i UNIVERSITY VAN AMSTERDAM
AMSTERDAM BUSINESS SCHOOL BSc Economics & Business
Specialisation Financial and Organization
The Research on Foreign Exchange Exposure in US Import
and Export Industries
Author: Hao Zhang Student Number: 10717838
Thesis Supervisor: Chih-Chung Ting Finish Date: 06-2018
ii Statement of Originality
This document is written by Hao Zhang who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
iii Abstract
It has been argued for a long time that whether the performance of the company would be affected by the exchange exposure. What motivates me to research it is that it has not been researched in such a classification before and under the background of the trade war, this research might provide a new aspect for countries involved to concern. It creates the possibility for countries to introduce a policy related to the exchange rate which has less or even no hurt to themselves since the policies on custom duty are also harmful to countries introducing them. This paper provides empirical evidence based on the data from US clothing companies and vehicle companies. Besides, whether this effect is unilateral would also be discussed in this paper. Our findings shows that the performance of the company, which reflected by stock return in this case, would be weakly affected by exchange exposure and the existence of unilateral influence still needs to be researched.
iv TABLE OF CONTENTS
STATEMENT OF ORIGINALITY………..ii
ABSTRACT……….iii
PART 1 Instruction………1
PART 2 Literature Review………4
2.1 Exchange rate exposure………4
2.2 The Effect of Exchange Exposure on Stock Return……….………4
2.3 Exchange Rate Exposure on Import and Export………..7
2.4 Model Setup..………...……..9
PART 3 Data……….……….11
PART 4 Result………14
PART 5 Conclusion……….16
REFERENCE……….17
APPENDIX Listing of Sample Companies………20
Clothing Industry……….20
1 Part 1: Introduction
With the development of E-commerce and globalization, there are increasing number of multinational companies. According to the data from World Integrated Trade Solution (WITS), the global import climbed from 0.617 trillion dollar in 1988 to 18.458 trillion dollar in 2014. The amount of cross-border transactions had a 3000% increase in around 30 years, which means that there is a large amount of money needs to be exchanged into the currency that has been confirmed in the contract. In the other words, exchange rate is becoming more important in recent decades.
As the largest and the second largest economies in the world, the United States and China have a close contact in trade and the amount of cross-border transactions also developed steeply in these 30 years. However, in April 2018, the Trade War between China and the United States has been launched by president Trump. Goods valued 60 billion dollar in High-tech industry are involved in the Trade War and United States is going to collect higher custom duty for these goods involved. The custom duty is the main factor that would be controlled in the trade war because it would have a direct effect on the price according to “international trade”. However, the direct effect is negative and it is for both countries. When the custom duty rate is announced to be increased, the cost for exporting side increases. It will hurt the economy of the exporting country because companies are less willing to export. At the meantime, the price in the foreign market has to be increased to guarantee that the exporting company is not making a loss and with the increase of price, there are few demand in the market of the importing country, especially for non-necessities which also hurts the economy for the importing country. Thus, for participants in the trade war, they need to find a way to restrict other participants which will not hurt themselves so we would focus on the exchange rate in this paper. The reason that I choose exchange rate rather than other factors which can be controlled by the government, such as export
volume, is that the exchange rate is the indispensable part of import and export. Besides, for the factors like export volume, restricting these factor in the trade war might only affect export companies but not import companies. In other words, the effect of other factor might
2 be not complete. Additionally, in the previous research, authors always researched the market as a whole but did not split the companies into import and export companies. This method might ignore the different effect from import and export companies that the change of exchange rate or exchange exposure may have on the stock return. Thus, the research question for this paper would be:
To what extend that foreign exchange exposure would affect the stock return of US companies and what is the different effect of the foreign exchange exposure that may
have from export and import industries on the stock return.
