Financial Aid, Borrowing Constraints, and College Attendance: Evidence from Structural Estimates

Michael P Keane
2002 The American Economic Review  
The idea that borrowing constraints may preclude low-income youth from attending college is widespread. Here I will review some recent work on the importance of borrowing constraints, and the impact of financial aid programs (i.e., guaranteed student loans and tuition subsidies) designed to counteract them. This work suggests that borrowing constraints have little effect on college attendance decisions. There is evidence that tuition subsidies play a modest role in reducing inequality in
more » ... ng and earnings. But most inequality appears to be driven by unequal human capital accumulation prior to college-going age. Empirical work on college attendance decisions has consistently produced two findings: 1) parents' income helps predict college attendance, even conditional on other observed characteristics of youth, and 2) the effect of tuition on probability of college attendance is greater for children from low-income families. There is some (mixed) evidence that: 3) the rate of return to education is higher for low-income youth. These findings are commonly viewed as strong evidence for borrowing constraints. Recent work by Stephen Cameron and James Heckman (1998) challenges these findings. Furthermore, recent work by Michael Keane and Kenneth Wolpin (2001) shows they are compatible with a model in which liquidity constraints have a negligible impact on college attendance rates. To understand this work, it is useful to consider a very simple discrete-choice model of college attendance decisions. The model has the following features: 1) Agents are infinitely lived. Time is discrete.
doi:10.1257/000282802320189429 fatcat:zw4lgsldnfhgvp6zlchnhq6u2q