Transcript – Socioeconomic Impacts of Local Fracking Booms: Household Wealth and College Attendance

I’m happy to describe research as part of the 2020 President’s Excellence Fund symposium. Our project studies the socioeconomic impacts of local fracking booms looking at household wealth and college attendance. In this project, we study an important question in a higher education policy. Nine states, as we all know, higher education go into college is one of the strongest channels to provide social mobility, a way to achieve the so-called American dream. But is access to higher education policy impaired by family wealth? According to recent data, only 12 percent of college students are from the bottom quintile the family income distribution, while 28 percent of college students are from the top quintile. So we study the research question does household wealth significantly affect the children’s college going behavior? We certainly expect that all else equal higher family wealth would make it more likely that children go to college. But it’s important that we know the size of this effect. Why is that? Well, there are a number of policies in place are currently being discussed about making public higher education institutions either tuition free or significantly discounted to low and modest income families. And this can be an expensive policy. And ultimately, the net benefit of such a policy is going to depend on how much this actually incentivizes kids to go to college who otherwise wouldn’t attend rather than serve effectively as a transfer of funds to families of kids who would attend college regardless. So this parameter, how responsive college attendance is to wealth is something that we need to know to inform higher education policy debates about tuition. One might think that this would be known already and fairly easy to answer, but it’s not. So why is that? First, let me say what is not the right way to answer this question.

One would not want to simply gather data from families on wealth and college attendance and measure correlations. Why not? Because, as we all know, correlation does not imply causality, because what we want to know is why wealth and wealth alone affects attendance. Wealth, of course, is likely to be related to a number of other factors and about a family such as parental education or the parent’s views towards education. So in order to isolate only the effects of wealth so that we can speak to how much money is a barrier to college attendance, we need a natural experiment and we need a natural experiment akin to a helicopter randomly dropping money on some households, but not on other households. And then as researchers looking at college attendance of kids in those different types of households, it’s actually difficult to find the right natural experiment to truly disentangle the effects of wealth from other factors. But we argue in this project that the fracking boom provides exactly that natural experiment. And we study this using detailed data from the state of North Dakota. What we’ve done is we’ve linked data for individual households on whether those households signed mineral rights leases to information from the North Dakota State Longitudinal Data System on the children’s college attendance of those households. And by defining the right households to serve as, quote, treatment and control groups, we can measure what the effect of wealth is on college attendance, controlling for a bunch of other different factors. In this poster, we describe in more detail our research approach, how our econometric specification allows us to define these treatment and control groups that allows us to separate correlation from causality. And it describes our preliminary findings.