The Exchange Rate Exposure Puzzle A Quantitative Study of Public Swedish, Norwegian and Danish Firms. Author: Academic Advisor: MSc in Tom Aabo Department of Finance Aarhus School of Business Aarhus University May 2008
ii
Declaration of Independent Work I,, hereby confirm that I have composed this work on my own, and that I did not use any other materials than what is quoted. Aarhus, May 8 th 2008 iii
Abstract To empirically establish a relationship between exchange rate fluctuations and firm value has been a challenging issue during the last twenty years, the theoretical foundations are evident but the empirical evidence is complex. This study compliments earlier research and examines the so-called exchange rate exposure for a sample of Swedish, Norwegian and Danish firms between 1999 and 2006. The number of firms with a significant exposure to a national trade-weighted index was lower than expected. In order to further extend the investigation, bilateral exchange rates were incorporated in the model which gave a more significant result. Besides, the outcome further indicated that a weighted-index and bilateral currencies are complimenting measures. The methodological aspects regarding exchange rate exposure have been heavily discussed in earlier research. Therefore, the impact from the choice of market portfolio and the observation frequency was carefully examined. The choice of market portfolio did not alter the results for the sample of Swedish, Norwegian and Danish firms whereas the observation frequency gave varying results. Furthermore, there was some evidence of a potential time-varying nature of exchange rate exposure while contemporaneous effects of exchange rate changes on stock returns is more important than lagged effects. Finally, recent research has emphasized the importance and relevance of measuring firmspecific exchange rate exposure. Therefore, firm-specific information was collected from annual reports in order to establish relevant individual currencies for each firm with respect to their influence on firm value. Even though the number of firms with significant exposure did not increase significantly after the inclusion of firm-specific exchange rates, a more comprehensible picture of the relationship between firm value and exchange rate fluctuations could be established. iv
Table of content Table of content...1 1. Introduction...4 1.1 Background...4 1.2 Problem Statement...5 1.3 Delimitations...7 1.4 Disposition...7 2. Literature Framework...8 2.1 Theoretical Review...8 2.1.1 The Macroeconomic Environment...8 2.1.2 Macroeconomic Risk and Exposure...9 2.1.3 The Exchange Rate...9 2.1.4 International Parity Relationships...10 2.1.4.1 Purchasing Power Parity (PPP)...10 2.1.4.2 Interest Rate Parity (IRP)...11 2.1.4.3 International Fisher Effect (IFE)...11 2.1.4.4 Forward Rate Unbiased (FRU)...11 2.1.5 Exchange Rate Risk and Exposure...12 2.1.5.1 Transaction Exposure...12 2.1.5.2 Translation Exposure...12 2.1.5.3 Economic Exposure...13 2.1.5.4 Limitations of Traditional Exposure Measures...13 2.1.6 Implications of Deprecations and Appreciations...14 2.1.7 Market Efficiency and Rational Expectations...15 2.1.8 Management of Exchange Rate Exposure...16 2.2 Empirical Review...17 2.2.1 Exchange Rate Exposure of US Firms...18 2.2.2 Exchange Rate Exposure outside US...22 2.2.3. Recent Studies...25 2.3 The Exposure Puzzle...27 3. Methodology...29 3.1 Model Argumentation and Variable Definitions...29 3.1.1 Trade-weighted Exchange Rate Index, a One-factor Model...30 3.1.2 Trade-weighted Exchange Rate Index, a Two-factor Model...31 3.1.3 Orthogonalization...32 3.1.4 Bilateral Exchange Rates...33 3.1.5 Intertemporal Stability of Exchange Rate Exposure...34 3.1.6 Lagged Relationship...34 3.2 Hypothesis Development...35 3.3 Application of Statistical Tests and Econometric Issues...37 3.3.1 Ordinary Least Squares Regression...37 3.3.2 Multicollinearity...40 3.3.3 Heteroscedasticity...40 3.3.4 Autocorrelation...42 4. Data...43 4.1 Sample Selection Procedure...43 4.1.1 Sweden...44 4.1.2 Norway...45 4.1.3 Denmark...45 1
4.2 Data Description...45 4.2.1 Stock Return Data...45 4.2.2 Exchange Rate Data...48 4.2.3 Firm-Specific Data...50 5 Empirical Part...52 5.1 Sweden...52 5.1.1 Trade-Weighted Index Exposure Estimates...52 5.1.2 Robustness Aspect I; Intertemporal Stability...56 5.1.3 Robustness Aspect II; Lagged Exposure Estimates...57 5.1.4 Firm-Specific Exposure Estimates...58 5.2 Norway...59 5.2.1 Trade-Weighted Index Exposure Estimates...59 5.2.2 Robustness Aspect I; Intertemporal Stability...62 5.2.3 Robustness Aspect II; Lagged Exposure Estimates...64 5.2.4 Firm-Specific Exposure Estimates...65 5.3 Denmark...66 5.3.1 Trade-Weighted Index Exposure Estimates...66 5.3.2 Robustness Aspect I; Intertemporal Stability...69 5.3.3 Robustness Aspect II; Lagged Exposure Estimates...71 5.3.4 Firm-Specific Exposure Estimates...72 5.4 Statistical Assumptions...72 5.4.1 Multicollinearity...73 5.4.2 Heteroscedasticity...73 5.4.3 Autocorrelation...74 6. Analysis...75 6.1 Exposure to the Trade-Weighted Index...75 6.2 Exposure to Bilateral Exchange Rates...77 6.3 Firm-Specific Exchange Rates...79 6.4 Robustness Analyses...82 6.4.1 Observation Frequency...82 6.4.2 Choice of Market Portfolio...83 6.4.3 Intertemporal Stability...84 6.4.4 Lagged Exchange Rate Exposure...85 7. Conclusion...87 8. Shortcomings and Future Research...89 9. Bibliography...90 9.1 Books and articles...90 9.2 Electronically sources...92 10. Appendix...93 10.1 Export share of GDP...93 10.2 Macroeconomic Uncertainty in a Firm Perspective...94 10.3 Parity Relationships in International Finance...95 10.4 Exposure Management Instruments...96 10.5 Overview of Empirical Findings...97 10.6 Equations...99 10.7 Hypothesis...100 10.8 Model Argumentation and Hypothesis Development...101 10.9 Overview Exchange Rates and Exchange Rate Agreements...102 10.10 Overview of Inflation Rates...103 10.11 Swedish Firm-specific Information...104 2
10.12 Norwegian Firm-Specific Information...107 10.13 Danish Firm-specific Information...108 10.14 Trade-Weighted Index...110 10.15 Exchange Rate Development...111 10.16 Overview of CD-ROM content...114 10.17 Firm-Specific Exchange Rate Exposure Sweden...115 10.18 Firm-Specific Exchange Rate Exposure Norway...119 10.19 Firm-Specific Exchange Rate Exposure Denmark...121 3
