1
The financial crisis and exchange rate exposure in
Eurozone countries
Jelle R. ter Horst Thesis MSc. BA - Finance Supervisor: dhr. K. Sklavos
Faculty of Economics and Business, Rijksuniversiteit Groningen
JEL-classification: C22, E58, F31, G01
Keywords: Exchange Rate Exposure, Euro, Financial Crisis, Eurozone
Abstract
This paper investigates the exchange rate exposure of Eurozone countries with the help of an Ordinary Least Squares regression model during the time span January 2001 to December 2010. By analyzing the stock returns of Eurozone multinationals, it is tested whether the financial crisis, originated in 2008, significantly influenced the exchange rate exposure in the Eurozone. The results show that the Eurozone countries were exposed to the exchange rate during the time span, but that the financial crisis did not have any effect on this exposure.
2 Eurozone countries. Adler and Dumas (1984) define exchange rate exposure as the amount of foreign currencies which represent the sensitivity of the future, real domestic currency (market) value of any physical or financial asset to random variations in the future domestic purchasing powers of these foreign currencies, at some specific future date. In practice, it follows that the more the Eurozone countries relatively trade with each other, the less exchange rate exposure there is for Eurozone firms and the more benefits participation in the Euro has.
One of the most influential costs of a monetary union is that countries lose the possibility to follow an independent monetary policy. This cost is best reflected in the case of a countrywide shock in demand or supply, which leads to exchange market pressure. Tanner (2002) defines exchange market pressure on a currency as its excess supply in the foreign exchange market, if monetary authorities did not try to influence the exchange rate. As a result, the trade balance of countries hit by a shock can become unstable and this can easily lead to an economic recession. According to Aizenman and Hutchison (2010) exchange market pressure can be resolved by currency depreciation, the interest rate is then used as main instrument to influence the exchange rate, or by the loss of international reserves. In a monetary union, country-specific monetary action is no longer possible and the member states have to rely on actions from the central bank to refuel their economies.
In September 2008 a global financial crisis originated in the United States, after the announcement that two large American firms would be nationalized to try to ensure the financial stability of the firms. One week later, it came to light that the Lehman Brothers bank filed for bankruptcy after being denied support by the Federal Reserve Bank. Following these two announcements, there was major instability on the global stock markets (Frattiani and Marchionne, 2009). This instability was the starting point of a negative snowball effect on the financial markets, resulting in a worldwide financial crisis.
3 Did the recent financial crisis significantly influence the exchange rate exposure in Eurozone countries?
By answering this question, this paper will bring new insights to the investigation of benefits and costs of being in a monetary union for the countries in the Eurozone. For this purpose, an Ordinary Least Squares regression model is used to analyze firm-specific return data in the period January 2001 to December 2010. If the regression model shows a significant change in exchange rate exposure after the origination of the financial crisis, this indicates an additional disadvantage of participation in the European Monetary Union. After all, the countries then suffer increased exchange rate exposure due to the loss of the possibility to follow an independent monetary policy.
To fully understand the purpose of this paper, the EMU-system and the role of the ECB in this system need some clarification. In 1992, twelve1 European countries signed the Maastricht Treaty, in which the foundation of the EMU was established. The main goal of the EMU is the convergence of the intra-European market with the help of a common currency that carries high price stability (De Grauwe, 2007). For this purpose the ECB was installed as an independent central bank. The goal of the ECB is to maintain price stability in the Eurozone by keeping its inflation rate below, but close to, 2% in the medium term. Without prejudice to the objective of price stability, the ECB wants to support the general economic policies in the Eurozone2. In contrast to for example the Federal Reserve Bank in the United States, the ECB assigns price stability as primary objective while the support of general economic policies is subordinate. As a result, the ECB only cares about economic policies when they do not interfere with the primary objective.
The relationship between the exchange rate of the Euro and the goal of the ECB to maintain price stability can be explained as follows. Changes in the exchange rate will normally affect inflation in three ways2. First, exchange rate movements may directly affect the domestic price of imported goods. If the exchange rate appreciates, the price of imported goods tends to fall, thus helping to reduce inflation directly, insofar as these products are directly used in consumption. Second, if these imports are used as inputs into the production process, lower prices for inputs might, over time, feed through into lower prices for final goods. Third, exchange rate developments may also have an effect via their impact on the competitiveness of domestically produced goods on international markets. If an appreciation in the
1
Belgium, Denmark, Germany, Greece, Spain, France, Ireland, Italy, Luxembourg, the Netherlands, Portugal and the United Kingdom.
2
4 exchange rate makes domestically produced goods less competitive in terms of their price on world markets, this tends to constrain external demand and thus reduce overall demand pressure in the economy. However, the strength of the relationship between the exchange rate and inflation depends heavily on the degree of openness of the particular economy to international trade. Exchange rate effects are in general of less influence in a large, relatively closed currency area like the Eurozone. Moreover, financial asset prices depend on many other factors in addition to monetary policy, and changes in the exchange rate are often dominated by these factors. That is why an exchange rate targeting strategy is not considered appropriate for the ECB, instead it generally uses the interest rate to keep prices stable.
I Literature review
A. Theoretical review
The problem of exchange rate exposure appears when firms trade products or financial services abroad. The price of these products or services is set up in a foreign currency and will have to be converted to the domestic currency to compute the real price. The rate of conversion is not fixed over time and is given by the exchange rate. Because of its variability, the exchange rate is a source of risk to firms trading outside the domestic country. According to Döhring (2008) there are three dimensions of exchange rate risk: transaction risk, economic risk and translation risk. Transaction risk refers to the impact of exchange rate changes on the value of committed cash flows. In contrast, economic risk is subject to the impact of exchange rate movements on the present value of uncertain future cash flows. Finally, translation risk is the risk of the variability in the valuation of foreign assets and liabilities on a multinational’s consolidated balance sheet, caused by exchange rate fluctuations. The exchange rate is typically four times as volatile as the interest rate and ten times as volatile as inflation (Jorion, 1990), which shows that it can be a major source of uncertainty. Doidge, Griffin and Williamson (2002) show that the breakdown of the Bretton Woods system in 1973 was followed by large real exchange rate movements. Furthermore, the strong increase in international equity investments in the last 30 years led to a higher demand and supply of foreign currencies. This combination should in theory be an indicator for high exchange rate exposure for firms all over the world.
5 cheaper in terms of the foreign currency, which increases foreign demand and consequently foreign sales revenues. On the other hand, importing goods will become more expensive and so net importing firms will be hurt by depreciation (He and Ng, 1998). The impact of a depreciation of the domestic currency on the profitability of various industries is shown in Table I.
Table I. The effect of a depreciation of the home currency on the profitability of a domestic industry (Bodnar and Gentry, 1993).
