The use of foreign currency hedging and its influence on
company value:
Evidence from the Canadian oil, gas and mining sector
Master Thesis
Date of submission: 09.01.2015
Name:
Adrian Maximilian Böckmann
Student number:
1985299
Study Program:
MSc International Financial Management
Email:
a.m.bockmann@student.rug.nl
Supervisor:
Dr. B. Scholtens
Abstract
This study is focused on the use of foreign currency hedging and its influence on company value. The methodology is broadly following Allayannis and Weston (2001) but the sample selection is different and the work is extended by a currency exposure and stock return analysis. The sample consists of 93 companies from the Canadian oil and gas as well as the mining industry and it is studied from the period 2005 to 2013 to test whether the use of foreign currency hedging increases company value as measured by Tobin’s Q. The logarithm of market value to the book value of assets is used as alternative measure. The results for the two company value proxies are different and the multivariate regression analysis provides results for the hedging variable that are not statistically significant. Therefore, the results of this study provide no evidence that currency hedging increases company value.
Contents
1. Introduction ... 4
2. Literature review ... 8
2.1 Risk management theory and hedging ... 8
2.3 Previous research on the use of hedging ... 10
3. Data and methodology ... 12
3.1 Sample ... 12
3.2 Model and methodology ... 14
3.3 Descriptive statistics ... 17
4. Results ... 23
4.1 The use of foreign currency hedging over time ... 23
4.2 Univariate test ... 23
4.3 Multivariate test ... 24
4.4 Exchange rate exposure and stock return ... 29
4.5 Summary of the results ... 32
5. Conclusion and Limitation ... 34
5.1 Conclusion ... 34
5.2 Limitation ... 36
6. References ... 37
1. Introduction
The use of foreign currency hedging contracts has increased over the last years and keeps expanding. The most recent statistics on the use of foreign currency contracts published by the Bank of International Settlements point out that the amount of foreign currency contracts increased from US$66.672 billion in June 2012 to US$73.121 billion in June 2013 and US$74.782 in June 2014 (Bank of International Settlements, 2014). On the basis of this example one might expect that companies increasingly rely on the use of foreign currency contracts as risk management instrument. According to Stulz (1996) the main objective of risk management is to reduce or to avoid financial inconvenience and manage situations that prevent the company to carry out its strategy. Modigliani and Miller (1958) state, that risk management is irrelevant and does not influence the value of the companies using it. Shareholders can manage risk on their own through the diversification of portfolios. However, several more recent studies provide evidence, that the use of foreign currency hedging as risk management strategy does increase growth opportunities (e.g. Nance, Smith, and Smithson, 1993 and Geczy, Minton, and Schrand, 1997).
Allayannis and Weston (2001) were the first who researched whether the use of foreign currency derivatives is connected to higher company market value. They use a sample of 720 large U.S. nonfinancial firms between 1990 and 1995 to test if the use of foreign currency derivatives is associated with higher firm value and find a positive association. On the other hand Jin & Jorion (2006) found contradicting results, while studying the U.S. oil and gas sector.
by using a sample of non-financial Canadian companies, from the oil and gas as well
as the mining industry.
Canada as country has been selected because the Canadian dollar is highly influenced by the U.S. dollar due to the high degree of economic interdependence between the two countries (Brown, 2006). Additionally, the exchange rate fluctuation of the Canadian dollar is strongly effected by the inflation rates as well as interest rates of Canada and the U.S. (Holden, 2007).
Canada is strongly connected to one of the world’s largest economies and there is a strong influence on the exchange rate and the exchange rate risk. Therefore, the sample promises to generate reliable results, testing the influence of currency hedging on company value. The oil and gas sector has been investigated before and has been selected because almost all companies involved can be classed as multinational companies, which are subject to foreign currency risk (Jin and Jorion, 2006 and Haushalter, 2000). The same is transferable on the mining industry, since it is mainly controlled by multinational acting corporations (Guerrieri, 1982). Because the companies of both industries are classed as multinational companies and both heavily rely on natural resources this study includes companies of both industries within the sample and treats them as homogeneous.
higher exposure towards currency fluctuations than companies which do not make use of foreign currency hedging based on the companies stock returns. By analyzing a sample of 93 companies over the years 2005 to 2013, a total of 837 observations emerged, used for both the analysis of the influence of hedging on company value as well as the exchange rate exposure and stock return analysis.
The main hypothesis tested for, is that companies which use foreign currency hedging are rewarded with higher company value. The alternative hypothesis is that companies, which use foreign currency hedging, are not rewarded with higher company value or even attain lower company value. Furthermore, it is hypothesized that hedging companies face a higher exchange rate exposure in comparison to non-hedging companies. Relying on the previous study of Allayannis and Weston (2001), Tobin’s Q is used as proxy for company value to test the main hypothesis. It is calculated as the ratio of total assets less the book value of equity plus the market value of equity to the book value of assets. To assure the reliability of the results, an additionally proxy for company value, defined as the logarithm of market value of the company to the book value of assets (Allayannis and Weston, 2001) is used. To test for the exchange rate exposure and stock return of hedgers and non-hedgers, an exposure analysis is adopted, relying on the study of Adler and Dumas (1984). Through the comparison of the currency exposure of means from hedgers and non-hedgers it is possible to identify, whether non-hedgers or non-non-hedgers are more sensitive towards currency exposure in comparison.
The first analysis of this paper is to test if the use of foreign currency hedging affects the company value in a positive way. The results found are diverse for Tobin’s Q and Log(!"
!"). However, since the results for the hedging variable for the analysis
with both proxies are not statistically significant and Tobin’s Q is treated as main proxy the conclusion is that there is no or a negative relationship between currency hedging and company value.
currency exposure in comparison to non-hedging companies in connection with lower stock return.