1. Introduction One of the most important prices in the international economy today is the exchange rate. It simplifies the conversion between prices into different currencies; the importance of exchange rate fluctuations for companies involved in foreign activities is hence substantial. It is a widely held view that exchange rates affect firm value and thereby creates uncertainty for multinationals all over the world. Since exchange rates can affect cash flows and stock prices of firms, the exposure to this risk is a key concern for investors, analysts and managers. The magnitude of the importance of exchange rate variability is also evidenced by the increasing efforts firms place on resource allocation, exchange risk management and business strategy management. 1 The returns of all assets do respond to changes in economic conditions, yet, the responses vary. 2 understanding of the impact of foreign exchange risk is an important element of both firm valuation and risk management. The 1.1 Background The Bretton Woods system 3 ended in the year of 1973. In the final years before the break up, the debate focused on the introduction of floating exchange rates and in particular their impact on the exchange rate fluctuations. 4 The volatility of both nominal and real exchange rates has increased substantially since then but empirical studies of exchange rates and their influence on international business activities have been complex to determine. In the beginning of the 1990s more than a few researchers considered it to be noteworthy that so little empirical researched had been undertaken within the field of exchange rate exposure. 5 Even though exchange rates are generally believed to have an effect on corporate performance, the specification of the mechanisms that influence the firm value is complex. A change in the exchange rate could, from a general point of view, affect the market value of the firm through its input prices, its output prices, the number of outputs sold, the inputs required and the discount rate to value these cash flows. Consequently, even a firm without foreign sales, input, assets and liabilities will be exposed to exchange rate volatility. 6 To establish an accurate relationship between exchange rates and corporate performance is a complicated assignment. The multinational firm with several product lines, operations in 1 Amihud (1994). 2 Fama and French (1989). 3 The Bretton Woods System was an international monetary management system that established financial and commercial rules for the world s major industrial countries. 4 See Copeland (2005), pp. 23-33, for a more thorough overview of the post-war currency market history. 5 See for example Jorion (1990) and Amihud (1994). 6 Amihud and Levich (1994), pp. 2-3. 4
numerous countries and competitive pressures from yet other nations is a demanding task to analyse. Not forgetting the categorization of exchange rates; daily nominal changes are rather easily hedged at future, forward, swap and option markets but long-term real changes involve challenging questions regarding market locations, product mix composition and production place choice. The most serious problem is the medium-term complications since it might be uncertain whether the exchange rate fluctuations are of a permanent or a temporary nature. The uncertainty is thus whether to use a financial hedge for a possible short-term movement or to use an operational change when the movements are considered to be permanent and real. 7 The lack of evidence in support of exchange rate exposure is confusing since concern with exchange rate fluctuations is widespread among corporations. Over the last two decades, the exchange rate exposure puzzle has been a challenging issue. The relationship between the sensitivity of the firm value to exchange rate changes has been examined substantially among researchers. Although much has been learned; neither a consensus of relevant parameters nor a unique model integrating all complexities has been assessed. A number of questions are still unanswered in the understanding of exchange rate exposure. 1.2 Problem Statement To measure foreign exchange exposure is an essential issue of international finance management since exchange rate movement is a main source of uncertainty for many corporations. But the existing empirical evidence is rather puzzled; several studies have failed to find a strong significant relationship between contemporaneous exchange rate volatility and the value of firms. More than a few studies have found the weak results and the subsequent insignificant correlations between exchange rate changes and stock return unpersuasive. The inability of early empirical findings to establish a relationship between the firm value and the exchange rate has resulted in numerous studies considering more rigorous issues. This study focuses on Sweden, Norway and Denmark; three countries with small domestic markets and a relatively large number of listed firms with substantial overseas activities. There are several reasons and motivations to investigate the Scandinavian 8 market, specifically Sweden, Norway and Denmark, and look for evidence of exchange rate exposure. Firstly, the Scandinavian countries are all small open economies with a considerable exporting activity. 9 7 Amihud and Levich (1994), pp. 4. 8 Scandinavia is defined as comprising Denmark, Norway and Sweden. See Columbia Encyclopaedia and Encyclopaedia Britannica for more information regarding the definitions of Scandinavia and the Nordic countries. 9 See Appendix 10.1 regarding information of the export as a percentage of GDP for Sweden, Norway, Denmark and USA. 5
