The effect of board composition on subsidiary performance
Differences between developed and emerging markets
By Marit Veldkamp
University of Groningen Faculty of Economics and Business Master Thesis IB&M (EBM719A20)
January 6th 2016
Abstract
Many companies look for opportunities abroad and currently many of them are looking into emerging markets. In this research we looked at the board characteristics; average age, board size and gender composition and their effect on subsidiary performance. We made a
distinction between emerging and developed markets and used it as an interaction effect. The sample included 126 subsidiaries (57 from emerging markets and 69 from developed markets) from companies listed in the Fortune 500. The data (for the year 2013) is collected by using Orbis, annual reports and proxy statements of the companies. We found that the ratio of women in the board, the average age and the size do not have an effect on subsidiary performance. Moreover we found no support for the interaction effect of emerging and developed markets. What we do found is that the type of market has a direct effect on subsidiary performance. The subsidiary performance is better in a developed market than in an emerging market. We give interesting practical and theoretical implications, as well as future research directions.
Keywords: average age, board size, gender composition, subsidiary performance, emerging
market, developed market
Research theme: The effect of board composition on subsidiary performance, differences
TABLE OF CONTENT
1. INTRODUCTION ... 3 2. LITERATURE REVIEW ... 8 2.1 Subsidiary performance ... 8 2.2 Board of directors ... 9 2.3 Board characteristics ... 10 2.3.1 Gender composition ... 10 2.3.2 Board size ... 12 2.3.3 Age diversity ... 132.4 Emerging markets and developed markets ... 14
2.5 Conceptual model ... 16
3 RESEARCH METHODOLOGY ... 17
3.1 Sample ... 17
3.2 Measurements of variables and data ... 18
3.2.1 Dependent variable ... 18
3.2.2 Independent variables ... 19
3.2.3 Moderator variable ... 19
3.2.4 Control variables ... 20
3.3 Methods ... 22
4 EMPERICAL RESULTS AND ANALYSES ... 24
4.1 Descriptive statistics ... 24
4.2 Pearson Correlation ... 24
4.3 Multiple regression analysis ... 27
5 DISCUSSION ... 32
5.1 Theoretical implications ... 32
5.2 Practical implications ... 34
5.3 Strengths, limitations and future research ... 35
6 CONCLUSION ... 37
ACKNOWLEDGEMENTS ... 38
1. INTRODUCTION
Many large multinationals have their subsidiaries in different countries. And still, more and more companies want to expand abroad. We could think that the main reason is that there is a market to increase their sales. So for many firms subsidiary performance becomes very important and is it part of their total revenue. A recent article from CNTV (2015) shows that a lot of companies like Apple and Amazon benefit a lot from their subsidiaries by using transfer pricing methods. This indicates that companies want to expand abroad and give their sales a boost. Some firms choose a developed country to establish their subsidiary and a lot of them are looking for emerging markets to establish their subsidiary. But what is the effect on performance?
The interesting part is what the differences are between subsidiary performance and firm performance. It is interesting because, the subsidiary is established in a foreign country and therefore they may cope with different cultures, rules and regulations which may have an influence on their performance. A lot of articles like Guest (2009) focused on firm
performance. Firm performance can be defined as the total of the home country and subsidiary results. According to Guest (2009), firm performance is measured according to three different types of measures; the profitability of a firm, Tobins Q and share returns. However, as many companies expand abroad, in particular subsidiary performance becomes an important topic in the current literature. Subsidiary performance is more specific for a certain subsidiary and may have other antecedents that influence it, then normal firm performance. Subsidiaries in foreign countries may cope with different institutions, which may influence their performance. Subsidiary performance includes the same measures as firm performance but is focused on the subsidiary and not on the company as a whole. Subsidiary performance shows how a subsidiary performs under control of the managers of the
One of the antecedents of performance is researched by Guest (2009), who examined for the UK the effect of board size on firm performance. They found that board size have a negative effect on firm performance. The board of directors is an important mechanism in a company. In many cases is the board formed with a mix of executives and non-executives. Executives mean that these members are also manager in the same company and have therefore insider information. Non-executives are independent and are often part of another company (De Andres, Azofra & Lopez, 2005). In this thesis we do not focus on top
management teams (TMT) as these focus more on the execution of the tasks (Williamson, 1984) and they do not take the major decisions in companies. We focus on the board of directors (tier 1), because the board of directors influences the major decisions taken by the manager when ownership from control is separated (Srinidhi, Gul & Tsui, 2011). The board of directors may also decide where and how to establish a subsidiary, and therefore we focus on the board of directors. Furthermore, if the performance is bad, the board of directors could intervene in what has to be done next. This could indicate that it also influences subsidiary performance, as major decisions are made by the board of directors. The board is an
instrument for internal governance to ensure effective mechanisms of monitoring, controlling and advising (Park & Shin, 2004) and the board is used by shareholders to monitor the top managers (Jensen, 1993). Furthermore, the board of directors has to deal with the incentives of the TMT. Moreover, the board of directors’ reduces the potential of conflicts (agency problems) between the TMT and their shareholders (Vance, 1983; Mace, 1971; Kosnik, 1987). All in all, clearly the board supervises and could influence major decisions. As such, the board of directors is likely to also influence subsidiary performance, for example where to establish a subsidiary.
But what board characteristics do influence subsidiary performance? Bianco,
Ciavarella and Signoretti (2013) found that several characteristics of the board influence the ability to give advice and monitor and as such might have an influence on subsidiary
composition and size (Gavrea & Stegerean, 2012; Guest, 2009; Rose; 2007) and therefore we focus on these three antecedents of the board. We choose these characteristics, because they are the major pillars and we have limited time to execute this research.
Research has shown that a higher average age of the board has a negative effect on performance. This is because the board should be innovative and vigilant (Bianco et al. 2013). This means that younger people are more innovative and take more risks. The report of Norby mentioned that a director should withdraw from the board of directors at the age of 70,
because they become too old (Rose, 2005).
Not only age has an influence on performance, but also the board size does influence the performance. Lipton and Lorsch (1992) stated that smaller boards are more effective than larger boards, because of the free riding of the director and coordination problems that occur in larger boards. These problems could have an influence on subsidiary performance, as the process runs not very smoothly. However, larger boards may contain more information as it includes more people.
Also board composition and then especially the ratio of women in a board is an interesting topic in academic literature. Currently more women are on top and in the board of large multinationals. A woman in the board of directors occurs more and more and sometimes it is used as a token by companies (Kanter, 1977). They do this by showing they are acting in a good way and give women a chance. In the past rarely women were present in the board of directors of a company. The question here is; does the presence of women influence the subsidiary performance? Sabatier (2015) found that more diverse boards (with respect to gender) have a positive effect on performance. They showed that more diverse boards should reduce corporate inefficiencies and therefore enable companies to reach their optimal
performance. They could not specify what the underlying mechanism was. As an explanation they referred to the article of Hillman, Cannella and Paetzold (2000), which stated that diversity may introduce new skills in the board, which in turn could make decision making more effective. When decision making becomes more effective, it might have an influence on subsidiary performance.
role in the type of market. What we see at the moment is that emerging markets are booming. Emerging markets can be defined as economies with high growth potential/huge economic development and economies that are in favour of a free market (Arnold & Quelch, 1998). Of course have companies subsidiaries in developed markets which are according to the Oxford English Dictionary (2015) defined as technically and economically advanced markets, which could indicate it is less risky.