This thesis aims to contribute to the existing literature by extending previous research under a different setting. Moreover, the focus lies on Canadian companies, which are actively taking part in the oil and gas sector, as well as the mining industry. Previous research concerning the relationship between company value and foreign currency hedging mainly focused on the U.S. (e.g. Allayannis and Weston, 2001 and Jin and Jorion, 2006). Additionally, previous research arrived at contradicting results Allayannis and Weston (2001) found a positive relationship between foreign currency hedging and company value while Jin and Jorion (2006) found a negative connection. This thesis will answer the main problem stated above, whether the use of foreign currency hedging does increase company value and should be used as risk management technique. It is of academic relevance because current literature and research will be expanded by a repetitive study building on a different country and industry sample and thereby creating new results, which will help to come up with additional information to make a decision on the main problem described. Additionally, through the use of an exchange rate exposure analysis further insides of the reason for the use of currency hedging and its effects are created. The thesis is relevant for society because it constructs additional information and data, to answer the question, whether companies should engage in currency hedging and use it as risk management technique. Furthermore, the thesis complies with all requirements necessary for an international financial management paper, as it studies foreign currency hedging, which is a financial topic with several implications for management, especially for risk management. Since the whole study is performed in industries mainly acting in an international environment and foreign currency hedging by itself is an international topic the requirement of an international relevant topic is complied as well.
the exchange rate exposure analysis. Section 5 focuses on the conclusion as well as limitations.
2. Literature review
The literature review is subdivided into two parts. The first part discusses the theoretical research, previous literature and theory related to risk management and especially hedging. The second part deals with previous empirical research on foreign currency hedging, mainly its influence on company value.
2.1 Risk management theory and hedging
The classic Modigliani and Miller theorem (1958) states that in a perfect market the capital structure of companies is irrelevant and the financial decisions of companies have no effect on company value. Hence, risk management policies are irrelevant and foreign currency hedging therefore cannot create company value. Mainly because private investors are assumed to have the same access to information and prices, they are able to manage, for example the currency risk by diversification of the portfolio without the need of the company to engage in hedging. To sum it up, Modigliani and Miller (1958) find, that companies financial decisions and hence their currency hedging decisions have no effect on company value in a perfect market. However, real markets face imperfections and several more recent studies state, that hedging can be used to shift risk and thereby create value (e.g. Froot, Scharfstein and Stein, 1993; Geczy, Minton, and Schrand, 1997 and Smith and Stulz, 1985).
Diverse authors state that risk management can be used to reduce taxes. Smith and Stulz (1985) suggest that the structure of tax might be positively influenced by the use of hedging. The authors found that through the use of hedging, taxes might be reduced and thereby company value can be created because of higher income. Leland (1998) argues, that based on lower pre-tax income through the use of hedging the debt of a company can be increased, leading to possible tax reduction because of interest deduction.
observed inefficiencies in the financial market by making the cost of capital proportional to the cash flow. The authors observed that companies use hedging to protect its cash flow from fluctuations and thereby reduce the underinvestment problem. Geczy, Minton and Schrand, 1997 found that companies that face greater growth opportunities and are subject to strong financial constraints are found to engage in currency hedging more often to keep the possibility to carry out potential valuable projects.
On the other hand there are suggestions, that risk management might be misused by managers (agents) who use it to maximize their own wealth on the costs of owners (principals) leading to a classic agency theory problem. For example, Stulz (1984) observed the behavior of company managers towards the use of hedging based on their attitude towards risk. Risk-averse managers with their income linked to company results were found to rely on hedging to reduce the currency exchange risk and thereby create value for themselves. More authors confirm the conclusion of Stulz (1984) and his theory of managerial risk aversion for instance Tufano (1996) investigated the gold and mining industry and found a positive relationship between the use of hedging and the amount of stock held by managers. Speaking in favor of these findings DeMarzo and Duffie (1995) relate to Modigliani and Miller and find that even though shareholders can protect themselves through hedging or diversification, the outcome of hedging is ideal only when executed by managers who have private information on the company’s profits.
2.3 Previous research on the use of hedging
The empirical research of hedging theories has been limited because of data and information on hedging activities, which rarely have been available until the beginning of the 1990s (Allayannis and Weston, 2001). Since the early 1990s, companies have been required to report the use of derivatives in the footnotes of their annual reports. Therefore, the possibilities of earlier studies have been limited to the use of survey data, which is harder to obtain and quite time consuming. Most recent studies rely on the use of a binary variable to indicate whether a company uses hedging or not (Allayannis and Weston, 2001; Allayannis, Lel, and Miller, 2012 and Jin and Jorion, 2006).
There are numerous empirical researches on hedging activities, uncovering contradicting results (e.g. Allayannis and Weston, 2001; Allayannis and Ofek, 2001; Graham and Rogers, 2002; Jin and Jorion, 2006 and Belghitar, Clark, and Mefteh 2013). Different authors examined why companies make use of hedging. Haushalter (2000) studied the hedging policies of oil and gas producers from the years 1992 to 1994 and found that a higher debt ratio increases hedging because companies with a higher debt ratio have greater risk to encounter financial distress. Those outcomes are concurrent with the findings presented by Smith and Stulz (1985), which have been discussed in the previous section. Allayannis and Ofek (2001) examined why companies make use of hedging and found that the main motivation for hedging is to reduce the exchange rate risk and not speculative purposes. Based on these findings the further focus of this literature review is on studies that examined the use of currency hedging to reduce exchange rate risk. Within those studies two groups can be identified. The first is identifying positive results for the use of hedging and the second identifies negative results for the use of hedging.
currency hedging positively affects and causes an increase in company value. Furthermore, the authors were able to obtain the level of foreign currency hedging for some of the companies studied. However, studying only the companies that reported the level of hedging reduced their sample by 37%. Therefore, the authors relied on the use of a binary dummy variable to study whether a company uses hedging or not. In a later study the results of Allayannis and Weston (2001) are confirmed. Allayannis, Lel and Miller (2012) study the influence of hedging on company value using a broad sample of countries from thirty-nine countries with significant exchange rate exposure and found a significant premium for hedging companies with internal company level or external country level governance. Clark and Mefteh, (2010) use a sample of 176 non-financial French companies during 2004 to show that the use of derivatives is a significant determinant of company value. Furthermore, they found that this is especially true for large companies and the effect is sensitive towards the company’s overall level of exposure. Hagelin and Pramborg (2004) examined a sample of companies from Sweden and found a significant reduction of foreign exchange rate exposure for companies making use of hedging. Through the reduction of foreign exchange rate exposure, company value is increased and previous findings are confirmed.
identified a negative influence on company value when an agency problem between shareholders and managers is present.
