Abstract
Government-backed income contingent student loans are increasingly being used to fund higher education. Until the outstanding balance is cleared, an income contingent repayment plan acts as an incremental marginal tax on earnings above a threshold. If this additional “tax” on earnings reduces the labor supply and hence the earnings of borrowers, this could reduce both loan repayments and tax receipts, increasing the cost of funding higher education. This paper investigates this under-studied topic by exploring bunching at various loan repayment thresholds between 2002 and 2014, using a novel, linked administrative dataset from the United Kingdom. Our findings suggest that the UK's income contingent repayment plan does not cause borrowers to reduce labor supply, at least for those with earnings near to the threshold.
Highlights
Government-backed income contingent student loans are increasingly being used to fund higher education
The focus of this paper, is whether the additional marginal tax on the income of indebted borrowers impacts their labor supply.5. Evidence on this particular topic is clearly important for informing the higher education funding debate – the costs of an income contingent loans (ICLs) might be un derstated if they result in a reduction in labor supply, for example, as this would reduce both loan repayments and tax receipts – yet is extremely limited, largely due to the challenges associated with iden tifying causal effects
Australian students who borrowed from the Higher Education Contribution Scheme (HECS) had to make repay ments of 4% of their taxable income once they earned above the threshold of around AUS $24,000
Summary
The question of how labor supply responds to marginal tax rates has long been of interest to labor economists and those interested in op timal tax policy. Gruber and Saez (2002) find that taxpayers who itemize (and who are likely to have the lowest frictions) are responsive to taxation, Kleven and Waseem (2013) find that the frictions are large in Pakistan, and Adam et al (2019) estimate that they could be worth as much as 9% of income based on evidence from the UK. They show that the individuals most responsive to taxes are company ownermanagers and the self-employed. We are studying a kink of 9ppts, which is at the lower end in terms of kink size relative to the previous literature. We explore the importance of opti mization frictions in our data, including looking at the selfemployed, but do not find any evidence to suggest that they are driving our results
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