When the domestic market is small; selling abroad becomes much more important. Therefore it is possible to assume that the Scandinavian countries has an extensive involvement in, and are very dependent on, international operations. The more open the country, the more firms business is related to foreign trade and consequently more exposed to exchange volatility. 10 Subsequently, exchange rate volatility should have a considerable impact on firm value. Investigating open economies is necessary in order to provide further insight into the complex relationship between firm value and exchange rate volatility. As Nydahl (1999) emphasizes, it is quite surprising that most of the empirical research until now has studying one of the least open economies in the OECD, namely USA. Secondly, this study is a combined analysis of all three Scandinavian countries. To examine the effect of exchange rate fluctuations on Swedish, Norwegian and Danish multinational firms is therefore both an interesting and attractive opportunity. The relationship between the Scandinavian stock markets and exchange rate changes have not been extensively analysed in the past. Besides, a simultaneous examination of Sweden, Norway and Denmark has not been conducted within earlier research, as far as the knowledge of the author. Finally, the Swedish krona and Norwegian krone are both small, freely floating currencies which accordingly should be more vulnerable to unanticipated fluctuations in foreign exchange rates. The Danish krone is pegged to the euro and do therefore not experience the same volatility with regard to unanticipated fluctuations in the euro. The main objective of the Central Bank of Denmark, Nationalbanken, is to keep the krone stable vis-à-vis the euro. However, the exchange rate between the Danish krone and other currencies is freely floating. 11 The main purpose of this study is to investigate whether the value of Swedish, Norwegian and Danish corporations are affected by contemporaneous and lagged unexpected fluctuations in the value of currencies; partly on an aggregate level and partly on a firm-specific level. Several objectives are therefore appropriate. Is there any predictable relationship between a trade-weighted index of foreign currency and stock returns on the Swedish, Norwegian and Danish market? Is there any predictable relationship between selected bilateral currencies and stock returns on the Swedish, Norwegian and Danish market? At a firm-specific level, is there a predictable relationship between firm-specific exchange rates and stock returns for Swedish, Norwegian and Danish firms? 10 Friberg and Nydahl (1999) find a significant positive relationship between exchange rate exposure on the stock market and the openness of the economy. 11 For more information regarding the Danish foreign exchange policy see: http://www.nationalbanken.dk. 6
Conducting a firm-specific analysis avoids the averaging effect that grouping of firms can experience. Using a disaggregated set of data can prove to enhance some of the aggregating problems in earlier studies. 12 A possible lagged relationship between exchange rates and firm value are also examined. To incorporate both a contemporaneous and a lagged correlation allows a more thorough analysis; the complex set of issues involved when estimating the relationship between the value of firms and exchange rate changes may have signs of a miss-pricing effect. If stock price adjustments not occur immediately, inclusion of lagged changes are appropriate. 13 1.3 Delimitations This study focuses only on Swedish, Norwegian and Danish firms. It is for that reason not possible to draw any general conclusions about the relationship between exchange rate exposure and firm value; the results are only relevant in the context of this geographical area. In addition, only public firms are included in the sample. The elimination of private firms may lead to a biased result but the data accessibility of private firms is insufficient. Furthermore, a number of earlier researchers have emphasized different firm characteristics that explain the nature of the exchange rate exposure. The determinants of foreign exchange rates are however not incorporated into this study and no attempts to link the sensitivity to exchange rate volatility to firm-specific variables will therefore be performed. The objective of this study is to investigate whether Swedish, Norwegian and Danish firms are exposed to exchange rate fluctuations and not the underlying determinants. Additionally, the Swedish, Norwegian and Danish firms are investigated separately. The intention is hence not to compare the exchange rate exposure in the three countries but to analyse the results individually. 1.4 Disposition The study is organized as follows. Chapter 2 presents the theoretical literature framework and earlier empirical evidence on exchange rate exposure. Chapter 3 discusses the methodology with model argumentation, variable definitions, hypothesis development and the application of statistical tests. Chapter 4 describes the sample selection procedure and the data. Chapter 5 consists of the empirical evidence. Chapter 6 analyzes the exchange rate exposure of Swedish, Norwegian and Danish firms, both on an aggregated and on a firm-specific level. Chapter 7 concludes and chapter 8 contains shortcomings further research. Chapter 9 and chapter 10 enclose the bibliography and the appendices respectively. 12 See for example Nydahl (1999), Jong et al (2006) and Ihrig (2001). 13 See for example Amihud (1994) and Bartov and Bodnar (1994). 7
2. Literature Framework This section addresses the literature framework and is divided into two separate parts; the theoretical review and the empirical review. The former lays the foundations of exchange rate exposure and the latter consist of relevant empirical research within the field of exchange rate. The empirical literature exposure has increased considerable during the last twenty years. This section highlights the main findings and conclusions and can be subdivided into three different categories; exchange rate exposure of US firms, exchange rate exposure outside the US and the most recent studies. The literature framework concludes with a brief section about the exposure puzzle. 2.1 Theoretical Review This chapter of exchange rates and international finance presents the important theoretical preliminaries. The introduction to the macroeconomic environment and macroeconomic risk and exposure is followed by the definition of an exchange rate and its characteristics. The key equilibrium international parity relationships will be established and the conventional measures of exchange rate risk and exposure will be discussed. The chapter continues with the implications from exchange rate deprecation and appreciation and the development of the efficient market hypothesis and rational expectations. The theoretical review is concluded with the management of exchange rate exposure. 2.1.1 The Macroeconomic Environment A higher degree of capital movement and the increased openness to international trade have made individual countries more vulnerable to real and monetary shocks. At the corporate level, this can only imply a greater concern regarding exchange rates, interest rates, inflation, and demand and competition conditions. Hence, the management of the firm, analysts and investors need to understand how the firm results are affected by the uncertainty involved in the macroeconomic development. Accordingly, it is not relevant to distinguish between domestic and international firms when examining the macroeconomic exposure. Every single firm is exposed to shocks and disturbances; it is the channels that may differ. 14 Three sets of factors establish a firm s exposure. Firstly, the macroeconomic structure, defined by the capital mobility and the speed of price adjustments, determines interest rates, exchange rates and inflation effects after a macroeconomic disturbance. Secondly, the authorities policy regimes influence the degree to which interest rates, exchange rates and inflation adjust to the disturbance 14 Oxelheim and Wihlborg (2005), pp. 11-12. 8
