that should give critics of the free market.
Specifically, Snowden takes up the issues the Easterlin paradox purports to establish, namely, the notion that there is a correlation between equality of wealth and level of happiness. He nicely frames the issue:
It has been suggested that people living in 'more equal' societies are happier than those who live in countries where the gap between rich and poor is wider. If so, it would mean that wealth redistribution is more important than wealth creation. By a happy coincidence, that is exactly what those who make such claims have always believed. (p. 100)
And indeed, the idea that people are happier in societies with more equal distributions of wealth seems to be a commonly held belief. But Snowden argues forcefully that this idea is simply a canard.
Snowden points out that, in fact, few studies have been done actually comparing inequality rates and happiness levels over time. Certainly, a scatterplot of happiness levels versus inequality levels (as measured by the Gini coefficient) by country shows no apparent correlation. High equality societies like Sweden and Norway show high national levels of happiness, but so do high inequality countries such as Singapore and the U.S. And, as Snowden notes, Arthur Brooks found earlier that the U.S. happiness level was essential flat from 1972 to 2004, while the Gini coefficient rose by nearly half.
If we take suicide as a proxy for unhappiness, again Snowden (citing a number of relevant studies) shows that there is no positive correlation between inequality and suicide. If anything, there is a negative association: suicide rates tend to be higher in countries with lower inequality.
Snowden then reviews several studies that show that differences in the relation between happiness and inequality vary by country. While some poor are comfortable with inequality (for example, in America), in Europe the reverse is true. This would seem to indicate that "perceptions of fairness and social mobility are more important than inequality itself' (p. 105).
After surveying a large number of studies, Snowden argues that the majority of them show no significant correlation (either positive or negative) between happiness and inequality. There is essentially no evidence supporting the claim that inequality leads to unhappiness in society. Even Richard Layard finally conceded this point. Layard is reduced to making the weaker claim that the poor are made happier than are the rich for every extra dollar gained, and then concluding that this justifies income redistribution measures--a point even echoed, as we saw above, by Sacks/Stevenson/Wolfers. To this, Snowden has an astute reply: the fact that a given amount of new income makes the poor happier than the rich is only a reason to conclude that we should focus on making the poor richer more quickly. It is not in the least a reason to conclude that economies with income redistribution schemes will help the poor.
Snowden then takes up the issue about how happiness relates to relative income. He points out that many writers--including many who should know better--conflate the concept of "relative income" (or what I call "local income inequality") and "income inequality" (what I call "global income inequality"). But income inequality is a notion that refers to the entire distribution of a country's income, whereas relative income refers to the income gap between specific individuals or groups.
The difference between these concepts is huge. As a lecturer, I might well be more disturbed by the disparity between my pay and that of a senior professor than between my pay and that of an NFL quarterback. And Snowden's literature search indicates that most studies have shown that happiness levels are certainly affected by relative income inequality, though studies vary in their assessment of the effect. Snowden cites at the lower end Layard's (7) estimate that one's unhappiness at a personal loss of 10 to 30 cents is equivalent to his unhappiness at seeing his neighbor gain $1, while other researchers (8) put the equivalence at one dollar to one dollar.
Snowden attributes this to two factors. One is that seeing people in my own "reference group" increase their wealth increases my own expectations, that is, income levels I thought were only for the rich are in fact open to me. The other is "status anxiety," that is, most people prefer earning less money if means they still earn more than the neighbors are earning.
Snowden cites here a study (9) where students at the Harvard School of Public Health were polled on whether they prefer to earn $50,000 when everyone else earns $25,000, or $100,000 when everyone else earns $200,000. Half the students preferred the first to the second option. A subsequent survey showed that those same students--public health students, recall--when asked if they would prefer having two painful dental operations while others endured four, to undergoing one painful procedure when everyone else endured none, nearly one-fifth said they would prefer the first option.