Activity Sign of effect
Non-traded good producer (-)
Exporter (+)
Importer (-)
Import competitor (+)
Foreign investor (+)
User of internationally-priced inputs (-)
Thus, exposure may reveal the simultaneous impact of monetary factors on exchange rates and stock prices. To the extent that monetary shocks affect firms differentially, cross-sectional variation could appear in the association between stock prices and exchange rates, even for purely domestic firms (Jorion, 1990).
This theory of exchange rate exposure can be applied to the system of the Eurozone. When there is for example a negative demand shock in Sweden and a positive one in Norway (an asymmetric shock), Sweden will use its interest rate to make the Swedish Krona depreciate against the Norwegian Krone. As a result, Sweden regains competitiveness. The situation changes considerably when it is applied to Eurozone countries. Now, there is no possibility for the member states to steer the exchange rate using domestic interest rates, since monetary action can only be performed by the central bank. The ECB will be paralyzed, since any action it undertakes to support a member state, will hurt another one. However, when this negative demand shock is a symmetric shock, all countries hit by the shock have the same interest in depreciating the domestic currency. The ECB will indeed follow a strategy to depreciate the Euro, as long as it does not harm the currency’s price stability (De Grauwe, 2007). In the Empirical review section, this statement will be evaluated. In any case it holds that the more asymmetric the current shock, caused by the financial crisis, in the Eurozone is, the more costs it brings in terms of exchange rate exposure.
6 that, in case of high inelastic consumer demand, producers can pass their exposure on to consumers. Döhring (2008) calls this invoicing. By invoicing in the domestic currency, an exporter is able to shift transaction risk to the customers abroad. One would therefore expect a strong interest from the side of exporters to invoice in the domestic currency. However, once economic risk and market structures are also taken into account, it becomes less certain that the exporter always has an interest in using the domestic currency for invoicing.
In addition to the theory of Döhring, Dumas (1978) states that multinational firms have the choice to either move their production to another country (which fully eliminates transaction risk and economic risk, but creates translation risk) or to hedge against exchange rate risk. There is, for a given rate of exposure, an optimal hedging decision, taking into account the presence of bankruptcy costs and market segmentation. Firms can follow two possible hedging strategies: operational hedging and financial hedging. In operational hedging, firms match the possible currency losses on their assets caused by exchange rate movements with currency gains on their liabilities. Financial hedging means that firms match the currency risk on their assets using forward contracts. Examples of these hedging options can be found in Appendix A. Clearly, the hedging operations of firms decrease their exchange rate exposure, which makes the relation between stock returns and the exchange rate less obvious. But, although hedging seems a logical decision in the presented examples, this is not always the case. The future value of assets and liabilities is frequently uncertain and may also be correlated with exchange rate changes. This makes that the degree on which firms are able to hedge is not clear. Brown (2001) states that the market view of firms is of great importance to their decision to hedge or to speculate. Moreover, Chowdhry (1995) finds that only a few firms hedge most of their assets and liabilities and that almost all hedges are short term. In the Empirical review section, relevant empirical studies that support or extend this theoretical background will be discussed.
B. Empirical review
7 related to national figures. This may not be the case and may therefore bias the results of empirical studies. Finally, the third possible reason is that firms have shielded themselves against exchange rate risk. Corporate operational and financial hedging activities can reduce the firm’s exposure. Döhring (2008) found that Eurozone non-financial firms systematically use financial derivatives to reduce transaction risk. Also, these firms use operational hedging to cope with economic risk, but in much lower levels. Translation risk management is less clear and seems to vary a lot across firms. Döhring concludes with stating that financial and operational hedging can often be seen as complements, but, even when used together, they do not make exchange rate risk disappear in total.
Irrespective of these remarks, the methodology of Adler and Dumas (1984) has been leading in recent papers. They made use of an Ordinary Least Squares (OLS) regression model to investigate exchange rate exposure cross-sectional at firm level. The paper shows that a firm’s exposure in terms of a certain foreign currency can be measured by the value of the foreign currency future contracts that investors must sell to minimize the variance of their hedging portfolio. The advantage of measuring exposure with a statistical regression technique is that it decomposes the probability distribution of a risky asset’s domestic currency price into two parts, namely one that is correlated with the exchange rate and a second that is independent (Adler and Dumas, 1984). The first component can be defined as exposed as its variability can be removed by hedging. For the second component, residual variability remains but is not exposed to exchange rate risk. Since its randomness is not correlated with any exchange rate, it cannot be further reduced by hedging instruments such as forward currency transactions. Jorion (1990) was one of the first papers to apply the OLS-model. He examined the relationship between foreign sales and exchange rate exposure for 287 multinationals in the United States. It appeared that exchange rate exposure increased with the amount of foreign sales, but the results were only weak. This is in line with the results of De Jong, Ligterink and Macrae (2006), who found that firm size is not a direct source of exposure.Typically, larger firms are more internationally oriented and therefore face more exposure than smaller firms. Larger firms also more often are multinationals. In contrast, smaller firms tend to be more domestically focused. Other studies, like Allayanis and Ofek (2001), did find a significant positive effect of size on exposure.
8 changes in the value of a specific currency over time and have observed the impact of past currency changes on firm performance, assets and liabilities. This happens usually at the end of a fiscal quarter. When investors use this information appropriately, they should be able to form unbiased expectations about the economic impact of the currency changes and incorporate this effect into their firm value estimates. Although Bartov and Bodnar found some support for their theory, they state that their results are not convincing because the relationship between currency movements and firm value is too complex to estimate in a regression model.
After the failure of most papers to find strong, significant results, criticism on the use of an OLS-model to locate exchange rate exposure started. Doidge, Griffin and Williamson (2006) state that the model is not suitable, in that it does not take into account i) the ability of firms to hedge against future conversion price uncertainty and ii) the general paucity of data sources. Indeed, the model does not account for hedging operations, so researchers can only speculate about the influence of hedging. In return, Wan (2006) claims that the OLS-model is the best model in estimating exchange rate exposure, because the return of a portfolio is statistically independent from fluctuations in exchange rate when currency exposure is fully covered.
Although research on exchange rate exposure has been extensive, the combination with a financial crisis and monetary unions has not been investigated earlier. Aizenman and Hutchison (2010) focused on the extent to which the current financial shock adversely affected the external position of emerging market economies. They found clear evidence that countries with a higher foreign operations/net sales ratio were more vulnerable to the financial crisis, in that they faced high exchange market pressure. As stated earlier, these emerging market countries tried to solve their trade balance problems during the crisis by preferring currency depreciation over the loss of international reserves. At times of collapsing global demand, countries are more willing to engage in competitive depreciation, as the downside of higher inflation is sharply mitigated by the global recession. In addition to Aizenman and Hutchison’s work, Berman, Berthou and Héricourt (2011) examined the Asian financial crisis in 1997 to 1998. The banking and currency crises in Asia at that moment generated a large negative demand shock for the French firms, who served this destination, in their sample. The conclusion of Berman, Berthou and Héricourt is that firms that showed more exchange rate exposure before the crisis to Asian currencies suffered a decrease of their domestic sales as compared to firms with less exposure or no exposure at all. It is obvious that multinational firms suffer various negative demand effects during financial crises and that their exchange rate exposure is a good predictor of the degree to which the crisis actually hits the firm.