In order to address the research question, the clothing industry would be chosen in this paper. There are two reasons that this industry is chosen. Firstly, the custom duty rate for clothing industry does not change in the Trade War. Secondly, according to the data from WITS, around 30% of import of clothing industry in the United States are from China which is the second largest importing industry for the United States (the largest import industry is affected by trade war). It contains services (outsourcings), raw materials, intermediates and final goods from Chinese brands. Besides, around 15% of clothing products from the United States are exported to China which shows that clothing industry is closely related to the exchange rate since there is a large amount of cross-border transactions. Vehicle Industry, which contains vehicle and parts, will also be discussed since that Vehicle Industry is the largest US exporting industry to China. By comparing the previous study e.g. Jorion (1990), Du et al. (2014), it is found that import or export might lead to different result of effect that exchange rate exposure has on the stock return. Thus, according to the result of both industries, we can have a rough comparison and figure out the differences that the change of exchange rate would cause between import and export industries. Since both industries are “large” and “extreme”, the result would be reliable and representative and the
conclusion can be applied in other situations similar to the trade war which need macro-control from the government side. At least, the research provides a new perspective for future researches in the relevant field that the output of import industries might be different from that of export industries even though other control factors are the same. 2000-2007
3 and 2010-2013 would be chosen. The reason that 2008 and 2009 are dropped is that I want to decrease the effect to stock return due to the financial crisis started in 2008. Besides, the reason that only US companies are researched here is that most companies of production in clothing industries are unlisted private companies. Thus, we cannot find the data of stock price for Chinese companies and we will only focus on US companies in this paper.
According to Anhar (2015) and Anwaar (2016), stock return can significantly reflect the performance of the company. Thus, when stock return increases, it can be concluded that the company was affected positively and vice versa. Besides, according to Jorian (1990), stock return would be significantly affected by the exchange rate exposure coefficient but it was focus on the whole US market. According to Liu and Shu (2003), in different industries, the scales of import and export are also different and the influence that would have can also be different which means the result of that research might not be applicable enough to all industries separately which motivates me to test it again only for clothing industry and vehicle industry to see whether it is still tenable.
The remainder of this paper is organized as follows. Part 2 reviews previous studies which analyzed the effect of the change of exchange rate on stock return and the model setup. Part 3 summarizes the data, part 4 presents the results. Part 5 discusses the conclusion, limitations, discussions and recommendations.
4 Part 2: Literature Review
2.1: Exchange rate exposure
According to Dumas (1978), Hodder (1982), Adler and Dumas (1984), foreign exchange rate exposure is the relationship between excess return and the change of exchange rate and this definition has been widely approved and used. Besides, Adler and Dumas (1984) also
pointed out that the concept of exposure is arbitrary that exchange rate and stock price are jointly and endogenously determined. In the research of Jorion (1990), a specific regression model of foreign exchange rate exposure has been introduced:
𝑟𝑖 = 𝛽0+ 𝛽1𝑟𝑥+ 𝛽2𝑟𝑚+ 𝜀 (1)
Where ri is the stock return, rm is the return of market index and rx is change of exchange
rate. In this model, β1 is the exchange exposure coefficient, which shows the effect that
the change of exchange rate on stock return. Since it reflects its degree that the change of exchange rate would affect the stock return, which might answer the research question in this paper, the change of exchange rate would be chosen as the dependent variable. This model combines all the definitions from the researchers mentioned above and according to a review which summarized all related studies about exchange exposure by Omar et al. (2017), this model is used most widely in the researches related to stock return and exchange exposure till now. The stock return in this model is the independent variable, which also matches what I am going to research so that this model would also be used in my paper. In addition, to research this effect, exchange rates is assumed to be exogenous to stock return and after that the definition can be transferred to the model above (He and NG, 1998)
2.2: The Effect of Exchange Exposure on Stock Return
In 1990, Jorion published a paper which did a research about the exchange rate exposure of multinational companies in the United States. There was no research, which was going to figure out the determinants of foreign exchange rate exposure, has been done before Jorion. Thus, he had a good start in the research at this field. First, he ran the regression of
5 𝑟𝑖 = 𝛽0+ 𝛽1𝑟𝑥+ 𝛽2𝑟𝑚+ 𝜀
By using the data during 1971-1987 with 287 sample companies. There are 3 sub-periods: 1971-1975, 1976-1980, 1981-1987 chosen, which showed that the foreign exchange rate exposure coefficient is close to zero and only 5% of the sample companies generated the significant result from the first regression model. There was a large standard deviation due to the cross-sectional differences so Jorion came up with a model with was estimated by generalized least squeares (GLS). Jorion selected 40 companies from the sample which have common features at the same time and this time, the result rejects the hypothesis that the exposure coefficient are all equal or equal to zero which means that exchange exposure do have some effect on the stock return.
