Problems with Linking Student Loans to Income

by Daniel L. Bennett

Ben Miller has a cool analysis of the Department of Education's proposed debt-to-income ratio that would limit the amount that a student can borrow for specific career programs according to BLS salary data for workers in the respective field. Miller has a great chart that shows what the maximum loan amounts would be for several vocational programs if the policy were currently in place. For instance, students studying to become a registered nurses would be limited to $28,212 in student loans, and students studying to become an auto mechanic would be limited to $15,003 in loans.

Tying student loan limits to earnings potential is an interesting idea, but there are some problems with the ED's proposal.

First, using wage data ignores other forms of compensation such as health care insurance, retirement benefits, vacation time and other fringe benefits. These are all part of the total compensation paid by employers to employees. Not including these benefits in any type of debt-to-income calculation is ignoring a significant portion of an employee's compensation.

By restricting the income portion of the ratio to the 25th percentile of wage earners, the government is ruling out that many entry level workers' compensation may rise within the first 10 years of their career (the standard loan repayment period). Simply using the bottom 25% of earners in a field ignores the actual experience of individuals in that field and instead relies on a group of people that may change as individuals move up the pay scale. In other words, the bottom 25% of earners may always contain the entry level workers who may very well move up to the 50% of earners in a few years as they gain experience in a given field. To be effective, the income portion of the ratio would need to account for the net present value of expected earnings over a 10 year period by tracking the earning of individuals in a given vocation over the first 10 years of their career, as group statistics are misleading.

Next, setting loan limits on the basis of national wage statistics would undermine local market conditions which may vary a great deal. Compensation for some occupations may vary considerably by region, due to either differences in cost of living or labor supply and demand variations.

Such a mechanism would also need constant monitoring as market conditions change frequently. Compensation of various professions rarely stay in equilibrium for very long as labor supply and demand conditions, as well as other factors (inflation, technological, regulatory, etc), change continuously in a manner that affects compensation. A once a year snapshot of wage levels would not be just for a labor market that changes so often and varies so widely.

This policy could also potentially open the door to price controls. If a loan limit were imposed, schools would have an incentive to set their tuition at the maximum allowable loan amount and in effect, control the prices that an institution is able to charge. This could result in adverse effects such as constructing a barrier to entry for some prospective students who may need to take out loans to cover their cost of living while in school. Reducing access to educational/career opportunities is surely not a desirable outcome.

Theoretically, I think there is some merit to the idea of student loan amounts being somehow linked to career earning expectations; however, such a mechanism would be much better left to the market to determine the terms, an idea that Dr. Vedder espoused in this blog.