حل پازل نوردهی: بسیاری از جنبه های مواجهه با نرخ ارز

Theory predicts sizeable exchange rate (FX) exposure for many firms. However, empirical research has not documented such exposures. To examine this discrepancy, we extend prior theoretical results to model a global firm's FX exposure and show empirically that firms pass through part of currency changes to customers and utilize both operational and financial hedges. For a typical sample firm, pass-through and operational hedging each reduce exposure by 10–15%. Financial hedging with foreign debt, and to a lesser extent FX derivatives, decreases exposure by about 40%. The combination of these factors reduces FX exposures to observed levels.

Given the globalization of many industries, foreign exchange rate (FX) fluctuations are a source of uncertainty for many corporations. Empirical studies show significant effects of exchange rate changes on firm cash flows, sales, and competitive positions in product markets (e.g., Hung, 1992; Williamson, 2001). Similarly, theoretical models (such as Bodnar, Dumas, and Marston, 2002) predict that many firms should have significant exchange rate exposures. However, empirical studies have tended to point out weak (or non-existent) relations between exchange rate changes and firms’ stock prices.1 In this paper, we examine the discrepancy between theoretical predictions and observed levels of exchange rate exposure in a broad cross-section of global corporations. Our analysis investigates how firms combine three different mechanisms at their disposal for mitigating exchange rate risk. First, firms can (to varying degrees) pass through to customers the changes in costs due to exchange rate movements. Second, firms can often affect their exchange rate exposure by choosing the location and currency of costs (e.g., where factories are located). Third, firms can utilize an array of financial products, such as foreign currency denominated (FC) debt and FX derivatives, as exchange rate risk management tools. Our results show that each of these factors plays an important role in mitigating observed exchange rate exposure, and together they account for the vast majority of the discrepancy between prior theoretical predictions and observed exposures. Our analysis has two primary parts. First, we expand the theoretical model of Bodnar, Dumas, and Marston (2002) (hereafter BDM) to examine the exchange rate exposures of a global firm that can compete and produce in both a foreign and local market. In the BDM model, the exporting firm cannot sell in its own market and the local firm cannot produce abroad. By assuming that global foreign exchange rate exposure is a weighted average of a firm's foreign exchange exposure in the foreign market and the domestic market, we can derive optimal pass-through decisions and the resulting foreign exchange exposures of global firms in globally competitive industries. Our model generates exposures as a function of market share, product substitutability, pass-through, sales and costs in foreign currency that are smaller than the original BDM model under most conditions, and in some cases, the model generates negative foreign exchange rate exposures. Overall, our global competition model allows for a richer, more realistic, set of FX exposures since it allows firms to sell and source both at home and abroad. Second, we analyze a sample of 1,150 manufacturing firms in 16 countries using our global competition model. We show that pass-through and operational hedging are important for reducing the level of exchange rate exposure. However, after accounting for pass-through and operational hedging, theoretical exposures are still larger than observed exposures on average.2 We show that firms with high theoretical exposures are both more likely to have FC debt and more likely to use FX derivatives. We also conduct an attribution analysis that estimates the magnitude of the reduction in exposure due to each channel for a typical firm. Depending on the level of product substitutability, pass-through reduces exposure by about 10–15%. Operational hedging reduces exposure by similar amounts, while financial risk management (FC debt and FX derivatives) accounts for about a further 40% reduction in exposure. Altogether, firms reduce their gross exchange rate exposure by about 70% via the three channels. This reduction results in average exposure values very similar to those estimated from regression models. Consequently, for reasonable parameter values, it is not possible to reject our global competition model after correcting for the estimated effects of financial risk management. Our analysis augments an empirical literature that examines FX exposure. A number of studies estimate the exchange rate sensitivity of stock prices and typically find small exposures. The results of this paper suggest that these observed estimates are reasonable once the three channels which firms have to mitigate exposure are considered. Other studies report differences in exposures across industry classes and countries (Campa and Goldberg, 1999; Bodnar and Gentry, 1993; Marston, 2001; Allayannis and Ihrig, 2001; Williamson, 2001).3 These findings are also consistent with the range of exposure estimates obtained from our model. A separate strand of literature examines exchange rate risk premiums. Several studies (Dumas and Solnik, 1995; De Santis and Gerard, 1998; and Carrieri, Errunza, and Majerbi, 2006) show significant risk premiums at the aggregate market level, but not the industry level. Francis, Hasan, and Hunter (2008) claim that this is not due to firms hedging currency risk but instead the result of model misspecification which they correct by estimating a conditional pricing model at the industry level. While seemingly contradictory to our results, our model and evidence are actually consistent with both views. For example, we find that, on average, many firms retain some exchange rate exposure which could carry a risk premium. However, we find significant negative correlations when we compare our measures of hedging with the Francis, Hasan, and Hunter (2008) estimates of the currency risk premium. This is consistent with evidence from our analysis (as well as others) that firms reduce risk with financial hedging. Recent research also examines the associations between exposure and proxies for operational hedging. For example, Carter, Pantzalis, and Simkins (2003) examine the use of derivatives and operational hedges on foreign exchange exposures and show that exposures vary, not only as to whether a firm is a net exporter or net importer, but also across weak and strong dollar states. Allayannis, Ihrig, and Weston (2001) find that operational hedging strategies only benefit shareholders when used in combination with financial hedging strategies. Dewenter, Higgins, and Simin (2005) find evidence that operating hedging may help explain low exchange rate sensitivities of firms during the 1994 peso crisis in Mexico and the 1997 devaluation of the Thai Baht. Other research has examined the relationship between FX exposure and various types of financial risk management. Allayannis and Ofek (2001), Wong (2000), Simkins and Laux (1996), and Hagelin and Pramborg (2004) report significant negative associations between foreign exchange exposures and the use of financial hedges. Nain (2004) shows that an unhedged firm's foreign exchange exposure increases with the extent of hedging in the industry. This paper adds to these areas of research by quantifying the impact of operational and financial hedging on exposures. While a large number of studies show that FX derivative use is common among global firms with foreign sales, others question the importance of derivatives for financial risk management. For example, Guay and Kothari (2003) show that the magnitude of derivative use is small compared to firm sales, assets, and income, thus, derivatives are a small piece of nonfinancial firms’ overall risk profile. Similarly, Brown, Crabb, and Haushalter (2006) show that potential gains from time-series variation in the derivative portfolios of gold mining firms are economically insignificant. The results from our analysis provide a potential resolution of this puzzle as well. Specifically, financial hedging with FC debt appears to have a larger effect on exposure than the use of FX derivatives though the use of both is widespread in our sample. The paper is organized as follows. Section 2 motivates our analysis with a brief case study of the global automotive industry. Section 3 develops our global competition model for foreign exchange rate exposures and pass-through based on Bodnar, Dumas, and Marston (2002). This global competition model is estimated for the sample of manufacturing firms in Section 4. Finally, Section 5 concludes.