So if feelings of status anxiety are held to buttress government policies aimed at equalizing incomes, would it not also support government policies to equalize pain?
In short, Snowden avers, there is no empirical proof that people in more equal societies are happier than those in less equal ones, nor is there any compelling analytical reason they should. He ends by quoting Diener and Biswas-Diener: "Thus, our advice is to avoid poverty, live in a rich country, and focus on goals other than material wealth." (10)
In the second part of the book, the chapters deal with the topic of happiness and government intervention. In Chapter 5, economist J. R. Shackleton looks at happiness literature as it bears on life in the workplace. In particular, he focuses on the question of whether recent findings on happiness economics justify increased governmental regulation of personnel practices in private industry.
He starts by noting that research documents what appears to be common sense, namely, that over the last century at least, work has become much safer, less exhausting, cleaner, more well-compensated, and healthier. By any objective measure, the quality of jobs has risen over time.
But, Shackleton notes, psychologists have focused on workers' mental states. This psychological literature tends to distinguish two basic senses of job-related well-being: "hedonic" well-being, meaning ongoing positive feelings (of pleasure, say), and "eudemonic" well-being, involving senses of purpose, meaning, personal growth, and social respect.
Such studies are based on surveys done using Likert scales for measuring job satisfaction (which range from "very dissatisfied" to "very satisfied"). They show various correlations, such as levels of job-satisfaction with higher pay, smaller workplaces, higher job security, higher job autonomy, less tight deadlines, and less performance monitoring. This, Shackleton notes, has led a number of researchers to posit a Manichean division between "good" jobs, which involve the positive qualities mentioned above, and "bad" jobs, which lack those qualities. Bad jobs are purported to be the jobs that cause excessive stress, potentially leading to health problems.
However, he notes, this view is oversimplified for a number of reasons. First, research shows that differences in self-reported job satisfaction often grow out of personal characteristics (such as age, gender, and ethnicity), personality traits, individual health factors, and even the business cycle (employees report higher happiness during economic booms).
More importantly, employing the notion of "compensating differentials," Shackleton argues that in a free labor market, any negative job factors will likely be compensated for by higher pay, and positive job factors will by lower pay, and he reviews some studies that document this. Pay differentials would explain why "bad jobs" characteristics (such as long hours) don't seem to reduce reported job satisfaction. So the idea of restricting work hours in hopes of increasing job satisfaction (such as France did) is dubious, as is the contrary notion of increasing minimum wages (which can kill off jobs many workers might prefer, such as unpaid or low wage internships for students).
Shackleton briefly reviews the evidence on the issue of whether the case can be made for employers to adopt practices explicitly aimed at increasing employee job satisfaction (on the basis of improved productivity), and argues that to the extent such measures work, they are likely already implemented. He concludes by reviewing whether the employment happiness literature lends much support for increased regulation of business, and again argues that it doesn't, especially considering how much unemployment is shown to lower happiness levels.
Chapter 6, by economist Christian Bjornskov, examines the relation between average well-being and the size and scope of government. Do people report higher levels of life satisfaction in countries where the government "does more" for them?
He starts by noting two reasons why economists and political scientists have traditionally assumed that there must be a positive correlation between government spending and happiness. First is the "standard working assumption" that politicians and government employees are kindly and disinterested purveyors of public goods (p. 160). Second is the "classical welfare economics" assumption that since the increase in well-being the poor experience from the gain of a given amount of money exceeds the decrease in well-being the rich experience from the loss of it, redistribution schemes will therefore increase national well-being (p. 161). (He attributes the second assumption to socialist economist Abba Lerner; he should have noted that the first is a Hegelian one.)
In reply, Bjornskov points to the massive amount of empirical work done over the last half-century in public-choice economics, which refutes the notion that the actors in government (politicians and bureaucrats) are disinterested and benevolent. He concludes that any sound consequentialist argument for increasing the size and scope of government will have to show that there is a positive correlation between government intervention and well-being.