Taking all the information into account, we come up with the following research question:
What is the effect of the headquarters’ board size, gender composition and average board age on subsidiary performance and how is this relation influenced by the type of market
(developed versus emerging markets)?
Peng (2004) mentioned that a key issue in international business literature is what way (like strategy) will lead to international success. Subsidiary performance can end in
international success or failure, as expanding abroad is an important area in current research. We think this research makes an important contribution, because we want to determine, what is needed for a headquarters board of directors to let subsidiaries operate successfully in an emerging or developed market. This highlights the importance of our research question. Furthermore, stated by Khanna and Palepu (2013), it is still important to distinguish between developed countries (like USA and The Netherlands) and emerging markets (like the BRIC countries). It is important because companies have to cope with the different institutions in those countries.
The term firm performance is not specific enough and therefore we focus on subsidiary performance.
The remainder of this paper is constructed as follows. The next section explains more about this topic in the literature review and includes a conceptual model and hypotheses. This part is followed by the methodology and data. Then we continue with the results. The
2. LITERATURE REVIEW
In this section we expand the literature of the dependent variable: subsidiary
performance, the independent variables: board size, gender composition and average board age and our moderator: type of market. In the end the conceptual model is shown.
2.1 Subsidiary performance
Already many articles researched firm performance (Guest, 2009). They used various measures for firm performance; return on assets, return on investment and return on equity. Subsidiary performance can be seen as a subjective index namely: making profits, loss or break-even (Lo & Lin, 2015).
Chan, Makino and Isobe (2010) came with the results that subsidiary performance is influenced by variables on subsidiary level (14-17%) and by variables on parent level (19%). These percentages are followed by the industry variables and the percentages left, explains other variables which may influence subsidiary performance. Therefore subsidiary
performance cannot be exchanged for normal firm performance, as these results from Chan et al (2010) showed in their research that the variables on parent level does influence the
subsidiaries’ performance.
Subsidiary performance is important on the level of the corporations as it can be a mechanism for co-ordinating and integrating in a certain foreign country (Chang & Taylor, 1999). The corporations have a certain degree of control over the subsidiary, to maintain they are acting in the corporations’ interest. When a country has good institutions, this is related to better performance of the subsidiary. Good institutions may facilitate the process of
transferring your advantages (from headquarter) to the subsidiary (Gugler, Mueller, Peev & Segalla, 2013). Research has shown that the advantage the company has, influences the performance of the subsidiary (Isobe, Makino & Montgomery, 2000; Tallman 1992). Therefore plays institutional distance also a role in international business. Institutional distance can be defined as the difference (relative) between the institutions in the host and home countries (Gaur & Lu, 2007). The study of Pattnaik, Chloe and Singh (2015)
highlighted that the institutional distance between the host and home country is an important factor that has an influence on subsidiary performance. They especially mention the
markets. Moreover they argue that this relative difference is more important than the absolute quality of these host countries institutions. When the company faces a monopolistic resource this depends also on the institutions in the host countries. Each country has their own
regulations and the government plays an important role, if it is willing to help or not. We can think of the labour, product and capital markets that influences subsidiary performance (Pattnaik et al. 2015). The resource based view states that the company benefits from its internal resources and therefore influences the performance of a firm. The company should have an advantage over their competitors to overcome the liability of foreignness; moreover the company needs to have monopolistic resources like for example a patent (Hymer, 1976). With the competitive advantage, the company can create market
opportunities in the host country (Tallman, 1992). People are a valuable resource for a firm and influences performance, as researched by Pattnaik et al. (2015). Human capital brings knowledge, talents and skills related to individuals (Coff, 1999). In this thesis we look at the board of directors and their influence on subsidiary performance. The board of directors consists also of a range of people. This can imply that the board of directors is a valuable resource of the company and therefore could influence subsidiary performance.
2.2 Board of directors
The board of directors can be interpreted differently. There are two types of boards; namely tier one and tier two boards of directors. The United Kingdom is a perfect example of a tier one board. This means that the company has one board with executives and
non-executives. A two tier board comprises of a dual system; a supervisory board and a
management board (Jungmann, 2006). So it really depends on the country what kind of board of directors a company has. As already mentioned before executives are defined as members of the board that are also managers in the same company and non-executives mean they are independent and do not have another function in that particular company (De Andres et al. 2005).
focus on a one tier board, as this is based on our sample, which is subtracted from US companies. US companies have a tier one board.
The board of directors deals with organization-wide decisions that may have
implications for organizations’ long term performance (Judge & Zeithaml, 1992). This means also that the board can have an influence on subsidiary performance. The board safeguards and represents the dispersed shareholders interest (Becht, Bolton & Roell, 2003), mitigate self-serving behaviour and for that reason enhances the value of the firm by decreasing agency costs (Shleifer & Vishny, 1997).
Various articles did research to the antecedents of the board of directors and their effect on performance. Dharmadasa, Gamage and Herath (2014) used the board
characteristics; board composition (number of men/women), board size (number total), board diversity (backgrounds) and board independence in their research. Other researchers like for example Chen (2015) used board size and Rose (2007) looked what the effect of female representation in a board of directors was. Gavrea et al. (2012) looked at the antecedents age and gender and their influence on firm performance. To really get an understanding on how these individual characteristics affect the subsidiary performance of companies in developed and emerging markets, we explain three of them in more details. We decided to use the gender composition, age diversity and board size, as these are, from our perspective, the main variables to explain the board characteristics.
2.3 Board characteristics 2.3.1 Gender composition
Gender composition is in this research defined as the ratio of women in a board. In October 2014, the percentage of women in the board of the largest listed firms in Europe was only 20.2% (European Commission, 2015). The number of women represented as a CEO in the Fortune 500 is in 2014 just 24, but this number has increased in the last couple of years. Mary Barra is the highest ranked women at position seven as CEO of General Motors (Fortune 2014). However, what is the effect of women in the board?
Huse, Nielsen and Hagen (2009) found out that gender diversity brings more creative discussions, then when there were just men in the board of directors for example.
innovation, creativity and may help to solve problems (Daily & Dalton, 2003). Nielsen and Huse (2010) mentioned that women need (to be successful in the board and really change something), as a prerequisite, traditional professional experience, because if they do not have this experience they have less influence in the decision-making process. If women have less influence in the decision making process, they could have less influence on the subsidiary performance.
