Domestic & Foreign Institutional Investor
Types, International Board Diversity, and
Firm Performance: an Agency Theory
Perspective
A study on non-financial Dutch listed firms at the Euronext
Amsterdam
Ilka Jurriaan van Vliet (S1742787)
University of Groningen
Domestic & Foreign
Institutional Investor Types,
International Board Diversity
and Firm Performance: an
Agency Theory Perspective
A study on non-financial Dutch listed firms at the Euronext
Amsterdam
I.J. van VlietUniversity of Groningen
Faculty of Economics & Business
Master thesis MSc. International Financial Management and International Business & Management 14-01-2017 Bloemstraat 47-24 9712 LC Groningen 0636128572 i.j.van.vliet@student.rug.nl S1742787
Words: 12.624 (excl. references)
First supervisor: Mr. P.J.M. Morgado
I ABSTRACT
This firm level study tries to connect the field of international business management to the field of international financial management. It examines the effect of direct institutional equity holdings on the target firm’s accounting and market performance. The study segregates domestic and foreign; traditional and alternative institutional ownership. It is unique in testing the moderating effects of nationality board diversity on the relationship between domestic and foreign institutional ownership. Building on agency theory, institutional investors exploit their shareholder rights via exit, voice and loyalty to align their objectives with the target firm’s management board and supervisory board. This reduces agency costs and should ultimately result in a win-win situation. The paper finds that institutional ownership does influence market performance. However, it concludes that ‘pressure-sensitive’ firms i.e. firms that are more likely to have business relations with the target firm, positively influence this market performance, whilst more ‘pressure-resistant’ institutions do not. Thus, domestic and traditional institutions have a higher probability of increasing the firm market performance than foreign and alternative institutional investors. International board diversity negatively moderates this relationship. Moreover, the study contributes by finding an anti-takeover mechanism that is not properly addressed in literature and concludes that rigid aggregation of institutional investors causes pitfalls e.g. by overstating alternative investors. This study sampled 58 non-financial listed firms at the Euronext Amsterdam over a time span of eight years, from fiscal year end 2008 to fiscal year end 2015.
II Preface
First of all, I would like to thank my supervisors for their willingness to guide me in the intense and sometimes exhaustive period of writing my thesis. Their guidance was straight-forward and honest from the start.
This combination of supervisors was ultimately exactly what I longed for, although I might not have known it at the direct time. Therefore, I hope that both supervisors will intellectually guide many more students, as they have a bright future ahead in science. I am grateful that I could be part of this team.
Secondly, I would like to thank my parents Andre and Priska, and partner Sophie for being there when it was needed. Although it often is taken for granted, I do not take it for granted.
Thirdly, I would like to thank the University of Groningen to facilitate me as a student.
Lastly, I would like to introduce the reader with a brief quote from one of the five hot minds in economics at a psychology-festival in Amsterdam:
“The father of economics Adam Smith famously argued that the wealth of nations comes from specialization. It doesn’t come from what you have in terms of gold, it doesn’t actually
come from anything else except for specialization; that you do this, and I do that. Now he forgot to mention… I think this was something that he thought was somewhat self-evident,
that in order to specialize we first of all have to be different.”
Tomáš Sedláček
TABLE OF CONTENTS Abstract ……….. I Preface ………... II 1. Introduction ……… 1 2. Theoretical Background ……….……….... 4 2.1 Definition ………. 4
2.1 Institutional Ownership Theory ……… 6
2.2 Institutional Ownership in a Global Context ……… 8
2.3 Institutional Ownership in The Netherlands ………. 9
2.4 International Board Diversity Theory ………... 10
3. Literature Review ………..……….…...……. 11
3.1 Institutional Ownership and Firm Performance ……..………. 11
3.1.1. Firm Performance ………..………... 12
3.1.1.1. Firm Agency Costs ……… 12
3.1.1.2. Firm Accounting Performance ……….……... 12
3.1.1.3. Firm Market Performance ………. 13
3.1.1.4. Firm Accounting and Market Performance ………..…... 13
3.1.2. Domestic Institutional Ownership ……… 14
3.1.3. Foreign Institutional Ownership ………... 15
3.1.4. Traditional Institutional Ownership ………. 15
3.1.5. Alternative Institutional Ownership ………. 16
3.2 International Board Diversity ……….…………..…..…….. 16
3.2.1. IBD and Domestic Institutional Ownership………... 16
3.2.2. IBD and Foreign Institutional Ownership ……… 18
4. Methodology ……….. 20
4.1 Data ……….. 20
4.2 Sample ………...…………... 21
4.3 Variables ……….. 22
1 1. Introduction
Institutional investors exert a vast influence in the global capital market. The Boston Consultancy Group (BCG) reports a global all-time high of $74 trillion of assets under institutional management (BCG, 2015). The role and influence has significantly grown over time, also in the Dutch capital market. Institutional investors, in a broad sense, can be seen as institutions that manage and invest other people’s pooled money. These investments come in conjunction with ownership. This makes it important to look at the role of investors as owners of firms. The corporate governing functions of institutional owners can reduce agency problems and subsequently improve firm performance (Shleifer & Vishny, 1986). The institutional investors can affect the management activities directly as owners of firms or indirectly through trading in securities of such firms (Ahmad & Jusoh, 2014; Gillan & Starks, 2003). Therefore, they can play a key role in monitoring the management of firms and their investment or divestments in firms can act as an important source of information to other shareholders (Demiralp, D’Mello, Schlingemann, & Subramaniam, 2011; Gillan & Starks, 2003). Hence, it becomes crucial to examine the role of institutional ownership in affecting the financial performance of firms.
