Various studies have shown that exchange rate exposure is empirically lower for US multinationals than theory predicts. We examine this discrepancy for the largest Eurozone companies from 1994 till 2010 where the introduction of the Euro could have led to significant exposure reduction. Furthermore, we examine the relationship between exchange rate exposure and transnationality which is a novelty to date. We find that few European multinationals seem to have a significant exchange rate exposure and surprisingly the overall risk profile of transnational companies is lower which indicates that the risk from exchange rate exposure is more than compensated by risk reducing effects resulting from a high degree of transnationality.
In the last two decades significant contributions have been made to explore the relationship between exchange rate exposures of multinational firms and the risk that the stock market assigns to those firms showing higher exchange rate exposures. Typically significant exposures where found only for relatively few companies of the analyzed sample (Jorion, 1990). Moreover, exchange rate exposures were much lower than theoretically expected from the model which Bodnar, Dumas and Marston (2002) developed to measure exposure elasticity. Furthermore, large firms show empirically higher exposure (Dominguez & Tesar, 2006) oblivious to the fact that they are in a better position to apply advanced risk management due to human resource advantages over smaller firms. In their research they found that exchange rate exposure is higher in European countries, which is rarely documented to date.
In the existing literature we find a clear lack of investigation of European companies. Jorion (1990), Bodnar (1993), He & Ng (1998) and Allayannis et al. (2001) only focused on US, Canada and/or Japan in their studies of exchange rate exposures among multinationals. Though Dominguez and Tesar (2006) suggested that exchange rate risk and degree of internationality correlate in European companies, no further research has been done afterwards about specifically non-US firms. With the majority of previous studies focusing primarily on US or Japanese multinationals, we will examine the effect of exchange rate movements on European blue chip companies being the constituents of the EURO STOXX 50 (ES50) index. We will examine the exposures in a timeframe that covers sufficient time before and after the Euro introduction, which happened in 1999. Furthermore, we include financial corporations in our study which to date has not been investigated, as it has been argued that exchange rate exposure is only supposed to be found in firms dealing with foreign currencies through imports and exports. However, we think that by taking all constituents of the ES50 we represent a more concise overview within our methodology of transnationality as underlying variable of exchange rate exposure. Although theory suggests that exchange rate exposure is very important to firm value, in line with previous research we expect to find only a weak correlation between stock price changes and exchange rate movements.
The outline of this paper is set-up as follows. In the next section we start with a literature review, wherein previous studies regarding this topic are analyzed including the benefits and pitfalls of their methods and the results are discussed upon. Secondly, hypotheses will be developed that reflect our expectations of the relationships between the different variables. Thirdly, the dataset and the ways in which this study is conducted are elaborated upon. Finally, an analysis of the outcomes of the statistical results is given, followed by the conclusions, limitations and suggestions for further research.
As said in the introduction, several authors have discussed and studied the exchange rate exposure of multinational firms. Jorion (1990) was one of the first to question why the association between exchange rates and the value of the firm has not been empirically studied upon. Jorion (1990) found significant differences in exchange rate risk correlated to the foreign involvement of the firms. He concluded that there is a (small) positive correlation between exchange rate exposure and foreign involvement, and that this exposure influenced the value of exposed US firms. His findings, however, are not aligned with theoretical exposure estimates as he found only 5.2 percent of the sample had a significant exchange rate exposure, which according to Bartram & Bodnar (2007) is just ‘little more than by pure chance’.
Firms that are internationally operating should, in theory, have higher exchange rate risk, since they are exposed to different currencies and are therefore more influenced by unexpected exchange rate changes. Unfortunately, empirical studies were never able to measure statistically significant exposure for most multinational firms and only weak relationships between exchange rate changes and the value of the firm, which is measured by the firm’s stock prices, were found. This problem is also called the ‘exchange rate exposure puzzle’ (Bartram and Bodnar, 2007). Its existence lies in the fact that companies with high levels of operations, assets or competition outside their foreign-currency domain are advised to hedge the exchange rate risk through operations, financing activities or both. We need to take this into account and control for the ‘exchange rate exposure puzzle’ when analyzing the exposure of the ES50 firms.
Adler and Dumas (1984) were the first that came up with a way how exchange rate exposure can be measured. The measure came from the following equation:
With P being the price, a a constant, B being the exchange rate and e is an random error term. The term s is the measure for exchange rate exposure which is the coefficient of the slope with the firm’s stock returns being the dependent variable against exchange rates.
Bartram et al. (2010) tried to come up with a solution for this puzzle by including exposure reducing actions as passing the bill, operational hedging and financial hedging into their analysis. They showed that firms with high theoretical exposure were also more likely to have more debt in foreign currency and participated in foreign exchange derivatives. Bartram et al. (2010) estimated the exposure with the formula of exchange rate exposure elasticity of Bodnar, Dumas and Marston (2002), expressed as:
Where h1 is the foreign currency denominated revenue as percentage of total revenue, h2 is foreign currency denominated costs as percentage of total costs and r is the firm’s profit margin. Bodnar and Marston stated that a firm’s pass-through behavior and exchange rate elasticity are closely related, because they are both functions of product substitutability. The model can be used for firms that produce and sell at home and abroad, and demonstrates that foreign exchange exposure elasticity is smaller for multinational firms that match their foreign currency revenues and costs. Therefore, they believed it was a good measure for the firm’s sensitivity to exchange rate measures without using stock price data.
In line with the theories of Bartram and Bodnar (2007) and Bartram et al. (2010) we believe that a firm which has larger operations outside its home country will have larger exchange rate exposure. However, in previous research the results tend to be much lower than expected. Therefore, we would like to support this theory by using another unit of analysis. Our focus shifts from the US, which was mainly used in previous literature, to the EU, which also has some interesting features since the introduction of a single currency in this area in 1999.