Certainly, he reminds us, there are theoretical reasons why such a correlation is apt to be nonexistent. There is Friedrich Hayek's information problem: How could politicians know what specific people or groups want? And assuming the politicians could even approximately know what the "average" citizen wants, there is still a "heterogeneity" problem: a large number of citizens will be over-provided with specific public goods, while others will be under-provided. Citing the work of Gordon Tullock, he argues that because consensus in a democracy on spending is typically reached by logrolling, any redistribution of public goods will be driven by special interest groups that will gain a disproportionate share of the redistribution.
Bjornskov gives a nice review of what factors have been shown to correlate with greater national average well-being (unemployment, quality and fairness of governmental institutions, degree of religiosity and social trust in the population), most or all of which seem not to be a result of government action. He then does a literature survey of the empirical studies of the relation between the size and scope of government and well-being (much of which he and his colleagues have done). Some research has found no relation between government size and average well-being, and some has found a strong negative correlation between governmental spending and average wellbeing--especially (and ironically) among the poorest citizens. Nor is there a positive correlation between redistributionist governmental policies and average well-being.
Bjornskov concludes by putting together his research and that of Sacks/Stevenson/Wolfers. If large, activist government doesn't result in increased well-being, but economic growth does, and if (as other studies he cites indicate) large, activist government results in lower economic growth, then the happiness economics research would seem in fact to support Adam Smith's classic public policy prescription of "peace, easy taxes, and a tolerable administration of justice."
The third part of the anthology examines whether happiness is delivered by government or the free market. In Chapter 9, law professor Marc De Vos critically examines the use of happiness measures to drive public policies. He starts by observing that the nature of happiness and how to achieve it were central questions of philosophy from the ancient Greeks onward, and the utilitarians made maximizing people's happiness the foundation of public policy. But in the setting up of the welfare state, the goal was traditionally a materialistic one: insuring populations against material privation (as opposed to making them happy). Yet more recently, the pursuit of happiness has been used to argue for expansion of the welfare state.
De Vos agrees that the recent research on happiness economics is valuable for several reasons. It "adds a quality dimension" to other quantitative economic measures, such as GDP growth, unemployment rates, educational test scores, and poverty rates (p. 182). It broadens economic understanding of human action beyond the view of people as rational egoists. And personal happiness is important as an issue of public concern.
But the author is skeptical that this means that the promotion of happiness is an appropriate area of policy making, much less that it should be the goal of policy. He has several reasons for this skepticism.
First, De Vos doubts that we have or will ever have reliable happiness data, which he terms "crude and unsophisticated" (p. 185). He lists sixteen different survey questions aimed at eliciting subjective estimates of a person's happiness; they range widely, including the Gallup World Poll question that asks the respondent to imagine a ladder with the rungs representing a successively better life. He remarks that this "data" is little better than the "data" of the psychoanalyst's therapy sessions, hopelessly subjective and relative to our personal biases (p. 186).
Second, he notes that with happiness data, there is always a problem in disaggregating causality from mere correlation.
Third, happiness data are ephemeral "snapshots" of people's feelings at a given time. Here De Vos makes a good point: if we are to drop economic growth as the focus of governmental policy because people become accustomed to higher wealth, why focus on happiness, when people also adapt to misfortune (and revert to their prior levels of happiness)?
Fourth, he notes the distinction between eudemonic and hedonic happiness, and argues that happiness scholars are too focused on measures of hedonic happiness. This, he avers, runs the risk of leading to policies that focus on "instant and often simple gratification,"--which is the very thing happiness proponents accuse traditional economists who focus on GDP of doing (p. 192).
Here De Vos makes a point one wishes more economists who advocate policies would acknowledge: in making a policy recommendation based on happiness data, one must make value judgments about the type of happiness (merely hedonic, or actually eudemonic) that one wants to promote. I would generalize the point: in any move beyond empirical economics to the realm of public policy, one inevitably crosses the "is/ought" divide (i.e., commits the naturalistic fallacy).