9 work out in practice? The ECB actually did intervene in the currency market on September 22nd and on November 3rd, 6th and 9th 2000 to strengthen the Euro against the US dollar. But since its introduction, the Euro is generally seen as a stable currency and is widely accepted in the international currency market. So despite the fact that the ECB is essentially independent of political influence and has the possibility to influence the value of the Euro, it is clear that the ECB does not actively intervene in the Euro exchange rate (De Grauwe, 2007). Furthermore, when analyzing the trade pattern in the Eurozone since the start of the Euro, it is clear that the role of the Euro in international markets has increased to some extent. Both intra trade as well as trade with non Eurozone countries increased significantly (Ottaviano, Taglioni and Di Mauro, 2007). In addition, the mechanism through which the Euro has boosted trade finds its origin in the fact that the existence of the Euro has lowered fixed and variable costs of exporting firms (Baldwin, 2006). Since intra trade decreases the exchange rate exposure and external trade increases it, no simple conclusion can be drawn based on the Eurozone trade pattern.
The findings of the discussed papers and the fact that monetary unions have not been subject to research in this field make the research question of this paper even more interesting. Since the ECB will certainly not accept increasing inflation patterns, but firms in the Eurozone need the help of monetary instruments, it can be expected to find increased exchange rate exposure for Eurozone firms during the financial crisis. An overview of the relevant empirical findings on exchange rate exposure is presented in Table II.
Table II. Overview of relevant empirical findings regarding exchange rate exposure. The most important results of the related papers are presented, in combination with the investigated countries and the time span used in the analyses.
Paper Countries Time span Conclusion
Aizenman and Hutchison (2010)
Emerging markets 2008-2009 Countries rely on currency depreciation rather than reserve loss to deal with
exchange market pressure Allayanis and Ofek
(2001)
World 1993 Strong negative correlation between
foreign currency derivative use and exchange rate exposure Bartov and Bodnar
(1994)
United States 1978-1989 No exchange rate exposure found
Bodnar and Gentry (1993)
United States, Canada and Japan
1979-1988 There is a large impact of exchange rate movements on industry returns Jongmoo and Prasad
(1995)
10 De Jong et al. (2006) The Netherlands 1994-1998 More than 50% of the sample exposed,
all would benefit from a depreciation Döhring (2008) Eurozone 2003-2006 Hedging with exchange rate derivatives
is very widespread, 50% of the exports is invoiced in the Euro Doidge et al. (2006) Eighteen countries 1975-1999 Firm size, the level of international
sales, foreign income and foreign assets are negatively related to exposure He and Ng (1998) Japan 1979-1993 Exposure positively related to export
ratio, liquidity, size and variables that are proxies for hedging needs and
negatively related to leverage Jorion (1990) United States 1971-1987 Exposure is positively related with the
degree of foreign involvement Muller and
Verschoor (2006)
Europe 1988-2002 Around 20% of firms exposed to dollar, yen or pound. Short-term exposure
seems to be well hedged Ottaviano et al.
(2007)
Eurozone 1989-2002 EMU positively affects the competitiveness of firms in the
Eurozone
Williamson (2001) United States and Japan 1973-1995 The exposure of firms changes as the structure of the industry and competition changes through time
II Methodology
This paper will use the OLS-regression model suggested by Adler and Dumas (1984), which is used in most related relevant literature. The OLS-model suggests that there is a linear relationship between stock returns and the exchange rate. The underlying basis of the model is the efficient market hypothesis. This hypothesis states that real-world financial markets are efficient, i.e. security prices fully reflect all available information (Shleifer, 2000). When the efficient market hypothesis holds, the model works adequately because the measured values of stock returns are not biased.
The basic model that is going to be estimated is the following:
11 Where
R
it is the ith company’s stock return on time t,R
st is the percentual change in the Euro trade-weighted exchange rate, measured as the Euro price of a basket of foreign currencies, on time t andR
mtis the market index return on time t. The market index return is added to adjust for possible, stock returns influencing, shocks in the market that are not caused by exchange rate fluctuations. However, due to the inclusion of the market return, the firms’ exchange rate exposure is being measured relative to the market. When
β
1 is equal to zero, this means that the sample’s exposure is equal to the exposure of the market and it does not mean that there is no exposure at all. The changes in the companies’ stock returns, the trade-weighted exchange rate and the market index return are calculated using simple returns (see equation (2)). The simple returns equation has the advantage over the continuously compounded returns equation that it can compute the return on a portfolio of stocks at one time.(2) ∆Rt = 1 1 − − − t t t R R R
12 Taking into account the research question and the relevant related literature, the following hypotheses can be obtained from the basic model:
0
H
:(3)
[
Rit =β
0 +β
1Rst +β
2Rmt +η
it]
beforestartcrisis =[
Rit =β
0 +β
1Rst +β
2Rmt +η
it]
afterstartcrisis1
H :
(4)
[
Rit =β
0 +β
1Rst +β
2Rmt +η
it]
beforestartcrisis <[
Rit =β
0 +β
1Rst +β
2Rmt +η
it]
afterstartcrisisSo when
H
0 can be rejected, the result of the financial crisis is that there is a change in exchange rate exposure for Eurozone firms since the start of the financial crisis. To investigate if there is a difference in exposure before and after the start of the financial crisis, a dummy variable will be included in the model. The dummy variable Dt will take on the value 0 before the start of the crisis and the value 1 afterwards. The inclusion of the dummy variable in the model leaves the following equation to be estimated:(5) Rit =
β
0 +β
1Rst +β
2DtRst +β
3Rmt +β
4DtRmt +η
itThe coefficient
β
2 now represents the change in exchange rate exposure since the start of the financial crisis. So if theβ
2-coefficient in equation (5) turns out to be significant, H0 can be rejected and it can be concluded that there is a significant change in exchange rate exposure for the largest firms in Eurozone countries since the start of the financial crisis. The value and significance of the coefficientsβ
3 andβ
4 are of little relevance in answering the research question, but these coefficients need to be included in the model to estimateβ
1 andβ
2 more purely.III Data
A. Data characteristics
13 considerable amount of time taken into account before and after the start of the financial crisis and the sample size will be large enough to do proper research. The paper sets the start of the financial crisis at September 7th 2008. That day it was announced that mortgage banks Freddie Mac en Fannie Mae would be nationalized to try to ensure the financial stability of the two firms. This announcement is generally seen as the starting point of the current financial crisis, because of the turbulence it caused afterwards on the world financial markets.