Sabatier (2015) mentioned that gender diversity is related to the characteristics of the firm and also to other connecting boards. Mostly, when a woman is appointed in the board, it is part of the long term strategy of a company. The author also found that a more diverse board achieves better performance. Besides, appointing women has a positive effect on diversity goals, corporate social responsibility and on economic growth. The increase in performance is also confirmed by the article of Adams and Ferreira (2009). They found more women in the board would increase performance. The article of Rhoades, Rechner and
Sundaramurthy (2003), found there is a little positive relationship between board composition and financial performance. The authors Gavrea et al. (2012) found different results; there is no relationship between gender and firm performance (measured in return on assets). Unfortunately they gave no clear argument why there is no relationship. The authors Matsa and Miller (2011) found there is a negative relationship between women in the board and performance. They did research to Norwegian (developed market) public companies which included women in their board and they showed including women have had a negative effect on the profitability of these Norwegian companies. They stated that when men are replaced by women they may be less experienced. Moreover in their sample, women were younger than men and this may explain why they are less experienced. Furthermore they used as the main argument that the labour costs increased because of higher employment (relatively) and fewer layoffs. Clearly, overall research has shown that the results with respect to board composition and its influence on performance are not straightforward.
the board of directors. A more gender diverse board may have also different ideas and brings more creative discussions and therefore they might create better ideas to the where and how to open a subsidiary. In fact, men and women are different. Furthermore found Silverman (2003) that women are more long-term oriented. To open a subsidiary is also a long term goal to achieve superior performance. Lastly it is part of CSR which shows a better image of the company. This created image could therefore have a positive effect on the performance of the subsidiary. We therefore come up with the following hypothesis:
H1 There is a positive relation between the ratio of women in the headquarters’ board and subsidiary performance
2.3.2 Board size
The board size is defined as the total number of people that is active in the board. The article of Jensen (1993) stated that the board is less effective when it increases in size. The reason is that it slows down the decision making process. Furthermore, board size can have an influence on the conflict within a team (Amason & Sapienza, 1997). Jensen (1993) showed a smaller board is more effective, as the CEO cannot function well if the board has more than seven members. Yermack (1996) stated that smaller boards take more risky managerial decisions, but also more value increasing decisions. The author found a negative effect between board size and firm performance by using US listed firms. The article of Chatterjee (2011) found also that the board size has a negative effect on performance in India. Because if the board increases, some costs are increasing as well. For example there will be a
communication gap and the coordination costs will increase.
However, the article of Golden and Zajac (2001) noticed that a smaller board has less confidence and has problems with making strategic changes. The article of Haniffa and Hudaib (2006) found that larger board provide more expertise, experience and diverse contracts which are all needed to increase performance. This additional experience and
contracts may be needed to establish a successful subsidiary abroad. This is also confirmed by Fama and Jensen (1983) which argued that complex organisations are prone to have a larger board, as these boards provide them with more information. To conclude, research shows contradictive results with regard to the relation between board size and performance.
boards have more information, knowledge and ideas. Moreover, they will all have different backgrounds which could help to make the right decision. Furthermore a larger board has more experience, and more experience could also lead to better decisions where to establish this subsidiary and how. So we argue that a larger board is associated with better subsidiary performance. We come up with the following hypothesis:
H2 There is a positive effect between the headquarters’ board size and subsidiary performance
2.3.3 Age diversity
Age highlights the experience someone has. We assume that older people are more experienced and have more knowledge than younger people. The literature about what the influence of the average age is in the board is scarce. But for example Bonn, Toru and Phillip (2004) found that a lower average age of the board has a significant positive effect on the market value of a firm, especially when it is compared to the book value. Botwinick (1977) and Burke and Light (1981) found that the cognitive abilities of older people decrease. We may think of their ability to learn, the way of reasoning and their memory. Hafsi and Turgut (2013) found also a negative relationship between corporate social performance and board age diversity. The article of Rose (2005) did the research for Danish companies. The author found that a higher average age has a negative effect on financial performance, as the members are less innovative when they are older.
Establishing a subsidiary abroad is a risk. You need to cope with the institutions in that particular country. Wholly owned subsidiaries are more risky than for example a joint
venture. Everything needs to be built from scratch. Younger people are more innovative and take more risk, so they are more prone to open a subsidiary with perhaps more innovative ideas. Furthermore, is innovation required, to become successful. The ability to learn is much higher under younger people. It may be that they are also more experienced with
internationalization which would also help to successfully establish a subsidiary abroad. So we think that taking risk and innovation is more important than the experience someone has. We therefore come up with the following hypothesis:
2.4 Emerging markets and developed markets
The moderator in this research is the type of market; emerging markets and developed markets. Emerging markets can be described as markets that use economic liberalization as their major growth pillar, characterized by high growth potential and low-income (Hoskisson, Eden, Lau & Wright, 2000). Several characteristics of emerging markets are interesting like; globalization, growth and liberalization. Other characteristics are poverty (low/middle income) and low market capitalization. The emerging markets have cheaper human capital and fewer resources as opposed to developed markets. For the whole community this could be a source or opportunity. Consumers can buy their products cheaper and for large
multinationals it is interesting because there are fewer competitors especially when the home market is saturated (Khanna et al. 2013). This can be seen as an opportunity for many
companies. The difference between emerging and developed markets is that in emerging markets the company has mainly a dominant shareholder which is the state or family. In that sense the board may require different requirements, as the institutions are weaker and the state is dominant. Chan et al (2010) found that the regional effects are stronger in an emerging market than in a developed market. Furthermore they argue that emerging markets are ethnically and culturally diverse.
According to the Oxford English Dictionary (2015) a developed country, means that these countries are technically and economically advanced and are also known as
industrialized nations. Developed markets face strong institutions. Chen (2015) classified institutions according to if the government gives a “helping hand” and contract enforcement. Already many articles found that institutions play an important role in economic growth. A subsidiary from a developed country faces technological advantages over the local firms in an emerging economy, but also faces high liabilities, because there is uncertainty (Li, Lin & Arya, 2008).
When we looked at the subsidiaries we thought it would be interesting to see if it matters if the subsidiary is located in an emerging or developed market. We aim for an interaction effect, as we think that board characteristics may have a different effect in
particular type of market (emerging or developed). This is based on the fact that there is some institutional distance between markets, which may have an influence on subsidiary
institutions. The author stated that when a country has good institutions, the subsidiary
performance will be better. Good means that when for example a patent is protected. So better institutions in the country where the subsidiary is located means better subsidiary
performance. For that reason we think that the effects of the above mentioned hypotheses will be more positive for a developed market as opposed to an emerging market.
We think that more women in a board would have a larger positive effect on the performance of a subsidiary in a developed market than in an emerging market. As mentioned above, play institutions an important role. It could be that the attitude towards women in developed markets institutions is better than in emerging markets. The reason could be that women are more accepted by developed markets institutions. In developed markets it is more common that women work and do not only take care of their children. Hewlett and Rashid (2010) stated that women in emerging markets face unique challenges. This may indicate that emerging markets institutions would not easily accept women. Furthermore, in emerging markets there is some kind of gender-bias. This all could imply that the roles in emerging markets are more traditional than in developed markets. Therefore we think that women could have more influence in developed markets. We therefore come up with the following
hypothesis:
H4 There is a positive relationship between the ratio of women in the board and subsidiary performance and this effect is stronger in developed markets as opposed to emerging markets
An emerging market has weaker institutions, so it is more risky to have a subsidiary in that country. When we have a larger board we have more experience, ideas and information and they can share all their information to perform well. We think that this information is needed, to establish successfully a subsidiary in an emerging market (and less needed in a developed market, as the risk is lower). Companies are dependent on the institutions in that country and they need to overcome the liability of foreignness or have a competitive advantage that is transferable. We think that the effect of board size on subsidiary
performance is stronger in an emerging market, as these markets face weaker institutions and are therefore more risky. We come up with the following hypothesis:
Older people take less risk and are less innovative. To open a subsidiary in an emerging market you need to take risk and you need to be innovative. This is important because emerging markets have weaker institutions. This means that they are less helpful and the rules for example are perhaps more focused on companies from their home country. So when someone is older the effect on subsidiary performance in an emerging market may be smaller. Furthermore, it could be that such a person will either open a subsidiary in a developed market. We therefore come up with the following hypothesis:
H6 There is a negative relationship between the average age of the headquarters board and subsidiary performance and this effect is stronger in emerging markets as opposed to developed markets.