In Chapter 8, economists Peter Boettke and Christopher Coyne bring insights from Austrian and public-choice economics to bear on the happiness debate. Boettke/Coyne define the Easterlin paradox narrowly as the alleged phenomenon that as incomes rise, beyond a certain point, average happiness remains flat. The two most common explanations offered are that people judge their wealth by relative rather than absolute terms, and the hedonic treadmill effect (which holds that any increase in wealth leads to an initial uptick in happiness, but it returns to its prior level as the person becomes accustomed to it). The authors note that progressive policy advocates who accept these views call for similar policies--especially steeply progressive taxes on income and heavy taxes of so-called "luxury goods"--because those progressive advocates view the greater wealth of some as a negative externality on everyone else.
But Boettke/Coyne pose several objections to happiness economics and the policy prescriptions it is used to justify. Regarding happiness economics, the authors (following Will Wilkinson (11)) note that the literature has at least four different elements of "happiness": life satisfaction, hedonic experiential quality, some third state besides life satisfaction and hedonic quality, and well-being (p. 208). Worse yet, people don't even agree what these elements are, and researchers seldom try to make clear which concept they are attempting to measure in their research.
Moreover, Boettke/Coyne argue that the data on happiness are context--and time-sensitive, making cross-respondent, cross-country, and cross-time comparisons dubious.
They also question the notion that status is a zero-sum good, that is, that the more status one person has, the less other people have. This is used by progressive policy advocates to urge the adoption of policies that lessen wealth inequality, since it causes status envy and thus unhappiness. The authors reply that this assumes that the amount of "status" is somehow fixed. But in a modern economy, old forms of status may lessen or disappear, while new and more varied ones emerge.
Regarding the policy proposals, Boettke/Coyne make a couple of common-sense points. Even if we view above-average income/wealth as a kind of "pollution" to be taxed, we have to remember that to tax income is to tax productive entrepreneurship, which would deter productive activity. This would cost us in increased wealth, hence in better standards of living, education, longevity, and so on. More generally--and here the authors invoke Hayek's knowledge problem--redistribution will always result in negative unintended consequences.
They also observe that in taxing productive work to provide public goods, it is unlikely the government will deliver these goods in an optimal manner. Moreover, if the hedonic treadmill thesis is correct, and people adjust to an increase in material wealth to return to their prior level of happiness, that would surely apply to any public goods as well.
Chapter 9, by economist and historian of philosophy Pedro Schwartz, takes on utilitarianism. Schwartz provides a useful survey of Layard's (and Bentham's) ethical philosophy. Schwartz argues three points. First, Layard's utilitarianism elevates (hedonic) happiness to the supreme value, which unduly narrows the field of normative economics. Second, by making happiness a public good, it confuses negative and positive rights and thus guarantees that civil rights will be violated. Third, this philosophy confuses the morality of small groups with that of society as a whole. Contentment is fine as the goal of small groups, such as the family, but a large society requires competition to flourish.
The author makes a couple of trenchant criticisms along the way. One is that Layard's sort of happiness philosophy "elevates envy to the category of a public virtue" (p. 236). Another is that if we really want to maximize the public happiness, we should enact cruel and unusual punishments, entirely outlaw immigration, and totally end free trade, for people seem to want these things. (Yes, in the long run, the nation would be impoverished, but income equality would be achieved!)
I think that Schwartz's criticisms make sense only if you take happiness to be hedonic. If we take happiness to be eudemonic, utilitarianism is not so easily dismissed, which may be why Mill moved away from Bentham's sort of hedonism and later utilitarians (such as G. E. Moore) gave it up entirely.