From 1980 to 2001, institutional investment on the Dutch market consisted entirely of domestic investment. In 2001, the market (€0.8 trillion total) experienced a turning point, as foreign institutional investors (FIIs) started to dominate. In the period 2001-2007 foreign institutional investments accelerated even more. After the financial crisis domestic institutional investors (DIIs) prevailed, because of largely absent foreign institutional investment. According to the Dutch Central Bureau of Statistics, in 2015 the total annual amount of institutional investment in The Netherlands accounted for €2.4 trillion, where pension funds are by far the largest investors and can be accounted for €1.2 trillion. Investment funds (excluding money market funds) invested €0.8 trillion, and insurance companies €0.4 trillion.
2 alignment hypothesis; and (iv) the efficiency abatement hypothesis. All these theories are connected to the agency theory of Jensen & Meckling (1976).
The efficient monitoring hypothesis expects a positive relationship as institutions can monitor managements of firms by discussing corporate plans with them, supporting or opposing managements’ decisions through voting for or against them. The institutional investors take on small marginal costs of monitoring via their economies of scale in research and monitoring activities (Demiralp et al., 2011). Hence, from an economic perspective even small shareholdings can be researched and monitored by institutional investors. Therefore, it is useful to examine institutional investments even when there is no controlling share in a firm. With diligent research before investing they provide capital to firms with great return on investment possibilities.
The conflict of interest-hypothesis predicts a negative relationship based on possible business relations with firms in which they have investments. It suggests that in latter situation institutional investors are forced to align themselves with the management of these firms to avoid jeopardizing their business ties (Pound, 1988). Thirdly, the strategic alignment hypothesis posits that institutional investors and management mutually expropriate corporate resources for their mutual benefits and thus act against the interests of other shareholders (Pound, 1988). This results in a negative relationship.
Lastly, the efficiency abatement hypothesis argues that institutional shareholders might be passive, and therefore they do not influence firm performance at all. They could have a myopic view and e.g. passively sell their stake in a poorly performing firm without discussing with the management (Duggal & Millar, 1999).
According to (Douma, George, & Kabir, 2006) DIIs are more likely to have long-term strategic ties with firms in which they invest and FIIs primarily focus on short-term capital gains of their investments. The difference in investment orientation and horizon could lead to different impacts on the invested firm’s performance. Hence, this study disaggregates institutional ownership in domestic and foreign institutional ownership.
3 in traditional and alternative institutional investors, as (Brav, Jiang, Partnoy, & Thomas, 2008; Ferreira & Matos, 2008) prove that they perform differently.
Ferreira & Matos (2008) are one of the few who perform such a study. This paper tries to extend their study by questioning the moderating role of international board diversity on the relationship between domestic and/or foreign institutional ownership and firm performance. Consequently, it examines whether a target firm of a DII or FII performs better or worse with an international diverse board.
Most relevant and recent research is conducted in one-tier board system markets e.g. the US, thereby neglecting the two-tier board systems e.g. The Netherlands. Moreover, the majority of studies seem to test the effect of solely CEOs and CFOs on firm performance. This research tries to fill that void by zooming in on the Dutch system, which is a two-tier board system. Therefore, it examines both the management (Raad van Bestuur) and the supervisory board (Raad van Commissarissen).
The mentioned hypotheses are all agency-mechanisms in which an institutional investor i.e. owner tries to exert influence over the target firm. Consequently, this paper questions: Does domestic and/or foreign institutional ownership have an effect on firm performance via reduced agency costs, and does international board diversity moderate this possible relationship? Moreover, does traditional and/or alternative institutional ownership have an effect on firm performance via reduced agency costs? The time span of this research spreads from the fiscal year end 2008 until fiscal year end 2015. Empirical evidence could have meaningful implications for market regulators, policy makers, investors and researchers.
4 Subsequent chapter presents the theoretical framework and background. Next, empirical findings are discussed in the literature review. The fourth chapter discusses the methodology and panel data. Afterwards, in the fifth chapter the results will be presented. Then, the sixth chapter will discuss limitations and suggestions for future research. Lastly, the final chapter contains the conclusion.
2. Theoretical Background
This chapter provides an overview of relevant theories and background information. Firstly, the relevant theories for my research will be discussed. Next, I will try to put institutional ownership in a global context. Lastly, I will specify the Dutch institutional ownership situation. Nevertheless, I want to start with defining institutional owners, as it is rather complex to frame institutional owners.