in question. Thirdly, firm-specific factors in the output and input markets determine the sensitivity of a firm s value and cash flows to changes in macroeconomic conditions. It has been argued that interest rates, exchange rates and inflation are the most important variables that link macroeconomic fluctuations and firm performance. Each firm must therefore determine which variables within each group that have the largest influence. 15 2.1.2 Macroeconomic Risk and Exposure In general, the concept of risk refers to the degree and probability of unanticipated changes that have an effect on a firm s value, cash flows and profitability. Risk is hence evaluated relative to expected changes in a specific variable. From an individual s point of view, risk is defined as the variance of the return of the portfolio. This is not only negative; it presents opportunities for both corporations and individuals. Better forecasting abilities and flexibility when the conditions of the business settings change are possible opportunities that risk management deals with. 16 Macroeconomic risk depends on the uncertainty in the environment for all firms, though the exposure of the individual firm is firm-specific. To describe and categorize the risk in the macroeconomic environment; interest rate risk, exchange rate risk and country risk are the interdependent variables. The exchange risk has received the most attention in the literature and the majority of approaches to managing this risk assume that the exchange rate is implicitly and explicitly independent of the variability of other risks. However, this is not always appropriate; two or more variables may not be two different risk factors. 17 This study will only focus on the exchange rate exposure. For an overview of the macroeconomic uncertainty and the relationship to a firm s cash flows see Appendix 10.2. 2.1.3 The Exchange Rate The exchange rate is simply a price. Throughout the rest of this study the exchange rate will be defined as the domestic currency price of foreign currency. Consequently, a rise in the exchange rate signifies a rise in the price of foreign currency, a relative cheapening of the domestic currency and hence a depreciation. The opposite reasoning holds for a fall in the exchange rate and a subsequent appreciation of the domestic currency. A bilateral exchange rate is thus a two-country context. When considering the general overall value of a specific currency it is appropriate to examine it as an index of its international value. The multilateral effective exchange rate is a weighted average of the domestic currency against foreign currencies. 18 15 Oxelheim and Wihlborg (2005), pp. 19-24. 16 Oxelheim and Wihlborg (2005), pp. 28-32. 17 Oxelheim and Wihlborg (2005), pp. 39-41. 18 Copeland (2005), pp. 3-8. 9
As with any other market, the exchange market is driven by supply and demand. Apart from speculators and foreign investors, exporters and importers supply and demand foreign currency. Exporters supply goods to foreign buyers and get in return either foreign currency directly or domestic currency that has been purchased overseas by the importer with foreign currency. The net effect is associated with a supply of foreign exchange. Inversely, an importer is associated with the demand for foreign currency. In the currency market, there is a complete symmetry between demand and supply as the agents involved often are the same institutions on both sides. However, in the short term, this is not the likely to be true for importers and exporters as they do not switch when exchange rates are unfavourable. 19 2.1.4 International Parity Relationships International business transactions occur in many different countries and the foreign exchange market makes it possible to trade in foreign currencies. To facilitate the management of foreign exchange risk, it is of great importance to understand the key equilibrium relations involving international prices, interest rates, inflation rates and spot versus forward exchange rates. Those fundamental propositions require a perfect capital market 20 and if no deviations exist, investing in a particular country or in a specific currency would not result in any excess return. 21 Deviations from the perfect capital market assumption imply that several other economic, political and cultural forces will influence the economic variables embedded in the following relationships. It is not surprising that the parity conditions do not hold at any point in time; uncertainty and changing expectations and beliefs all affect the equilibrium conditions. 22 Appendix 10.3 for an overview of the parity relationships in international finance and the relevant equations. 2.1.4.1 Purchasing Power Parity (PPP) The purchasing power parity is an expression of one of the most fundamental laws in economics, the law of one price. If the market is efficient, identical tradable goods and financial assets should be equal worldwide when the prices are exchange-adjusted. It is the method of using the long-run equilibrium exchange rate between two currencies to equalize the currencies purchasing power in their domestic countries for a given basket of goods. Hence, the purchasing power parity doctrine states that individuals will value currencies for what they will buy. Furthermore, the purchasing power parity predicts that the equilibrium relationship between two currencies reflects market forces and underlying economic conditions; any deviations from the parity relationship See 19 Copeland (2005), pp. 11-13. 20 I.e. numerous buyers and sellers, no taxes, no information symmetries or transaction costs and no government controls. 21 Copeland et al (2005), pp. 830. 22 ibid 10