Amihud (1994), Bodnar and Gentry (1993) concluded the similar result with Jorion (1990), which is that the stock return would be affected by the change of exchange rate. However, Gendreau (1994), Bartov and Bodar (1994) had different opinions. Gendreau pointed that it is not convincing by applying the weak result. Bartov and Bodar concluded an opposite result about the relationship between exchange rate change and the stock return. On the other hand, Levi (1994) said that the differences between research results were caused by the difficulty to stably obtain the foreign exchange rate exposure. Thus, in 1998, He and Ng did a research which was about the foreign exchange rate exposure of Japanese multinationals, which will be discussed in the following.
Because the weak evidence concluded from Jorion’s study was based on international data and it needs to be investigated further, He and Ng chose to do the research of Japanese market to find more evidence. There are several reasons that the authors chose Japanese market and two of these reasons are important. First, Japan was the second largest economy at that time, which means that the result from previous study can be examined with the result of this paper convincingly. Second, Japanese market is more susceptible to the changes of exchange rate because of the diversity of industries.
6 The same regression model from Jorion (1990) has been used here. For the samples, He and Ng followed the method of Jorion and 171 Japanese multinational companies which have the export ratio (exports divided by total sales) larger than 10 percent are picked as sample companies. The data during 1979-1986 and 1987-1993 of these 171 companies would be tested by the regression model. Additionally, there are two economic factors were
considered. The first one is the market risk, and the second one is the currency risk factor. There are 23% of sample companies generate significant result and it also rejects the null hypotheses of the stock return will not be affected by the change of exchange rate.
By comparing two papers, authors in both papers used Jorion’s two factor model to test whether the stock return would be affected by the exchange exposure. However, only 5% of the sample companies generated significant results in the research of Jorion which was 23% in the research of He and Ng and as it is mentioned above, some researchers thought the result that Jorion concluded was too weak but what is the reason that the result became more significant in the research of He and Ng. For this reason, I did more readings on close researches. In 2014, Huston and Laing published a paper which is also focus on US
multinational companies, and there are only 5.2% of sample companies generate significant result by using the 953 US firms during 1999-2006 and in this paper, author also used Jorion’s two factors model to verify. In the same year, Du, Hu and Wu did a research on Taiwanese multinational companies, there are 31% of sample companies significantly expose to exchange rate. In the research of Du, He, and Wu, there are several models used in the article. Except Jorion’s two factors model, author also used also used one factor model𝑟𝑖 =
𝛽0+ 𝛽1𝑟𝑥+ 𝜀 , three factors model 𝑟𝑖 = 𝛽0+ 𝛽1𝑟𝑥+ 𝛽2𝑆𝑀𝐵 + 𝛽3𝐻𝑀𝐿 + 𝜀 and four
factors model 𝑟𝑖 = 𝛽0+ 𝛽1𝑟𝑥+ 𝛽2𝑆𝑀𝐵 + 𝛽3𝐻𝑀𝐿 + 𝛽4𝑀𝐾𝑇 + 𝜀 (where the variable SMB
is the difference between the returns on diversified portfolios of small stocks and big stock; HML is the difference between the returns on diversified portfolios of high book-to-market (value) stocks and low book-to-market stocks and MKT is the excess market return), which are transformed from basic model of CAPM 𝑟𝑖 = 𝛽0+ 𝛽1𝑟𝑚+ 𝜀 and Fama & French Model
𝑟𝑖 = 𝛽0+ 𝛽1𝑆𝑀𝐵 + 𝛽2𝐻𝑀𝐿 + 𝛽3𝑀𝐾𝑇 + 𝜀 or referred from previous studies, for example,
7 contains variable “change of exchange rate” by checking R square in order to find out the model that explains the most of the relation between the change of exchange rate and stock return and they found that four factors model has the highest R square.