B. Stock returns
In previous papers on this topic, monthly stock returns have been used the most. According to Wan (2006), using daily data is less suitable because they erase more noise in the relationship between the exchange rate and the stock returns and the use of quarterly or yearly stock returns fabricates too little data to work with. The firms involved in the sample are the 200 largest Eurozone firms, based on their market values3. A firm’s market value is the price at which its assets would trade in an efficient and competitive auction setting. The market values are measured at the end of each year in the time span, so the sample contains different firms every year4. This selection method has been used before by for example Flannery and James (1984), who state that market value can be a good predictor of exchange rate exposure and indeed Choi and Prasad (1995) found a significant positive effect of firm size on exchange rate exposure. Firms with higher market values are expected to trade more internationally and thus have a better chance of running exposure to exchange rates. Moreover, Jorion (1990) claims that a firm’s real assets will be affected in value by exchange rate movements, whatever their location. Thus also purely domestic firms, like utilities, may be affected by exchange rate movements through effects on aggregate demand or on the cost of traded inputs; domestic firms that sell goods competing with imports will be exposed to exchange rate movements as well. The complete list of investigated firms can be found in Appendix B. In Table III below, the descriptive statistics of the stock returns, market index return and the trade-weighted exchange rate are set out.
3
Financial Times: Euro 500 4
14
Table III. Descriptive statistics. The descriptive statistics of the monthly observed stock returns, market return and trade-weighted exchange rate during the time span January 2001 to December 2010 are presented.
Stock returns Market return Exchange rate
Mean 0.009 0.008 0.002 Median 0.014 0.014 0.003 Maximum 0.177 0.147 0.076 Minimum -0.141 -0.188 -0.056 Std. Dev. 0.053 0.056 0.019 Skewness -0.492 -0.635 -0.005 Kurtosis 4.163 4.406 5.478 Jarque-Bera 11.6 17.9 30.7 Probability 0.003 0.000 0.000 Sum 1.118 0.921 0.251 Sum Sq. Dev. 0.331 0.372 0.042 Observations 120 120 120
From Table III some interesting conclusions can be drawn. First, the maximum and minimum for both the stock returns and the market return are remarkable. In April 2009, the firms in the sample together gained a return of 17.7%. As can be seen in Appendix C these high returns were not only gained in the Eurozone, the S&P 500 also went up 8% during this month.5 Furthermore, in November 2008 both the stock returns and the market return reached their lowest point. The second half of 2008 was a period of high turbulence in the financial markets and the negative returns during this period are most certainly caused by the loss of consumer confidence after the start of the financial crisis. So the high and low values of the minima and maxima can be explained and should not be interpreted as outliers. Second, the mean change in the trade-weighted exchange rate is positive, which means that the Euro depreciated on average against the basket of foreign currencies (see Appendix D). This should in theory help exporting firms and hurt importing firms. Finally, the stock returns slightly outperformed the market return, which means that the sampled large Eurozone firms performed somewhat better than the global sample of large firms, used to compute the market return. From the Correlation matrix in Appendix E it can also be concluded that the stock returns almost go together with the market index return. These variables show a high correlation, which could be expected since the firms in the sample form part of the market. The exchange rate shows low correlation with the market index return, so multicollinearity problems in the model are avoided.
Looking at the list of firms in Appendix B it is notable that there are for example relatively little Greek firms in the sample, while many German firms are recorded. Because the Eurozone countries are in
5
15 a monetary union, but certainly not in a political one, exchange rate exposure could very well differ among them. Accounting standards, import fees and other political decisions can affect the trade balances of firms to such a degree that significant changes in exchange rate exposure can be expected to be found. For this reason a new sample of firms is created. The firms in this sample are all listed on the national stock exchange of the specific country, which means that their stocks are extensively traded. According to the theory of Choi and Prasad (1995) these firms are expected to be more sensitive to exchange rate movements than non-listed firms. This sample can be used to examine whether there are country-specific differences in exchange rate exposure in the Eurozone. Again, firms without a listing during the whole sample period are filtered out, leaving 248 firms in the final sample. The descriptive statistics of the investigated countries are set out in Appendix F. Most remarkable is the difference in mean return among countries. Austrian firms gained an average return of 1.5%, whereas Greek firms underperformed on average with a mean return of only 0.4%. To test whether this new sample of firms faced significant exchange rate exposure during the time span or a significant change in exchange rate exposure since the start of the financial crisis, equation (6) will be estimated. Equation (6) differs from equation (5) in that
ijt
R is the ith company’s stock return in country j on time t.
(6) Rijt =
β
0 +β
1Rst +β
2DtRst +β
3Rmt +β
4DtRmt +η
it16 other industrials industries, insignificant results were found. According to Bodnar and Gentry (1993) exchange rate fluctuations can have a substantial impact on the profitability of domestic industries. Price changes caused by movements in the exchange rate may change the terms of competition with foreign firms for domestic exporters and import competitors, alter input prices for industries that use internationally-priced inputs or firms that import for resale and change the value of assets denominated in foreign currencies. Because of this diverse set of influencing variables, exchange rate movements should affect some industries differently than others (see Table I). However, it must be noted that the impact of exchange rate movements is more difficult to determine when firms undertake activities (as presented in Table I) with different signs. The industry comparison should clearly be performed over a static sample of firms, so the 200 largest Eurozone firms in 2011 are sorted by industry6. Industries with at least six firms are investigated and all firms need to have a listing at a national stock-exchange during the whole sample period. This selection method leaves fourteen industries in the final sample. The descriptive statistics of the investigated industries are set out in Appendix F.
C. Market return
According to De Jong, Ligterink and Macrae (2006), it is common in studies on exchange rate exposure to add a market index to the estimation model to reduce noise. This way most of the factors that can influence stock returns, not being the exchange rate, are filtered out of the regression equation and the exchange rate exposure can be estimated more purely. Bodnar and Wong (1999) recommend the use of an equal weighted market index, which has the advantage over a value-weighted index that it has a much lower correlation with the used trade-weighted exchange rate index. This correlation can, after all, negatively influence the reliability of the test results for exchange rate exposure. Admittedly, the value weighted market return removes the more negative effect of exchange rates on larger companies (Bodnar and Wong, 2003). Larger companies are more likely to experience a higher negative cash flow reaction to domestic currency appreciations than other firms because of a higher degree of international activity. But due to the fact that this paper’s sample exclusively exists out of large and internationally active firms, the argument is not relevant for this paper. The equal weighted market index that will be used is the Standard & Poor’s Global 1200. This index captures approximately 70% of the world’s capital markets and is a composite of 31 local markets from seven headline indices. In other studies, like Doidge, Griffin and Williamson (2006) and Jorion (1990), the market index is a reflection of the performance of the largest firms in the investigated countries. However, the sample in this paper is constructed in such a way that it is a market index for the Eurozone in itself. Therefore, the use of a global market index like the Standard & Poor’s Global 1200 is a satisfying alternative. After all, the financial markets were globally influenced
6
17 by the financial crisis and from the correlation matrix in Appendix E it can be seen that the returns of the Eurozone firms follow the market return for almost 90%. In the country comparison and the industry comparison, the total sample of 200 firms can be considered as the market return, since it captures all investigated countries and industries.