The discussed studies provide evidence for both groups mentioned beforehand, there are positive as well as negative influences through the use of hedging. Furthermore, it can be estimated, that currency hedging affects company value positively as well as negatively. The findings of previous research are highly contradicting to the theoretical perspective as well as to the empirical perspective. There is no mutual agreement on the questions whether risk management techniques, especially hedging is effective and should be employed as risk management technique on the theoretical side. Along, the empirical results are mixed and provide reason to assume a positive as well as a negative influence through the use of currency hedging. Specifically the influence of currency hedging on the company value provides mixed results and is unsettled from previous literature. Because there is no common agreement on the question whether foreign currency hedging increases company value and should therefore be employed as risk management technique, this thesis addresses this question. Particularly, to discover whether the use of foreign currency hedging impacts company value a set of non-financial companies listed on the Toronto Stock Exchange (TSX) from the oil and gas as well as mining industry is examined. This is of interest because it will answer the main problem stated, if the use of foreign currency hedging increases company value and should be used as risk management technique. Additionally, it will contribute to existing literature by expanding current findings by a repetitive study building on a sample from a different country and different industries delivering new results, which will provide new information to make a decision on the main problem stated. This thesis is of societal relevance because it delivers information and data to answer the question, whether companies should make use of foreign currency hedging as risk management technique or not.
3. Data and methodology
3.1 Sample
multinational companies, which are subject to foreign currency risk as shown in earlier studies (Jin and Jorion, 2006, Haushalter, 2000 and Guerrieri, 1982). Furthermore, both industries heavily rely on natural resources and are officially grouped as one business enterprise sector (Statistics Canada, 2014). Based on this information, the two groups are treated as homogenous. The companies included in the sample are all headquartered in Canada and additionally have been listed on the Toronto Stock Exchange (TSX). Besides, companies having missing data on important financial and accounting data (e.g. total assets, total equity or market value) have been excluded from the sample. Finally, 93 companies have been included in the sample to arrive at a total of 837 company-year observations from 2005 to 2013. The period from 2005 to 2013 was selected to discover a more recent sample of observations in comparison to Allayannis and Weston (2001) who studied the years 1990 to 1995. Additionally, the time period was extended to nine years to arrive at a greater and more reliable sample.
change in exchange rates of the Canadian and US dollar to be able to perform the exchange rate exposure and stock return analysis.
However, none of the available databases contained information on the use of foreign currency hedging. Therefore, the information on the usage of foreign currency hedging was collected from the annual reports of each company for each year of the sample. Companies that reported the use of foreign currency hedging in the footnotes of their annual reports have been defined as hedgers. On the other hand, companies that reported that no foreign currency hedging was employed were defined as non-hedgers. Different to the study of Allayannis and Weston from 2001 the notional value on the hedging contracts cannot be obtained because of lack of available data. Therefore, no measurement for the level of foreign currency hedging is obtained but its only tested if company uses currency hedging or not through a binary dummy variable.
3.2 Model and methodology
To test for the main hypothesis, whether the use of foreign currency hedging has an impact on the company value of Canadian firms in the gas and oil as well as mining industry and if it should be used as risk management technique, the following regression models will be used:
The empirical model, which is used to analyze the influence of hedging behavior on firm value is based on the model of Allayannis and Weston (2001) and has the following form:
Model 1:
1. 𝑇𝑜𝑏𝑖𝑛!𝑠 𝑄
!" = 𝛼!"+ 𝛽! 𝐻𝑒𝑑𝑔𝑖𝑛𝑔!"+ 𝛾!𝑋!"+ 𝜀!"
Where:
stands for the control variables that will be discussed in greater detail in the following section. ε is the error term.
To test for the robustness of the results arrived from Tobin’s Q and the connected model 1 an alternative proxy for company value defined as the Logarithm of market value to book value of assets is used in the corresponding model 2 as done by Allayannis and Weston (2001).
Model 2:
2. 𝐿𝑜𝑔 !"
!" !" = 𝛼!"+ 𝛽! 𝐻𝑒𝑑𝑔𝑖𝑛𝑔!" + 𝛾!𝑋!" + 𝜀!"
Where: Log
(
!"!"
)
is defined as the proxy for firm value as alternative measure for Tobin’s Q,α is the constant, β is the coefficient of the equation. Hedging is a binary dummy variable, which indicates, whether the observed company use foreign currency derivatives in the corresponding year or not. For companies, that use currency hedging 1 is used, for the opposite 0 is used. X stands for the control variables that will be discussed in greater detail in the following section. ε is the error term.
The dependent variables for the two models are calculated as follows.
For model 1, Tobin’s Q is calculated as defined by Allayannis and Weston (2001) and Allayannis, Lel and Miller (2012) as the ratio of total assets less the book value of equity plus the market value of equity to the book value of assets. Market value has been calculated multiplying share price by the number of outstanding shares.
For model 2, Log
(
!"!"
)
is calculated as the logarithm of market value of the companyto the book value of assets. This alternative measure is consistent with Allayannis and Weston (2001) who used it to control for the results derived with Tobin’s Q. This alternative measure is used because it measures the company value as Tobin’s Q does and therefore can be employed to test for the robustness of previous results of the analysis.
for speculative purposes an additional dummy variable that accounts for companies using hedging for speculative purposes is not needed within this model.
As introduced by Allayannis and Weston (2001) several control variables need to be implemented to exclude the effect of the main variables that could have an impact on the company value proxies Tobin’s Q and the logarithm of market value to book value assets. Therefore, the following control variables will be implemented.
Size is controlled by implementing a proxy calculated as logarithm of total assets, as described by Mackay and Philips (2005). Since it is assumed that due to high costs large companies are more likely to engage in currency hedging activities (Nance, Smith, and Smithson, 1993). Building on Allayannis and Weston (2001) the control variable for size is expected to have a negative influence on company value. Access to financial markets is controlled through the use of a dummy variable, which is 1 for companies paying dividends and 0 otherwise (Allayannis and Weston, 2001). This is necessary, since companies without access to financial markets are expected to have higher company value according to Serveas (1996) and Allayannis and Weston (2001) because only highest net positive value projects are engaged in. The capital structure of the company may be related to its value. Therefore, a leverage variable defined as long-term debt divided by shareholders equity is used, which is expected to have a negative influence on company value (Allayannis and Weston, 2001).
Profitability is expected to have a positive influence on company value and the proxies used because a profitable company is more likely to trade at premium and will therefore be controlled through the use of return on assets calculated as net income to total assets (Allayannis and Weston, 2001).