2.3: Exchange Rate Exposure on Import and Export
Jorion (1990) was the first one that related the exchange exposure coefficient with a factor of export. In the GLS process from the first part, he found that the group of 40 samples that has the high and the most significant exposure is the one with highest percentage of foreign operation and that is also the reason that he picked export volume as the determinant of the exchange exposure coefficient to verify. With the result from previous sessions, Jorion built a new model following in order to research the relationship between the exposure coefficient and the export volume:
β = 𝛼0+ 𝛼1𝑋𝑥+ 𝜇 (𝟐)
Whereβis theβ1 from the regression model (1), which is the foreign exchange rate
exposure coefficient and Xx is the percentage of export occupying in the total sales.
Moreover, Jorion combined the equation (1) and (2) because all the betas are estimated over the same sample period and model (2) is difficult to be regressed individually. Thus, he got:
𝑟𝑖 = 𝛽0+ (𝛼0+ 𝛼1𝑋𝑥)𝑟𝑥+ 𝛽2𝑟𝑚+ 𝜀 (𝟑)
Besides running the regression with the data from GLS, Jorion also ran the data divided by oil companies and non-oil companies. At the end, Jorion found that the foreign exchange rate exposure is positively and reliably related to the scale of foreign operations.
The regressions of Jorion were clear in his research and the relation between variables were also explained. However, the sample size is not large enough in my opinion. There were only 19 non-oil companies were counted. Besides, different industries have different sensitivity and reaction to the change of exchange rate. According to Jorion, export volume has a positive relationship this the exchange rate exposure so conversely, import volume should have a negative relationship with that. However, the effects to import volume cannot be
8 observed from the research of Jorion. Thus, in my paper, I would extend the samples to 100 companies and I would pick companies from both import industries and export industries so that the effects from both industries would be concerned.
The second part of He and Ng (1998) also mentioned the export volume. The second regression that He and Ng did is to test that what are the determinants for the foreign exchange rate coefficient. Except export ratio, company size, dividend payout ratio, quick ratio, book-to-market equity value and long-term debt to its market value of equity were considered. Besides, comparing to Jorion, He and Ng made a change of the model to differ the effects from exporting and importing companies (with positive and negative beta respectively). The expression this this model is shown as:
𝛽̃𝑖𝑥 = 𝛼0𝐷 + 𝛼1𝐷𝑙𝑜𝑔𝑆𝑖𝑧𝑒 + 𝛼2𝐷Expr + 𝛼3𝐷𝐷𝐼𝑉 + 𝛼4𝐷𝑄𝑅 + 𝛼5𝐷𝐵𝑀 + 𝛼6𝐷𝐷𝐸 + 𝛼𝑑0(1 − 𝐷) +··· +𝛼𝑑6(1 − 𝐷)𝐷𝐸 + 𝜀
In this expression, D is a dummy variable, for exporting companies, D=1. According to the result, He and Ng concluded that export volume would significantly and positively affect the exchange exposure coefficient.
Besides, there is one other important point mentioned by He and Ng, positive and negative of beta, which is related to export and import. They pointed that importing companies would benefit from the appreciation of the home currency, which leads to a negative
exposure coefficient in the regression model. According the result of the Jorion’s two factors model, He and Ng concluded that for the company with positive exposure coefficient,
depreciation has a positive impact on the stock returns. However, for companies with negative exposure coefficient, depreciation also have a positive impact, which is not explained in the paper but generally speaking, it can be concluded that the stock return would be affected by the change of exchange rate.
Compare to the research of Jorion (1990), He and Ng improved the research, especially by introducing the positive and negative characteristic of beta and used the dummy variable in the second regression which divided import and export companies. However, according to
9 the data that He and Ng found in the paper, there were very few importing companies. This is also the reason that they could not explain the phenomenon that for importing
companies, the effect on stock return was same as the exporting companies. Besides, in the research of He and Ng, they also do not divide companies by industries, which means the result might be not that significant due to the differentiations in sensitivity to the change of exchange rate. Moreover, the companies that He and Ng picked were companies with export ratio larger than 10%. If the samples were chosen in this way, many import companies would be ignore, which has been shown in this article. For instance, a company may have large import of raw material or outsourcing but only sell its products to domestic market. Thus, I would use other method to choose samples and to generate more significant results, I choose and focus on clothing and vehicle industries in my paper. As the data shown, the goods that U.S. clothing companies imported from China are much more than that they exported. The effect from a negative beta could be figured out in some degrees in my paper.