D. Exchange rate
The relevant exchange rate for the multinational firms is the real exchange rate change between the firm’s home market and the foreign markets in which the firm has significant sales and competition (Williamson, 2001). The rate in the home country of the competitors is used to evaluate the relative exposure to a change in the competitors’ home currency. Most related literature, like Frankel (2009), makes use of a trade-weighted exchange rate to analyze exchange rate exposure. A trade-weighted exchange rate measures the import and export numbers of the countries using a specific currency. Suppose that a country’s trade balance shows that 80% of the international trades are performed with the United States and 20% with Japan. The trade-weighted exchange rate then composes a basket of exchange rates which combines the 80% actual exchange rate of the domestic currency with the US Dollar and the 20% actual exchange rate of the domestic currency with the Japanese Yen.
Although the trade-weighted exchange rate is used in most literature, some studies are skeptic about the variable. Williamson (2001) states that the main shortcoming of using this rate is the lack of power when companies are only exposed to a small number of currencies. He made a firm-specific or industry-specific exchange rate for every firm included in his sample to increase the power. The problem with this approach is that the trade figure of firms changes over time (Frankel, 2009). When for example a company starts to trade with a new country during the sample time, the exchange rate of the domestic currency with the currency of this new country needs to be included in the basket from that moment on. It is obviously not realistic to create such a real, firm-specific exchange rate for every firm in the sample at every point in time. In addition, the use of a trade-weighted index avoids the problem of multicollinearity that arises because the exchange rates of individual countries are fixed relative to each other (Jorion, 1990). So even though it is not a perfect solution, Frankel (2009) advises to use the trade-weighted exchange rate over a firm-specific rate.
Following the arguments of Frankel (2009), this paper will use the Eurozone trade-weighted effective exchange rate constructed by the Bank of England7. The fact that this index is also used by the IMF makes it a reliable index to use. The index composes a weighted basket of twelve extra Eurozone currencies, with the US Dollar and the British sterling being the most heavily weighted currencies in the basket (see Appendix D). An appreciation of the Euro against this basket of currencies creates a lower
7
18 index, see equation D1 in Appendix D. Because the index will be, on average, more stable than the individual Euro exchange rates, it will tend to be more volatile than the individual country competitiveness indicators (Bank of England, 1999). After all, a high proportion of country trade takes place with other Eurozone states at fixed nominal exchange rates, decreasing movements in the individual countries indices.
IV Results
19
Table IV. Results. The estimates of the OLS-regression model
it mt t mt st t st it R DR R DR
R =
β
0 +β
1 +β
2 +β
3 +β
4 +η
, as suggested by Adler and Dumas (1984), arepresented. The ith company’s stock return on time t is given by
R
it,R
st is the percentual change in the trade-weighted exchange rate, measured as the Euro price of the basket of foreign currencies, on time t andR
mt is the market index return on time t. The coefficientβ
1 is an estimate of the degree of exchange rate exposure for the 200 largest Eurozone firms during the time span January 2001 to December 2010 andβ
2 is an estimate of the change in exchange rate exposure for the same Eurozone firms since the start of the financial crisis. The coefficientsβ
3 and β4 are estimates for the exposure of the Eurozone firms to the Standard and Poor’s Global 1200 market index return and the change in this exposure since the start of the financial crisis respectively. The dummy variable Dttakes on the value 0 before the start of the crisis and the value 1 afterwards. Moreover, the t-statistics of the White-heteroskedasticity test and the Ramsey RESET test, the Durbin Watson statistic and the Jarque-Bera statistic are presented. These parameters represent a check to test whether the model fits to normality standards.
t-statistic Coefficient Standard error
Degree of exchange rate exposure (
β
1)-2.081 -0.348** 0.167
Change in exchange rate exposure (β2)
0.380 0.094 0.247
Exposure to the market index return (
β
3)16.371 0.950* 0.058
Change in exposure to the market index return (β4)
-1.351 -0.080 0.080
White-test (F-statistic) 0.656
Ramsey RESET test 0.365
Durbin-Watson statistic 2.153 Jarque-Bera statistic 13.343*
Note: * significant at 1% level ** significant at 5% level *** significant at 10% level
There are some important conclusions that can be drawn from Table IV. First, the coefficient β1
20 investigation of the Eurozone’s trade balance shows that indeed the Eurozone countries were importing more than they were exporting since 2004, except in 2009.8 Allayanis (1997) found that a 1% appreciation of the home currency reduces an industry’s value by 0.46% through exports and increases the value by 0.37% through imports, concluding that this particular industry is net exporting. This result is contrasting with most related literature, like De Jong, Ligterink and Macrae (2006) and Doidge, Griffin and Williamson (2006). These studies found positive exposures for firms in the Netherlands and the United States. However, some studies did find negative exposures, although this was never the case for all firms in the sample. Jongmoo and Prasad (1995) found negative exposure for 40% of the exposed firms in their sample, whereas Williamson (2001) states that the sign of exchange rate exposure depends on the state of the industry.
Second, the coefficient
β
2 is not significant. It follows that there is no significant change inexchange rate exposure for the Eurozone firms since the start of the financial crisis. This result can have several causes. First, the firms may have increased their hedging activities since the financial crisis, because they wanted to eliminate the risk of running increased exposure. According to Döhring (2008) and Muller and Verschoor (2006) most firms hedge against exchange rate exposure, but it is hard to measure the intensity of hedging activities. Another reason that the β2-coefficient is not significant can be that the exposure found during the sample period is negative. It is discussed earlier that currency depreciation can be a remedy to improve a country’s trade balance. If the ECB indeed took steps to depreciate the Euro, this would overall lead to lower stock returns of the Eurozone firms. Finally, as mentioned in the introduction, in case of a symmetric shock the exchange rate exposure of firms in countries undergoing the shock is not expected to change. The more asymmetric the shock, the more costs the shock will bring in terms of exchange rate exposure. The fact that no change in exposure is found since the financial crisis originated, might support the theory that the financial crisis is a very symmetric shock for the Eurozone countries.
Although it is an unexpected result that no change in exposure has been found, this does not have to hold for single Eurozone countries. Political differences can cause different exposures among countries (Bodnar and Gentry, 1993). That is why the OLS-model in equation (6) has been estimated for every Eurozone country separately. The results of these regressions are presented in Table V.
8
21
Table V. Results of country comparison. The estimates of the OLS-regression model
it mt t mt st t st ijt R D R R D R
R =
β
0 +β
1 +β
2 +β
3 +β
4 +η
are presented, where Rijt is the ith company’s stock return in country j on time t,R
st is the percentual change in the trade-weighted exchange rate, measured as the Euro price of the basket of foreign currencies, on time t andR
mt is the market index return on time t. The coefficientβ1 is an estimate of the degree of exchange rate exposure for the largest listed firms in each country during the time span January 2001 to December 2010 and β2 is an estimate of the change in exchange rate exposure for the same firms since the start of the financial crisis. The coefficientsβ
3 and β4 are estimates for the exposure of the domestic firms to the 200 largest Eurozone firms and the change in this exposure since the start of the financial crisis respectively.The dummy variable Dt takes on the value 0 before the start of the crisis and the value 1 afterwards. Standard errors are in the parentheses.22 The Netherlands -0.624* (0.212) -0.039 (0.318) 1.180* (0.075) -0.120*** (0.105) Note: * significant at 1% level ** significant at 5% level *** significant at 10% level.