2.1 Definition
The definition of institutional owners is scarce, and if, shortly explained in the literature. Most papers (e.g. Dahlquist & Robertsson, 2001; Douma et al., 2006; Elyasiani & Jia, 2010) define institutional owners as: “institutions that manage and invest other people’s money”. This definition is rather broad and denies more complex cases e.g. sovereign wealth funds who are seen as ultimate owners, whilst they are de facto state ownership agencies. Another example is a private equity fund where the managing partner co-invests with the limited partners.
I find the most elaborate explanation at the OECD (2014: 96) which states that institutional owners are: “organized as legal entities, where the exact legal form, however varies widely and covers everything from straightforward profit maximizing joint stock companies (e.g. closed-end investment companies) to limited liability partnerships (e.g. private equity firms) and incorporation by special statute (e.g. sovereign wealth funds). They either act independently or are part of a larger company or conglomerate.” This is, for example, the case for mutual funds that are often subsidiaries of banks and insurance companies.
5 Grey institutions comprises of bank trusts, insurance companies, and other institutions (e.g., pension funds, endowments). These types of institutions are likely to avoid conflicts with the management and vote in favor of the management. In general they are more ‘pressure-sensitive’ to corporate management and considered to be ‘passive’ (Ferreira & Matos, 2008).
On the other hand, independent institutions include shares held by mutual fund managers and investment advisers. Independent institutions are more likely to collect information, face less regulatory restrictions and have less potential for business relationships with the target firm. These institutions are more likely to monitor corporate management and are ‘pressure-resistant’ and considered to be ‘active’ (Ferreira & Matos, 2008).
However, I argue that the distinction between grey and independent has become rapidly out-dated and it does not capture the current dynamics of institutional investors. The institutional owner forms and entities have transformed over time. In my opinion, the OECD (2014: 96) provides me with a better and updated definition.
Despite the complexity of framing institutional owners, the OECD states that traditional types of institutional investors comprise of pension funds, investment funds and insurance companies. Alternative types are hedge funds, private equity firms, exchange-traded funds and sovereign wealth funds (OECD, 2014: 96).
As these types of institutional investors are applicable in a global context, it might be meaningful to use their definition likewise in the Dutch situation. Therefore, I distinguish my sample in traditional and alternative institutional investors, as defined by the OECD.
This study finds Orbis an appropriate database for analysis of ownership types, as it distinguishes between a rather large amount of institutional owners (https://help.bvdinfo.com/mergedProjects/WHDOTNETOWNERSHIP_EN/Home.htm), which in turn can be aggregated to traditional and alternative owners.
The last step before ultimately defining traditional and alternative owners is to link the Orbis-types to the definition of the OECD. I will elaborate on latter matching in the methodology section. For now, I move on to the theory section.
6 institutional shareholder is an institutional investor. Therefore, I use these words interchangeably.
2.2 Institutional Ownership Theory
In the introduction I proposed the efficient monitoring hypothesis; the conflict of interest hypothesis; the strategic alignment hypothesis and the efficiency abatement hypothesis. Latter hypotheses are all connected to a single theory: the agency theory by Jensen & Meckling (1976).
This paper credits McNulty & Nordberg (2016) for creating a clear and structured perspective on institutional ownership via the agency theory. The theory explains that ownership can be viewed as a bundle of rights where alignment of objectives between agent and principal is crucial. In this case the agent is the target firm who might want to work rather in his or her own interest, than working for the principal i.e. institutional investor, who controls the pooled money and e.g. anticipates maximization of the shareholder value in the long term. Latter contradicting objectives create so-called ‘agency costs’.
An example an agency cost is information asymmetry between both parties, which could lead to a ‘moral hazard’. The firm manager might know that investing in a particular company or project is a bad idea, but decides to follow through thinking about the short-term gains, thereby deteriorating long-term shareholder value. Moreover, neglecting the institutional investor its objectives and preferences. This is a clear opposite of the stewardship theory, which posits that agents have more will to do good for the principal than earn more themselves. This trade-off briefly introduces problems concerning institutional investment, investment policy and firm performance.
By setting out this example I want to prove that agency theory is connected to all hypotheses. The efficient monitoring hypothesis is met by the possible information asymmetry between the institutional owner and the target firm in which it invests. An example could be that the institutional investor proposes certain investment or financing opportunities to the target firm e.g. give out low interest loans to the firm in which it invests.
7 firm promises to change its investment policy, but it doesn’t, which adversely effects firm performance. This could result in a conflict of interest between the agent and principal.
The strategic alignment hypothesis argues that the principal and agent could reach implicit agreements on the mutual expropriation of corporate resources e.g. taking care that each of them receives bonuses upon certain investment projects, which could be detrimental for the firm performance. The efficiency abatement hypothesis states that the principal i.e. institutional investor is so passive that it has no influence on the agent’s behavior and performance.
Nonetheless, these hypotheses might not be as clear-cut as they seem. When it comes to institutional ownership there are different degrees and objectives that influence these hypotheses. Institutional ownership in terms of degrees is mostly explained in relation to the amount of activism i.e. intenseness.
