Analysis of Two Higher Education Funding Systems under Uncertainty ∗

We present a theoretical model to analyze the effects of two higher education loan schemes on individual schooling decisions, in a world where graduate earnings are stochastic. The paper is structured in two parts. The first part is static, we assume a single lifetime shock on graduate incomes and we compare the individual expected utilities under a mortgage loan (ML) and an income contingent loan (ICL). We find that when agents are risk-neutral they always strictly prefer an ICL. When they are risk-averse, for increasing uncertainty the expected utility under an ICL system is higher. The second part is dynamic, we assume a graduate income following a geometric Brownian motion. Applying the Euler-Maruyama method to generate the incomes, we develop a numerical iterative solution to compute the average utilities under the two funding schemes. We find that an ICL scheme is preferred to a ML scheme when the initial level of the income is low and the uncertainty on future wages is increasing. However, if the individuals are very risk-averse and receive a low initial wage, the differences between the two funding systems decline and a ML becomes the most advantageous system. JEL Classification: D81, I22, H80.