It is clear from Table V that not all Eurozone countries faced exchange rate exposure during the 2001 to 2010 time span. Only in Finland and the Netherlands significant exposure can be observed. This is in line with the results of Muller and Verschoor (2006), among others, who found that only a few European firms were exposed to the US Dollar, UK Pound or Japanese Yen. Both Finland and the Netherlands show a negative sign of the β1-coefficient, which confirms the earlier finding that European firms are probably net importers when trading outside the Eurozone. The very significant coefficient of the Netherlands is no surprise looking at the country’s trade balance. A large part of its exports takes place with Eurozone countries, while the country imports mostly from outside the Eurozone.9 Remarkable is that for several Eurozone countries a significant change in exposure has taken place after the start of the financial crisis. In all cases the coefficient has a positive sign. This means firms gained, more than they did before, from a depreciation of the Euro. Moreover, the sign of the β2-coefficient has in most cases the opposite sign of the β1-coefficient. This indicates that the exchange rate exposure has decreased for the particular countries since the start of the financial crisis, which is against the hypothesis of this paper. However, since firms in the Eurozone countries seem to be net importing, this result does not contrast with economic intuition. It can be the case that firms changed their trade figures due to the crisis and preferred exporting over importing more than they did before. This follows the theory of Bodnar and Gentry (1993) and De Grauwe (2007). More research needs to be performed in the field of Eurozone firms’ trade figures to check this statement. The industry comparison as described in the Data-section can add insights to this topic too. The degree of exposure and the change of exposure since the start of the financial crisis have been tested for several industries using equation (6). The results of the estimation are set out in Table VI.
9
23
Table VI. Results of industry comparison. The estimates of the OLS-regression model
it mt t mt st t st ijt R DR R DR
R =
β
0 +β
1 +β
2 +β
3 +β
4 +η
are presented, where Rijt is the ith company’s stock return in industry j on time t,R
st is the percentual change in the trade-weighted exchange rate, measured as the Euro price of the basket of foreign currencies, on time t andR
mt is the market index return on time t. The coefficientβ1 is an estimate of the degree of exchange rate exposure for the largest Eurozone firms in the specific industry during the time span January 2001 to December 2010 and β2 is an estimate of the change in exchange rate exposure for the same firms since the start of the financial crisis. The coefficientsβ
3 and β4 are estimates for the exposure of the firms in the industry to the 200 largest Eurozone firms and the change in this exposure since the start of the financial crisis respectively. The dummy variable Dt takes on the value 0 before the start of the crisis and the value 1 afterwards. Standard errors are in the parentheses.Industry β1 β2
β
3 β4 Automobiles -0.849** (0.384) -0.105 (0.575) 1.191* (0.136) -0.016 (0.191) Banks -0.160 (0.329) -0.064 (0.492) 1.267* (0.091) 0.460* (0.146) Chemicals -0.396 (0.257) 0.292 (0.385) 1.110* (0.067) 0.113 (0.108) Electricity -0.046 (0.206) 0.052 (0.308) 0.653* (0.061) 0.175*** (0.098) Financial services -0.394*** (0.225) -0.429 (0.337) 0.729* (0.064) 0.216** (0.103) Food producers -0.025 (0.201) -0.584*** (0.301) 0.817* (0.063) -0.301* (0.101) Gas, water and24 Oil & gas 0.407
(0.332) 0.152 (0.397) 0.885* (0.098) 0.106 (0.158) Personal goods -0.425*** (0.226) -0.283 (0.338) 1.036* (0.065) -0.104 (0.104) Telecommunications -0.774*** (0.424) 1.062*** (0.635) 1.099* (0.139) -0.650* (0.224) Note: * significant at 1% level ** significant at 5% level *** significant at 10% level.
25
V Conclusion
The main goal of this thesis is to investigate whether the financial crisis has significantly influenced the exchange rate exposure of firms in Eurozone countries. For this purpose, an OLS-regression model, as proposed by Adler and Dumas (1984), is estimated. The model tests the exchange rate exposure of the 200 largest Eurozone firms, based on their market value, during the time span January 2001 to December 2010 and whether or not this exchange rate exposure has changed for the firms after the start of the financial crisis in September 2008. To estimate these variables more purely, the model corrects for market movements using the Standard & Poor’s Global 1200 index. The relevant exchange rate measure to be analyzed is the trade-weighted Eurozone effective exchange rate constructed by the Bank of England. The hypothesis, as formulated in the Methodology section, is that the Eurozone firms faced increased exchange rate exposure after the start of the financial crisis. This hypothesis is based on the theorem that a crisis can cause exchange market pressure which can easily lead to exchange rate exposure (Tanner, 2002 and Aizenman and Hutchison, 2010).
The OLS-regression estimates indicate that the firms in the sample faced significant exchange rate exposure during the time span and that this exposure is negative. This means that firms gain when the Euro appreciates against the basket of currencies in the Eurozone effective exchange rate and lose in case of a depreciation. Since the Euro has depreciated on average during the investigated time span, the firms have lost value due to exchange rate exposure. This finding contradicts with most literature, like De Jong, Ligterink and Macrae (2006), who found positive exchange rate exposure in their analysis. However, some studies, like Jongmoo and Prasad (1995), find negative exposure for parts of the total sample. Several studies fail to find significant exposure at all. Muller and Verschoor (2006), for example, find exchange rate exposure for only 20% of the European firms and Bartov and Bodnar (1994) do not find any exposure in their sample of firms in the United States. The most cited reason for the failure of papers to track down exchange rate exposure is firms’ possibilities to use hedging operations. Firms can and do hedge their exposure by using operational or financial hedges (Döhring, 2008), although it is hard to measure the quantity of these hedging activities (Muller and Verschoor, 2006).
26 financial crisis brings in terms of exchange rate exposure (De Grauwe, 2007). After all, the ECB can undertake monetary action that helps all Eurozone countries lower their exchange rate exposure in case of a perfect symmetric shock.
In addition to the investigation of the 200 largest Eurozone firms’ exposure, a country and industry comparison is performed. It appears that just a few Eurozone countries faced exchange rate exposure during the time span. However, five of them (Belgium, France, Germany, Italy and Spain) saw a significant change in exposure after the start of the crisis. For most countries, this exposure was decreasing. According to the theory of Bodnar and Gentry (1993), it could well be that these countries adjusted their trade balances and put more weight on exporting during the crisis. From the industry comparison, it can be concluded that several industries faced significant exchange rate exposure between January 2001 and December 2010. This is a confirmation of the theory of Berman, Berthou and Héricourt (2011), that the degree of foreign involvement influences exposure. Bodnar and Gentry (1993) already stated that when firms undertake more than one of the activities pointed out in Table 1, exposure is not expected to be found. Most industries, however, are likely to be net importing or net exporting as a whole, because firms within an industry usually are concerned with the same sort of activities (Muller and Verschoor, 2006). It follows that when industries are exposed, this is most likely the result of high foreign involvement.
