Accounting for Debt Service: The Painful Legacy of Credit Booms

Abstract

This paper documents the main channel through which credit booms affect real economic activity in the future. As a matter of simple accounting, credit booms generate a predictable increase in future debt service that transfers spending power from borrowers to lenders. We document this dynamic pattern in a panel of 17 countries from 1980 to 2015 and identify a robust lead-lag relationship of about 3 years between the peak of credit booms and the peak in debt service. We develop a method to decompose what fraction of future real effects of credit booms is explained by debt service and show that debt service almost fully accounts for several puzzling findings in the recent empirical literature: that high growth in credit predicts low output growth in the future, deeper recessions, and a greater likelihood of financial crises. Explicitly accounting for debt service not only sheds light on the channel behind these findings but also generates stronger empirical relationships. We hope that our results will provide useful guidance for future efforts to model credit cycles.