2.5 Conceptual model
To summarize all the information, the conceptual model is shown in figure 1.
The effect of board characteristics on subsidiary performance
3 RESEARCH METHODOLOGY
This part contains all the information regarding our sample, measurement of the variables and an explanation how we want to test our hypotheses.
3.1 Sample
We took the Fortune 500 as our source of companies. This list contains the largest US-based companies. Fortune 500 was chosen US-based on data availability/accessibility. The US requires companies to give information about their nominees and if they considered diversity (US Securities & Exchange Commission, 2012). We started by doing research for the first 114 companies of the list which should result ideally in sample of 228 subsidiaries (114 emerging and 114 developed). But especially Orbis showed that not all the data was available regarding the performance of their subsidiaries, or the company had no subsidiaries at all. To get about the same emerging and developed subsidiaries in the sample we ended with 57 subsidiaries located in an emerging market and 69 subsidiaries in a developed market. So in total we had a sample of 126 companies with the performance in the year 2013. 54
multinationals had a subsidiary in an emerging and in a developed market. 3 multinationals had their subsidiary in just an emerging market and 15 multinationals had their subsidiary in just a developed market. With these decisions we follow the article of Ararat et al. (2015) which took also a sample of about 100 companies, and also took for the firm performance one year. This article measured the effect of board diversity on firm performance in emerging markets. Thus this can be seen as similar to our research.
The sample needed to satisfy some criteria. First of all the data needed to be available online or in a database, as contacting the firm was not possible, due to time limitations. Secondly the firms needed to be headquartered in a developed country. In sum we made the decision that the companies must be headquartered in the same country. The choice for a developed country is based on the assumption, that these companies have this data available and most large companies are headquartered in a developed country. Moreover it is a
always available for 2014, we decided to use 2013 as our measurement year. Initially the last requirement was that the company has a subsidiary in an emerging market and also in a developed market, to make the sample more comparable.
However, when collecting the data, it came across that the last requirement did not apply for many companies. As we had to deal with time limitations we decided to include also the companies who had either a subsidiary in an emerging market or a developed market. In total we had 54 companies with an emerging and developed market subsidiary. Moreover there were 3 with just an emerging market subsidiary and 15 with a developed market subsidiary. Finally we did not select any subsidiaries from the United States, as these
subsidiaries need to be located in the host country. The home country is already in het US, so subsidiaries in the US are not included in the sample.
3.2 Measurements of variables and data 3.2.1 Dependent variable
We looked for useful measurements of subsidiary performance. Nguyen (2011) stated that there is no agreement on how to measure subsidiary performance. A frequently used measure is that of Lo et al. (2015), which used different categories for measuring the performance of a subsidiary, namely; incurring loss, break-even or making profits. As our sample shows only positive amounts, this is not a useful way to measure it and we would rather choose an interval measurement. Gugler et al. (2013) used the measure profit after tax divided by the total assets. They mark here that it is often not a reliable measure as
accountants manipulate those figures. Moreover the article of Chang, Gong and Peng (2012) used the measures return on investment and return on equity as subsidiary performance measures.
3.2.2 Independent variables
Following the article of Jhunjhunwala and Mishra (2012) we use the total number of women divided by the board size as the measure to explain the ratio of women in the board. We divided it also by board size, as not every board has the same size. So we looked for the share women have in the board. We measured for every company how many women are active in the board during the year 2013. What is shown is that women represent just a very small part of the total number of the board of directors, as measured according to the tier 1 structure in the United States.
The board size will be measured according to the total number of directors. Yermack (1996) used a log because the author has shown that the relation between performance and board is not linear but convex and thus they had to adjust for skewness. However, because the number of directors does not vary that much, we do not use the log in this research, but use the total number of directors.
To measure the average board age we follow the article of Cochran, Wartick and Wood (1984). They took the mean age of the age of the board of directors of each firm. In our research we do the same. We took from every director the age for 2013 and divided it by the number of directors, to get the average age.
The data about the characteristics of the board was found in the annual reports (of 2013) of all the companies listed in the Fortune 500. As this sample consists of large
companies, all companies had an annual report available on their website. In many cases, the names and a picture of the board was reflected in the annual reports of these companies. With respect to the age of the board members, additional information was found in the proxy statements (of 2013) of those companies. When this information was not available here, Bloomberg.com was an important source to complete the missing data. Bloomberg is a source which contains information about companies, board structures and other information.
3.2.3 Moderator variable
markets subsidiaries in our sample we chose for countries like Brazil, Chile, India, China, Hong Kong, Romania, Nigeria and Israel. Developed countries are in our sample countries like: US, European countries, Australia, Japan and Canada. For all the countries we decided if it is a western country or not. The country of the subsidiaries was listed in Orbis.
Subsequently we created a dummy variable to distinguish between subsidiaries that have their physical presence in an emerging market or a developed market. In this case 0 would be developed market and 1 would be emerging market. Bahadir, Bharadwaj & Srivastava (2015) made a similar distinction between developed and emerging markets. Furthermore they used about the same countries in their sample.
3.2.4 Control variables
The control variables that are used are: MNC age, subsidiary and MNC size (Chang et al. 2012). Another control variable that is often used in international business articles when measuring performance is type of industry.
MNC age is measured according to the article of Chang et al. (2012). Age is defined as the total of years of operation of the company. It is important because it is mentioned that when a firm ages they become more complex, also in management. Newer organisations have less formalized structures and therefore may benefit more from for example creativity and innovation. For that reason we want to control for MNC age.
Subsidiary size is the total number of employees in the subsidiary. This is also used as a control in the article of Gupta and Govindarajan (2000) and Chang et al (2012). Size is a frequently used control variable in international business research. Size may have an influence on the results as normally a larger company achieves larger revenues for example.
The MNC size will be measured according to the log number of employees. Lo et al. (2015) used the same control variable in their research. It is already stated that there may be some correlation between board size and firm size (Chatterjee, 2011). Log is not necessary in our case and therefore we use the normal number of employees.
industries. To control for industry is important, because there may be different levels of interaction between customers and employees, and also among employees (Ali, Kulik & Metz, 2011; Godthelp & Glunk, 2003). Those levels of interaction differ per industry and could therefore influence performance.
The last control variable that is used in this research is MNC performance. As these differ per firm we want to overcome these problems that MNC performance will have an influence on the results, especially when it comes to the difference between larger and smaller companies. MNC performance will be measured according to the total revenue of the MNC. Table 1 shows all variables that are used.
