Foreign currency borrowing, balance sheet shocks, and outcomes for firm employment and investment

Emerging markets firms frequently borrow in foreign currency. This practice, which perhaps expands available funding, makes the firm vulnerable to a depreciation which would increase their debt burden and lower net worth. The resulting damage to the balance sheet can lead to reduced access to credit, and consequently a real decline in economic activity. These balance sheet shocks and their impact on borrowing and investment can be difficult to identify empirically, but are potentially important to the dynamics of the economy. In this paper, I examine the real and financial effects of exchange rate related balance sheet shocks using a unique dataset of 134 listed firms from Mexico over 2008q1-2015q2. This dataset combines loan level data with firm level balance sheet data and foreign currency (FX) exposures. The data is collected from quarterly financial reports submitted by the firms to the Mexican Stock Exchange (BMV). 112 firms were included in my primary analysis. I test how borrowing by firms with a currency mismatch changes in the wake of a large, exogenous depreciation episode, both on the intensive margin (volume and price of borrowing) and the extensive margin (obtaining new loans and forming new banking relationships), using fixed effect regressions on panel data implementing a difference-in-difference estimator. The matched nature of the loan level data enables the inclusion of bank-time and firm-bank fixed effects, which control for confounding changes in credit availability due to supply factors and time-invariant firm-bank match specific factors. Additionally controlling for exports, imports, and sector-time fixed effects accounts for other firm specific effects that may be correlated with the exchange rate change. Using this identification, I find that, following a large depreciation, non-exporters with a large FX exposure decrease their borrowing in FX relative to less exposed firms, while exporters with a large FX exposure increase their borrowing in peso. I then construct measures of firm and bank specific effects for credit growth from the loan level data, and use those to estimate real firm level outcomes due to the balance sheet shock. At the firm level, preliminary results suggest that non-exporters with a large exposure have decreased investment following the depreciation, while all highly exposed firms have decreased employment.