One drawback of this estimable anthology is that there is a certain tension between the strains of criticism offered in the selections. One strand (represented by Omerod, De Vos, and Boettke/Coyne) seems to hold that the happiness data--at least as presently collected--is simply so unreliable, so statistically noisy (i.e., so governed by random results), that it is useless as an evidential basis from which policy conclusions of any stripe can be drawn. The other strain (represented by Sacks/Stevenson/Wolfers and Bjornskov) holds that while happiness data, properly analyzed, certainly is reliable enough to draw policy conclusions, the problem is that the egalitarian policy prescriptions that have been drawn from it are either not supported or outright refuted by that data.
So are we simply to reject happiness measures as being fatally noisy and thus uninformative, or embrace them and cheerfully use them to argue against egalitarian policies? This tension isn't explicitly addressed in the anthology, although Sacks/Stevenson/Wolfers touch on this obliquely when they note in passing that psychologists (including Kahneman and Diener) have shown that this subjective data strongly correlates with intersubjective reports and objective physiological measures.
However, Stevenson and Wolfers have addressed this point in an interview they did with Russ Roberts for the online journal EconTalk. (12) Wolfers points out that while happiness data seem "noisy," it is noisy in a way that correlates with the noisy GDP data. This seems difficult to explain if the two data sets were purely or mainly a result of random noise. In fact, as Wolfers notes, the observed strong correlation between income and happiness must be in reality even stronger if both data sets are as noisy as claimed.
A second critical point is that several of the readings (including the Sacks/Stevenson/Wolfers paper) seem to accept the Lerner thesis that transfers of money from the wealthier to the less wealthy will increase happiness on net. This is because gaining a given amount of money correlates with a greater increase in well-being of a poor person than the decrease of well-being correlated with the loss of that same amount of money among the rich. However, that crucially assumes that giving money to the poor will cause them to be happier. But happiness studies of lottery winners show that they typically show no permanent (i.e., long-term) increase in self-reported wellbeing despite the increase in wealth. This suggests that it is not any money that correlates with increased well-being, but earned money. An obvious explanation for this would be that it is meaningful work that produces both wealth and well-being.
Put another way, a possible explanation for the correlation between income and life satisfaction may be found in a virtue theory of happiness rather than a hedonistic one. Perhaps the reason higher income correlates with life satisfaction is that people--especially in an epistemic (i.e., knowledge-based service) economy--usually have to exercise their virtues (especially their intellectual virtues) to get higher income. That is, while income and life satisfaction correlate, it is the exercise of virtue that causes both. So merely transferring money--that is, giving unearned money--to the poor will not increase their life satisfaction, because that money is not earned virtuously. This would explain Bjomskov's findings that there is no positive correlation between redistributionist policies and happiness.
Another critical point is worth making. The editor would have done well to solicit a contribution from an economic historian to sketch out the broad historical background from which the Easterlin impulse--the impulse to denigrate the amazing economic flourishing brought about by the modern capitalistic enterprise--derives much of its appeal.
Specifically, there is a strain of anti-capitalist criticism that has been recurrent for centuries, going back to the 1950s with the work of John Kenneth Galbraith, to the mid to late 1800s with the work of Karl Marx, all the way back to the mid to late 1700s with the work of Jean-Jacques Rousseau. This is the strain of Romantic anti-materialism, which is a visceral revulsion against the focus on individual wealth (especially "consumer goods") involved in the Industrial Revolution.
This visceral disgust at industrialization (and the consumer economy it enables) is typically articulated as the thesis that growing material wealth leaves people greedy, spiritually shallow, and ultimately unhappy. In fact, a word was coined to name this alleged spiritual or moral disease: "affluenza" (a portmanteau word combining "affluence" and "influenza").
The basic tenets of Romantic anti-materialism were laid out by Rousseau in his two immensely influential early essays, Discourse on the Arts and Sciences (1750) and Discourse on the Origin of Inequality (1753). These tenets are that primitive life (i.e., life in a "state of nature") is superior to modern life, that the civilized (industrialized) life is characterized by disgusting greed and degrading inequality, and that the root of these evils is the existence of private property. (13)
I would suggest that the reason the work by Easterlin and others has attracted so much support is that Romantic anti-materialism has been a constant and compelling ideology from Rousseau's day to the present, and that this recent work seems primafacie to provide "empirical proof' of it.
