Pierre-André Chiappori is a leader of the Family Inequality network and E. Rowan and Barbara Steinschneider Professor of Economics at Columbia University. His research focuses on household behavior, risk, insurance and contract theory, general equilibrium, and mathematical economics.

Please describe your area of study and how it relates to current policy discussions surrounding inequality.

My main topic is related to family economics. There are many aspects of this. One aspect is the decisions made within the family, in particular regarding children’s education, what we call investment in human capital. Many issues are related to this topic, for instance, matching — who marries who, by education, by human capital and so on. The relationship with inequality is very deep. Most of the inequality now in the U.S. is between educated and less educated people; the decision to invest in children’s education, therefore, has a crucial role. And of course, family is the primary place where those kinds of decisions are made. An important concern is the presence of mechanisms that tend to increase inequality in a systematic way. We know from empirical work that has been done on investments in human capital (in particular by Jim Heckman at the University of Chicago, together with Flavio Cunha and other people) that the level of human capital depends on a number of factors: schooling, of course, but also parent’s education, the natural ability of the child, and parent’s investment in the child, including early (before even starting school). After a few years, those various factors are complements. What this means is that education is good for everyone but is especially profitable for people who are already ahead because of the other factors. In other words, if there is some initial inequality between children at that age, the mechanism will tend to increase inequality because the incentive to invest will be larger for those kids who are already advantaged. This suggests that if we want to fight inequality — and I think we should — the interventions should be made very early. 

I am also working on matching in the marriage market. One thing we see is that especially at the top of the distribution by education, there is more assortative matching now than in the past, which in our jargon means that while there has always been a tendency for educated people to marry educated people, it’s even more the case now than a few decades ago. We have models that explain this by the increased importance of investment in children’s education. A crucial insight is that although marriage has a host of purposes (and benefits), one of the most important is children; not only fertility per se, but also equipping children with a high level of human capital, which is known to be important for the child’s future. And our models predict that the growing importance of investing in children, together with the crucial importance of parents’ education, should result in an increase in assortative matching. 

What areas in the study of inequality are most in need of new research?

Definitely early formation of inequality; and by early, I mean between during the very first years of the child. Our perception here is that the kind of inequality that has been built up by the age of 10 will be extremely difficult to reverse. If we want to fight inequality, it’s probably much more efficient to fight it very early than wait until high school or college – or even later. 

What advice do you have for emerging scholars in your field?

The same advice I would give to any young economist. A.) Take theory seriously. There is no way you can look at the data without having clear ideas about the theoretical issues involved. B.) Take data seriously. Theory which is not based on a precise knowledge of facts is more likely to be misguided. And if data disagree with your theory, data tend to be stubbornly right. C.) Take econometrics seriously. The way you look at the data must be consistent with theory and must use state-of-the-art technique to really implement the ideas coming from theory in the most efficient way. You should be both conscious of the empirical challenges and equipped to address them.