Going to War and Going to College: Tuition costs

Our estimates of the net effect of military service and the availability of subsidies for education on collegiate attainment among veterans of World War II speak to two long-standing questions: (1) what was the effect of World War II and the G.I. Bill on educational attainment and (2) how the availability of direct subsidies for college costs are likely to affect investments in education. To be clear, our estimates even if we were to present a single point rather than a range do not resolve either issue entirely. Obviously, given the nature of our study, we can not estimate what the effect of World War II alone would have been, nor can we estimate the direct effect of the subsidies built into the G.I. Bill. Moreover, in terms of the effect of World War II on post secondary educational attainment, our estimates are partial equilibrium estimates. As such, they reflect the change in collegiate attainment associated with World War II military service and the availability of generous benefits through the G.I. Bill under the assumption that the return to college and the supply side of higher education also remained fixed. Conventional supply and demand logic suggests that the any expansion in the pool of college educated manpower would be accompanied by a fall in the relative wages for this type of labor. Similarly, subsidies for college tuition would be likely to drive up college costs, given anything but perfectly elastic supply among colleges and universities. Thus, the actual effect of World War II together with the G.I. Bill on educational attainment is likely to be smaller than the effect of service on a single individual. This reasoning only reinforces the impression obtained from Figure 2: while the G.I. Bill may very well have had a noticeable effect of post secondary educational attainment, the war was less of a watershed than it is sometimes depicted.
In terms of measuring the effect large scale subsidies on post secondary school educational attainment, we wish to stress that our measures reflect the effect of the disruption of the war and the availability of aid, not just the availability of aid (as would be the case in the traditional estimates of changes in the Pell grant on enrollment). Regardless, it is of interest to compare the net effect of military service and the G.I. Bill to more recent estimates of the behavioral response to changes in the cost of higher education. The G.I. Bill dramatically reduced the cost of attending college. Not only did the benefits effectively cover tuition, but the generous stipend dramatically reduced the opportunity costs associated with college attendance. Kane (1994) reports some of the most credible contemporary estimates of the effects of college costs on student enrollments. He uses the cross-state variation in the changes in tuition to identify the effect of college costs on enrollment rates. Kane’s estimates imply an overall cost elasticity of college enrollment of about 0.5 for the typical student. Our preferred estimates imply an elasticity of the effect of World War II of years of completed education for veterans of 0.4. In terms of college completion, the total cost elasticity is elasticity is 0.5. We regard the behavioral effects of World War II on the educational choices of veterans to be quite similar to the contemporary response of students to changes in tuition costs.