The results of this paper add some interesting insights to the existing financial knowledge. At the time of writing, the Euro is subject to an intensive debate. It is no longer the stable currency as was the purpose when the Eurozone was established. Several countries are in an acute bankruptcy crisis and the foundation of a new Euro with only financially strong member states starts to gain advocates. What started as a widely supported experiment twelve years ago, has turned into a possible disaster. This paper adds to the recent debate about the benefits and costs of the Euro and of being in the Eurozone. It is one of the major benefits of the Euro that it decreases exchange rate exposure for firms in participating countries, but this paper shows that the largest Eurozone firms still face significant exchange rate exposure. The cost that large demand or supply shocks, as the financial crisis that originated in 2008 is, bring in terms of exchange rate exposure can be a downside to a monetary union. However, this paper finds no evidence that the financial crisis is of influence to the exposure within the Eurozone.
28
References
Adler, Michael and Bernard Dumas, 1984, Exposure to Currency Risk: Definition and Measurement, Financial Management 13, 41-50
Aizenman, Joshua and Michael Hutchison, 2010, Exchange Market Pressure and Absorption by International Reserves: Emerging Markets and Fear of Reserve Loss During the 2008-09 Crisis, working paper
Allayanis, George, 1997, The Time-Variation of the Exchange-Rate Exposure: An Industry Analysis, working paper
Allayanis, George and Eli Ofek, 2001, Exchange Rate Exposure, Hedging and the Use of Foreign Currency Derivatives, Journal of International Money and Finance 20, 273-296
Baldwin, Richard, 2006, The Euro’s Trade Effects, ECB working paper 594
Bartov, Eli and Gordon Bodnar, 1994, Firm Valuation, Earnings Expectation and the Exchange Rate Exposure Effect, Journal of Finance 44, 1755-1785
Berman, Nicolas, Berthou, Antoine and Jérôme Héricourt, 2011, Export Dynamics and Sales at Home, working paper
Bodnar, Gordon, Dumas, Bernard and Richard Marston, 2002, Pass-through and Exposure, Journal of Finance 57, 199-231
Bodnar, Gordon and William Gentry, 1993, Exchange Rate Exposure and Industry Characteristics: Evidence from Canada, Japan and the USA, Journal of International Money and Finance 12, 25-49
Bodnar, Gordon and Franco Wong, 1999, Estimating Exchange Rate Exposures: Some Weighty Issues, working paper
29 Brown, Gregory, 2001, Managing Foreign Exchange Risk with Derivatives, Journal of Financial Economics 60, 401-448
Choi, Jongmoo and Anita Prasad, 1995, Exchange Risk Sensitivity and its Determinants: A Firm and Industry Analysis of US Multinationals, Financial Management 24, 77-88
Chowdhry, Bhagwan, 1995, Corporate Hedging of Exchange Rate Risk When Foreign Currency Cash Flow is Uncertain, Management Science 41, 1083-1090
De Grauwe, Paul, 2007, Economics of Monetary Union, 7th Edition (Oxford University Press, New York, NY)
De Jong, Abe, Ligterink, Jeroen and Victor Macrae, 2006, A Firm-specific Analysis of the Exchange-rate Exposure of Dutch Firms, Journal of International Financial Management & Accounting 17, 1-28
Döhring, Björn, 2008, Hedging and Invoicing Strategies to Reduce Exchange Rate Exposure: A Euro-area Perspective, European Economy – Economic Papers 299
Doidge, Craig, Griffin, John and Rohan Williamson, 2002, Does Exchange Rate Exposure Matter?, working paper
Doidge, Craig, Griffin, John and Rohan Williamson, 2006, Measuring the Economic Importance of Exchange Rate Exposure, Journal of Empirical Finance 13, 550-57
Dumas, Bernard, 1978, The Theory of the Trading Firm Revisited, Journal of Finance 33, 1019-1029
Flannery, Mark and Christopher James, 1984, The Effect of Interest Rate Changes on the Common Stock Returns of Financial Institutions, Journal of Finance 39, 1141-1153
30 Frattiani, Michele and Francesco Marchionne, 2009, The Role of Banks in the Subprime Financial Crisis, working paper
He, Jia and Lilian Ng, 1998, The Foreign Exchange Exposure of Japanese Multinational Corporations, Journal of Finance 53, 733-753
Jongmoo Jay and Anita Prasad, 1995, Exchange Risk Sensitivity and Its Determinants: A Firm and Industry Analysis of U.S. Multinationals, Financial Management 24, 77-88
Jorion, Philippe, 1990, The Exchange-Rate Exposure of U.S. Multinationals, Journal of Business 63, 331-345
Muller, Aline and Willem Verschoor, 2006, European Foreign Exchange Risk Exposure, European Financial Management 12, 195-220.
Ottaviano, Gianmarco, Taglioni, Daria and Fillipo di Mauro, 2007, Deeper, Wider and More Competitive? Monetary Integration, Eastern Enlargement and Competitiveness in the European Union, working paper
Shleifer, Andrei, 2000, Inefficient Markets, An Introduction to Behavioral Finance, 1st Edition (Oxford University Press, New York, NY)
Tanner, Evan, 2002, Exchange Market Pressure, Currency Crises and Monetary Policy: Additional Evidence from Emerging Markets, working paper
Wan, Xiao, 2006, Exchange Rate Exposure of Chinese Firms, International Journal of Business and Management 1, 95-102
31
Appendix A Hedging activities
Assume a bank invests €200,-, with an interest rate of 5% in 50% European loans (interest rate 6%) and 50% US loans (interest rate 7%), with an exchange rate of €1/$. This investment should yield a positive return. But at the end of the loan, the Dollar has depreciated to €0.9/$. Now the bank makes a loss:
Assets:
Start of the loan Interest rate End of the loan Return
€100 6% €106 6%
€100 = $100 7% $107 = €96.3 -3.7%
Average return = 1.15% < 5%
When the bank uses operational hedging, it funds 50% of its assets with US loans (interest rate 6%):
Liabilities:
Start of the loan Interest rate End of the loan Return
€100 5% €105 -5%
€100 = $100 6% $106 = €95.4 4.6%
Average return = -0.2%
Now the bank has a positive ROI of 1.15% - 0.2% = 0.95%, whilst without hedging it would have made a loss. In financial hedging, firms match the currency risk on their assets using forward contracts. Assume the same bank invests its €200,- in European and US loans. Now it buys a forward contract that allows the bank to sell the $100 against the forward rate €0.98/$. Independent of the state of the exchange rate, the bank makes the following return on its assets:
Assets:
Start of the loan Interest rate End of the loan Return
€100 6% €106 6%
€100 = $100 7% $107 = €104.86 4.86%
32
Appendix B Largest firms in the Eurozone in the period January 2001 – December 2010
Table BI. Firms in the sample.