Institutional activism can be defined as “actions taken by shareholders (institutions) with the explicit intention of influencing corporations’ policies and practices” (Goranova & Ryan, 2014: 1232). Activist shareholders pressure management to improve performance and shareholder value (Gillan & Starks, 2000). Activism contrasts with passive ownership, which involves holding the shares, collecting dividends and perhaps voting, but in an inattentive way; and trading. In their activism an institutional investor can consider three actions upon ownership: exit through selling the shares, loyalty by holding the shares, and voice via shareholder activism (Hirschman, 1970). Exit and loyalty are pretty straightforward. Voice i.e. ownership engagement, however, is clearly the most ambiguous one.
The formal and common way to exert voice is the annual shareholders meeting. During this meeting it is possible to account management for their actions. The annual meeting(s) serves three purposes. Firstly, the meeting is held to inform shareholders about the financial situation. Secondly, decisions that need consent from the shareholders, such as the approval of the annual account, election of members of the supervisory board or board of directors or liquidations, mergers and acquisitions. Thirdly, there is a goal to discuss important issues (Jong, Mertens, & Roosenboom, 2006).
8 proxy as institutional investors have the power to influence even when they have small shareholdings. They have power in their scale resulting in diminishing influencing and costs, which stimulates to influence company policies and management.
Institutional ownership in terms of objectives is better explained by looking at the global context in which it operates, as objectives seem to differ largely per continent, country and of course per institutional investor. Therefore, I would like to introduce a new paragraph.
2.3 Institutional Ownership in a Global Context
Institutional ownership in the U.S. and U.K. has been researched extensively. Outside of the U.S. and U.K. the information is rather limited (Poulsen, Strand, & Thomsen, 2010). The U.S. and U.K. can be characterized by their market-oriented corporate governance systems, and common law, which bolster minority shareholder protection (La Porta, Lopez-De-Silanes, & Shleifer, 1999). For example, in the U.K. (http://www.legislation.gov.uk/uksi/2009/1632/contents/made) shareholders with five percent or more of the total vote can call for meetings and can suggest resolutions. Therefore, large shareholdings are uncommon in the U.S. and U.K., which leads to most shareholdings being dispersed and under five percent (Aguilera & Jackson, 2003).
On the contrary, continental Europe characterizes by the network-oriented corporate governance systems and builds on stakeholder-collaboration. Continental Europe has weak legal minority shareholder protection. For example, in the Netherlands shareholders with at least ten percent of the outstanding shares can call for a general meeting. Moreover, shareholders with at least one percent of outstanding shares can submit proposals to the forthcoming general meeting (Jong, Mertens, & Roosenboom, 2006).
European block holders (owner of a large amount of shares) e.g. owners or families could exploit this fact and control the board directly. In general, this translates to most institutional shareholders in the U.S. and U.K. aiming for short-term returns, whilst European shareholders aim for long-term continuity (Aguilera & Jackson, 2003).
9 institutional owners. Moreover, it could affect the type of ownership that is present in the current Dutch institutional investment market. Latter forms a motivation to conduct this new study.
2.4 Institutional Ownership in The Netherlands
Next to general key differences in corporate governance systems around the world, there is a rather specific idiosyncratic part applicable to the Dutch system, which might have been overlooked by other studies. There is a special foundation (‘stichting administratiekantoor’ that freely translates to ‘foundation administrative office’) in The
Netherlands which firms use as anti-takeover mechanism
(http://wetten.overheid.nl/BWBR0030033/2011-06-15). This foundation buys shares from the respective firm and converts them to certificates. These certificates are controlled by the foundation to gather voting rights of the respective shareholders. In turn they receive dividend on their certificates for their voting rights. The foundation is often founded by the firm itself and in hands of people that have a long-term vision for the company. Example cases are:
- a firm who wants to reward its’ employees with shares in the company, so they have benefits from their hard work. An employer would then choose for certificates, because some interests are the opposite of the employer e.g. the height of the salaries.
- family disagreements on the strategy of the firm. As a resolution the shares are put in the foundation where they convert to certificates without voting rights. A neutral board is hired for the control of the foundation.
- inheritance of the firm. Imagine a parent having two children, one is capable of making decisions on behalf of the firm, the other is not. A solution would be to give the capable one the shares and the other one certificates.
- a small shareholder tries to take control of the firm, as other shareholders did not attend the shareholder meeting. An example of a relevant case is Unilever N.V. who has determined that the voting rights of shareholders that do not attend the shareholder meeting are automatically transferred to the foundation.
10 observe these anti-takeover mechanisms. However, in the study of Ferreira & Matos (2008) I did not find this. In my study this could lead to overstating of alternative institutions. In the methodology section I reveal how I dealt with this problem.
2.4 International Board Diversity Theory
In the international business management literature, diversity is often underpinned with the Upper Echelon theory by Hambrick & Mason (1984). This theory states that top-managers or supervisors view strategic decisions through their own highly personalized lenses. This view determines their dealing with opportunities, threats, alternatives and likelihoods of various outcomes. Although it is kind of ambiguous, the author himself and fellow researchers (Finkelstein, Hambrick, & Cannella, 2009) have confirmed upper-echelon theory based on a dozen of studies.