In conclusion, this anthology should be of great use to ethical theorists and political philosophers. For ethical theorists, the nature of happiness is a crucial component of the field of ethics. Happiness has been philosophically analyzed for millennia, of course, but it has been only relatively recently that happiness has been the subject of intense empirical work, chiefly among economists and experimental psychologists. All ethical theorists ought to be familiar with this empirical research, and this anthology is an accessible survey of the relevant recent work done by economists.
Political philosophers will find that the arguments and data this anthology presents provide ample reason to be skeptical of proposals for redistributionist and anti-growth policies justified on the grounds that they will make people happier. There may be other more compelling reasons to call for such policies--and, then again, there may not be--but the facile invocation of the Easterlin Paradox clearly will no longer suffice.
Gary James Jason
California State University, Fullerton
(1) Philip Booth, ... and the Pursuit of Happiness: Wellbeing and the Role of Government (London: Institute of Economic Affairs, 2012).
(2) Helen Johns and Paul Omerod, Happiness, Economics, and Public Policy (London: Institute of Economic Affairs, 2007). For a review of the book, see Gary Jason, Independent Review. 14, no. 3 (Winter 2010), pp. 458-60.
(3) For investigating techniques to help establish sound statistical inferences from these sorts of data sets--"cointegration vector" techniques--Clive Granger and Robert Engle won the 2003 Nobel Prize in economics. The reader can review the Wikipedia entry for more details, accessed online at: http://en.wikipedia.org/wiki/Cointegration.
(40 Daniel Kahneman and Angus Deaton, "High Incomes Improve Evaluation of Life, but Not Emotional Well-Being," Proceedings of the National Academy of Science 107, no. 38 (2010), accessed online at: http://www.pnas.org/content/107/38/16489.full?sid=4b9b9633-871d-41eb-b0ee-9781b022c6c3.
(5) Betsey Stevenson and Justin Wolfers, "Economic Growth and Subjective Well-Being: Reassessing the Easterlin Paradox," Brookings Papers on Economic Activity 1 (2008), pp. 1-87.
(60 See David Leonhardt, "Maybe Money Does Buy Happiness After All," New York Times, April 16, 2008, accessed online at: http://www.nvtimes.com/2008/04/16/business/16leonhardt.html? r=0.
(7) Richard Layard, Happiness: Lessons from a New Science (London: Penguin, 2005), pp. 46 and 252.
(8) Snowden cites: (a) Ada Ferrer-i-Carbonell, "Income and Well-Being: An Empirical Analysis of the Comparison Income Effect," Journal of Public Economics 89 (2005), pp. 997-1019; (b) Erzo Luttmer, "Neighbors as Negatives: Relative Earnings and Well-Being," The Quarterly Journal of Economics 120, no. 3 (2005), pp. 963-1002.
(9) Sara Solnick and Robert Hemenway, "Is More Always Better?: A Survey on Positional Concerns," Journal of Economic Behavior and Organization 37 (1998), pp. 373-83.
(10) Ed Diener and Robert Biswas-Diener, "Will Money Increase Subjective Well-Being?" Social Indicators Research 37 (2009), pp. 119-54.
(11) Will Wilkinson, "In Pursuit of Happiness Research. Is It Reliable? What Does It Imply for Policy?" Cato Institute Policy Analysis no. 590 (Washington, DC: Cato Institute, 2007).
(12) See: http://www.econtalk.org/archives/2013/Stevenson_and_W.html.
(13) Rousseau did not hesitate to draw out public policy prescriptions from his ideology, prescriptions startlingly similar to those of today's progressives, most notably that income taxes should be steeply progressive and that there should be heavy taxes on "luxury goods."