A2A BNP Paribas Eutelsat Communications Legrand SAP
Abertis Bollore Exor Linde SBM Offshore
ABN Amro Boskalis Westminster Ferrovial L'Oreal Schering
Acciona Bouygues Fiat Luxottica Schneider Electric
Accor Brenntag Finmeccanica LVMH Scor
Acerinox Brisa Fomento Man SEB
ACS Activ.Constr.y.Serv. Bulgari Fonciere des Regions Mapfre SES
Adidas Bureau Veritas Fortis Mediaset Siemens
ADP Buzzi Unicem Fortum Mediobanca Snam Rete Gas
Aegon Cairn Energy France Telekom Merck Kgaa Societe Generale
Ageas Cap Gemini Fraport Metro Sodexho Alliance
AGF Carrefour Fresenius Medical Care Metso Sodexo
Ahold Casino Guichard-Perrachon Fugro Michelin Software
AIB Celesio Galp Energia Millicom International Solvay
Air France-KLM CGG Veritas Gamesa Technologica Mobistar Sparkassen Immo-Invest
Air Liquide Christian Dior Gas Natural Munich Re STMicroelectronics
Akzo Nobel Cie Nationale a Portfeuille Gaz de France National Bank of Greece Stora Enso
Alcatel-Lucent Cimentos De Portland GBL Natixis Strabag
Alleanza Cintra Concesiones GDF Suez Neste Oil Suedzucker
Allianz CNP Assurances Gea Group Neuf Cegetel Suez Environment
Almancora Coca-Cola Hellenic Bottling Gecina Nokia Technip
Alpha Bank Colruyt Gemalto Nokian Rentaat Tele2
Alstom Commerzbank Generali Numico Telecinco
Altadis Continental Groupe Eurotunnel OMV Telecom Italia
Altana Corio Grupo Inmocaral Opap Telefonica
Amadeus IT Cosmote Mobile Telecom Hal Trust OTE-Hellenic Telc. Telekom Austria
Andritz Credit Agricole Hannover Ruck Outokumpu Telenet
Anglo Irish Bank CRH Heidelbergcement Pagesjaunes Terna
Anheuser-Busch Inbev Criteria CaixaCorp Heineken Parmalat Thales
APRR Daimler Hellenic Telecommunications Pernod-Ricard ThyssenKrupp
Arcelor Mittal Danone Henkel Petroleos (Cepsa) TNT
Arkema Dassault Systemes Hermes International Peugeot Tognum
ASML Holding Delhaize Hochtief Philips Electronics Total
Atlantia Deutsche Bank Hugo Boss Porsche AML UBI Banca
Autoroutes Du Sud France Deutsche Boerse Hypo Real Estate Portugal Telecom UCB
Autostrada Deutsche Lufthansa Iberdrola PPR Umicore
Axa Deutsche Post Iberdrola Renovables Prysmian Unibail-Rodamco
Axel Springer Deutsche Telekom Icade Public Power UniCredit
Banca Antonveneta Dexia Iliad Publicis Unilever
Banca Carige Dragon Oil Imerys Qiagen Union fenosa
Banca CR Firenze DSM Immoeast Raiffeisen Intern. Bank Unipol
Banca Monte Dei Paschi E.On Immofinanz Randstad UPM-Kymmene
Banca Nazionale de Lavoro EADS Inditex RAS Valeo
Banche Popolari Unite Edenred Infineon Technologies Red Electrica de Espana Vallourec
Banco de Sabadell EDF ING Reed Elsevier Veolia Environnement
Banco de Valencia EDP Renovaveis Intesa Sanpaolo Renault Verbundgesellschaft
Banco Esperito Santo EFG Eurobank Ergasias Irish Life & Permanent Repsol-YPF Vienna Insurance
Banco Popular Espanol Eiffage JC Decaux Rexel Vinci
Banco Santander Elan Jeromino Martins Rodamco Europe Vivendi
Bank of Ireland Enagas K + S RWE VNU
Bank of Piraeus Endesa Kabel Deutschland Ryanair Holdings Voestalpine
Bankinter ENEL KBC Group Sacyr Vallehermoso Volkswagen
Basf ENEL Green Power Kerry Group Safran Vopak
Bayer Energias de Portugal Kesko Saint-Gobain Wacker Chemie
BBVA Eni Klepierre Saipem Wartsila
BCP Eramet Kone Salzgitter Wendel Investissement
Beiersdorf Erste Group KPN Sampo Wolters Kluwer
Bekaert Essilor International Lafarge San Paolo IMI Zardoya Otis
Belgacom Eurazeo Lagardere Groupe Sandvik
33
Appendix C Standard & Poor’s 500 Stock Market Sector Performance
34
Appendix D Composure of the Euro trade-weighted exchange rate index
Table DI. Eurozone effective exchange rate weights. This table presents the weight of the different exchange rates used in the Eurozone effective exchange rate index, constructed by the Bank of England.
Country Rounded weight
Australia 0.9 Canada 1.9 Denmark 3.3 Greece 1.6 Japan 16.9 New Zealand 0.1 Norway 1.8 Sweden 6.6 Switzerland 10.6 United Kingdom 23.1 United States 23.0 Hong Kong 1.8 Singapore 2.2 South Korea 3.2 Taiwan 2.9 TOTAL 100.0
Figure 2. Eurozone effective exchange rate graph. This figure presents the value of the Euro effective exchange rate index during the time span January 2000 to December 2010.
Eurozone effective exchange rate equation
(D1) Eurozone Effective Exchange rate Index =
∑
countriesj weight j j Euro currency ) ( 0 20 40 60 80 100 120 140 1 /1 /2 0 0 0 8 /1 /2 0 0 0 3 /1 /2 0 0 1 1 0 /1 /2 0 0 1 5 /1 /2 0 0 2 1 2 /1 /2 0 0 2 7 /1 /2 0 0 3 2 /1 /2 0 0 4 9 /1 /2 0 0 4 4 /1 /2 0 0 5 1 1 /1 /2 0 0 5 6 /1 /2 0 0 6 1 /1 /2 0 0 7 8 /1 /2 0 0 7 3 /1 /2 0 0 8 1 0 /1 /2 0 0 8 5 /1 /2 0 0 9 1 2 /1 /2 0 0 9 7 /1 /2 0 1 0 B a se p o in ts Date
35
Appendix E Correlation matrix
Table EI. Correlation matrix. This table presents the correlation coefficients between the 200 largest Eurozone firms, the Standard & Poor’s Global 1200 and the Eurozone trade-weighted effective exchange rate index during the time span January 2001 to December 2010.
Stocks Market Exchange rate
Stocks 1 0.89 0.11
Market 0.89 1 0.28