From the perspective of upper echelon, one could argue that strategies can be determined by demographic variables. Cultural patterns of thinking, feeling and acting are acquired when a person is most influenced by learning, which is in the early childhood. Formal and informal institutions guide a person in their formative years in dealing with uncertainty and taking appropriate actions i.e. decision making. These patterns are deeply rooted and once there, they are unlikely to change via experiences. National culture has been found to have an enduring impact on managers’ mind-sets and response to strategic issues (Crossland & Hambrick, 2007). These effects are thus likely to prolong when joining a board in a foreign country.
In general researchers advocate that diversity in boards enhances quality and efficiency of decision-making and in that increases shareholder value (Stephenson, 2004; van der Walt & Ingley, 2003). I find some creative theoretical perspectives on the possible moderating relationship. Firstly, Carleton, Nelson, & Weisbach (1998) reason from an agency theory perspective that diversity can be linked to greater independence of the boards. They reason that a higher diversity implies a larger discrepancy between the board and the top management. This diversity leads in general to the board not identifying with the top management. Thus, it is less costly for them to intervene the top management. Hence, it enhances the monitoring role of the boards.
11 directors is important (Fama & Jensen, 1983). Diversity enhances the quality of the discussion by incorporating more perspectives and higher creativity (Carter, D’Souza, Simkins, & Simpson, 2010). Therefore, including the supervisory board in my analysis is essential.
From a resource dependency perspective, a larger diversity means a bigger variety of external resources and better embeddedness in the corporate network. This could lead to access to scarce management talent, better expertise and access to financial funds, or politic and legislative institutions (Pfeffer & Salancik, 1978). Please visit (http://www.commissiecorporategovernance.nl/download/?id=804) for more elaborate explanations on these theoretical perspectives.
The theory of Zaheer (1995) might also help to deduct hypotheses. The theory states that firms have problems in transferring their home-country capabilities abroad. In doing so, they have to overcome the liability of foreignness. They can also overcome this liability by copying successful local firms. Therefore, I expect the monitoring and research skills to diminish when controlling their target firm abroad. On the contrary, I expect the monitoring and research skills to improve when foreign institutional owners can identify and communicate with their board more easily i.e. boards with high nationality diversity.
3. Literature Review
In this chapter I will firstly discuss the empirical findings of several relevant studies that are relevant to the main relationship between institutional ownership and firm performance. In between I derive several hypotheses (deductive approach). Secondly, I will zoom in on my moderator international board diversity i.e. nationality board diversity. I close this chapter by presenting a conceptual model and an overview of relevant studies.
3.1 Institutional Ownership and Firm Performance
12 In one of the earlier studies Brickley, Lease, & Smith (1988) zoom in on the voting rights of institutional investors. They find that institutional investors participate in voting more actively than non-substantial owners. They also observe difference in voting patterns per type of investor. Institutional investors such as mutual funds, pension funds, foundations and endowments, which are not influenced by the management of firms, vote actively against the managements, if the proposal has the potential to reduce shareholders’ wealth. On the contrary, institutional investors such as insurance companies, banks and trusts, which are likely to have business relationships with the firm, do not oppose the management. Latter results support the monitoring hypothesis.
In the same year there has also been scientific support for the strategic alignment hypothesis. Pound (1988) finds that institutional investors work together with insiders and managers against the shareholders interest. Pound (1988) finds that in firms that have high institutional ownership the entrenched management cannot successfully be removed as the institutional investors vote for the entrenched management.
3.1.1. Firm Performance 3.1.1.1. Firm Agency Costs
As introduced, possible performance variation from institutional ownership stems from agency costs. In the theory section I connected all hypotheses to the agency theory. As this is an important mechanism for institutional owners to influence the target firm, how can I observe these agency costs?
Ferreira & Matos (2008) argue that institutional investors try to reduce investment i.e. capital expenditures at the target firm. They anticipate that traditional and domestic institutional investors are more likely to have business relations with the target firm and therefore are less likely to interfere with the investment policy of the target firm. On the contrary, foreign and alternative institutions are less likely to have business relations with the firm and therefore try to reduce investment policy of the target firm. I blend latter propositions throughout the next section in my hypotheses.
3.1.1.2. Firm Accounting Performance
13 100 S&P firms in the 1990s. They measure operating performance by cash flow return on assets and institutional ownership by the percentage of institutional shareholding, the number of institutional stockholders, their board membership and the type of institutional investor. They find that types of institutional investors that are less likely to have a business relationship with the firm increase operating performance. Moreover, they conclude that institutional investors, who have business ties with the firm, do not have an impact on the operating cash flow of the firm. Latter suggests institutional investors with business ties do not actively monitor the management in order to protect their business relations. Their main finding is a positive significant relationship between the percentage of institutional shareholdings and operating cash flow returns.
3.1.1.3. Firm Market Performance
The paper of Gompers & Metrick (2001) shows that institutional investors have a tendency to invest in large liquid stocks, that have low past returns. They find that institutional investment dramatically increases over the period 1980 to 1996, where the price appreciation of these stocks translate to positive returns.