Even though Allayannis and Weston (2001) included Industrial Effects and Credit Rating in their study those two variables have not been included. Industrial effects have not been used because the sample studied includes companies acting in closely to similar sectors and Credit Rating because of the lack of data available. The exclusion of these two control variables is consistent with Jin and Jorion (2006) who studied the U.S. oil and gas industry with regard to the influence of hedging on company value. Model 1 and Model 2 are both analyzed through using ordinary least squares regression analysis first without control variables and second with control variables to control for possible influences on company value.
by Adler and Dumas (1984). The exposure sensitivity is defined as the change in the market value of the company resulting from a unit change in the exchange rate (Bodnar and Wong, 2003). With assistance of their model Adler and Dumas (1984) were able to measure the total exposure elasticity. The total exposure of a company includes two effects. First, the average change on the present value of cash flow caused by exchange rate movement. Second, the non-exchange rate effect, which is partly idiosyncratic and partly a macroeconomic effect, that influences the value of all companies as for instance changes in the market risk premium (Bodnar and Wong, 2003). To control for macroeconomic influences and to reduce the residual variance of the regression, compared to the equation without the macroeconomic variable, the market portfolio return is included as control variable (Bodnar and Wong, 2003).
Model 3:
3. 𝑅! = 𝛼! + 𝛾!𝐹𝑋 + 𝛽!𝑅!+ 𝜀
Where:
R is the stock return for company i, FX is the percentage change from Can$/US$ in the exchange rate variable on a yearly basis, RM is the return on the domestic market portfolio. γ is the exchange rate exposure elasticity of company i.
The extended version of the model including the macroeconomic control variable is used, as it is more reliable and generally preferred by researchers (see, e.g. Allayannis and Ofek, 2001; Bodnar and Wong, 2003). Model 3 is analyzed by comparing the means of the outcome for hedgers to the means of the outcome for non-hedgers.
3.3 Descriptive statistics
whole sample studied. However, Allayannis and Weston (2001) distinguished between companies with foreign sales and without foreign sales. Because in their analysis multinational companies with foreign sales are assumed to hedge more often in comparison to companies not engaging in multinational business. Allayannis and Weston (2001) studied all industries and therefore differentiated between companies with and without foreign sales, to identify multinational acting corporations. In this study it is not necessary to differentiate because the focus lies on companies from the oil and gas as well as the mining industry, which are acting in a multinational environment and are therefore assumed to be multinational in any case.
Therefore, the overall percentage of companies engaging in foreign currency hedging within this study should also be compared to the findings of Allayannis and Weston (2001) for companies with foreign sales. Allayannis and Weston (2001) found, that 60% of companies where engaging in foreign currency hedging while accounting for foreign sales. This paper needs to be compared to the sample of Allayannis and Weston (2001) with foreign sales because as described in the previous sections the companies within this study are assumed to act in a multinational environment.
Tobin’s Q is 1.93 on average, which indicates that companies have an overrated market value in general. A positive Tobin’s Q, which is higher than 1, identifies an overrated market value while a Tobin’s Q lower than 1 indicates an underrated market value (Allayannis and Weston, 2001). In comparison the average Tobin’s Q in the study of Allayannis and Weston (2001) amounts to only 1.20. The alternative measure for Tobin’s Q is positive and confirms the overrated market value of the companies within the sample. The mean for the Log(!"!") is 12.97. On average 30% of the companies included in the sample are paying dividends. Profitability was calculated using the ratio of net income to total assets. It is important to mention that on average the companies within the sample are not profitable providing a mean of -0.40. The reason for this negative number is that many companies within the sample have not been profitable over the years and reported high negative net income. The mean value for leverage is only 0.18, the reason is that many companies did not make use of any debt. This might be related to differences in strategy of companies in Canada or in the oil and gas as well as the mining industry.
hedgers, 2.13 compared to 1.57, which indicates that hedging is not enhancing company value and therefore stands against the hypothesis that company value is positively related to the use of foreign currency hedging. This finding is contradicting to the results presented by Allayannis and Weston (2001) but in line with the findings of (Jin and Jorion, 2006). However, the comparison of the means for the Log(!"!") indicates contrary findings than Tobin’s Q. The mean for hedgers is 14.41, while the mean for non-hedgers amounts to 12.14. This confirms the findings of Allayannis and Weston (2001) and indicates that no conclusion can be drawn at this point. The dummy variable for the payment of dividends reports a mean value of 0.19 for non-hedgers and 0.49 for non-hedgers, indicating that on average non-hedgers are more likely to pay dividends and have access to financial markets than non-hedgers. In line with the findings of Allayannis and Weston (2001) companies that use currency hedging are larger in size, with a mean of 14.46 compared to 11.99 for companies that not use currency hedging. However, hedging companies are more profitable than non-hedging companies, with a mean of 0.012 compared to -0.07 and have higher leverage 0.30 compared to 0.12. These results are somewhat contradicting to earlier studies, especially because companies engaging in hedging are more profitable but have less company value compared to non-hedging companies. The reason for these contradicting results might be explained by the high minimum for profitability of non-hedgers -2.23 in comparison to only -0.83 for non-hedgers.
Table 1 A-C Descriptive Statistics
N Minimum Maximum Mean Std.
Deviation
A.) All Companies
This table presents the descriptive statistics for model 1 and model 2. The sample includes 837 observations from companies active in the oil and gas as well as mining industry over the years 2005-2013. The data units used to calculate the different variables are in millions. Panel A presents the results for all companies within the sample. Panel B presents the results for all hedging companies within the sample. Panel C presents the results for all non-hedging companies within the sample. Tobin’s Q is calculated asthe ratio of total assets less the book value of equity plus the market value of equity to the book value of assets.
Table 2 A-C provides the descriptive statistics for all 837 companies from the oil and gas as well as the mining industry for the second analysis employed with the corresponding model 3. This is used to measure the exchange rate exposure of the companies within the sample with respect regarding the stock return. The mean of the return of the company stocks is 0.0269 indicating that on average the companies provide return. As already found in the previous section, on average 37% of the companies within the sample engage in currency hedging. The overall exchange rate exposure is 0.9367 in mean for all hedgers and non-hedgers taken together.
is greater in comparison to non-hedging companies. The findings of this model can be used to confirm and extend the findings of the previous analysis
Table 2 A-C Descriptive Statistics
N Minimum Maximum Mean Std.