The data for most control variables is subtracted from Orbis. Orbis shows the figures for the control variables, like firm age, the industry of the company and subsidiary size. It also completed the gap when the figures for firm performance and firm size could not be found in the annual reports of the companies.
Overview main variables
Variable Proxy
Dependent variable
Subsidiary performance Log operating revenue
Independent variables
Ratio of women board Number of women divided by board size
Board size Number of directors in board
Average board age All ages divided by the board size
Moderator variable
Type of market 0= developed market 1= emerging market
Control variables
MNC age Total years of cooperation
Subsidiary size Number of employees subsidiary
MNC size Number of employees MNC
MNC industry 0= other 1= Industrial
MNC performance Total revenue
3.3 Methods
For the first three hypotheses we use a simple linear regression analysis. The basis for testing these hypotheses will be 𝒀 = 𝜶 + 𝜷 𝑿 + 𝜺
Hypothesis 1: 𝑺𝒖𝒃𝒔𝒊𝒅𝒊𝒂𝒓𝒚 𝒑𝒆𝒓𝒇𝒐𝒓𝒎𝒂𝒏𝒄𝒆 = 𝜶 + 𝜷𝟏 𝒓𝒂𝒕𝒊𝒐 𝒐𝒇 𝒘𝒐𝒎𝒆𝒏 + 𝜷𝟐 𝑺𝒖𝒃𝒔𝒊𝒅𝒊𝒂𝒓𝒚 𝒔𝒊𝒛𝒆 + 𝜷𝟑 𝑴𝑵𝑪 𝒂𝒈𝒆 + 𝜷𝟒𝑴𝑵𝑪 𝒔𝒊𝒛𝒆 + 𝜷𝟓 𝑴𝑵𝑪 𝒊𝒏𝒅𝒖𝒔𝒕𝒓𝒚 + 𝜷𝟔 𝑴𝑵𝑪 𝒑𝒆𝒓𝒇𝒐𝒓𝒎𝒂𝒏𝒄𝒆 + 𝜺 Hypothesis 2: 𝑺𝒖𝒃𝒔𝒊𝒅𝒊𝒂𝒓𝒚 𝒑𝒆𝒓𝒇𝒐𝒓𝒎𝒂𝒏𝒄𝒆 = 𝜶 + 𝜷𝟏𝒃𝒐𝒂𝒓𝒅 𝒔𝒊𝒛𝒆 + 𝜷𝟐 𝑺𝒖𝒃𝒔𝒊𝒅𝒊𝒂𝒓𝒚 𝒔𝒊𝒛𝒆 + 𝜷𝟑 𝑴𝑵𝑪 𝒂𝒈𝒆 + 𝜷𝟒𝑴𝑵𝑪 𝒔𝒊𝒛𝒆 + 𝜷𝟓 𝑴𝑵𝑪 𝒊𝒏𝒅𝒖𝒔𝒕𝒓𝒚 + 𝜷𝟔𝑴𝑵𝑪 𝒑𝒆𝒓𝒇𝒐𝒓𝒎𝒂𝒏𝒄𝒆 + 𝜺 Hypothesis 3: 𝑺𝒖𝒃𝒔𝒊𝒅𝒊𝒂𝒓𝒚 𝒑𝒆𝒓𝒇𝒐𝒓𝒎𝒂𝒏𝒄𝒆 = 𝜶 + 𝜷𝟏 𝒂𝒗𝒆𝒓𝒂𝒈𝒆 𝒃𝒐𝒂𝒓𝒅 𝒂𝒈𝒆 + 𝜷𝟐 𝑺𝒖𝒃𝒔𝒊𝒅𝒊𝒂𝒓𝒚 𝒔𝒊𝒛𝒆 + 𝜷𝟑 𝑴𝑵𝑪 𝒂𝒈𝒆 + 𝜷𝟒𝑴𝑵𝑪 𝒔𝒊𝒛𝒆 + 𝜷𝟓 𝑴𝑵𝑪 𝒊𝒏𝒅𝒖𝒔𝒕𝒓𝒚 + 𝜷𝟔𝑴𝑵𝑪 𝒑𝒆𝒓𝒇𝒐𝒓𝒎𝒂𝒏𝒄𝒆 + 𝜺
The basis for testing the hypotheses with the moderator will be 𝒀 = 𝜷𝟎+ 𝜷𝟏𝑿 +
𝜷𝟐𝒁 + 𝜷𝟑𝑿𝒁 + 𝜺 where Z will be the developed market or emerging market and 𝜺 is the residual. Interaction effects are important, because when you have only X and Y, there is a linear relationship, but this is different when Z exists. 𝜷𝟑 is the predictor x moderator and is
4 EMPERICAL RESULTS AND ANALYSES
In this part of the study we test our hypotheses. We started with a Pearson correlation test. Furthermore we did a multiple regression analysis, and a multiple regression with the interaction effect; the type of market.
4.1 Descriptive statistics
To get an overview of the main variables, we start with the descriptive statistics in table 2. What we see is that there are some missing variables especially in the total of employees of the MNC. Moreover we can see that about 20% of the board members are women. The average board size in our sample is 12.17. Besides is the average age of the board defined as about 62.56. Another interesting aspect is that the average of the MNC is quite old, namely 99.89. For subsidiary performance we took the log, because there are great variations for subsidiary performance in our sample.
Descriptive statistics
Variable Mean St. Deviation N
Ratio of women .20 .08 126
Board size 12.17 2.21 126
Average board age 62.56 2.67 126
Log subsidiary performance 19.00 2.63 126 Subsidiary size 3667.34 18054.10 125 MNC performance 79799158730.16 7.3476940299.33 126 MNC size 152248 281713.25 122 MNC age 99.89 52.19 126 MNC Industry .82 .39 126 Type of market .45 .50 126
Table 2 overview variables
4.2 Pearson Correlation
a negative correlation. For example in table 3 we see that the ratio of women and average board age have a negative correlation (r = -.19). This means an increase in the ratio of women in the board will lower the average board age. The first number in table 3 is the Pearson’s correlation. The second number in the table is the significance level. The last number in the table is the number of observations. The results can be seen in table 3.
From the correlation table we can draw some conclusions. We see that the ratio of women has a significant positive relationship with MNC age (r = .28, p = <0.01). This means an increase in the ratio of women, the older the company is (the higher the MNC age).
Subsidiary performance has a significant positive relationship with MNC performance (r = .19, p = <0.05), subsidiary size (r = .32, p = <0.01) and industry ( r = .15, p = <0.05). Those results mean an increase in subsidiary performance means also an increase in MNC
performance and subsidiary size. Moreover an increase in subsidiary performance means that subsidiary performance is higher in industrial industries than in other industries.
Board size has significant positive correlations with subsidiary size (r = .22, p = <0.01), MNC performance (r = .30, p = <0.01), MNC size (r = .36, p = <0.01) and MNC age (r = .28, p = <0.01). This means than an increase in board size also increases, subsidiary size, MNC revenue, MNC size and MNC age. The table shows as well that subsidiary performance has a significant negative relationship with type of market (r = -.19, p = <0.05). This means that subsidiary performance is higher in a developed market than in an emerging market.