2.3: Model Setup
According to all researches above, I found that for countries with more import than export, like US, the result of the regression is weak but in countries and areas like Japan and Taiwan, the results are more significant. Consequently, I would like to research that whether the import or export would affect the result of regression model which contains stock return and exchange exposure coefficient. In my paper, I would use two models above to examine the relation between stock return, which reflects companies’ performance, and exchange exposure. The first model I would use is Jorion’ s two factors model 𝑟𝑖 = 𝛽0 + 𝛽1𝑟𝑥+
𝛽2𝑟𝑚+ 𝜀, which is also the most widely used in relevant studies and the other model I would use is the four factors model𝑟𝑖 = 𝛽0+ 𝛽1𝑟𝑥+ 𝛽2𝑆𝑀𝐵 + 𝛽3𝐻𝑀𝐿 + 𝛽4𝑀𝐾𝑇 + 𝜀 since in the
article of Du, He and Wu (2014), this model explains the most. Two models are substitutes since the independent variable and the variable of change of exchange rate are the same and control variables in two models describe the market situation. We use 2 models here is to confirm that the result is reliable. As it is mentioned, both models are substitutes so that the results from both models should be the same in the major direction, e.g. the positive and
10 negative of beta. If the positive and negative of beta are different from both models, the exact effect from the change of exchange rate would be ambiguous. The method that I would use these two models is to run the data from clothing and vehicle industries
separately. The result of regression could test the null hypothesis that the stock return will not be affected by the exchange exposure. Simultaneously, a comparison of the results from 2 extreme industries can be done, which can be used to test the second null hypothesis that export industries generate better results than import industries.
11 Part 3: DATA
In this paper, top 50 companies from US clothing and vehicle industry are chosen (see Appendix Listing of clothing and vehicle companies in the sample). The research period would be 2000-2007 and 2010-2013, which is 10 years in total. All data could be find in the Center for Research in Security Prices (CRSP). The exchange rate used in this paper is defined as RMB/USD which is different from the paper of Jorion (1992) and He & Ng (1998) and this is the reason that the positive and negative of coefficient is opposite to that in those two papers.
According to previous studies of Lymer and Tallberg (1997), Suzukii (1980), Guthrie et al. (2006) and Bosiu et al. (2017), “top” companies can be defined and chosen according to the market capitalization, which is equal to price multiplies shares outstanding. In the period I chose, over a half of companies in the stock market do not exist during the whole period. There are several reasons that these companies are not existed. The first reason is
companies broke up. The second reason could be the companies transfer from public to private. The other reason that companies disappears in the stock market is that there is a merger or acquisition happening. After the M&S process, the asset would be transferred to another company (might be in the stock market of other countries). For these reasons, the market capitalizations for all companies cannot be found in the database CRSP directly so I used the method that price multiplies shares outstanding in 2000. Besides, to ensure that the companies in sample are useful and to avoid the companies scale is too small even if it is in top 50, the companies with market capitalization smaller than 150,000,000 dollar would not be considered. The reason that the method to choose sample companies by export ratio, which has been mentioned in the previous part, is not used here is because only depending on export ratio might ignore large import companies which have zero or little export. Moreover, non US companies in top 50 would also be dropped because many companies appear on the stock market in the US but their headquarters are still in their origin countries. For these companies, export and import are usually counted in their origin countries. The stock return in the Jorion’s two factors model need also be calculated since it is not available in the CRSP. The formula to compute the stock return is same as computing accounting rate
12 of return (ARR), which is r =𝑝2𝑝−𝑝1
1 . After the processes above, the summary of the data is
shown as following:
TABLE 1 - Descriptive Statistics of Sample in Clothing Industry
Variable Obs Mean Std. Dev. Min Max
Date 5,668 20000131 20131231 Price 5,665 29.88615 23.07171 0.94 234.58 Shares Outstanding 5,666 91888.73 135761.8 6594 903759 Stock Return 5,652 0.012155 0.137825 -0.806175 1.296774 Excess Return 5,656 0.004176 0.041699 -0.1072 0.1135 SML 5,656 0.003780 0.036845 -0.1728 0.2214 HML 5,656 0.005140 0.031331 -0.1057 0. 1290 Change of Ex-rate 5,655 -0.001634 0.002910 -0.0173312 0.0061826 Market Return 5,658 0.005717 0.039841 -0.108965 0.109014
TABLE 2 - Descriptive Statistics of Sample in Vehicle Industry
Variable Obs Mean Std. Dev. Min Max
Date 6,880 20000131 20131231 Price 6,877 29.08909 20.05472 -26.485 136.27 Shares Outstanding 6,878 145364.6 396638.9 10089 3873585 Stock Return 6,824 -0.003245 0.211537 -3.666667 2.7 Excess Return 6,823 0.001878 0.046555 -0.1723 0.1135 SML 6,823 0.004191 0.036218 -0.1728 0.2214 HML 6,823 0.004851 0.034109 -0.1110 0. 1290 Change of Ex-rate 6,823 -0.001787 0.003391 -0.0173312 0.0061826 Market Return 6,823 0.003258 0.044756 -0.166980 0.109014
13 As it is summarized above, there are very large standard deviations in variables Price, Shares Outstanding and Stock Return. The large standard deviations are caused by the
differentiations between companies. Consequently, the Jorion’s two factors model and four factors model in this paper would be run in company level in order to avoid the inaccuracy due to large standard deviations.
