In praise of ‘messy economics’

Economists needn’t be afraid to fight inequality; it’s the efficient thing to do

Credit: Michael Byers

Marianne Bertrand

Marianne Bertrand | Jun 26, 2018

Sections Economics

Collections Social Impact

As I prepared to deliver the commencement address at the University of Chicago’s convocation in June, I experienced a great deal of stress, and not simply about my performance. I had some concerns as to how well an economist—an economist working in a business school, no less—was going to be received. 

The broader public does not trust economists. A 2017 YouGov poll revealed that only 25 percent of people in the United Kingdom trust us. That is in contrast with an 82 percent trust level in doctors, and 71 percent trust in historians. It is only thanks to politicians that economists escape the bottom of the list. 

Some of this distrust, and sometimes outright dislike, comes, I believe, from a misunderstanding of what economists study and teach. For too many, economics is equated to a toolbox of tricks, masquerading as a science, that teaches people and companies how to make as much money as possible, at whatever cost to society. 

That is not what economics is about. I hope it goes without saying that misleading and cheating should never be part of anyone’s job description, including economists’. Nothing good happens to our economy or our society when some businesses flourish because they cheat while other businesses struggle because they remain honest. 

Rather, economics studies how people and companies make choices under constraints. These constraints derive from the fact that the resources we have as a society are limited. There is only so much physical capital, human capital, and time to spread around, and there are only so many natural resources. Given this, economists tend to focus much of their attention on efficiency considerations: given the limited resources we have, how do we best allocate them to maximize people’s well-being? We love markets because markets—with appropriate safeguards—can be truly transformative in this quest for efficiency. 

More-sophisticated critics of economics dislike how our discipline dehumanizes people. We treat people, the argument goes, as hyperrational, utility-maximizing robots and base our conclusions about what is or is not efficient on models that are fundamentally flawed, given the unrealistic assumptions they draw about how people make choices. 

This hyperrational view of human decision-making does not reflect the field of economics today. To put it humbly, economists have realized their mistakes and now embrace, rather than dismiss, the other social sciences. After multiple Nobel prizes celebrating behavioral economics, it is now apparent even to outsiders that when it comes to how people make decisions, economics has moved toward a more realistic view: that they are lazy, have self-control problems, and are really not that good at statistics. 

Beyond the transformational imports economics has received from cognitive psychology via the work of Chicago Booth’s Richard H. Thaler and others, sociology and social psychology have also reshaped the field. Indeed, our understanding of human behavior becomes more complete when we are willing to accept that people care about what others think of them, that their well-being might be a reflection of not just how much money and leisure they have but of how they fare compared to others, or how their choices and circumstances map with their social identity (as fourth-generation coal miners who have been raised to view themselves as the main providers for their family, for example, or as African American boys trying to fit in in the most violent neighborhoods of Chicago). 

It is true that the economics we practice today, with all these insights from psychology and sociology, is less disciplined than it was in the past. It is messier, less modelcentric and theory driven. But there is no doubt in my mind that this diminished rigor has come with the benefit of greater realism and an increased likelihood that we are getting some of it right. 

It is definitely in this messier type of economics that I feel comfortable. 

Starting from the assumption that talent is equally distributed at birth between men and women, it follows that we as a society are leaving money on the table by employing women as only 5 percent of Fortune 500 CEOs. 

Another criticism of economics is related to economists’ supposed lack of concern for distributional issues: while economists care about efficiency and value creation in society, they do not care about who benefits from the value that is created. 

And indeed, what is most efficient may sometimes (but not always) create greater inequalities. Globalization and trade are certainly today’s leading examples of the trade-offs between efficiency and distribution.

It is true that some economists have strongly argued against economists studying distributional effects. One of my Nobel Prize–winning colleagues famously said that “nothing [is] as poisonous” to sound economics as “to focus on questions of distribution,” questions that might be better left to other disciplines, or to political institutions. 

However, many economists are in fact very much concerned with distribution and inequality. I am only one of a large contingent of scholars in economics departments, at public-policy schools, and at business schools who have been devoting much of their careers to measuring inequalities, understanding causes, and studying solutions. 

In a nutshell, many economists are “minding the gap.” 

To be clear, I do not think that making the case that inequality should be of concern to economics requires diminishing the focus on efficiency. Inequality also hurts efficiency. 

Let me explain. 

In some of my research, I have been measuring gender and racial inequalities in the labor market, and though my findings have been reported elsewhere, some facts are startling enough to be worth repeating:

  • Only 25 percent of college-educated women working full time in the United States in 2010 had earnings above the median of similarly educated men working full time; only 6 percent had earnings that put them in the top 20 percent of the men’s distribution. 
  • In some work I did on racial inequality, I used a field experiment to demonstrate that job applicants with African American–sounding names received 50 percent fewer callbacks for interviews than did job applicants with identical résumés but white-sounding names. Since I completed this work, it has been replicated in many countries around the world, with each of these new studies documenting similar patterns of adverse labor-market outcomes for the minority group in that country: North Africans in France, indigenous people of Australia, Mongolians and Tibetans in China. The list goes on. A recent study that mined hundreds of millions of US tax records revealed that large income disparities exist between white and black men even when they are raised in homes with the same income and the same family structure. 