However, recently Edelen et al. (2016) find negative effects for stock returns of institutional investors. The main reason is that institutional investors buy overvalued stocks. They do find a positive return at the quarterly horizon, however at the one year-horizon they find a strongly negative relationship. The authors indicate: “Our evidence largely rules out explanations based on flow and limits-of-arbitrage, but is more consistent with agency-induced preferences for stock characteristics that relate to poor long-run performance.” This supports the strategic alignment hypothesis.
3.1.1.4. Firm Accounting and Market Performance
Another study that supports the monitoring hypothesis is the study of Elyasiani & Jia (2010) who find that institutional shareholding has a positive impact on firm performance. They define ownership using ownership persistence, as in holding the shares, over time. Moreover, they use industry-adjusted return on assets (ROA, net income/book value of assets) as the measure of firm performance and industry-adjusted Tobin’s Q as a measure for firm performance. My research uses firm accounting and market performance measures. Therefore, I hypothesize:
14 Hypothesis 1b. Institutional ownership increases firm market performance.
Hypothesis 1c. Institutional ownership reduces agency costs i.e. capital expenditures.
3.1.2. Domestic institutional ownership
There is empirical evidence (Dahlquist & Robertsson, 2001; Grinblatt & Keloharju, 2000; Kang & Stulz, 1997) that domestic and foreign institutional investors differ in their investment behavior and investment performance. Dahlquist & Robertsson (2001) find that DIIs have an information advantage due to home bias. Costs and efforts make them more likely to participate in shareholder meetings, voting and decision making. Thus, they are more able to align their interests with the firm they invested in. This seems to be correct for market measures, moreover intuitively seems right, however surprisingly empirical evidence from more renowned journals suggests otherwise.
Thanatawee (2002) for example disaggregates domestic and foreign institutional ownership and finds that domestic institutional ownership has a positive effect on the firm value of Thai firms (measured by Tobin’s Q). Ahmad & Jusoh (2014) use three years’ panel data of 730 Malaysian firms and use two types of market measures to examine firm performance: Tobin’s Q and share price. Their results also show a positive relationship.
As discussed, other studies (Douma et al., 2006; Ferreira & Matos, 2008) that use accounting measures e.g. operating performance as a measure of firm performance find opposite i.e. negative results. Ferreira & Matos (2008) find significant negative effects of domestic institutional ownership on the net profit margin for 11,224 non-U.S. firms with a total of 38,604 observations. However, they find a positive significant result for return on assets (ROA) as a dependent variable. Moreover, they find a significant negative effect on the capital expenditure of managers. Douma et al. (2006) investigate 1,005 Indian listed firms and they identify a significant negative relationship between domestic institutional ownership and return on assets. They find support for the strategic alignment and the conflict of interest hypothesis. Therefore, I hypothesize:
Hypothesis 2a. Domestic institutional ownership has a negative effect on firm accounting performance.
15 Hypothesis 2c. Domestic institutional ownership significantly influence agency costs i.e. capital expenditures.
3.1.3. Foreign institutional ownership
According to Nashier and Gupta (2016) FIIs invest in firms purely for return on investment based on their fund’s objectives and are less likely to have business relations with the firms. The study of Douma et al. (2006) concludes that foreign institutional ownership has a positive impact on Tobin’s Q. Moreover, Sarkar & Sarkar (2000) document that firms with a higher foreign institutional ownership have better firm accounting performance.
Moreover, Ferreira & Matos (2008) conclude that foreign, and independent (in my research: alternative) institutions enhance shareholder value and operating performance, because they have fewer business relationships with the firm. They find robust results, including potential endogeneity effects of institutional ownership on firm performance. Latter leads to the following hypothesis:
Hypothesis 3a. Foreign institutional ownership has a positive effect on firm accounting performance.
Hypothesis 3b. Foreign institutional ownership has a positive effect on firm market performance.
Hypothesis 3c. Foreign institutional ownership reduces agency costs i.e. capital expenditures.
3.1.4. Traditional institutional ownership
The subgroups of traditional institutional owners have been researched individually. An interesting piece of work by Chirinko (1999) on the firm performance of 94 Dutch firms non-financial firms from 1992-1996 finds a significant positive effect of insurance companies and pension fund share ownership on firm performance (please visit, http://www.rug.nl/research/portal/publications/corporate-control-mechanisms-voting-and-
cash-flow-rights-and-the-performance-of-dutch-firms(510898b6-a7a5-40cf-bcd3-fff379cc36ed).html) Thus, I hypothesize:
Hypothesis 4a. Traditional institutional ownership increases firm accounting performance.
16 Hypothesis 4c. Traditional institutional ownership significantly influences agency costs i.e. capital expenditures.
3.1.5. Alternative institutional ownership
Several studies (Achleitner, Betzer, & Gider, 2010; Brav et al., 2008; Ferreira & Matos, 2008; Voußem, Schäffer, & Schweizer, 2015) conclude that alternative institutional investors are better-positioned than traditional institutional investors to increase performance, via corporate governance at the portfolio company, as they are: (1) allowed to strongly link fund manager compensation to investment performance, which reduces agency costs (2) privately organized investment firms equipped with large capital resources (which allows them to have majority ownership in companies) who employ professional fund managers to maximize investment returns, (3) allowed to (due to a reduced degree of regulation) make severe use of derivative securities and debt financing (Edelen et al., 2016; Guo, Hotchkiss, & Song, 2011; Harris, Jenkinson, & Kaplan, 2014). Thus, I derive the following hypothesis:
Hypothesis 5a. Alternative institutional ownership increases firm accounting performance.
