At the onset of the Covid-19 pandemic, Peru experienced a sharp increase in local-currency loans and a significant decline in foreign-currency loans among firms. While the literature has examined the real and risk-taking effects of pandemic credit policies, their impact on de-dollarization remains unexplored. This paper provides the first causal evidence on how public credit guarantees affect firms’ currency composition of debt. Using microdata from Peru’s Credit Register, we exploit variation in access to Reactiva loans through a difference-in-differences design, complemented by a triple-differences framework distinguishing exporters and highly dollarized firms. We find that firms that received Reactiva loans experienced a significant reduction in credit dollarization, with an average treatment effect of –18.8 percentage points on dollar-loan growth and –24.8 points on the dollarization ratio. Effects were strongest among firms with high pre-pandemic dollar exposure. The effect follows a U-shaped pattern: a sharp initial shift toward domestic-currency borrowing during the program, followed by a partial reversal as policy incentives expired. We also document heterogeneous effects across firm sizes, with stronger de-dollarization among small and medium-sized firms and more persistent effects among firms with initially high dollar exposure. The persistence of lower dollarization ratios suggests that this effect may have reflected not only a mechanical portfolio rebalancing, but also some behavioral adjustment in firms’ borrowing preferences. A simple dynamic model based on Bayesian persuasion rationalizes this U-shaped pattern, highlighting how temporary policy incentives can produce lasting changes in firms’ financing behavior.
Renzo Pardo-Figueroa (Sun,) studied this question.