Previous studies have examined in detail independent channels that give rise to exchange rate exposure, but little academic research has investigated the interrelationships between these determinants. For example, the associations between exchange rate exposure, product market competition, and corporate risk management are not well understood. This paper takes a step toward filling this gap by examining these relations for a global sample of nonfinancial companies. Specifically, we assume corporate financial managers can use pricing policies, operational hedging (e.g., international allocation of production costs), and financial hedging strategies in order to mitigate the effects of currency fluctuations. Moreover, these decisions are not made under the assumption that the firm operates in isolation, but rather after a detailed analysis of the industry's situation and the actions of industry peers. Thus, by reflecting the broader economic context the firm is operating in and the nature of competition it is facing, firm-level decisions make foreign exchange exposure endogenous. Our primary results suggest that each of the three channels for mitigating exchange rate exposure is likely to be important for our sample of large global manufacturing firms. Exchange rate pass-through and operational hedging are each responsible for about a 10–15% reduction in exposure, and financial risk management accounts for about another 37–43% decrease in exposure relative to a hypothetical firm that cannot mitigate exposure at all. Altogether, firms reduce their gross exchange rate exposure by about 70% via the three channels to levels measured from equity prices. Consequently, our analysis suggests that the relatively weak foreign exchange rate exposure, which has been considered a puzzle by some, is to be expected once accounting for all the relevant factors. In particular, firms implicitly appear aware of their gross exposures and adjust their operations and financing activities. Our results point to a need for further research examining the precise role of financial risk management in globally competitive corporations.31 Overall, our study contributes to the literature in several ways. First, we take a more comprehensive approach to studying exchange rate exposure. For example, the model we derive allows for global firms in an imperfectly competitive global market. Second, our sample of firms using financial risk management to mitigate exchange rate exposure is among the largest examined to date. Finally, and most importantly, we are able to shed light on a puzzle in financial economics by carefully decomposing the close relationship between theoretical and observed foreign exchange rate exposures.