Pearson correlation (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (1) Ratio of women 1 (126) .08 .20 (126) -.19* .02 (126) -.05 .28 (126) .08 .20 (125) .05 .30 (126) .14 .07 (122) .28** .00 (126) -.06 .26 (126) -.02 .40 (126) (2) Board size .08 .20 (126) 1 (126) .01 .47 (126) .09 .15 (126) .22** .01 (125) .30** .00 (126) .36** .00 (122) .28** .00 (126) -.11 .10 (126) .03 .36 (126) (3) Average board age -.19* .02 (126) .01 .47 (126) 1 (126) -.01 .47 (126) -.07 .23 (125) .04 .35 (126) -.09 .16 (122) .02 .42 (126) -.13 .08 (126) -.02 .41 (126) (4) Log Subsidiary performance .07 .28 (126) .09 .15 (126) -.01 .47 (126) 1 (126) .33** .00 (125) .19* .01 (126) .06 .27 (122) -.02 .40 (126) .15* .05 (126) -.19* .02 (126) (5) Subsidiary size .08 .20 (125) .22** .01 (125) -.07 .23 (125) .32** .00 (125) 1 (125) .41** .00 (125) .57** .00 (121) -.05 .31 (125) .06 .25 (125) -.02 .42 (125) (6) MNC performance .05 .30 (126) .30** .00 (126) .04 .35 (126) .20* .01 (126) .41** .00 (125) 1 (126) .71** .00 (122) -.12 .08 (126) .10 .13 (126) -.05 .29 (126) (7) MNC size .14 .07 (122) .36** .00 (122) -.09 .16 (122) .06 .27 (122) .57** .00 (121) .71** .00 (122) 1 (122) -.12 .10 (122) .06 .27 (122) .03 .78 (126) (8) MNC age .28** .00 (126) .28** .00 (126) .02 .42 (126) -.02 .40 (126) -.05 .31 (125) -.12 .08 (126) -.12 .10 (122) 1 (126) -.12 .09 (126) .03 .39 (126) (9) MNC Industry -.06 .26 (126) -.11 .10 (126) -.13 .08 (126) .15* .05 (126) .06 .25 (125) .10 .13 (126) .06 .27 (122) -.12 .09 (126) 1 (126) -.07 .23 (126) (10) Type of market -.02 .40 (126) .03 .36 (126) -.02 .42 (126) -.19* .02 (126) -.02 .42 (125) -.05 .29 (126) .01 .44 (122) .03 .39 (126) -.07 .23 (126) 1 (126)
4.3 Multiple regression analysis
As we work with continuous data, we had to do a regression analysis to test our hypotheses H1, H2 and H3. In this section we looked at the outcome of our outcome variable (subsidiary performance) which is expected to be influenced by our independent variables. To get a clear overview of our test results and our hypotheses, we will explain them
independently in more detail.
What we see in table 4, model 1 is that the control variables we used, because other authors (Chang et al. 2012; Lin et al. 2015) have used them, are confirmed by our research. The same conclusions can be drawn for model 2 with regard to the control variables. We see that subsidiary size (β = 6.15, p= <0.01) and MNC performance (β = 1.16, p = <0.01 and β = 1.15, p <0.01) have a significant positive effect on subsidiary performance. This means the more employees a subsidiary has and the better the MNC performance, the higher the
subsidiary performance. MNC size has a significant negative effect (β = -4.00, p -= <0.01 and β = -3.90, p = <0.01). This means the higher the MNC size, the lower the subsidiary
performance. Furthermore our control variables can count for 21% of the variation in
subsidiary performance (R Square = .21). Model 2 contains the independent variable the ratio of women in the board. The results are not significant and therefore must H1 be rejected (R2 = .21 and β = -1.14, p = .64). This means we found no support that there is a positive
relationship between the ratio of women in the board and subsidiary performance.
Regression results H1
Model 1 Model 2
Steps and variables β SE β SE
Intercept 17.59** (.75) 17.80** (.88) Control Subsidiary size 6.15** (.00) 6.15** (.00) MNC performance 1.16** (.00) 1.15** (.00) MNC size -4.00** (.00) -3.90** (.00) MNC age .00 (.00) .00 (.00) MNC industry 1.09 (.60) 1.08 (.61) Main effects Ratio of women -1.14 (2.85) R Square .21 .21 ∆ R Square .00
Table 5 model 1 shows the control variables. The interpretation of the results of the control variables can be found above. Model 2 shows the effect of the independent variable board size on subsidiary performance. The results are not significant (R2= 0.22 and β = -.11, p
= .35). This means H2 is not confirmed by this research. So we found no support that there is
a positive effect between the headquarters board size and subsidiary performance.
Regression results H2
Model 1 Model 2
Steps and variables β SE β SE
Intercept 17.59** (.75) 16.44** (1.44) Control Subsidiary size 6.15** (.00) 6.13** (.00) MNC performance 1.16** (.00) 1.13** (.00) MNC size -4.00** (.00) -4.26** (.00) MNC age .00 (.00) .00 (.01) MNC industry 1.09 (.60) 1.17 (.61) Main effects Board size -.11 (.11) R Square .21 .22 ∆ R Square .01
Table 5 regression results H2 ** p <0.01
Table 6 shows the results with respect to the third hypothesis. Here we measured the effect of the average board age on subsidiary performance. The results for H3 are not
Regression results H3
Model 1 Model 2
Steps and variables β SE β SE
Intercept 17.59** (.75) 17.14 ** (5.35)
Control
Number of employees Subs 6.15** (.00) 6.15** (.00)
Total revenue 1.16** (.00) 1.16** (.00)
Number of employees MNC -4.00** (.00) -3.98** (.00)
Year of incorporation .00 (.00) .00 (.00)
Type of industry 1.09 (.60) 1.10 (.61)
Main effects
Average board age .01 (.08)
R Square .21 .21
∆ R Square .00
Table 6 regression results H3 ** p <0.01
For the hypotheses H4, H5 and H6 we used a regression with moderation. This means we had to standardize the variables. Moreover we had to test the interactions terms. The moderator in this research was the type of market, which was developed market or emerging market. We see that in the models below, the same control variables are significant as in the prior regressions.
Table 7 shows the results of the regression with moderation. The interaction term in this model is the ratio of women in the board x the type of market. The results show that the type of market has a significant negative effect on subsidiary performance (β =-.99, p=
<0.05). This means that subsidiary performance is higher in a developed market than in an
emerging market. The interaction effect turns out to be not significant (β = .27, p = .54, R2 = .25). This means that H4 must be rejected. This means that we found no support for a
Regression results H4
Model 1 Model 2 Model 3
Step and variables Β SE Β SE Β SE
Intercept 17.59** (.75) 18.04** (.78) 18.05** (.79) Control Employees Subs 6.15** (.00) 6.01** (.00) 6.04** (.00) Total revenue 1.16** (.00) 1.08* (.00) 1.07* (.00) Employees MNC -4.00** (.00) -3.67** (.00) -3.67** (.00) Year of incorporation .00 (.00) .00 (.00) .00 (.00) Type of industry 1.09 (.60) .99 (.60) .99 (.60) Main effects Ratio of women -.13 (.23) -.24 (.29) Type of market -.99* (.44) -.99* (.44) Two-way Interaction WomenXTypemarket .27 (.44) R Square .21 .25 .25 ∆ R Square .04 .00
Table 7 regression results H4 ** p <0.01 * p <0.05
In table 8, the regression results with respect to H5 are presented. The interaction term in this model is board size x the type of market. The type of market is again significant. The interaction effect turns out to be not significant (β = .27, p = .54, R2 =.25). This means that H5 must be rejected. So we found no support that there is a positive relationship between the board size and subsidiary performance and this effect is stronger in emerging markets as opposed to developed markets.
