Category: Education

By Glen Weyl, here are two talks, Imperfect Platform Competition and Concordance Among Holdouts.  Each video is over an hour long and presents one of Glen's research papers.

Why doesn't every paper have one of these?  Is it because most papers don't matter?

Up through the 1980s, the Philippines offered a relatively level playing field for non-profit and for-profit institutions of higher education.  What was the result?:

Unlike Filipino non-profits, the for-profits typically did not have entrance examinations, and accepted any student who has completed a secondary education and can pay the relevant fees (Zwaenepoel 1975, pp.163-4).  From a survey of Manila institutions, the for-profit institutions had an average student to fulltime faculty ratio of 27:1, whereas the non-profit religious institutions had an average ratio of 19:1 (Miao 1971, pp.71-2). For-profit institutions tend to invest in classrooms to accommodate large enrollments, rather than investing in library facilities, book holdings, or laboratory facilities. Furthermore, Filipino for-profit institutions tend to limit their class offerings to low-cost, labor-intensive classes, such as teacher education and commerce (Zwaenepoel 1975, pp.322, 342, 348, 587).  As of 1970, nonsectarian institutions (typically for-profits) spent four percent of their total budget on sites, equipment, and facilities, whereas sectarian institutions (typically non-profits) spent a much higher 12.41 percent (Isidro and Ramos 1973, p.157).  As of 1971, for-profits held an average of 2.58 books per student, whereas non-profits held an average of 8.9 books per student (Zwaenepoel 1975, pp.347-8).

Filipino for-profits also produce a different kind of education. Students from for-profit institutions tend to take standardized vocational exams in much greater number, although they pass them at a lower rate.  These facts reflect both the vocational emphasis of for-profits as well as the lower academic reputation of their students.  Based on a sample of institutions from the Manila area (from 1963 and 1968), students from nonprofit religious institutions pass these standardized tests at an average rate of 38 percent, whereas students from for-profit institutions pass the same tests at a lower rate of 18 percent.  For-profits, however, produce a much greater number of students taking the tests, and therefore pass a much greater number of students through the tests.  Students at for-profits are approximately ten times more likely to take the tests.  Adjusting for the lower pass rate from for-profits, the for-profits are putting about five times the number of students through the tests as the non-profits, even though for-profits educated no more than three-fifths of all Filipino students at the time (Miao 1971, p. 207).

This is broadly similar to the patterns we see in the United States.  You might conclude that the for-profit status is more useful when an external test takes care of the signaling, and that a non-profit status is required when no test certifies quality.  But why exactly should that be the case?  Is the non-profit institution, by being so jealous of reputation, perks, and donations itself, a better producer of signals?  If signaling yields so much private value, why can't a private for-profit make a sufficiently strong commitment to a credible signal?

My paper has been published in this book.

According to a new paper from Paco Martorell and Damon Clark, not so much:

Although economists acknowledge that various indicators of educational attainment (e.g., highest grade completed, credentials earned) might serve as signals of a worker’s productivity, the practical importance of education-based signaling is not clear. In this paper we estimate the signaling value of a high school diploma, the most commonly held credential in the U.S. To do so, we compare the earnings of workers that barely passed and barely failed high school exit exams, standardized tests that, in some states, students must pass to earn a high school diploma. Since these groups should, on average, look the same to firms (the only difference being that “barely passers” have a diploma while “barely failers” do not), this earnings comparison should identify the signaling value of the diploma. Using linked administrative data on earnings and education from two states that use high school exit exams (Florida and Texas), we estimate that a diploma has little effect on earnings. For both states, we can reject that individuals with a diploma earn eight percent more than otherwise-identical individuals without one; combining the state-specific estimates, we can reject signaling values larger than five or six percent. While these confidence intervals include economically important signaling values, they exclude both the raw earnings difference between workers with and without a diploma and the regression-adjusted estimates reported in the previous literature.

The full paper you will find here.  This to me looks like a fairly clean test.  I still think signaling matters a great deal, but that is distinct from thinking that a high school diploma is the relevant form of such signaling.

I'll transplant the links in from Mark Thoma:

Here's my response, along with responses by Alberto Alesina, Laurence Kotlikoff, Michael Pettis, Gilles Saint-Paul, Paul Seabright, Richard Baldwin, Michael Bordo, Guillermo Calvo, Tyler Cowen, Harold James, David Laibson, and Eswar Prasad [All Responses].

I realized I forgot to talk about the need for more economic history, but that is discussed by several others so at least the point got made.