Deviation
A.) All Companies
Ri 837 0.00 1.43 0.0269 0.12005 Hedging 837 0 1 0.37 0.482 RM 837 0.84 1.35 0.1192 0.01479 CFX 837 0.82 1.01 0.9367 0.06930 B.) Hedgers Ri 306 0.00 0.31 0.0221 0.04437 Hedging 306 1 1 1.00 0.000 RM 306 0.08 0.13 0.1198 0.01466 CFX 306 0.82 1.01 0.9482 0.06886 C.) Non-Hedgers Ri 531 0.00 1.43 0.0297 0.14689 Hedging 531 0 0 0.00 0.000 RM 531 0.08 0.13 0.1189 0.01487 CFX 531 0.82 1.01 0.9358 0.06960
4. Results
4.1 The use of foreign currency hedging over time
Building on Allayannis and Weston (2001), this paper first observes the use of foreign currency hedging over the sample period from 2005 to 2013. Table A attached in the appendices provides the statistics of how many companies engaged in hedging in the years 2005 to 2013. This paper does not report the notional value of the foreign currency contracts because of lack of information. Therefore, the level of foreign currency hedging is not included. Consistent with the findings of Allayannis and Weston (2001) the percentage of companies making use of foreign currency hedging increases over time. In 2005 only 35% of the companies included in the sample used foreign currency hedging, in 2013 the number of hedgers increased to 38, resulting in 41% for the whole sample. In the study of Allayannis and Weston (2001) the percentage of hedgers increased from 32% in 1990 to 40% in 1995. The increase of hedgers in this paper is slightly lower compared to their findings 6% to 8%. From the year 2007 with 39% the number of hedgers decreased to 34% in 2008 and 32% in 2009 before rising again in 2010. This might be related to the recent financial crisis with its height in 2008.
4.2 Univariate test
exposure (Allayannis and Weston, 2001). In turn this might increase the requirement for companies to make use of foreign currency hedging.
Table 3 provides the results of the mean of Tobin’s Q for non-hedging companies, hedging companies and the difference between the two means. The mean company value of non-hedgers is 2.13 and the mean value for hedgers is only 1.57. The value of non-hedgers is significantly greater by 0.56 in comparison to their counterpart. There is no hedging premium as stated by Allayannis and Weston (2001) who found a hedging premium through their whole sample and therefore, there is no evidence that the use of foreign currency hedging can increase company value. Based on the first results the hypothesis tested should be rejected because the use of foreign currency hedging seems to either have no or a negative influence on company value. These results are in alignment with the findings provided by Jin and Jorion (2006) who found no influence of foreign currency hedging on company value.
Table 3: Comparison of Tobin’s Q: Hedgers and Non-Hedgers
Mean N Std. Deviation
Non Hedgers 2.1333 531 2.41822
Hedgers 1.5709 306 0.87979
Difference 0.5624***
This table presents the results of the comparison of the means for Tobin’s Q for hedgers and non-hedgers. The sample includes 837 observations from companies active in the oil and gas as well as mining industry over the years 2005-2013. The data units used to calculate the different variables are in millions. ***, **, * denote significance at the 1%, 5%, and 10% levels, respectively.
4.3 Multivariate test
test whether the results are unbiased and reliable, which is presented in table 6. In the third step the same multivariate analysis using the same control variables is employed but the dependent variable Tobin’s Q is replaced by the previously described alternative measure. The results for the Log(!"
!") are presented in the tables 5 without
control variables and in 7 with control variables.
Since the overall model is slightly following the model developed by Allayannis and Weston (2001) the following control variables are implemented: 1.) Access to financial markets measured by using a dummy variable as explained in the model and methodology section is expected to be negatively related to Tobin’s Q; 2.) Profitability is expected to be positively related to the company value proxy; 3.) Size is expected to be negatively related to company value as supposed by Allayannis and Westons (2001); 4.) Leverage has is expected to be negatively related to company value; 5) Time effects are captured using year dummies. Even though Allayannis and Weston (2001) additionally controlled for Credit Rating and Industrial effects, those controls are not included as explained and motivated in the previous section.
Table 4 presents the results for the ordinary least square analysis with Tobin’s Q as dependent and hedging as independent variable without any control variables. The result shows a negative hedging coefficient, which indicates that the company value of hedgers is lower in comparison to the company value of non-hedgers. Under an economic perspective the use of hedging instruments would therefore be pointless and would not provide benefits. This negative influence additionally confirms the findings of the previous subsection and is consistent with the findings of Jin and Jorion (2006) who found no positive influence of hedging on company value in their study of the U.S. oil and gas sector. However, the results are again contradicting to the findings of Allayannis and Weston (2001). It needs to be mentioned that the R Square is quite low 0.018. The analysis was additionally performed with the logarithm of Tobin’s Q as done by Allayannis and Weston (2001) but the R Square presented was lower and therefore it was not included in this paper.
Table 4 OLS Tobin’s Q without control variables
Dependent variable: Tobin’s Q Coefficient Std. Error t
Observations 837 (Constant) 2.133*** 0.087 24.596 Hedging -0.562*** 0.143 -3.921 𝑅! 0.018 Std. Error of the Estimate 1.99862
This table presents the results for the ordinary least square analysis without control variables for model 1. The sample includes 837 observations from companies active in the oil and gas as well as mining industry over the years 2005-2013. The data units used to calculate the different variables are in millions. Tobin’s Q is calculated as the ratio of total assets less the book value of equity plus the market value of equity to the book value of assets is the dependent variable. Hedging is the independent variable defined as binary dummy variable indicating hedgers with 1 and non-hedgers with 0. ***, **, * denote significance at the 1%, 5%, and 10% levels, respectively.
The table 5 provides the results of the alternative measure. Consistent to the findings in the previous subsection the Log(!"!") is different to results of Tobin’s Q. and indicates a positive influence of hedging on company value.
Table 5 OLS Log(𝐌𝐕𝐁𝐕) without control variables
Dependent variable: Log
(
!"!")
Coefficient Std. Error tObservations 837 Constant 12.144*** 0.90 135.442 Hedging 2.266*** 0.148 15.283 𝑅! 0.219 Std. Error of the Estimate 2.06619
dummy variable indicating hedgers with 1 and non-hedgers with 0. ***, **, * denote significance at the 1%, 5%, and 10% levels, respectively.
