Governing Schools Like Colleges, Part II
In the prior post, I described an argument that we should govern our schools like our colleges. In breaking this down, I pointed out that colleges have different objectives than schools (e.g., equity) and this affects the degree to which market-oriented reforms work. In this post, I take on one piece of that argument--the conventional wisdom that our colleges are better than our schools. Specifically, I provide evidence that colleges may not generate better results, that the college funding model increases costs, that college productivity is declining, and that the higher education market has not solved the information problem.
(1) Limited evidence of better results. The conventional wisdom that colleges are better than schools comes, for example, from the substantial and growing economic return to college, the fact that U.S. colleges and universities dominate world rankings and that Nobel Prize winners are disproportionately from U.S. universities. Also, students from around the world clamor to pay the high (and fast-growing) tuition rates at U.S. universities; roughly one million students from other nations attend U.S. colleges every year.
But the evidence is much weaker than it seems. First, the returns to college aren't comparable to the returns to high school, especially given that schools are the primary pipeline for colleges--if schools are worse than colleges, why would colleges pay so much attention to students' high school outcomes in admissions? Also, while U.S. colleges have strong reputations in less developed countries where most visiting students come from, our reputation for college quality is fairly weak in Europe. Finally, the evidence on international rankings and Nobel Prize winners focuses on the research produced by colleges.
There is actually little direct evidence that U.S. colleges provide a superior academic experience for students. In fact, evidence from the important book Academically Adrift suggests that college students learn relatively little during college as measured by general tests of cognitive and verbal ability. Now, I'm skeptical of the test score part of this argument--because the tests they analyzed were not at all aligned with the material being taught. The bigger and more convincing concern is that students are spending less time studying. But I bring up the test score pattern because, when free market advocates argue for school choice policies, they often talk about stagnant test scores relative to dramatic increases in per pupil expenditure. In fact, if we created the same misleading graph with spending and test score trends for higher education, it would probably look very similar to K-12.
So, there is not much evidence that colleges generate better academic results.
(2) Rising costs. In my recent back and forth with Cato's Jason Bedrick, he pointed to the rising cost of K-12 education. That's also true of higher education, though to a lesser extent. From 1970-2000, expenditures increased by about 2.5 percent per year, after adjusting for inflation.
One reason for rising costs is the way in which government subsidies for college are distributed tends to push costs up. Since federal aid and loans are somewhat proportional to total tuition, students are insulated from tuition increases, especially in the short run but also the long run. There is evidence that this market-driven approach is partly what drives up costs. Indeed, if libertarians such as Cato's Bedrick got their way and school vouchers were provided regardless of family income, and schools were allowed to top off vouchers with higher tuition, one of the likely immediate effects would be for private schools to raise tuition and total costs even further, making them even less accessible than they are now, not to mention encouraging further isolation of low-income students in certain schools.
It's not just that costs would rise from the above pressure, but that a larger share of resources would probably end up going to non-academic resources. The fact that U.S. colleges are spending more and more on great entertainment complexes, and other factors not directly related to instruction, is driven by a combination of market demand and market imperfections. In New Orleans, we have also seen some evidence of time and money going to attract K-12 students in ways unrelated to academic quality.
It's important to emphasize that cost increases in both sectors is driven to a substantial degree by the so-called cost-disease--education institutions have fewer paths to increase productivity but have to continually increase salaries to keep up with sectors where productivity improvement is more feasible. But the point here is that colleges seem to suffer from the same problem and that extending the college funding model to schools would not help with cost containment.
(3) Declining productivity of higher education. The above discussion separates the discussion of outcomes from costs, but it's more useful to put them together. We can do that by focusing on the combination of college completion and college costs.
As additional evidence of effectiveness, note that college completion rates are declining. Given rising costs, this means productivity, in terms of academic degrees, is clearly declining as well. The figure below, from an article I wrote a few years ago, combines cost growth with trends in the number of college graduates. If U.S. colleges are so great, why has productivity in degree production been declining?
The situation is more ambiguous at the high school level. While costs are rising faster, high school graduation rates have apparently been rising. (I didn't try to create the equivalent figure for comparison, in part because there is considerable debate about the meaning and validity of high school graduation numbers.)
In any event, the declining productivity in higher education is not a great sign when we think about applying these policies to K-12.
(4) Poor information. A market-driven system does not function well without good information. In the Cato panel, Bedrick argued that government information in the K-12 sector "crowded out" information from the private sector and in a later blog post pointed to U.S. News and World Report college ratings, along with those of other magazines, as evidence that the market will provide good information, if given the chance. This is partly true--the magazine ratings do come from private markets. The problem is that these college ratings are awful indicators of college performance. K-12 could use better measures, but there is little to suggest that the market will provide them. Information is a public good, especially when being put to public purposes of education. This is exactly where the government can be helpful and the Obama Administration was wise to step in with its College Scorecards. These aren't perfect either, but are better than what the market would provide on its own.
The point here is not to paint higher education in a negative light, and certainly not to suggest that colleges should operate like K-12 schools. Rather, as I argued in the prior posts, education markets are just inherently uncompetitive even when seemingly free market policies are present.
Bottom line: The argument that our success in higher education should lead us to apply the same policies to K-12 still seems highly debatable.
Douglas N. Harris is a Professor of Economics, the Schleider Foundation Chair in Public Education, and Director of the Education Research Alliance for New Orleans at Tulane University
Follow him on Twitter: @douglasnharris