14 Part 4: Result
By using the data and methodologies from previous parts, the following result could be generated:
TABLE 3 - Summary Statistics of Beta with Jorion’s Two Factors Model The data in this table is based on the regression model: 𝑟𝑖 = 𝛽0+ 𝛽1𝑟𝑥+
𝛽2𝑟𝑚+ 𝜀
Industry N Min Median Max N** N* N- N+
Clothing 50 -95.812 2.335 10.863 5 7 11 39
Vehicle 50 -2259.5 0.35 332.897 0 4 20 30
Where ** is the significant level of 5%, *is the significant level of 10%, N+ is the number of companies that beta is positive and N-is the number of companies that beta is negative
TABLE 4 - Summary Statistics of Beta with Four Factors Model The data in this table is based on the regression model: 𝑟𝑖 = 𝛽0+ 𝛽1𝑟𝑥+
𝛽2𝑆𝑀𝐵 + 𝛽3𝐻𝑀𝐿 + 𝛽4𝑀𝐾𝑇 + 𝜀
Industry N Min Median Max N** N* N- N+
Clothing 50 -15.957 2.335 74.424 5 10 16 34
Vehicle 50 -1997.76 -0.56 197.115 1 3 31 19
Where ** is the significant level of 5%, *is the significant level of 10%, N+ is the number of companies that beta is positive and N-is the number of companies that beta is negative
There are two results can be produced from the tables above. Firstly, the stock return, which reflects the performance of company, would be affected by the exchange exposure but the result is also weak in the industries and period that I researched. There are only 5%
companies generate significant result at 5% significant level with the Jorion’s two factors model and 6% sample companies generate significant result with the four factors model. This result reject the first null hypotheses that the stock return will not be affected by the exchange exposure.
15 The second finding is that Vehicle (export) industry generates much fewer significant result than clothing (import) industry (100% fewer by using Jorion’s two factor model and 80% fewer by using four factors model). This result does rejects the second null hypotheses that vehicle industry should generate more significant result than clothing industry. There are several reasons which might lead to this result. The first reason might be that there are 12 months in the dataset with no change in exchange rate and in these 12 months. The coefficient cannot be computed but the standard deviation would still be affected and this would affect the calculation of t-value. The second reason might be that financial crisis still makes influences to sample companies even though year 2008 and 2009 have been
dropped. In the data part, it can be found that there are more observations in vehicle industry than that in clothing industry and there is the difference because that clothing companies exist shorter so that many companies in clothing industry are not affected by financial crisis because these companies quitted stock market before the financial crisis. Last but not the least, the method to choose sample companies is not appropriate. According to He and Ng (1998), import and export companies can be judged by beta. However, the companies I pick do not show it obviously, which means there might be some companies with low foreign operation level or some companies that cannot represent the whole industry (clothing companies with more export or vehicle companies with more import) were chosen.
There is an additional finding which is that four factors model does explain more than Jorion’s two factors model. The R-square of all 100 sample companies generates from both method were compared and in all cases, four factors model generates better R-squares and this proofs the point of view which was introduced in Du et al. (2014).
