Student loans are becoming a major challenge for the United States. An estimated 43 million borrowers owe about $1.6 trillion dollars, meaning a significant fraction of U.S. households is burdened with debt that is not dischargeable in bankruptcy. The problem has taken on increased political and economic significance since the Biden Administration announced a federal student-loan forgiveness program that opponents have criticized as a $400-billion social welfare program for the well-to-do.

In a new paper, we take a novel approach to student loan forgiveness by empirically exploring whether it has important economic benefits that could at least partially offset its cost to U.S. taxpayers. We address the relationship between student loan reductions and the nature and extent of entrepreneurial activities and entrepreneurial firm outcomes. We make use of the adoption of no-loan financial aid policies by many U.S. universities as a natural experiment to conduct this analysis. We also estimate the entrepreneurial benefits of the Biden Administration’s student loan forgiveness program.

We address the following research questions in this paper. First, how does the adoption of a no-loan financial aid policy (rather than a standard student loan policy) affect the propensity of graduates to become entrepreneurs rather than salaried employees? Second, how does a switch to a no-loan financial aid policy affect risk-taking by entrepreneurial firms started by students graduating under such a policy? Third, what is the propensity of entrepreneurial firms founded by students graduating under a no-loan financial aid policy to receive funding from venture capitalists (VCs)? Do VCs make large investments in such firms in lump sums rather than in stages, a sign that they view these firms as high quality? Fourth, how do entrepreneurial firms started by students graduating under a no-loan financial aid policy perform relative to firms started by students graduating under a standard student loan policy, as measured by sales, jobs created, innovation, and trademarks granted for new products?

We develop our hypotheses by relying on the theory of risk-aversion. We expect that student loans – and the payments they require – will dissuade  recent graduates from taking risks as  entrepreneurs. We therefore expect proportionally more students graduating under a no-loan financial aid policy to start entrepreneurial ventures than students graduating under a standard student loan policy. Further, given their lower risk aversion, we expect ventures started by students graduating under a no-loan financial aid policy to have riskier outcomes.

Based on the above, we expect entrepreneurs with smaller amounts of student loans to undertake riskier ventures that will have higher expected payoffs. Assuming that VCs are aware of the above and given that VCs objective is to invest in firms (projects) with the highest expected cash flows, we expect the entrepreneurial ventures started by students graduating under a no-loans financial aid policy to have a greater propensity to receive VC backing. Further, the amount invested by VCs in such ventures will be larger and the extent of staging by VCs of such investment will be lower (i.e., the number of rounds over which the total investment amount is distributed will be smaller). Finally, we expect the entrepreneurial ventures started by students graduating under a no-loan financial aid policy to perform better than those started by students graduating under standard student loans.

We combine data from various sources, including Crunchbase for data on entrepreneurs and VCs, the Department of Education’s Integrated Post-Secondary Education Database System (IPEDS) for data on financial aid policies at universities, and VentureXpert for data on venture backing. Our overall sample consists of 144 universities, 5,736 entrepreneurs, and 10,844 entrepreneurial firms for the years between 1987 to 2012.

We use difference in differences (DiD) analyses to test our hypotheses. We start by providing evidence that the no-loan financial aid policies are associated with a statistically significant reduction in the fraction of students graduating from an institution with student loans. We then analyze the effect of student loans on the propensity of recent graduates to start entrepreneurial ventures. We show that no-loan financial aid policies are positively related to the propensity of recent graduating students to start a firm within three years after graduating from college.

Next, we test the effect of student loans on the riskiness of outcomes of entrepreneurial firms started by recent graduates. We show that entrepreneurial firms founded by students graduating under a no-loan financial aid policy are more likely to exit through an IPO. However, we also show that start-up ventures of entrepreneurs that graduate under a no-loan financial aid policy are more likely to fail. These results suggest that entrepreneurial firms started by students with lower levels of student loans have riskier outcomes, consistent with our hypotheses based on risk-aversion.

We also test the effect of student loans held by student entrepreneurs on access to venture capital. We show that VCs are more likely to invest in startups started by students graduating from the universities with no-loan financial aid policies. Next, we show that graduates from universities with no-loan financial aid policies receive a greater amount of VC investment for their entrepreneurial ventures. This effect is economically large and corresponds to an increase of $8 million invested by venture capitalists on average. We also show that students graduating under a no-loan financial aid policy receive VC investments with less staging.

We also find that entrepreneurial ventures founded by students graduating under a no-loan financial aid policy perform better than such ventures founded by students graduating under standard student loans. This is true for a number of performance measures: sales, jobs created, innovation output, and trademarks granted. For the sample of ventures, we show that firm sales are 132 percent higher and firm employment 105 percent higher for ventures started by entrepreneurs graduating from no-loan institutions. We also find that entrepreneurial ventures started by such students (recent graduates) produce greater innovation and introduce more new products.

We also provide additional evidence of risk aversion as a motivation by conducting a cross-sectional analysis based on the income levels of students. We split our sample into groups based on the percentage of students at a university who are from  families whose income is below $30,000 per year, measured over two cohorts prior to the policy year. We show that the impact of no-loan financial aid policies on the fraction of entrepreneurship, access to VC-backing, firm sales and employment, and new product introduction are both statistically and economically stronger for the schools with a higher percentage of low-income students.

Finally, we discuss the entrepreneurial benefits of the Biden Administration’s student-loan forgiveness program. We calculate the sales and employment benefits of loan forgiveness in our sample by estimating the additional sales and jobs  firms started by entrepreneurs would have created if, instead of graduating from schools offering standard loans, the entrepreneurs had not had any student loans or had gone to schools with no-loan financial aid policies. Using the above estimates and assuming proportional benefits of student loan forgiveness along with other assumptions, we estimate the entrepreneurial benefits of the Biden Administration’s student loan forgiveness program. For example, assuming a growth rate of sales of 5 percent and a discount rate of 15 percent, and assuming that one-third of the student loan outstanding is forgiven, we estimate the present value of the additional sales generated over 10 years due to the Biden Administration’s loan student loan forgiveness program to be $178.8 billion. Alternatively, the present value of the additional sales comes out to be $305.5 billion.

Our findings have important policy implications. While generating the above entrepreneurial benefits is unlikely to have been the primary goal of the student loan forgiveness program, it should be counted as part of the overall package of economic and welfare benefits arising from the  program. Further, the costs of the program, while substantial, need to be balanced against its economic benefits.

This post comes to us from professors Thomas Chemmanur at Boston College’s Carroll School of Management, Karthik Krishnan at Northeastern University, Harshit Rajaiya at the University of Ottawa, and Pinshuo Wang at the University of South Florida. It is based on their recent article, “The Effect of Student Loans on Entrepreneurial Firm Risk-taking, Performance, and Access to Venture Capital with Implications for Biden Administration’s Student Loan Forgiveness Program,” available here.