will tend to be self-correcting. Nonetheless, estimating the purchasing power parity is a complex task since different countries not only differ in a uniform price level; the inhabitants also typically consume different baskets of goods. 23 2.1.4.2 Interest Rate Parity (IRP) The interest rate parity implies that the ratio of forward and spot exchange rates will equal the ratio of foreign and domestic nominal interest rates. If the parity does not hold, covered interest arbitrage will move the market back to equilibrium. This arbitrage condition says that the returns from borrowing in the domestic currency, exchange to foreign currency and investing in interestbearing instruments of the foreign currency and simultaneously buy a future contract to convert the foreign currency to domestic currency at the end of the investment period should equal the returns from a similar interest-bearing investment in the domestic currency. If the returns are different, investors can in theory make a risk-free return through arbitrage. The interest rate parity simply states that spot and future prices of a bilateral exchange rate incorporate the interest differential between two currencies. 24 2.1.4.3 International Fisher Effect (IFE) Domestic prices changes cause a difference between nominal and real interest rates in a single country. When pairs of countries are involved, the international fisher effect illustrates the relationship between interest rate differences across countries and the expected exchange rate changes. Simply, the international fisher effect indicates that differences in nominal exchange rates will have an impact on expected foreign exchange rates. Similar to the interest rate parity, under the assumption about perfect capital markets, the international fisher effect is built on arbitrage. Accordingly, the expected future spot exchange rate must equal the percent interest differential. 25 2.1.4.4 Forward Rate Unbiased (FRU) The international fisher effect holds under perfect capital market conditions. The forward exchange discount or premium will therefore equal the expected percentage change in the exchange rates. Nevertheless, the risk premium and deviations in expectations may cause inequalities between short-term movements in expected exchange rates changes and percent changes in the forward discount or premium. If the differences are small, the forward rate is generally an unbiased predictor of the future spot rate. 26 23 Copeland et al (2005), pp. 831-832. 24 Copeland et al (2005), pp. 832-834. 25 Copeland et al (2005), pp. 835-836. 26 Copeland et al (2005), pp. 836. 11
2.1.5 Exchange Rate Risk and Exposure A major source of macroeconomic uncertainty affecting open economy corporations is the exchange rate uncertainty. A firm is exposed to exchange rate risk when the firm value is affected by fluctuations in one or several currencies. The exposure effect can be both direct and indirect. Exposure is measured by the sensitivity of the systematic relationship between the exchange rate and the real value of a particular asset. A systematic relationship indicates a tendency for the variables to change in a fairly predictable way; the factors are consequently on average related to each other. As a consequence, a firm s exchange rate exposure is equal to how much the value of the firm will be affected by a change in the exchange rate in question. 27 It is conventional to classify exchange rate exposure into three types; transaction exposure, translation exposure, and economic exposure. 28 2.1.5.1 Transaction Exposure Transaction exposure is the risk that the exchange rate may change during the short interval of time when a foreign currency transaction is entered into and when it is settled. The exposure is defined as the sensitivity to unexpected exchange rate changes of the domestic currency values of the firm s contractual cash flows that are denominated in foreign currencies. Consequently, it is fixed-price contracts in a world with randomly changing exchange rates. The terms of these transactions are established and settled at a given point in time and their exposure can easily be measured by the accounting systems. Whenever the firm has receivables or payables denominated in a foreign currency it is subject to transaction exposure. 29 2.1.5.2 Translation Exposure The translation exposure is an accounting concept; it is the net position on the balance sheet in a foreign currency translated at the present exchange rate. In reality, the translation exposure arises when a multinational firm consolidates its balance sheet in an annual or quarterly report; the financial statements of foreign operations are converted from the local currency to the reporting currency. If exchange rates have changed since the previous reporting period, translating a subsidiary s assets and liabilities denominated in local currency results in foreign gains or losses. Different accounting standards recommend diverse translation methods and translation rates. 30 This is however beyond the scope of this report. 31 27 Levi (1990), pp. 185-186. 28 Eun and Resnick (2004), pp. 283. 29 Eun and Resnick (2004), pp. 303-304. 30 Oxelheim and Whilborg (2005), 62-66. 31 As Oxelheim and Whilborg (2005) remarks, the actual choice of translation method/rule is not very important when the market participants are informed about it. 12
