In 1992, the Royal Swedish Academy of Sciences awarded its prize in economics to University of Chicago economist Gary Becker. The prize was for “having extended the domain of microeconomic analysis to a wide range of human behaviour and interaction, including nonmarket behaviour.” Becker’s work was built on the notion that areas historically reserved to other social sciences, particularly sociology, could in fact benefit from the application of economic methodologies. Beginning with his 1955 PhD dissertation, The Economics of Racial Discrimination, Becker made contributions to our understanding of crime and punishment, the household, the family, and even organ donation.

Becker’s radically simple claim—that economics has something to tell us about areas other than GDP and price-setting—changed the face of his profession. It converted economics from a narrow discipline into the wide-ranging, methodologically robust social science it is today. At the same time, it also gave economics what might, by detractors, be seen as a certain conceitedness. Whereas once the field was limited in scope, now it seemed able to offer explanations for everything. Is there really no area where economics cannot speak?

Such is the question that undergirds The Wealth of Religions, the recent book from Harvard economics professors Rachel M. McCleary and Robert J. Barro. The book is, more or less, a collection of the pair’s past academic work on the “economics of religion,” meaning the application of economic methodologies to sociological questions about religion. Of necessity, McCleary and Barro’s answer to the Beckerian question is yes, economics has something to tell us about religion. But what?

The inquiries with which the book is concerned are of three kinds. The first is how religion affects economics. That means, for example, empirically revisiting Max Weber’s famous thesis that Protestantism drove the rise of capitalism. It also means an extensive discussion of the effect Islam’s different elements—the Hajj, the Ramadan fast, and so on—have on the comparative economic success of predominantly Islamic nations. The second is how economics affects religion, including a careful examination of all of the evidence on the “secularization hypothesis,” the idea that as countries grow more rich, they become less religious. The third is general-interest religion/economics questions, such as the pair’s exploration of patterns in Catholic saint-making using economic tools.

As far as this goes, The Wealth of Religions can be an enlightening, if occasionally overly academic, read. One imagines it as a sort of Freakanomics for religion, if perhaps lacking Dubner and Levitt’s characteristic wit. In fact, this not-quite Freakanomics approach is a problem with the book: It seems unsure whether it is a compilation of serious academic work or a popularization of that same work, unclear as to its intended audience.

At the same time, The Wealth of Religions invites us, almost begs us, to ask a question not about religion, but about economics itself. McCleary and Barro clearly believe that economics has something to tell us about religion as a sociological phenomenon—they insist more than once that religion is “sui generis,” that religion per se is a legitimate object of economic inquiry. And their book is evidence that there is indeed a “there” there. Their findings in saint-making are particularly interesting, suggesting that the process has been used to shore up Catholicism in countries where it is threatened by emerging Protestantism. This sort of insight is made possible by the data-driven approach with which McCleary and Barro are clearly competent.

Yet McCleary and Barro also offer us evidence that economic methodology, applied in Beckerian fashion, runs up against real limits where religion is concerned. As much as they tell us, the most interesting part of the book is what McCleary and Barro are forced to leave unsaid.

Take, for example, their exploration of the effects of state religion on religious adherence. Barro and McCleary adopt a comparative international approach, using panel data from three years—1900, 1970, and 2000—on whether or not each of the world’s many nations has a state religion. Their goal, in part, is to weigh a contest between Adam Smith and David Hume, with the latter claiming that state religions can promote religious belief (or non-belief), and the former claiming they discourage belief. 

Barro and McCleary end up reaching a synthesis of the two positions, but one is forced to wonder about the generalizability of their data. When Hume and Smith wrote about religion in the 18^th century, were they writing about the same kind of thing that McCleary and Barro are in the 21^st century, or even about the kind of thing experienced by religious believers at the turn of the 20^th? This matters, because if the data are bounded in the 20^th century, then we cannot necessarily generalize from them to other religious eras—we can only make claims about what, broadly speaking, the effects of state religions were over the 20^th century.

A similar objection attaches to the pair’s examination of the determinants of religiosity (chapter two). They use data from the World Values Survey (WVS) to examine what causes nations to be more or less religious, giving particular attention to the secularization hypothesis. They find that the WVS does support a negative relationship between median income and religiosity, confirming the secularization hypothesis. But the WVS has only been administered since 1981—it is hard, therefore, to draw from it a conclusion about the determinants of religion, a category that by its nature shifts and moves on a timeline of millennia as much as decades.

This argument—that the data on which McCleary and Barro are forced to rely are limited in their scope and therefore applicability—may seem like a trifling objection. But it leads us toward a deeper point: If religion has a multi-millennia history, but the data only tell us about a small, late fraction of that history, then the conclusions of a data-driven, economistic methodology will necessarily bias our understanding of what “religion” constitutes to the period for which we have data. 

This matters because the kind of thing that the word “religion” references has shifted dramatically over the centuries, and is likely to shift again in future. In particular, essentially all of Barro and McCleary’s work covers changes that happened after or during the process of religion’s slow retreat from the public square. Yet some western nations now seem intent on undoing that retreat. If Barro and McCleary can only speak to an era of religious retrenchment, then their conclusions are profoundly limited in their future predictive validity.

This brings us to a second, definitional objection to The Wealth of Religions. McCleary and Barro are studying religion, but what, exactly, qualifies a set of beliefs or practices as a religion? One of the pair’s strongest claims is that religions determine economic outcomes in large part through their propagation of beliefs in “[o]therworldly compensators,” i.e. an “afterlife,” state of nirvana, etc. to which our earthly actions are causally related. But this analysis seems to falter at Judaism, which the authors note has substantial economic effects, but is “famously ambiguous” about the afterlife. Similarly, we might suggest that religions are organized ideologies concerned with the divine, but that throws into question Confucianism’s status as a religion, something even Barro and McCleary seem uncertain of.

Religions are a bit like Potter Stewart’s famous formulation for obscenity: We know them when we see them. But a lack of uniformity across the category makes it hard to make generalizations about the determinants and effects of religion per se, rather than religions severally. In Barro and McCleary’s defense, this may be why the book is entitled The Wealth of Religions, plural. But if so, it might be better entitled The Wealth of Protestantism and Islam and Hinduism and So Forth.

As far as it goes, the book does a reasonable job of examining the economic causes and effects of particular religions. But its limits in discussing religion per se—limits imposed not by the authors, who are clearly good social scientists, but by the data and the subject matter—ought to be understood as a challenge to Beckerian “economics of everything.” The author’s comparative international conclusions are at best radically limited to religion in the 20^th century, applicable until they are not. Take the secularization hypothesis, for example: infamously, it seems to hold true everywhere except the United States, in which religiosity is on-par with nations with a quarter of the GDP per capita.

The Wealth of Religions, then, can be understood as an indictment of the epistemic priority awarded to economics in the social sciences. Where the data are good, and where the question of concern is of the correct scope, economics can be an incredibly powerful tool. But for the study of religion—as a sociological, never mind theological, phenomenon—the book leaves one feeling that the more qualitative methods of the sociologist, anthropologist, historian, and archeologist are better suited to the task.

Does economics have something to tell us about religion? Probably, the reader of The Wealth of Religions will conclude, but far less than it has to tell us about other issues. The book is worth picking up for its case studies of Protestantism and Islam in particular, but readers who would like to have the mysteries of theology explained to them in fewer than 200 pages are better off looking elsewhere.