16 Part 5: Conclusion
In conclusion, there are two hypotheses according to the research question. The first hypotheses is that the stock return of companies would not be affected by the change of exchange rate. The second hypotheses is that the effect of exchange exposure would be more obvious for export industry than that for import industry. Based on results above, the performance of a company can be weakly affected by exchange exposure and for import industries, which rejects the first hypotheses. Accordingly, in the Trade War, the United States and China can introduce policies related to exchange rate but not only focus on the custom duty since it is harmful to both countries as we analyzed in the introduction part. The second hypotheses is rejected in my paper because the data from this paper is not perfectly chosen as it is discussed but the second research question still needs to be researched further since this paper provides a new direction of comparing the different effect from import and export industries for the researches in the relevant fields. It is important to do further researches on the research questions, because previous studies and we have found that the exchange exposure could affect the stock return. If this effect could be researched and found that as unilateral, it would be helpful for countries in the Trade War to win a lead. Another finding in this paper is that the four factors model explains more than Jorion’s model so that it can be more widely used in the future study.
The limitation is that sample companies exists for different period in the stock return so sample companies may be affected by other external factors (like financial crisis) differently which might lead to an inaccuracy of the result but in this case. However, there is no weekly data could be found for this research so the data could only be chosen in this way.
17 Reference
Adler, M., & Dumas, B. (1984). Exposure to currency risk: definition and measurement. Financial management, 41-50.
Amihud, Y. (1994). Exchange rates and the valuation of equity shares. Exchange rates and
corporate performance, 11, 49-59.
Anhar, M. (2015). Indicators of Company Performance, Investors’ Expectation and Investment Risk in Predicting Stock Return at Indonesia Stock Exchange. Indicators, 5.
Anwaar, M. (2016). Impact of Firms Performance on Stock Returns (Evidence from Listed Companies of FTSE-100 Index London, UK). Global Journal of Management And Business
Research.
Bartov, E., & Bodnar, G. M. (1994). Firm valuation, earnings expectations, and the exchange‐ rate exposure effect. The journal of Finance, 49(5), 1755-1785.
Bodnar, G. M., & Gentry, W. M. (1993). Exchange rate exposure and industry characteristics: evidence from Canada, Japan, and the USA. Journal of international Money and
Finance, 12(1), 29-45.
Bosiu, T., Nhundu, N., Paelo, A., Thosago, M. O., & Vilakazi, T. (2017). Growth and Strategies of Large and Leading Firms-Top 50 Firms on the Johannesburg Stock Exchange.
Dumas, B. (1978). The theory of the trading firm revisited. The Journal of Finance, 33(3), 1019-1030.
Du, D., Hu, O., & Wu, H. (2014). Emerging market currency exposure: Taiwan. Journal of
18
Francis, B. B., Hasan, I., & Hunter, D. M. (2008). Can hedging tell the full story? Reconciling differences in United States aggregate-and industry-level exchange rate risk
premium. Journal of Financial economics, 90(2), 169-196.
Gendreau, B. (1994). Comments on exchange rates, the macroeconomic environment, and the firm. Exchange Rates and Corporate Performance, 67-71.
Guthrie, J., Petty, R., & Ricceri, F. (2006). The voluntary reporting of intellectual capital: Comparing evidence from Hong Kong and Australia. Journal of Intellectual Capital, 7(2), 254-271.
Liu, X., & Shu, C. (2003). Determinants of export performance: evidence from Chinese industries. Economics of Planning, 36(1), 45-67.
Marston, C., & Polei, A. (2004). Corporate reporting on the Internet by German companies. International Journal of Accounting Information Systems, 5(3), 285-311.
He, J., & Ng, L. K. (1998). The foreign exchange exposure of Japanese multinational corporations. The Journal of Finance, 53(2), 733-753.
Hodder, J. E. (1982). Exposure to exchange-rate movements. Journal of international
Economics, 13(3-4), 375-386.
Hutson, E., & Laing, E. (2014). Foreign exchange exposure and multinationality. Journal of
Banking & Finance, 43, 97-113.
International trade. (2018). Retrieved from https://www.britannica.com/topic/international-trade
19
Jorion, P. (1990). The exchange-rate exposure of US multinationals. Journal of business, 331-345.