2.1.5.3 Economic Exposure The combined effect of transaction exposure and operating exposure is often referred to as economic exposure. Operating exposure is the implications from exchange rate volatility for operating cash flows of importing and exporting companies. In general, an importing firm loses from a depreciation of the home currency and gains from an appreciation. When the domestic currency depreciates the importing goods becomes more expensive and when the domestic currency appreciates importing goods becomes cheaper. The opposite effect is evident for an exporting firm. 32 Hence, fluctuating exchange rates can seriously alter the relative competitive position of corporations in domestic and foreign markets trough the affect on their operating cash flows. 33 Unlike the exposure on assets and liabilities on the accounting statements, operating exposure depends on the effect of randomly changing exchange rates on a firm s competitive position. 34 The economic exposure is measured over longer time horizons than both transaction and translation exposure. As exchange rate fluctuations alter the relative price of various goods sold in different countries, they affect indirectly the firm s economic environment and the future development possibilities. If the value of the firm is defined as the present value of expected future cash flows then the degree to which exchange rate volatility affects the value of the firm is measured by the economic exposure. The firm value could be affected through existing contracts, transaction exposure, or through changes in the value of revenues and costs, operating exposure. The economic exposure thus depends on the operations of the firm and consequently, only few firms will be unaffected by currency fluctuations. It is also apparent that the exposure will vary considerably across firms. 35 2.1.5.4 Limitations of Traditional Exposure Measures The traditional approach to measure the macroeconomic exposure, namely the translation and economic exposure has both limitations and weaknesses. Firstly, it deals with only one macroeconomic risk and therefore disregards the simultaneous adjustments that the exchange rate, interest rate and inflation often experience when the macroeconomic conditions and policies change. Secondly, the exchange rate exposure may not be stable over time. Thirdly, conventional measures often capture the direct exposure on accounting values. Exchange rates movements affect a firm s cash flows in a variety of ways that may not be observable in accounting data. While the direct exposure, that is to say transaction, translation and contractual exposure, can be easily hedged, indirect exposure provides considerable variability in cash flows for a large number 32 Levi (1990), pp. 440-466. 33 See 2.1.6 for a more thorough description about the implications of deprecations and appreciations. 34 Eun and Resnick (2004), pp. 288-289. 35 Oxelheim and Wihlborg (2005), pp. 51-62. 13
of corporations worldwide. Fourthly, the traditional approach does not distinguish between anticipated and unanticipated exchange rate fluctuations. The exposure management emphasizes unforeseen changes and not expected changes. 36 The traditional definition of exchange rate exposure from accounting information does not appear to be appropriate. Two more novel approaches to measure exchange rate exposure have been proposed, the capital market approach and the cash flow approach. By using historical data in a regression analysis, the capital market approach measures the exchange rate exposure. The cash flow approach is more comprehensive since it takes into consideration several macroeconomic and firm- and industry-specific variables. 37 Different models have been presented during the last three decades. The relationship between firm value and exchange rate fluctuations depends on a large number of variables; import and export structure, involvement in foreign operations, currency denomination of competitors and the competitiveness of the firm s input and output market. A consensus concerning the most relevant variables and a unique model that integrates all complexities does not exist so far. 38 A more detailed discussion regarding different model aspects is found in the methodology section. 2.1.6 Implications of Deprecations and Appreciations The relative price of domestic and foreign goods that affect both the present and the future expected cash flows of firms with international activities changes after exchange rate movements. When the domestic currency appreciates, exporting goods become more expensive in terms of the foreign currency which may lead to a drop in foreign demand, foreign sales revenue or even both. As a consequence, an exporting firm s value would be affected negatively. An importing firm experiences the opposite; when the domestic currency appreciate, imports become cheaper in terms of the domestic currency. It is possible that that these two contrary effects cancel out the total effect. But a general point of view is that import prices respond less than proportionally to exchange rate changes; the value of an importing firm should thereby experience a small direct effect. Domestic firms may also be affected by changes in the exchange rate, either due to changes in aggregated demand or because of the cost of imported inputs. Additionally, domestic firms competing against importing firms will be exposed to exchange rate fluctuations. 39 If the domestic GDP increases, the domestic currency is, ceteris paribus, expected to appreciate. The expectations of a strengthening domestic currency are incorporated into the current 36 Oxelheim and Wihlborg (2005), pp. 80-81. 37 Oxelheim and Wihlborg (2005), pp. 99-122. 38 For a more complete understanding of the development of a theoretical model see Muller and Verschoor (2006b). 39 Levi (1990, pp. 441-460. 14
exchange rate. Hence, if an appreciating domestic currency is associated with a stronger economy then the reduced demand for an exporting firm is at least to some extent offset by the increased demand for its products in the strong domestic economy. Similarly, the opposite relationship is prevailing when the domestic currency weakens. An importing firm experiences an additive effect when the domestic currency appreciates due to the higher price for its products and the higher domestic demand. When the domestic currency depreciates, the inverse relationship is present. 40,41 2.1.7 Market Efficiency and Rational Expectations In his classic statement of the efficient market hypothesis, Fama (1970) defined an efficient market as one in which prices fully reflect all available information. If the market price fails to incorporate all available information, unexploited profit opportunities must exist. 