Hypothesis 5b. Alternative institutional ownership increases firm market performance.
Hypothesis 5c. Alternative institutional ownership reduces agency costs i.e. capital expenditures.
3.2 International Board Diversity
3.2.1. IBD and Domestic Institutional Ownership
In general there is an extensive amount of research that examines the relationship between board diversity and financial performance. Unfortunately, for this research, most of this is focused on gender diversity. There might be several explanations for the scarce presence of literature on international board diversity i.e. nationality board diversity and firm performance. Firstly, there might be a problem of endogeneity, as there might be a reversed causal relation between latter two. Moreover, nationality might not predict a person’s behaviour. However, as discussed in the theory section it is reasonable to assume that nationalities could influence the efficient monitoring and conflicts on interest hypotheses.
17 researchers find mixed results. They do find that racial diverse teams have more creative ideas, but also have higher turnover (Milliken & Martins, 1996).
In a study of Cox, Lobel, & McLeod (1991) Asian, Black and Hispanic Americans were confronted with the prisoner’s dilemma, where participants could choose to cooperate or compete with another party. They found that people from more collectivist backgrounds are more likely to cooperate than participants who have individualistic cultural traditions. This briefly outlines that ethnicity, and maybe even nationality matters in decision-making.
Frijns, Dodd, & Cimerova (2016) examine the impact of cultural diversity in boards of directors on firm performance. They construct a measure of national cultural diversity by calculating the average of cultural distances between board members using Hofstede’s cultural framework. In their findings they conclude on a negative effect of cultural distance on firm performance, measured by Tobin’s Q and ROA. They control for foreignness and find that the negative impact of cultural diversity on firm performance is mitigated by the complexity of the firm and the size of foreign sales and operations.
However, the study that covers the most similarities with this study is Nielsen & Nielsen's (2013). They examine the effect of nationality diversity in boards and management on firm performance. They find a significant positive effect. They find an even stronger effect when the teams are highly internationalized. Moreover, firm and industry conditions are a shaping factor. So, therefore I expect a positive influence of international board diversity.
In addition, I do not have a priori agency cost expectations on the influence of international board diversity on domestic institutional ownership. Therefore, I hypothesize:
Hypothesis 6a. The relation between domestic institutional ownership and firm accounting performance mitigates when having an international diverse management- and supervisory board.
18
3.2.3. IBD and Foreign Institutional Ownership
Literature concerning the moderating effect of international board diversity on the relationship between domestic or foreign institutional ownership and firm performance is non-existent. Therefore, I have to rely on theoretical insights. Based on the resource dependency and transaction cost perspective one could argue that it enhances firm performance. The theory of Zaheer (1995) finds pro’s and cons on this matter, but in total the pro’s seem to outweigh the cons. In addition, I do not have a priori agency cost expectations on the influence of international board diversity on foreign institutional ownership. Hence, I hypothesize:
Hypothesis 7a. The relation between foreign institutional ownership and firm accounting performance strengthens when having an international diverse management- and supervisory board.
Hypothesis 7b. The relation between foreign institutional ownership and firm market performance strengthens when having an international diverse management- and supervisory board.
Figure 1 Conceptual Model
19
Table 1
Other relevant studies on Institutional Ownership and Firm Performance (without activism)
Author (Year + Journal) Independent Variable Dependent Variable Method Main Findings Region
Campbell et al. (2008) Institutional Ownership (holdings) Cumulative Abnormal Returns + Fama French Cumulative Abnormal Returns (-) U.S.
Journal of Finance Tobin’s Q Regression Negative
Choi et al. (2017) Institutional Ownership (holdings) Risk-Adjusted Returns Fama French Concentrated investment (+) World
Journal of Financial Economics increases returns
Edelen, Ince & Kadlec (2016) Institutional Ownership (changes) Cumulative Abnormal Returns + Fama French Cumulative Abnormal Returns (--) U.S.
Journal of Financial Economics Strongly negative
Gompers, Metrick (2001) Institutional Ownership (changes) Cumulative Abnormal Returns + Fama Macbeth Strong Positive Cumulative (++) U.S.
Quarterly Journal of Economics Market-To-Book Ratios Abnormal Returns
Nofsinger, Sias (1999) Institutional Ownership (changes) Cumulative Abnormal Returns Fama Macbeth Strong Positive Cumulative (++) U.S.
Journal of Finance Abnormal Returns
Vliet (2017) Direct Institutional Ownership Accounting and Market Panel Data (see conclusion) Holland (holdings) Performance, Capital Expenditures Regression
Other relevant studies on Institutional Ownership and Firm Performance (with activism)
Appel, Gormley and Kiem (2016) Institutional Activism Cumulative Abnormal Returns + OLS Cumulative Abnormal Returns (+) U.S.