Regression results H5
Model 1 Model 2 Model 3
Step and variables Β SE Β SE Β SE
Intercept 17.59** (.75) 18.27** (.79) 18.28** (.79) Control Employees Subs 6.15** (.00) 6.00** (.00) 6.11** (.00) Total revenue 1.16** (.00) 1.06* (.00) 1.05* (.00) Employees MNC -4.00** (.00) -4.05** (.00) -4.09** (.00) Year of incorporation .00 (.00) .00 (.00) .00 (.01) Type of industry 1.09 (.60) 1.08 (.60) 1.09 (.60) Main effects Board size .24 (.25) -.12 (.31) Type of market -.98* (.44) -.98* (.44) Two-way Interaction BoardsizeXTypemarket .27 (.43) R Square .21 .25 .25 ∆ R Square .04 .00
Table 9 shows the results of the interaction term average board age x type of market. The type of market is again significant. The interaction effect turns out to be not significant (β = .25, p = .56, R2 = .25). This means that H6 must be rejected. So we found no support that there is a negative relationship between the average age of the headquarters board and subsidiary performance and this effect is stronger in emerging markets as opposed to developed markets.
Regression results H6
Model 1 Model 2 Model 3
Step and variables Β SE Β SE Β SE
Intercept 17.59** (.75) 18.11** (.78) 18.10** (.79) Control Employees Subs 6.15** (.00) 6.02** (.00) 5.96** (.00) Total revenue 1.16** (.00) 1.09* (.00) 1.11* (.00) Employees MNC -4.00** (.00) -3.77** (.00) -3.77** (.00) Year of incorporation .00 (.00) .00 (.00) .00 (.00) Type of industry 1.09 (.60) 1.00 (.60) 1.01 (.61) Main effects
Average board age .01 (.22) -.10 (.29)
Type of market -.98* (.44) -.98* (.44)
Two-way Interaction
AverageageXTypemarket .25 (.43)
R Square .21 .24 .25
∆ R Square .03 .00
5 DISCUSSION
In this part we discuss and interpret the results. We do this by explaining the theoretical and practical implications, and indicate the strengths, limitations and future research directions. We found no support, that the average board age, ratio of women in the board and the board size has an effect on subsidiary performance. Moreover there are no significant results that this effect is moderated by the type of market. What we do found is that the type of market (developed or emerging) has an effect on subsidiary performance, such that subsidiary performance is higher in developed markets as opposed to emerging markets. 5.1 Theoretical implications
We found no relation between the representation of women in the board and
subsidiary performance. Rose (2007) found the same, namely that there is no relation between women in the board and firm performance. This result is also confirmed by the study of Gavrea et al (2012) which conducted the research for firm performance. So this means that despite women bringing creativity, experience and values, it does not directly mean an increase in subsidiary performance. In that way women are not a prerequisite to acquire performance increases and therefore we may state that gender does not play a role. In some cases women are appointed to the board as a token (Kanter, 1977). Furthermore, it could be that it is just morally desirable to appoint women. Therefore we think that women are not appointed because they achieve better results, but just to show they are acting in a good way for example. Rose (2007) gave a plausible reason why gender does not influence the
performance. They stated that there is some process of socialisation. This means the unconventional board members adopt the norms and behaviour of the conventional board members, and so gender does not play a role.
The relationship between subsidiary performance and board size is not confirmed in our research. This result is also confirmed by the research of Pathan and Skully (2010) for companies in the banking sector. They relate the outcome to the inefficiency argument of Boone, Field, Karpoff and Raheja (2007). Which means it does not maximize the
shareholders’ value. Size brings more variety in people with respect to background, but this is also not needed to achieve better subsidiary performance. Larger boards have poorer
1993). Unfortunately there is no clear agreement what the reason is, why board size does not influence subsidiary performance.
Besides is the relationship between board age and performance not significant. Gavrea et al (2012) did research in the Romanian market by using firms listed on the Bucharest stock exchange. The authors found also no evidence that the average board age has a significant effect on firm performance. We may state that average age is related to experience
(Jhunjhunwala et al. 2012); however we see that more work experience in the board is therefore not needed to achieve a better subsidiary performance. There is no clear agreement what the reason is why the average age does not influence subsidiary performance. However, Rose (2005) stated that is hard to identify any board variables (in general) that influence performance. Furthermore, they argue that the board is only important when the company has financial problems or there are threats. This could explain why board age does not have an influence on subsidiary performance in our case.
Regarding the different board characteristics; board age and board size, the authors gave no clear explanation what the reason is, that there is no relation between those
characteristics and subsidiary performance. The demographics like age and gender cannot be influenced. We think it is more important what a person’s capabilities and skills are. The article of Rose (2007) stated that sufficient human capital and skills is more important than educational background. This is more important because then they can understand the information that is given by the board. So we may argue that they should look more at the skills, education and experience of the board members and not at the demographics like gender and age.
An interesting finding is that the type of market does influence subsidiary
Currently many large companies expand into emerging markets. One reason is the cheap labour. However, our research found out that the subsidiary performance in developed markets is much better.
5.2 Practical implications
The board is an important mechanism in every company. What we showed is that companies do not have to bother so much about how to structure their board with respect to demographics of the board, i.e. what ages, gender and size when it comes to subsidiary performance. Many research focused on how to structure their board, but our results showed that the age, women and size do not have an effect on subsidiary performance.
Of course to represent women can be a part of their social corporate performance as researched by Sabatier (2015). Many companies want to show they are acting in a good way by appointing women, and creating a certain image. Moreover to have women in the board is an ethical assumption. Everyone deserves an equal opportunity. You may not exclude people based on gender. Firms are responsible for individuals’ equal treatment (Van der Walt & Ingley, 2003). Another argument is that women bring economic value to the company. If the company does not select valuable individuals, like women, they may forgo to get access to resources that are important for the firm (Burke, 2000; Westphal & Milton, 2000). A resource could be their background and therefore they could provide a unique perspective on a
strategic issue (Burke, 2000).
are more successful in a developed market. Developed markets are more technically and economically advanced (Oxford English Dictionary, 2015). These markets have good institutions which will help them to be successful (Gugler et al. 2013).
Companies should take these results of this research into account when opening a subsidiary abroad. Cheaper labour does not always mean better performance. This cheaper labour can be found in emerging markets. However, our research showed that subsidiary performance of developed markets is better than in emerging markets. Therefore we suggest that companies need to do research, and also look what the role of those institutions in a particular country is.