42 The rational expectations hypothesis states that the market s expectation or consensus forecast of the future value of a variable equals the mathematical expected value, conditional on all available information. If expectations are rational, wrong forecasts will occur but the errors will be random. A rational expectation in an efficient market is a return just sufficient to compensate for the time value of money and risk. 43 A price is said to follow a pure random walk if the change in the price from one period to another is purely random and displays no pattern over time. In competitive markets, with rational risk-neutral investors 44, returns are unpredictable and prices follow random walks. If past price changes could be used to predict future price changes, market participants could make easy profits. In a competitive market these easy profits would not last long; when investors try to take advantage of the opportunity, prices will adjust immediately until any abnormal profit disappears. Price patterns will no longer exist and the price changes in one period are independent of changes in the next; the price follows a random walk. 45 Economists usually define three levels of market efficiency characterized by the degree of information that is reflected in the prices. The weak-form of market efficiency implies that prices reflect the information in historical prices. When markets are efficient in the semi-strong-form, prices reflect all past prices and all other published information. Finally, in the strong-form of 40 Wachtel (1989), pp. 262-289. 41 Levi (1990), pp. 101-117. 42 Fama (1970). 43 Copeland (2005), pp.315-327. 44 A rational investor maximizes his utility. See for example Copeland et al (2005), pp. 45-69 about utility theory given uncertainty. 45 Brealey and Myers (2003), pp. 347-351. 15
market efficiency, prices fully reflect all available information. Ever since Maurice Kendall s 46 discovery in 1953 that prices follow a random walk, tests of the efficient-market hypothesis have been conducted in a large scale. 47 Early researchers concluded that the hypothesis of efficient market was a noteworthy good description of the reality. But, evidence of puzzles and anomalies arose. 48 Some researchers argued that the answers and explanations to these anomalies laid in the behavioural psychology. People are not considered to be fully rational with respect to their attitudes to risk and their beliefs about probabilities. Another important factor for an anomaly to persist is the role of professional investors and limitation to arbitrage. Less than fully rational investors trade against arbitrageurs whose resources are limited by risk aversion, short horizon and agency problems. Since investors not are immune to irrational behaviour, and arbitrage may be limited, behavioural finance is a possible explanation for puzzles and anomalies. 49 In the view of behavioural finance, economic theory does not lead us to expect financial markets to be efficient. The systematic and significant deviations from efficiency are expected to persist for long periods of time. The formation of the efficient market hypothesis is built on three arguments; investors are believed to be rational and to value securities rationally, if irrational investors exist, their trades are random and cancel out each other, if the trades do not cancel out each other, rational arbitrageurs eliminate the influences on prices. The efficient market hypothesis is thus first and foremost a consequence of equilibrium in competitive markets with fully rational investors. But even the existence of irrational investors is consistent with the hypothesis. However, when studying the human behaviour in competitive financial markets, behavioural finance rests on limitation to arbitrage and investor sentiment. Arbitrage is limited due to lack of perfect security substitutes and because prices not converge instantaneously to fundamental values. Investor sentiment explains how investors form their valuations and beliefs and hence their demand for specific securities. 50 2.1.8 Management of Exchange Rate Exposure With foreign exchange rate risk and exposure defined, it is suitable to consider the techniques to reduce and avoid risk and exposure. The objective of the firm s management is it to maximize the value of the firm. In a perfect market firms cannot increase their value by hedging; corporate diversification is not necessary, since individual investors can diversify risk themselves. Investors do not face higher transaction costs and information costs than corporations. Hence, firms should have a risk-neutral attitude and chose the asset (liability) with the highest (lowest) 46 See Kendal and Hill (1953)for more information regarding randomness and price patterns. 47 Brealey and Myers (2003), pp. 351. 48 Brealey and Myers (2003), pp. 358-360. 49 Schleifer (2000), pp. 1-2. 50 Schleifer (2000), pp. 2-27. 16
expected return (interest cost) and not consider the variance of the return (cost) in their decisions. 51 However, several arguments suggest that managers, rather than shareholders, should hedge foreign exchange exposure when various market imperfections exist. Firstly, information asymmetry implies that the management knows more about the firm s exposure position than shareholders. Secondly, the transaction costs involved to acquire various hedges are much lower for a firm than for an individual investor. Thirdly, when the default costs are significant, corporate hedging reduces the probability of default and simultaneously leads to higher credit ranking and lower financing costs. Fourthly, stable before-tax earnings lead to lower corporate taxes than volatile before-tax earnings when corporate taxes are progressive. 52 Two different hedging methods exist, internal and external tools. Internal instruments involve adjustments to business operations or financing plans. When internal hedging is limited, firms enter into external contracts with financial institutions or other firms. Hence, the starting point for the management of exchange exposure is to identify the firm s exposure of commercial activities which leaves the financial exposure aside. 53 Since the choice of hedging method is beyond the scope of this study, Appendix 10.4 contains the most common exposure management instruments. 2.2 Empirical Review Using the capital market approach to measure exchange rate exposure was first mentioned by Dumas (1978) and Adler and Dumas (1984). They argued that hitherto, the research on corporate exposure to exchange rate volatility had been based on the firm and its managers. Instead, Adler and Dumas (1984), proposed an alternative perspective which was adjusted to the interest of both stockholders and analysts. Risk and uncertainty is a question concerning randomness or unexpected exchange rate fluctuations; currency risk is not exposure, rather it is the probability that the current domestic purchasing power of the domestic or the foreign currency will differ from its anticipated value at a specific point in the future. Currency exposure on the other hand is defined as what one has at risk. Levi (1990) proposed a principally different view on exposure. He focused on the unpredictability of the value of assets, liabilities and operational incomes due to uncertainty in exchange rates, not on the uncertainty of the exchange rates themselves. This implies that the exchange risk is dependent on both the exposure and the variation in the exchange rate. 51 Levi (1990), pp. 225. 52 Eun and Resnick (2004), pp. 314-318. 53 Oxelheim and Wihlborg (2005), pp. 156-193. 17
