Publication Type

Working Paper

Publication Date



I study the role of need-based aid from selective universities in closing the achievement gap between rich and poor high school students. I focus on the incentive aspect of need-based aid that can change high school students’ effort choices. The impact of increasing need-based aid depends on the extent of borrowing constraints and how competition affects the relative performance of low- and high-income students. I develop a structural model of students’ learning, application, and admission processes, and estimate it with the Education Longitudinal Study of 2002, a nationally representative sample. I use a geographic variation in costs of attending selective and nonselective universities to control selection biases driven by an unobservable characteristic correlated with family income. I find that 3.8% of high-ability low-income students do not apply to selective universities because of borrowing constraints. If selective universities reduce the net cost for students from the bottom quintile of the income distribution by 10% by increasing need-based aid, it decreases the effort gap, as measured by the number of Advanced Placement (AP) classes taken, by 17.6%, the achievement gap, as measured by the SAT score, by 3.6%, and the wage gap by 5.9% between students from the top and bottom quintile of the income distribution and with top-quintile initial test scores. Need-based aid can close the achievement gap better than merit-based aid, which requires the same budget, if it is provided by selective universities. They have similar impacts on the aggregate achievement level.


Behavioral Economics | Education Economics

Research Areas


First Page


Last Page


Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Additional URL