5.3 Strengths, limitations and future research
This research shows several strengths. We showed that even though a lot has been written about the board size, the average age and the presence of women in the board, these do not have an influence on subsidiary performance. We were able to draw this conclusion, because we found the necessary information. We were able to do quantitative research with a sample of 126 companies. In our opinion is the most valuable strength is that we found the performance and size of the developed and emerging markets. For 54 companies we found a subsidiary in both type of markets.
A first limitation is that the sample includes several countries. To make it more transparent, further research should include the same country for an emerging and developed market subsidiary. So for example we take just subsidiaries from India for the emerging markets and The Netherlands for the developed market. Due to time limitations and data accessibility this was not possible during this research. It may be that the results differ per country, because every country has its own culture. For example the article of Rose (2007) found that the representations of women differ per country. Denmark has for example a low representation. According to this information, it could influence also subsidiary performance. There are even countries that have a quota for gender diversity in the board, to appoint more women in the board (Branson, 2012).
we may think that the results for the regression may differ. We did not use this measure, because the data needed was not available.
When evaluating the results there are perhaps other aspects that determine the success of a board and therefore may influence subsidiary performance. We may think that it is more important what a person has studied and what kind of person it is. Is it for example a leader; is it the kind of person that takes risk? Literature on for example character, networks, is, as far as we know, very scarce. These kinds of influences have nothing to do with gender, age and size of the board. Therefore it is interesting for further research. Moreover the values a person has and maybe the network someone is in may have an influence on subsidiary performance. This is also argued by Rose (2007) which mentioned, that human capital is more important than for example educational background. This could be because the work which is done by the board does not require this specific background. The article of Zahra and Pearce (1989) argued that the independence of the board is also important when it comes to performance. They mentioned that many outsiders in the board would increase their effectiveness.
We suggest doing more research in the area of subsidiary performance and the differences between emerging and developed markets. Moreover, is it important to do more research in what the roles of those institutions are. Chan et al. (2010) found that 19% of the variables influencing subsidiary performance are on parent level. This means there are other areas that may have an influence on this performance. As many companies want to expand aboard, subsidiary performance becomes an important topic. Moreover, there may be other characteristics of the board that do have an influence on subsidiary performance. We could think of their experience/background.
6 CONCLUSION
In the final part of this research we combine all the information and try to give some lessons and what we have learned from this research.
Unfortunately we found no support for our hypotheses. This means that board size, the ratio of women and average board age do not have an influence on subsidiary performance. As many companies focus on how to composite their board, we showed that these
demographics may not matter that much. Also the type of market as an interaction effect has no significant effect on subsidiary performance.
What we do found in this research is that the type of market has an influence on subsidiary performance. We found out that subsidiary performance is higher in developed markets than in emerging markets. From which we may conclude that developed markets achieve better performance. So we can learn from it that you never need to do what everyone else is doing; namely investing in emerging markets, because of for example cheap labour. You should also consider developed markets, with better institutions and which may achieve better performance. Furthermore, demographics are shown not to be very important in
ACKNOWLEDGEMENTS
By conducting this research I learned how to do research, collect data and work with SPSS. Even though it was not always easy, I am proud of the result that has been presented here. I expanded my knowledge in the area of the board characteristics and subsidiary performance. It was interesting to see how those large companies of the fortune 500 differ in board composition. What I find disappointing is that the hypotheses were not significant, even though I had good reasons why they were not significant.
REFERENCES
Adams, R., & Ferreira, D. 2009. Women in the boardroom and their impact on governance and performance. Journal of financial economics, 94(2): 291-309. Amason, A.C. 1996. Distinguishing the effects of functional and dysfunctional
conflict on strategic decision-making: resolving a paradox for top management teams.
Academy of management journal, 39(1): 123-148.
Amason, A.C., & Sapienza, H.J. 1997. The effects of top management team size and interaction norms on cognitive and affective conflict. Journal of management, 23(4): 495-516.
Ali, M., Kulik, C. T., & Metz, I. 2011. The gender diversity performance relationship in services and manufacturing organizations. International Journal of Human
Resource Management, 22: 1464–1485.
Ararat, M., Aksu, M., & Cetin, A.T. 2015. How board diversity affects firm performance in emerging markets: evidence on channels in controlled firms.
Corporate governance: an international review, 23(2): 83-103.
Arnold, D.J., & Quelch, J.A. 1998. New strategies in emerging economies. Sloan
management review, 40: 7-20.
Bahadir, S.C., Bharadwaj, S.G., & Srivastava, R.K. 2015. Marketing mix and brand salesin global markets: examining the contingent role of country-market
characteristics. Journal of international business studies, 46: 596-619. Becht, M., Bolton, P. & Roell. A. 2003. Corporate governance and control.
Amsterdam: Elsevier B.V.
Bianco, M., Ciavarella , A. and Signoretti. R. 2013. "Women on Corporate Boards in Italy", Occasional papers. Number 174, Questioni di Economia e Finanza, Bank of
Italy and the Eurosystem.
Bonn, I., Toru, Y., & Phillip, 2004. Effects of board structures on firm performance: A comparison between Japan and Australia. Asian business and management, 3(1): 105-125.
Boone, A.L., Field, C.L., Karpoff, J.M., Raheja, C.G., 2007. The determinants of corporate board size and composition: an empirical analysis. Journal of Financial
Botwinick, J. 1977. Aging and behaviour. New York: springer.
Branson, D. M. (2012). Initiatives to place women on corporate boards of directors: A global snapshot. Journal of Corporation Law, 37, 793–814.
Burke, D.M., & Light, L.L. 1981. Memory and aging: the role of retrieval processes. Psychological bulletin, 90: 513-546.
Burke, R. 2000. Company size, board size and the number of women directors. In R. J. Burke & M. C. Mattis (Eds.), Women on corporate boards of directors: International challenges and opportunities. Dordrecht: Kluwer Academic Publishers.
Chan, C. M., Makino, S., & Isobe, T. 2010. Does sub-national region matter? Foreign affiliate performance in the United States and China. Strategic Management Journal, 31(11): 1226–1243.
Chang, Y.Y., Gong, Y., & Peng, M.W. 2012. Expatriate knowledge transfer,
subsidiary absorptive capacity and subsidiary performance. Academy of management
journal, 55: 927-948.
Chang,E. & Taylor, M.S. 1999. Control in multinational corporations: the case of Korean manufacturing subsidiaries. Journal of management, 25(4): 541-565. Chatterjee, S.D. 2011. Board composition and performance in Indian firms: a
comparative analysis empirical. The international journal of management science
and information technology, 1(2): 1-15.
Chen, T. 2015. Institutions, board structure, and corporate performance: evidence from Chinese firms. Journal of corporate finance, 32: 217-237.
Claesens, S., & Yurtoglu, B. 2013. Corporate governance in emerging markets: a survey. Emerging markets review, 15: 1-33.
Cntv. Kun, L. 2015. Transfer pricing makes big splash on global taxes.
http://english.cntv.cn/2015/10/28/ARTI1446023289912521.shtml [accessed: October, 29th 2015]
Cochran, P.L., Wartick, S.L., & Wood, R.A. 1984. The average age of boards and financial performance, revisited. Journal of business and economics, 23(4): 57-63. Coff, R. 1999. When competitive advantage doesn’t lead to performance: the