KAUSHAL, K., & SINGH, S. (2017). USING ANT LION ALGORITHM FOR PORTFOLIO
OPTIMIZATION ON THE BASIS OF TIME PERIOD OF INVESTMENT. CLEAR International Journal
of Research in Commerce & Management, 8(9).
Lymer, A., & Tallberg, A. (1997, April). Corporate reporting and the Internet–a survey and commentary on the use of the WWW in corporate reporting in the UK and Finland. In 20th
Annual Congress of the European Accounting Association, Graz, Austria (pp. 101-110).
Omar, A. B., Mohammad, K. N. B. T., & Ahmad, N. B. (2017). Exposure to foreign exchange rate risk: A review of empirical evidences. Journal of Insurance and Financial
Management, 2(5).
Suzuki, Y. (1980). The strategy and structure of top 100 Japanese industrial enterprises 1950–1970. Strategic Management Journal, 1(3), 265-291.
20 Appendix
Listing of Sample Companies Clothing Industry:
CATO CORP NEW FACTORY 2 U INC
PHILLIPS VAN HEUSEN CORP P V H CORP
GENESCO INC
CHARMING SHOPPES INC CINTAS CORP
STRIDE RITE CORP DRESS BARN INC
ASCENA RETAIL GROUP INC WOLVERINE WORLD WIDE INC V F CORP
KELLWOOD COMPANY NIKE INC
GAP INC LIMITED INC
BURLINGTON COAT FACTORY WRHS COR CLAIRES STORES INC
QUIKSILVER INC
JONES APPAREL GROUP INC ANNTAYLOR STORES CORP BRAUNS FASHIONS CORP MENS WEARHOUSE INC BUCKLE INC
21 CHICOS FAS INC
GYMBOREE CORP
PACIFIC SUNWEAR OF CA INC TALBOTS INC
COLE KENNETH PRODUCTIONS INC PAYLESS SHOESOURCE INC
ABERCROMBIE & FITCH CO WILSONS LEATHER EXPERTS POLO RALPH LAUREN CORP CHILDRENS PLACE RTL STORES INC GILDAN ACTIVEWEAR INC
BEBE STORES INC TOO INC
CHARLOTTE RUSSE HOLDING INC AEROPOSTALE INC
WARNACO GROUP INC NEW YORK & CO INC VOLCOM INC
TRUE RELIGION APPAREL INC UNDER ARMOUR INC
CROCS INC
J CREW GROUP INC
REEBOK INTERNATIONAL LTD ROSS STORES INC
22 Vehicle Industry:
ASTEC INDUSTRIES INC DANA CORP
GENERAL MOTORS CORP
NAVISTAR INTERNATIONAL CORP BRIGGS & STRATTON CORP WESTVACO CORP
CASCADE CORP CLARCOR INC
FORD MOTOR CO DEL GENTEX CORP
FLEETWOOD ENTERPRISES INC GENUINE PARTS CO
FEDERAL SIGNAL CORP WINNEBAGO INDUSTRIES INC MANITOWOC CO INC
SUPERIOR INDUSTRIES INTL INC MODINE MANUFACTURING CO COACHMEN INDUSTRIES INC TEREX CORP NEW
BALDOR ELECTRIC CO REGAL BELOIT CORP PACCAR INC
HARLEY DAVIDSON INC HARMAN INTL INDS INC NEW THOR INDUSTRIES INC
INSURANCE AUTO AUCTIONS INC WABASH NATIONAL CORP
23 BORG WARNER AUTOMOTIVE INC
MONACO COACH CORP COPART INC
LEAR CORP
TOWER AUTOMOTIVE INC TESMA INTERNATIONAL INC COLONELS INTERNATIONAL INC
ROCKWELL INTERNATIONAL CORP NEW AUTOLIV INC
AMETEK INC NEW
MERITOR AUTOMOTIVE INC STONERIDGE INC
MIDAS INC
AMERICAN AXLE & MFG HLGDS INC DELPHI AUTOMOTIVE SYSTEMS CORP VISTEON CORP
HAYES LEMMERZ INTERNATIONAL INC LAIDLAW INTERNATIONAL INC
T R W AUTOMOTIVE HOLDINGS CORP COMMERCIAL VEHICLE GROUP INC GENTEK INC
