An important puzzle in international macroeconomics is the exchange rate disconnect puzzle. Based on recent empirical literature by Mussa (1986), Baxter and Stockman (1989), and Flood and Rose (1995), high exchange rate volatility under floating rates appear not to be related to the high volatility of other macroeconomic variables. However, there is little systematic research examining whether exchange rates affect real quantities at the microeconomic or firm level. In this paper, we examine the exchange rate disconnect puzzle using Japanese firm level data from 1982 to 1997, for 105 firms in the 14 largest export industries at the 4-digit level. We find that export volumes at the firm level are highly responsive to exchange rates. Depending on the industry, a one percent appreciation of the domestic currency results in an average decline in export volumes of 0.02 to 2.9 percent. One explanation given for the small estimated export elasticities in the aforementioned macro-aggregate empirical literature is that prices are sticky in the buyer's currency. From our estimates, we also find export prices in terms of the buyer's currency are sticky. Thus, the increased responsiveness of exports to exchange rate fluctuations in our model is not related to changes in international relative prices. Rather, in our paper, the responsiveness of exports to exchange rate fluctuations arises from a loosening of balance sheet constraints. Suppose that a depreciation in the exporter's currency is positively correlated with a relaxation of balance sheet constrants. With relaxed balance sheet constraints, the exporter with the depreciating currency is then simply able to produce more, regardless of the inflexibiliy of foreign prices. In most industries in our sample-10 out of 14 industries--a currency depreciation is correlated with a relaxation of financing constraints. For these industries, a currency depreciation will be related to a strong expansion in exports, through the relaxation of balance sheet constraints. There are 4 industries in which a currency depreciation is correlated with tightening of financing constraints. However, firms in these industries are able to substantially offset the adverse impace on balance sheets of aggregate shocks through hedging activities. Thus, we observe a positive relationship between a currency depreciation and exports, even in these industries. Our findings of large and significant export elasticities imply that there is no exchange rate disconnect at the firm level. Since our sample of Japanese exporters covers over 90 percent of total Japanese manufacturing exports, the discrepancy between the results obtained at the macro-aggregate level and at the firm level is not simply an artifact of incomplete sample coverage. Rather, the discrepancy suggests that it may be important to include non-linearities related to financing constraints when modeling the relationship between exchange rates and export volumes.