Table 6 presents the results of the ordinary least squares regression with the above described control variables added. The R Square increases, which points out the overall improvement of the regression by the use of the above mentioned control variables, even though it is still quite low with 0.055. The F-Statistic however is significant. All variables are statistically significant, except the binary hedging dummy variable and leverage. The insignificance of the binary hedging dummy variable and leverage is contrary to expectations. The binary dummy variable used to control for the access to financial markets is positively related to company value, which stands in contrast to previous stated expectations and as well to Allayannis and Weston (2001) who found a negative relationship. These findings lead to the conclusion that companies with access to financial markets have a higher company value contrary to the expectations of Serveas (1996). The author states that companies with access to financial markets have less company value because companies without access to financial markets only engage in high valued net positive value projects. A possible explanation could be that those companies with access to financial markets have the possibility to engage in all value enhancing projects without the need to select and thereby miss possible opportunities. Surprisingly, profitability has a negative influence on company value, which is again contradicting to the findings of Allayannis and Weston (2001) and the expectations stated in the data and methodology section. The reason for this extraordinary result might be traced back to the quite high number of unprofitable companies within the sample. The results for the control variable for size is in line with the stated expectations and with the findings of Allayannis and Weston (2001) and show a negative influence on company value. Leverage is negatively related to company value as previously expected, however it is only statistically significant on the 1% level.
Table 6 OLS Tobin's Q with controls variables
Dependent variable: Tobin’s Q Coefficient Std. Error t
Observations 837 Constant 4.317*** 0.568 7.596 Hedging -0.144 0.168 -0.857 Dividend 0.565*** 0.201 2.806 Profitability -0.797** 0.341 -2.340 Log Size -0.192*** 0.048 -4.053 Leverage -0.299* 0.158 -1.893 F-Statistic 9.673*** 𝑅! 0.055 Std. Error of the Estimates 1.96800
This table presents the results for the ordinary least square analysis with control variables for model 1. The sample includes 837 observations from companies active in the oil and gas as well as mining industry over the years 2005-2013. The data units used to calculate the different variables are in millions. Tobin’s Q is calculated as the ratio of total assets less the book value of equity plus the market value of equity to the book value of assets is the dependent variable. Hedging is the independent variable defined as binary dummy variable indicating hedgers with 1 and non-hedgers with 0. Dividend is defined as dummy variable indicating 1 for companies with access to financial markets and 0 otherwise. Size is calculated as logarithm of total assets. Leverage is long-term debt divided by shareholders equity. Profitability is the ratio of net income to total assets. ***, **, * denote significance at the 1%, 5%, and 10% levels, respectively.
The results for the ordinary least squares analysis of the alternative measure with the same control variables as used for Tobin’s Q can be found in table 7. The R square of 0.797 is much higher than the one for Tobin’s Q and all control variables are statistically significant. This indicates that the alternative measure provides a statistically better model. However, the hedging variable indicates a slightly positive influence on the dependent variable Log(!"!"). This result is contradicting to the finding of model 1 but its not significant. Therefore, it is not possible to reason that currency hedging positively influences company value.
currency hedging such as Belghitar, Clark, and Mefteh, (2013) and Jin and Jorion (2004). Furthermore, the results stand in contrast to the findings of Allayannis and Weston (2001), Allayannis, Lel, and Miller (2012) and Hagelin and Pramborg (2004) who found a significant positive influence on company value. Hence, the conclusion remains as described in the previous analysis.
Table 7 OLS Log(𝐌𝐕𝐁𝐕) with control variables
Dependent variable: Log
(
!"!"
)
Coefficient Std. Error t Observations 837 Constant 1.583*** 0.305 5.190 Hedging 0.024 0.090 0.262 Dividend 0.367*** 0.108 3.402 Profitability 0.414** 0.183 2.266 Log Size 0.882*** 0.026 34.013 Leverage -0.455*** 0.085 -5.375 F-Statistic 650.964*** 𝑅! 0.797 Std. Error of the Estimates 1.05605This table presents the results for the ordinary least square analysis with control variables for model 2. The sample includes 837 observations from companies active in the oil and gas as well as mining industry over the years 2005-2013. The data units used to calculate the different variables are in millions. Log(!"
!") is the dependent variable. Hedging is the independent variable defined as binary
dummy variable indicating hedgers with 1 and non-hedgers with 0. Dividend is defined as dummy variable indicating 1 for companies with access to financial markets and 0 otherwise. Size is calculated as logarithm of total assets. Leverage is long-term debt divided by shareholders equity. Profitability is the ratio of net income to total assets. ***, **, * denote significance at the 1%, 5%, and 10% levels, respectively.
4.4 Exchange rate exposure and stock return
significant. Therefore, the results presented mainly rely on the comparison of the means from the descriptive statistics of the three groups presented in table 8.
Table 8 Comparison of Return, Market risk and Exchange rate exposure: Hedgers and Non-Hedgers
Mean N Std. Deviation Ri Hedgers 0.0221 306 0.04437 Ri Non-‐Hedgers 0.0297 531 0.14689 Difference -‐0.0076*** RM Hedgers 0.1198 306 0.01466 RM Non-‐Hedgers 0.1189 531 0.01487 Difference 0.0009*** CFX Hedgers 0.9482 306 0.06886 CFX Non-‐Hedgers 0.9358 531 0.06960 Difference 0.0124***
This table presents the results of the comparison of the means from the exchange rate exposure and stock return analysis and the corresponding model 3. The sample includes 837 observations from companies active in the oil and gas as well as mining industry over the years 2005-2013. The data units used to calculate the different variables are in millions. The stock return (Ri), the market risk (Rm) and the Exchange rate exposure (CFX) are presented respectively. ***, **, * denote significance at the 1%, 5%, and 10% levels, respectively.
