اندازه گیری اهمیت اقتصادی مواجهه با نرخ ارز

This paper re-examines the nature and the economic significance of the exchange rate to firm value relation using a database of non-financial firms from over 18 countries. Our main contribution is to apply a portfolio approach to investigate the economic importance of exposure. We find that firms with high international sales outperform those with no international sales during periods of large currency depreciations by 0.72% per month, whereas they underperform by 1.10% per month during periods of large currency appreciations. In contrast to the previous literature, our evidence shows that exchange rate movements can have an economically significant impact on firm value.

While finance theory, firm-level survey results, and common intuition strongly support the notion that firm value is sensitive to exchange rate movements, empirical support is mixed. Studies that examine exchange rate exposure generally find some evidence of a relation between exposure and its theoretical determinants but that the economic importance of this relation is small. In this paper, we use firm-level data from 18 countries to systematically examine the nature of exposure around the world. To assess the economic magnitude of exchange rate exposure, we use a portfolio approach that is new to the exposure literature. Our main result is that exchange rates play an economically sizeable role in explaining stock returns. Prior studies in the empirical exposure literature have primarily focused on estimating exposure in a regression framework and testing whether the estimated exposure betas are consistent with the theoretical determinants of exposure. Jorion (1990) finds evidence of significant exchange rate exposure and shows that the level of foreign sales is the main determinant of exchange rate exposure for large U.S. multinational firms. However, Amihud (1994) and Bartov and Bodnar (1994) find no evidence of contemporaneous exposure for U.S. multinationals, although Bartov and Bodnar do show that U.S. firms respond to past quarterly exchange rate movements. Using a sample of Japanese firms, He and Ng (1998) uncover a strong contemporaneous relation between foreign sales and exposure but find no evidence of a lagged relation. Dominguez and Tesar, 2001 and Dominguez and Tesar, 2006 find a link between foreign activity and exposure in a sample of firms from eight non-U.S. countries, including Japan. The various findings regarding the nature of the exposure relation highlight the need for a systematic comparison of exchange rate exposure across time, countries, and determinants. To this end, we expand the investigation of the nature of the relation between exposure and firm characteristics by using unique firm-level data with broad coverage across markets. Our study begins with the linear regression framework that is traditionally used in the exposure literature. Consistent with the previous literature, we find that the number of firms that are exposed to exchange rate movements is greater than what can be attributed to chance, but that exchange rate movements are often not statistically or economically important. We then evaluate whether exchange rate exposure varies in a manner consistent with firm-level characteristics such as international sales, foreign income, foreign assets, and firm size and find that international sales is most reliably related to exposure. A common finding in the empirical exposure literature is that exchange rate movements do not explain a large proportion of the variation in stock returns. Although it is not the focus of the analyses, Jorion (1990) and Bartov and Bodnar (1994) find that exchange rates have low explanatory power (as measured by R2) for explaining individual stock returns. Griffin and Stulz (2001) demonstrate that in a variety of settings, exchange rate movements explain only a small amount of movement in international industry (and U.S. individual) stock returns. Our results based on the regression framework provide a broader confirmation of previous evidence that exchange rate movements do not explain much of the variation in individual firms' stock returns. The main contribution of this paper is to employ a portfolio approach to measure the economic importance of exposure. Evidence that suggests the economic importance of exposure is small is based on calculating the fraction of the variation of firms' stock returns that are related to exchange rate movements. For some applications, such as hedging, a firm-level perspective may be relevant. However, from the perspective of a portfolio manager, an investor who holds a diversified portfolio, or simply an economist who wishes to assess the average relation between firm value movements and exchange rates, the relevant issue is whether exchange rate movements affect the returns on certain groups of stocks more than others. To evaluate the economic impact of exchange rate movements on stock returns, we form portfolios that are long in firms with high international sales and short in firms with no international sales. This approach, which has not been used in the existing exposure literature, has two main advantages. First, it focuses on returns rather than changes in the adjusted R2. If exchange rates impact firms with high international sales and firms with no international sales differently, then the difference in returns between these groups of firms should be an informative gauge of the impact of exchange rates on firm value. Second, the regression framework assumes that exposures are linear and constant, which is unlikely to be true in many cases. The portfolio approach allows exposures to be both non-linear and time varying. Consistent with theory, we find that during periods of large currency depreciations (appreciations), firms with high international sales outperform (underperform) those with no international sales in 14 of 18 (16 of 18) countries. Although the magnitude of these effects varies widely across countries, during periods of large currency depreciations, the average difference in returns between the high and no international sales portfolios is 0.72% per month, whereas during currency appreciations these same firms underperform by an economically and statistically significant − 1.10% per month. When we use these portfolios in a regression, we find that, overall, a 1% appreciation in the home currency leads to a 0.21% loss in firm value for firms with high international sales as compared to firms with no international sales. We also find that these patterns are present in both high and low book-to-market equity firms but are concentrated in large firms. This suggests that our findings are not driven by some risk or behavioral explanation related to book-to-market equity or small cap firms. One problem with our portfolio sorts on international sales is that some firms with international sales have offsetting exposures if they also have operations abroad. Further, firms with no international sales may face significant exposures if they face foreign competitors. To address this issue, we directly form portfolios based on firms' past (estimated) exchange rate exposures. Such an analysis allows us to gauge whether our sorts on international sales truly capture firms with exchange rate exposure. The results of the direct sorts on past exposures are consistent with the results based on international sales sorts: during periods of large currency depreciations, the average difference in returns is 0.22% per month, whereas during currency appreciations the difference is an economically and statistically significant − 1.40% per month. Our findings provide evidence that exposure does vary systematically with international activity and that these relations are important for understanding variation in stock returns. Our results are particularly strong since we are only able to measure exposure net of firms' operational and financial hedging activities. These findings have implications for the international asset pricing literature in that showing that exposure broadly affects groups of stocks makes it possible that exposure can also be priced. The results should also be of interest in the many applications that seek to understand the sources of cross-sectional and time-series variation in stock returns. The remainder of the paper is organized as follows. Section 1 connects our approach to the theoretical and empirical literature that examines the relation between firm value and exchange rate movements. Section 2 describes the data, shows some basic properties of its coverage, and displays basic firm-level regression results for all firms. Section 3 relates exposure betas to its determinants through cross-sectional regression analysis. Section 4 presents portfolio returns during different periods of currency movements for portfolios that are long firms with high international sales and short firms with no international sales. Section 5 briefly examines some remaining issues related to exposure, and Section 6 concludes.