The following review of the empirical research on exchange rate exposure is divided into three different parts; the empirical research has more or less developed in three different stages. The early findings based on US firms, were founded on fundamental exchange rate exposure. Gradually more sophisticated models with refinements of variable specification developed and the focus shifted at the same time to non-us firms. The final part presents the most recent results. This empirical review is concluded with some commentaries on the exposure puzzle and its implications for prior and future research. See Appendix 10.5 for an overview of the empirical findings. 2.2.1 Exchange Rate Exposure of US Firms Estimating the exchange rate exposure began with rather simple models where US firm s stock returns were regressed on the market return and the exchange rate return. In order to investigate the exposure of US multinationals to foreign exchange fluctuations between 1971 and 1987, Jorion (1990) performed a cross-sectional analysis. This pioneer study reported an insignificant relationship at the five percent confidence level between the firm value of 287 firms and exchange rate movements. Additionally, Jorion (1990) focused on the determinants of the exchange rate exposure where firms with larger percentage of foreign operations experienced a positive, but very small, correlation with the stock return and the value of the US dollar. In a subsequent paper, Jorion (1991) also presented insignificant exposure for 20 value-weighted industry portfolios. Amihud (1994) examined contemporaneous as well as lagged effects on exchange rates changes on 32 US exporting firms values between 1982 and 1988. He estimated the relative monthly changes in an exchange rate index of fifteen industrial-country currencies against the US dollar on the equally-weighted return of a portfolio of American exporting firms and controlled for the return on an equally-weighted market portfolio. The estimation results on the effect of firm value from both nominal and real exchange rate changes were insignificantly different from zero; thereby no contemporaneous relationship seemed to exist. Amihud (1994) also tested whether a correlation between the exchange rate and the ratio of exports to sales existed but no significant results were found. Bodnar and Gentry (1993), composed industry portfolios in order to find exchange rate exposure for Canadian, Japanese and US firms. This type of classification is only useful if the greater part of firms has the same international linkage, or more correctly, have equally exporting, importing and foreign investment patterns. Otherwise, when firms within a given industry do not have a uniform international relation, the impact from domestic currency fluctuations on firm value may 18
be reduced. Bodnar and Gentry (1993) found similar results for all three countries, between 20 and 35 percent of the industries had a statistically significant exchange rate exposure. Furthermore, the impact of exchange rate fluctuations was larger for Canada and Japan than for the USA. The results also indicated that industry export ratios are associated with negative exchange rate exposure and industry import ratios are associated with positive exchange rate exposure which is according to the theory. Bartov and Bodnar (1994) suggested two possible reasons for the earlier documentation of insignificant contemporaneous relationship by Jorion (1990) and Amihud (1994). One potential explanation for the limited success of prior studies was based on a prevailing miss-pricing on the market. Bartov and Bodnar (1994) argued that investors make systematic errors when estimating the complex relationship between exchange rate exposure and firm value. This miss-pricing implies that the stock price adjustment takes time which suggests an inclusion of lagged changes in the exchange rate. The second potential explanation was the research design, particular in the sample selection procedures. Bartov and Bodnar (1994) conclude that Jorion (1990) and Amihud (1994) only considered samples of US firms with reported foreign operations and large export activity respectively. This selection procedure does not necessarily involve firms with substantial currency exposure since large exporters are more likely to undertake hedging activities. As a result of the sample selection procedure, Bartov and Bodnar (1994) considered that endogenous hedging practice made the detection of a significantly relationship between the exchange rate and the firm value unlikely. To correct for the sample selection difficulties they used a method which was based on identifying firms with a significant reported impact from changes in the value of the US dollar on the financial statement. Furthermore, the sample was restricted to firms with a negative correlation between reported foreign currency adjustments and the equivalent volatility in the US dollar. However, this restriction could actually provide nonuseful information since the reported foreign currency adjustments have been criticised for not representing the economic impact of currency changes on the firm in an accurate way. But Bartov and Bodnar (1994) stated that this should be a minor implication given that their sample selection criteria was designed to distinguish firms with large and consistent adjustments in their financial statements which improves the probability of useful information. However, Bartov and Bodnar (1994) concluded that previous studies failure to document a significant contemporaneous relationship did not appear to be the result of sample selection bias. Bartov and Bodnar s (1994) examination of a data set of 208 US firms between 1978 and 1989 showed 19