Table 9 Exchange Rate Exposure and Stock Return Analysis
Dependent variable: Ri Coefficient Std. Error t
Observations 837 Constant -0.021 0.057 -0.378 Hedging -.008 .009 -0.905 MI 0.116 0.380 0.304 Change FX 0.040 0.081 0.492 F-Statistic 0.587 𝑅! 0.002 Std. Error of the Estimates 0.12013
4.5 Summary of the results
First, the descriptive statistics presented in the data and methodology section are shown for all companies and separate for hedgers and non-hedgers respectively. The comparison of Tobin’s Q of hedgers and non-hedgers lead to the conclusion that companies that engage in foreign currency hedging are not rewarded with higher company value. In fact the Tobin’s Q of 1.57 for hedgers compared to 2.13. On the other hand the Log(!"
!") presents a mean of 14.41 for hedgers and 12.14 for
non-hedgers, which leads to the opposite conclusion than stated for Tobin’s Q. From these contradicting results no explicit conclusion whether currency hedging might have a positive influence on company value can be drawn.
These findings are different then the ones presented by Allayannis and Weston (2001) who found a positive relationship. Moreover, the descriptive statistics show that hedging companies are more likely to have access to financial markets, which is contrary to the expectations stated before and might be explained by the greater ability to finance possible projects that have a positive effect on company value. Furthermore, hedging companies are more profitable in comparison to non-hedging companies with a mean profitability of 0.012 and -0.07 respectively. As mentioned before, this might be explained through the high number of unprofitable companies within the sample. Additionally, hedging companies are greater in size with a mean of 14.46 compared to 11.99 for the control variable size and have more leverage with a mean of 0.30 in comparison to 0.12.
Second, it is tested if the use of currency hedging increases over time as found by Allayannis and Weston (2001). The results show, that over the sample period from 2005 to 2013 the percentage of companies using currency hedging increases steadily by 6% from 35% to 41%, with some decline in the years 2007 and 2008.
difference by the comparison of means between hedgers and non-hedgers amounts to 0.56.
Fourth, the overall hypothesis that companies that use foreign currency hedging are rewarded with higher company value is tested by employing an ordinary least square analysis. In the first setting the regression is run without any control variables. The results indicate a negative relationship between the use of foreign currency hedging and company value for Tobin’s Q. The use of the alternative measure, the Log(!"!") provide different findings, resulting in a slightly positive influence on company value. The reason for this contrary results might be related to the methodology of the calculation for the company value proxies.
In a second setting the ordinary least square analysis is performed with various control variables that might have an impact on company value. The results presented for Tobin’s Q and the alternative measure of company value lead to the same results as described for the regression without control variables. However both measures provide non-significant results for the independent hedging variable. Therefore, no positive influence on company value from the use of currency hedging can be concluded. Based on these findings hedging is unreasonable under an economic point of view because there is no value creation.
is not significant for both company value proxies. In fact, the results for the main proxy Tobin’s Q indicate no or even a negative influence, which is consistent with the findings of Jin and Jorion (2006) because companies that make use of currency hedging have a lower Tobin’s Q than companies that make no use of currency hedging. Therefore, the main problem whether companies should make use of foreign currency hedging as risk management technique should be negated, which is in line with the findings of Modigliani and Miller (1958).
Fifth, an exchange rate exposure and return analysis was employed to test for the currency exposure of hedging companies in comparison to non-hedging companies and the stock return. Through the comparison of means the following results could be obtained. Hedging companies face a significant higher currency exposure in comparison to their counterpart with means of 0.9482 in comparison to 0.9358. At the same time companies engaging in currency hedging activities provide a significant lower stock return, 0.0221 against 0.0297 in mean for companies not engaging in hedging. By using the results of model 1 and model 2 it can additionally be stated that hedging companies have lower company value than non-hedging companies. Therefore, the findings of Choi and Prasad (1995) presented in the literature review section can be confirmed. The authors found a relationship between company value and exchange rate exposure.
5. Conclusion and Limitation
5.1 Conclusion
line with the study of Allayannis and Weston (2001) the control variable for company size is negatively related with company value. Contradicting to earlier studies (Allayannis and Weston, 2001) profitability is negatively related to company value, which might be explained by the great number of unprofitable companies within the sample. As suspected and in alignment with Allayannis and Weston (2001) the influence of leverage is negatively related to company value.
This thesis cannot confirm the main conclusion of the paper from Allayannis and Weston (2001) that foreign currency hedging increases company value. Because the proxies for company value Tobin’s Q and Log(!"!") provide different results. Besides both regressions report non-significant results for the hedging variable
Moreover, the results of the main proxy Tobin’s Q provide evidence that hedging is negatively related to company value but these findings are not statistically significant. Therefore, the results of this paper are inconsistent with the theories that state hedging increases company value as for example Nance, Smith and Smithson (1993), Allayannis, Lel and Miller, (2012) or Hagelin and Pramborg (2004). Instead, the results are in line with the findings presented by Jin and Jorion (2006) and show no positive relation to company value. It also confirms the study of Belghitar, Clark, and Mefteh (2013) who found that the use of foreign currency derivatives can reduce overall foreign exchange exposure but is not positively related to company value. The results of the exchange rate exposure and stock return analysis provide evidence that companies engage in hedging to reduce their exposure but no evidence could be found that these hedging activities have any positive influence on stock return or enhance the market value of the companies using it. Additionally, the findings of Choi and Prasad (1995) that exchange rate exposure is related to company value can also be confirmed on the basis of the exchange rate exposure analysis.
positive relationship between hedging and the amount of stock held by corporate managers. Additionally, a negative relationship with company value was found in the presence of an agency problem between shareholders and managers. Even though this agency problem cannot be observed in this study, the overall results speak in favor of the theories and findings provided by Jin and Jorion (2006), Belghitar, Clark and Mefteh (2013), Tufano (1996), Stulz (1984) and Modigliani and Miller (1958) stating that currency hedging is not positively related to company value and should not be used as risk management technique.
Concluding, the findings of this thesis contribute to the international financial management literature because the context studied is in alignment with all three requirements for an international financial management thesis, as it studies foreign currency hedging, which is a financial topic with several implications for risk management and because the whole study is performed in industries mainly acting in an international environment. The findings are of academic relevance because they confirm the findings and results of Jin and Jorion (2006); Belghitar, Clark and Mefteh (2013); Tufano (1996); Stulz (1984) and Modigliani and Miller (1958) and additionally provide new and recent data as well as information concerning the influence of currency hedging on company value within two multinational industries. Furthermore, this thesis is of societal relevance because based on the results the main problem stated in the introduction can be answered and provides evidence that companies should not engage in foreign currency hedging as risk management technique or at least should really carefully assess it before making use of it.
