Applying falsification to popular ideas
I like to think that the reason social sciences are and should be part of general education is that they can teach students to check easy answers with variations on “if that were true …” Monday’s piece by Steven Mintz about infrastructure delays is a nice example. It tests various explanations for the glacial pace of infrastructure improvement in the U.S. Is it due to overregulation and unionized labor? If that were true, we’d see greater delays in Europe than we do here. In fact, the delays are greater here. The explanation doesn’t work.
(For those keeping score at home, this is the layperson’s version of Karl Popper’s notion of falsification. Popper noted that it’s hard to prove things conclusively, but it’s much easier to disprove them. There’s no shortage of critiques of the theory, but for gen ed purposes, it’s pretty solid.)
Over the last 10 years or so, a version of that conversation has happened every summer. Why are students waiting longer to register this year than last year? The usual default response is either “human nature” or “students procrastinate.” Neither explanation works. Did human nature change from last year to this year? Is procrastination new?
The ability to dismiss knee-jerk errors is valuable. It prevents wasted effort and false fatalism. If human nature is at fault, there’s nothing to be done. But if, say, there’s a new quirk in financial aid, we might be able to address that. Moving too quickly to fatalism can be self-fulfilling.
The latest data on applications to extremely expensive and selective universities offer some worthwhile falsification of a widely held belief. We hear often that the cost of college is prohibitive. We also hear often that student loan burdens are terrible, and folk wisdom suggests that the loan burdens are a function of cost increases.
If that were true, we’d expect to see the most expensive colleges suffering dramatic enrollment declines and the least expensive ones resorting to wait lists. We’d also expect to see the highest student loan default rates among the students with the highest balances.
Colleges that charge over $80,000 per year, which is higher than the national median annual household income, have set records for the number of applications. Their acceptance rates are in the single digits. Meanwhile, community colleges that charge less than 10 percent of what the elites do have suffered dramatic enrollment declines over the last decade (although they seem to have leveled off this year). If high cost were the salient barrier, we’d expect to see the exact opposite of what we’re actually seeing.
With loan balances, something similar is happening. If the issue was the amount students owe, we’d expect to see students with the lowest balances having the lowest default rates. In fact, borrowers who owe less than $5,000 have the highest default rates. The knee-jerk intuitive explanation is exactly wrong.
If the obvious story is exactly wrong, it’s time to look at other possible stories. For example, income and wealth polarization among families would fit the facts much more closely: more people are squeezed on the low end, while more occupy the high end and are willing and able to do what it takes to stay there. The inverse relationship between loan balances and default rates makes more sense when we realize that defaults are much more common among students who drop out without a credential. The highest loan balances are among the medical school graduates of the world, who tend to make enough to pay them back.
Getting the story right helps get the solution right. Stuffing everybody into Harvard wouldn’t work. Making accessible colleges more worthy of their students actually might.