Forbearance vs. Interest Rates: Tests of Liquidity and Strategic Default Triggers in a Randomized Debt Relief Experiment

I study debt relief policies in a large-scale field experiment with a unique 2-by-2-by-2 design. I then provide novel tests of models in which default is triggered by solvency, liquidity, and strategic behavior. In contrast to solvency driving decisions, modifications orthogonal to face value affect whether and when to default. Forbearance only has short-term effects, whereas rate reductions have immediate effects that persist. In contrast to liquidity being the sole trigger, reduction in payments has a weak association with default. Forbearance reduces payments twice as much as rate reductions, whereas delinquencies are more responsive to a rate reduction. Compatible with strategic behavior, unexpected news about a dollar increase in future payments increases defaults by as much as a 30-cent increase in current payments. Compatible with the endogeneity of triggers, whether a borrower is Ricardian—merely postponing forbearance is ineffective, behavior is sensitive to future payments, and defaults are strategic—is tightly linked to balance sheets. Among alternatives, results are most compatible with the interpretation whereby every default is strategic, with the endogenous trigger being influenced by liquidity and distress, as in Campbell and Cocco (2015). Findings provide a unified explanation for disagreeing results from previous debt relief studies and have implications for modeling loan modifications and the pass-through of interest rates.