Growth in University Cost Relative to the 99%


In continuing with the line of thought from my previous post on the factors that affect economic and ethnic diversity at Universities, there are three sets of statistics that have come out recently, highlighted by Matt Yglesias, Matthew Levitt, Brad Plumer and others, that paint a pretty ugly picture about wealth and education in the US. First, we have the CBO study that showed the widening gap in earnings between the “Top 1%” and all other wage earners. Then we have the continued exponential increase of college costs, and the declining value of a college degree in terms of wages. (I’ve pulled the earnings data from the CBO study, tuition information from the College Board’s Advocacy and Policy Center, and wages from the National Center for Education Statistics):

These figures give me pause, particularly as I think about my own role both as an educator and as a future college parent. It’s hard to imagine why the median cost of college tuition would be increasing at a rate that far outstrips the growth in household income by even the top 1% of households, at the same time that the median wage for full-time college grads ages 25-43 is dropping relative to inflation over the same period. For one, this indicates that even if paying for college is in part aspirational — based on the expectation of an increase in quintile of earnings that the graduate will have access to — that pricing is out of step with the actual increase in household incomes, and completely out of step with changes in wages. One possibility is that the total value of grants and scholarships brings the effective rate of increase for university tuition down. But factoring in the average value of grants and scholarships, we can see that the cost of college is (counterintuitively) actually increasing at a faster rate:

This suggests that the rate at which grants and scholarships increase does not match the rate of increase in tuition and fees — which should translate into a higher rate of growth in educational debt. To give an idea of what this looks like at the most selective and least selective public and private college, respectively, I’ve broken out the total cost and value of loans and fees at the different institutions for the last four years:

Across all universities, at least over the last four years, the rate of increase in tuition, fees, and total grants in aid are in the same ball park (the least selective public universities are an outlier because small increases in average amount of aid have a disproportionate effect on a small initial baseline). But over a longer period of time, the rate of increase in the cost of education fat outstrips grants, tax credits, and scholarships. This implies that educational debt is bearing and increasing share of the costs.

It is the question of debt which might hold some insight here. After all, while the cost of college tuition has been increasing, the rates for educational loans have been dropping on a somewhat consistent basis since the eighties. In order to try and figure this out, I assumed an average educational loan rate of prime + three percent, setting the prime rate based on the year of matriculation, and then calculated the total loan cost and the annual payment for a 20 year loan that only covered four years of tuition and fees at each university, based on normalized dollars:

I think the chart strongly implies that over all, the rate of increase in student debt should track the overall rate of increase in tuition and fees. There are exceptions, particularly over the last few years, which counterbalance the cost of college by decreasing the cost of financing. But to compare this to the CBO analysis that looks at household income by percentage, (and restricting the plot to the years I’ve been looking at educational data for), the cost of education is increasing at a far more rapid rate, even adjusting for changes in financing of student debt:

This means that, on the one hand, the administration’s decision to forgive more federal student debt helps address a pressing problem, particularly given the current economic climate. But the larger problem is that universities are, in effect, driving a vast increase in the amount of debt that students are carrying. From the perpective of pressures on cost, reducing the costs of federal debt should only reduce pressure, and, I’m guessing, and might serve to magnify the share of private debt that students are carrying. Yet as this chart from The Economist shows, the vast majority of student debt is still public. At the end of the day, I don’t see any problem with vast federal subsidy of both public and private education. But I wonder if running this subsidy through student loans is the most efficient and effective way to accomplish this goal. This complicates Ronald Ehrenberg’s argument that the rising cost of tuition is in part due to a drop in direct federal aid to universities. When we look at federal aid to students, this looks more like a reapportionment. And it’s also hard for me to believe that this chart from the Economist captures ancillary forms of private debt — on student and family credit cards, through mortgage liens, etc. At the very least, it seems like provisioning higher education and subsidies in this fashion increases financial risk for students and their families, and makes their finances much more sensitive to changes in the economy.

There are other possibilities for why students and families continue to shell out so much money. While I don’t want to take the time to work this out, I’m sure that possession of a college degree, particularly at an elite college, increases the likelihood that a worker will have access to a strong employer-based healthcare plan. From this perspective, increased educational debt in the short term could lead to less financial burden in the long term, and less financial instability. And of course, there’s always Bourdieu. But it is hard for me to see how the cultural capital of a college degree could be increasing at such a rapid rate (and the law of diminishing returns and scarcity effects suggest otherwise).