5.2 Limitation
proxies provide different results additional proxies should be implemented to find more clear and reliable results. Furthermore, this thesis relies on the results from the analysis with Tobin’s Q to a great extend because it relies on the paper of Allayannis and Weston (2001). However, the overall results of the alternative measure Log(!"
!")
used in model 2 provide better results as indicated by the 𝑅! of 0.797 in comparison
to 0.055 for model 1. Therefore, it would make sense to closely investigate the differences of the two proxies for future research.
A possible improvement of this thesis could be to observe and include the hedging ratio and thereby observe the level of foreign currency hedging from companies within the sample. Even though Allayannis and Weston (2001) have been able to obtain the level of foreign currency hedging for a great amount of companies within their sample, including only those companies that report the level of hedging reduced their sample by 39%. Therefore, the authors also relied on the use of a binary variable to perform their regression analysis. However, it would be of great interest to study the level of currency hedging to obtain more specific results. Future researchers could include such a ratio and additionally expand the sample. Furthermore, different multinational industries could be focused on to test for the robustness of the results from the oil and gas as well as the mining industry.
6. References
Adler, M., and Dumas, B. (1984). Exposure to Currency Risk: Definition and Measurement. Financial Management , 13, 41-50.
Allayannis, G., and Ofek, E. (2001). Exchange Rate Exposure, Hedging and the use of foreign currency derivatives. Journal of International Money and Finance, 20, 273-296.
Allayannis, G., and Weston, J. P. (2001). The Use of Foreign Currency Derivatives and Firm Market Value. The Review of Financial Studies , 14, 243-276.
Allayannis, G., Lel, U., and Miller, D. P. (2012). The use of foreign currency derivatives, corporate governance, and firm value around the world. Journal of
Bank of International Settlements. (2014). Triennial Central Bank Survey of foreign
exchange and derivative market activity. Basel: Bank of International
Settlements.
Belghitar, Y., Clark, E., and Mefteh, S. (2013). Foreign Currency Derivative use and Shareholder Value. International Review of Financial Analysis , 29, 283-293. Bodnar, G. M., and Franco Wong, M. H. (2003). Estimating Exchange Rate Exposure: Issues in Model Structure. Financial Management , 32 (1), 35- 67. Brown, B. (2006). Economists and the Financial Markets. London: Routledge.
Choi, J. J., and Prasad, A. M. (1995). Exchange Rate Sensitivity and its Determinants: A Firm and Industry Analysis of U.S. multinational. Financial Management , 24 (3), 77-88.
Clark, E., and Mefteh, S. (2010). Foreign Currency Derivative Use, Firm Value and the Effect of the Exposure Profile: Evidence from France. International
Journal of Business , 15 (2), 183-196.
DeMarzo, P., and Duffie, D. (1995). Corporate Incentives for Hedging and Hedge Accounting. The Review of Financial Studies , 8, 743-711.
Froot, K., Scharfstein, D., and Stein, J. (1993). Risk Management: Coordinating Corporate Investment and Financing Policie. Journal of Finance , 48, 1958.
Geczy, C., Minton, B., and Schrand, C. (1997). Why Firms Hedge in Response to Tax Incentives? Journal of Finance , 52, 1324-1354.
Graham, J., and Rogers, D. (2002). Do firms hedge in response to tax incentives? Journal of Finance , 57 (2), 815-839.
Guerrieri, P. (1982). Multinational corporations and the mining industry. Lo
Spettatore Internazionale , 17 (1), 51-60.
Hagelin, N., and Pramborg, B. (2004). Hedging Foreign Exchange Exposure: Risk Reduction from Transaction and Translation Hedging. Journal of
International Financial Management & Accounting , 15 (1), 1-20.
Haushalter, D. (2000). Financing Policy, Basis Risk, and Corporate Hedging: Evidence from Oil and Gas Producers. Journal of Finance , 55, 107-152.
Holden, M. (2007). Explaining the Rise in the Canadian Dollar. Ottawa: Library of Parliament Canada.
Leland, H. E. (1998). Agency Cost, Risk Management, and Capital Structure. Journal
of Finance , 53, 1213-1243.
Lindenberg, E., & Ross, S. (1981). Tobin's q ratio and industrial organization. Journal
of Business , 54, 1-32.
MacKay, P., and Phillips, G. M. (2005). How does Industry affect Firm Financial Structure? Review of Financial Studies , 18, 1433-1466.
Modigliani, F., and Miller, M. (1958). The cost of Capital, corporation finance and the theory of investment. American Economic Review , 48, 261-297.
Nance, D., Smith, C., and Smithson, C. (1993). On the Determinants of Corporate Hedging. Journal of Finance , 48, 267-284.
Serveas, H. (1996). The value of diversification during the conglomerate merger wave. Journal of Finance , 51, 1201-1225.
Smith, C., and Stulz, R. (1985). The Determinants of Firms Hedging Policies. Journal
of Finance and Qualitative Analysis , 20 (4), 391-405.
Statistics Canada. (2014, 09 18). http://www.statcan.gc.ca/start-debut-eng.html. Retrieved 12 29, 2014, from http://www.statcan.gc.ca/start-debut-eng.html: n151c-eng.htm
Stulz, R. (1984). Optimal hedging policies. Journal of Financial and Quantiative
Analysis , 19, 127-140.
Stulz, R. (1996). Rethink Risk Management. Journal of Applied Corporate Finance ,
9, 8-24.
Tufano, P. (1996). Who Manages Risk? An Empirical Examination of Risk Management Practices in the Gold and Mining Industry. Journal of Finance , 51, 1097-1137.
Yermack, D. (1996). Higher market valuation of companies with a small board of directors. Journal of Financial Economics , 40, 185-211
7. Appendices
Appendix A: Foreign Currency Hedging over time
2005 2006 2007 2008 2009 2010 2011 2012 2013 Hedgers 33 33 36 32 30 32 34 37 38 Non Hedgers 59 59 56 60 62 60 58 55 54 Percentage of Hedgers 35% 35% 39% 34% 32% 34% 37% 40% 41%
Appendix B: The exchange rate CAD$/US$ over time
0.00 0.20 0.40 0.60 0.80 1.00 1.20 2005 2006 2007 2008 2009 2010 2011 2012 2013
