We study the international transmission of US monetary policy shocks and a strong US dollar. We show that monetary tightening is linked to a higher risk premium, only in emerging markets, measured by deviations from uncovered interest parity. An appreciation of the US dollar, on the other hand, does not lead to higher risk premia anywhere, even though countries' currencies depreciate vis-à-vis the dollar. Our interpretation is that monetary policy shocks are a better proxy for global financial conditions than the exchange rate movements that may be capturing more fundamental shocks.

I study the role of trade credit—supplier financing via costly delayed payments—in determining the investment response of foreign firms to US monetary policy. Using firm-level data from several countries, I find that trade credit declines by less than bank credit following a US monetary tightening. Firms that rely more on trade credit than on bank credit reduce investment by less. To rationalize these findings, I develop a two-sector open economy model with heterogeneous firms, trade and bank credit. Trade credit arises as a risk-sharing mechanism between firms with different risk preferences along the supply chain, for which I provide direct empirical evidence. This mechanism dampens the transmission of US interest rate shocks to trade credit relative to bank credit, which is directly affected by changes in banks’ international funding costs. The presence of trade credit in the model reduces the sensitivity of aggregate investment and output to US monetary policy and increases their average level.

We study the role of firms' balance sheet currency mismatches---the excess of US dollar liabilities over dollar assets---in factor misallocation. Using matched firm-loan data from Uruguay for 2012-2019, a period marked by large exchange rate depreciations, we show that currency mismatches are prevalent and persistent. They are concentrated among firms with higher productivity, faster growth, and larger capital stock, and among foreign-owned firms. We find that currency mismatches account for no more than 7\% of the dispersion in the marginal revenue products of capital and labor, suggesting they are a limited source of factor misallocation.