This paper examines the nature and the economic magnitude of exchange rate exposure using a unique firm-level database that covers many countries. Using time-series regressions over 5-year windows, we find that more firms are exposed to exchange rate movements than can be attributed to chance; however, exchange rates do not explain a large portion of the variation in individual firm stock returns. We estimate cross-sectional regressions of exchange rate betas on determinants of exposure and find that firm size, the level of international sales, foreign income, and foreign assets are all significantly negatively related to exposure. We document that the use of portfolios dramatically improves our ability to observe the economic importance of exchange rate exposure. We evaluate the average magnitude of these exposure effects by examining the relative performance of firms with high international sales as compared to those with no international sales during different periods of currency movements. Firms with high international sales outperform those with no international sales in periods of currency depreciations, but underperform during periods of currency appreciations. These patterns are pervasive across countries and are economically important. In 16 of 18 countries, firms with high international sales underperform those with no international sales during periods of currency appreciations by an average of 1.10% per month. We find that our results cannot be explained by book-to-market equity risk but are more concentrated in large firms which may have more exposure. Additionally, the patterns we document with international sales are evident when we form portfolios according to a firm's past foreign exchange exposure beta. These findings, combined with the individual stock regression results, show that while exposure may only explain a small proportion of the variation in stock returns for a particular firm, this effect is pervasive across firms with international activities; hence, exchange rates are important for explaining cross-sectional differences in stock returns. Overall our results provide evidence that exchange rate movements do affect firm value in a manner consistent with theory and that exchange rate movements have an economically significant impact on differences in average stock returns. The portfolio return difference between firms with high and no international sales (or exposure betas) is often greater than 1% a month and is even higher in some countries. These findings should be of interest to policy makers who wish to understand the effects of relative exchange rate movements on certain sectors of the economy and to investors who under- or overweight multinational corporations in their portfolios. It should be promising to consider the impact of exchange rate movements in portfolio optimization, value-at-risk, performance attribution, and other analyses that seek to understand major sources of co-variation among stock returns.