There are many reasons all these findings are troubling. Most outside of economics would say that these outcomes are simply unfair, a violation of equal rights. But it is not just a matter of fairness and justice and rights. It is also a matter of efficiency. 

Starting from the assumption that talent is equally distributed at birth between men and women, and ruling out the assumption that women are born with a comparative advantage in loading the dishwasher, changing diapers, or driving kids to soccer games, it follows that we as a society are leaving money on the table by—to use the most extreme example—employing women as only 5 percent of Fortune 500 CEOs. 

The same logic about the efficiency costs of gender inequality obviously also extends to racial inequality. 

My Chicago Booth colleagues Erik Hurst and Chang-Tai Hsieh, together with Charles I. Jones and Peter J. Klenow of Stanford, have made this point forcefully in some of their recent work, where they argue that a quarter of the growth the US has experienced between 1960 and 2010 can be explained by the lowering over time of the barriers to entry for women and African Americans into occupations in which they have been historically underrepresented. 

Furthermore—and maybe more disturbing for those who believe economists should focus solely on efficiency and leave distribution questions to others—the income inequality that may be created by (seemingly) more-efficient policies may ultimately hurt efficiency if this income inequality is accompanied by lower social mobility, as the data seem to suggest. Unequal societies might be losing out by preventing themselves from tapping into the talent of those who were born into the lower end of the income distribution. 

The messier economics that I practice helps me understand why it is women that most often take the labor-market hit.

What economics has given me in my quest to measure gender and racial inequalities is a healthy dose of skepticism toward easy explanations for these inequalities. Economists like to start from the precept that there is “no free lunch.” It is a useful precept. Why would companies that are trying to do the best they can with limited resources not hire the talented women and minorities whose talent is apparently undervalued by the market? For sure, employers should be jumping on this opportunity, and this would naturally reduce the gender and racial gaps in earnings. Maybe there are sexist and racist employers out there, but these employers would not survive in a competitive marketplace, as they are paying more for the same talent by favoring white men in their hiring decisions. 

Taking this skepticism to the data has led me to question a one-size-fits-all explanation for women’s and racial minorities’ underperformance in the labor market. 

When it comes to racial inequalities, I believe it is difficult to look at the body of research that has accumulated without acknowledging the central role of prejudice. But probably much more common than explicit racism are the unconscious implicit biases that induce even the most well-intentioned employer to discriminate. 

In contrast, I do not believe that sexism in the workplace is the leading explanation for why many women’s careers do not match those of their husbands. Instead, I believe that the main explanation is childbearing and child care. Evidence that has now accumulated across countries reveals that college-educated women’s careers essentially track those of their husbands until the birth of the couple’s first child. Mothers take a massive earnings hit right after that first birth, and their income never recovers. 

Old-school economics helps me make sense of this finding: the thinking is that couples face time constraints and benefit from having each spouse specialize, one in the labor market and one in child care and home production. 

The messier economics that I practice helps me understand why it is the woman more often that takes the labor-market hit, even among couples where the allocation of labor-market talent would suggest the woman should work and the man should stay home. The root is in slow-moving gender-identity norms transmitted from one generation to the next, norms that are lagging today’s distribution of human capital. These are norms about where men belong (primarily in the workplace), about which spouse can best provide for the children, or about whether a husband can truly be a real man if he is outearned by his wife. 

Let me be very clear: I am in no way saying that sexual discrimination in the workplace does not exist. Through my service work over the last year in my own professional association, I have heard many—too many—stories of women whose careers were derailed by sexist colleagues. 

This said, I doubt that solely eradicating gender discrimination in the workplace will erase the gender gap in earnings. It would be a mistake to direct all efforts, whether corporate or public policy, to fixing the work environment when, I would argue, a key driver of the gender gap is rooted in the home or, more precisely, in the intersection of the home and the workplace. 

Young women who are finishing college or just starting their careers should mind that gap as they begin building their professional and personal lives. They should realize that trade-offs between career and family still exist, even though those trade-offs will undoubtedly be less pronounced in their generation than they were in mine. 

It would be presumptuous, I think, to simply assume that one can have it all (career and family) without being thoughtful about the choices you make, in particular in the marriage market. Women, if you want to have it all, make sure to check your date’s appetite for diaper duty and willingness to let you shine as the brightest star at work if that is what you want for yourself. That, as much as any policy change, will help us move toward a more equal—and, for us economists, more efficient—world.

Marianne Bertrand is Chris P. Dialynas Distinguished Service Professor of Economics and faculty director of the Rustandy Center for Social Sector Innovation at Chicago Booth. This essay is adapted from her address at the 531st Convocation at the University of Chicago on June 9, 2018.