Journal of Financial Economics Operating Performance Regression Operating Performance (+)
Brav et al. (2008) Hedge Fund Activism Cumulative Abnormal Returns + Wilcoxon Signed Cumulative Abnormal Returns (+) U.S.
Journal of Finance Operating Performance Rank Sum Operating Performance (+)
Smith (1996) Pension Fund Activism Cumulative Abnormal Returns + Wilcoxon Signed Cumulative Abnormal Returns (+) U.S.
Journal of Finance Operating Performance Rank Sum Operating Performance Insignificant
Vouβem et al. (2015) Institutional Activism Market-To-Book Ratios Continuous Time Insignificant Market-To-Book Germany
Journal Managerial Governance Event History Ratios
20 4. Methodology
A quantitative research and deductive approach is used for this study. The philosophy of this research is realism. Thus, the focus in this explanatory study will be on trying to explain within a context. Three processes are important throughout the study: collection, coding, and analysis of data. In line with other similar research I will use secondary data. My archival research uses a mono method. It is a longitudinal study as it concerns a time span of eight years. Next to time varying data it is also cross-sectional, as it examines a variety of firms. Therefore, this study examines a relationship across the same firms over multiple points in time, i.e. panel data. I have missing data, so it concerns an unbalanced panel dataset.
For testing the main relationship I prefer the Pooled Ordinary Least Squares-method, however according to Demsetz and Lehn (1985) the relationship between institutional ownership and firm performance is endogenous. The general consensus from previous studies is that firm performance is influenced by institutional ownership, however reversely firm performance might also attract institutional investors. Therefore, it is preferable to use robustness checks in order to reduce endogeneity, as ordinary least squares (OLS) might be inconsistent. However, due to time constraints I will proceed with Pooled OLS estimation only.
4.1 Data
In order to retrieve all data I use a wide array of information sources. Table 2 provides an overview of the used data sources. I use Eviews to compute the data.
Table 2
Overview of Data and Sources
Data Source
Firm selection Female board index (2008 & 2015) Institutional ownership Orbis (Bureau van Dijk)
Firm performance Thomson Reuter’s Datastream Nationality board diversity BoardEx
Control variables Thomson Reuter’s Datastream & Orbis (Bureau van Dijk)
21 4.2 Sample
In order to reduce the amount of unavailable data the sample requirements for firms is that they have to be (1) listed Dutch firms at the Euronext Amsterdam (2) active from the beginning of 2008 until the start of 2016, or longer. To come to a conclusive sample list the following steps are undertaken:
1. Firms listed on the Euronext Amsterdam of fiscal year end 2008 (http://www.mluckerath.nl/uploads/FemaleBoardIndex2008.pdf) and fiscal year end 2015 (https://www.tias.edu/docs/default-source/Kennisartikelen/femaleboardindex2015.pdf?sfvrsn=0) have been compared. For the fiscal year 2008-2009 I find 122 listed firms to be active and listed on the Euronext Amsterdam, whilst 83 listed firms were active during the year 2015-2016.
2. From these lists 67 firms are similar in name or survived with a different firm name during the period 2008-2016. The other 54 firms were delisted e.g. Nutreco; Simac, bankrupt e.g. InnoConcepts; Van Der Moolen Holding, or acquired e.g. Tele Atlas; Hagemeyer.
In international financial management literature it is common to exclude financial firms, because of their severely different characteristics and business models. I do the same, consistent with a specific Dutch study by Chirinko (1999). The SIC-codes (6000-6999) were excluded. Subsequently, I exclude these firms from the list (13% of the initial sample). This leaves us to a final sample of 58 firms. For the final list of firms please see Appendix A.
22 Table 3
Industry Distribution
Industry Description Industry SIC-codes (primary) Number of Firms
Manufacturing 2000-3999 26
Services 7000-8999 14
Wholesale Trade 5000-5199 6
Construction 1500-1799 5
Retail Trade 5200-5999 3
Transportation & Public Utilities 4000-4999 3
Agriculture, Forestry, Fishing 0100-0999 1
Total: 58
Source: Vliet (2017)
The percentage of manufacturing firms in the sample amounts to 45%, the rest amounts to 55%.
4.3 Variables
4.3.1. Institutional Ownership and Firm Performance
To test whether institutional ownership has an effect on firm performance, and in which forms, I adopt the majority of Ferreira & Matos (2008: 531) variables. I use the exact same domestic (DOM_INSTI) and foreign (FOR_INSTI) institutional ownership variables, except I replace grey and independent institutional ownership for traditional (TRAD_INSTI) and alternative (ALT_INSTI) institutional ownership. Moreover, I do not differentiate between U.S. and non-U.S. institutions. Firstly, I need to conclude on my definition of institutional owners.
The OECD defines traditional types of institutional investors as pension funds, investment funds and insurance companies. Alternative types are hedge funds, private equity firms, exchange-traded funds and sovereign wealth funds (OECD, 2014: 96). Ferreira & Matos (2008) defines grey institutional investors as of bank trusts, insurance companies, and other institutions (e.g., pension funds, endowments). Independent institutions are mutual funds and investment firms.
23 investors identify by the letters J (= foundation/research institute), P (=private equity firms), S (= public authorities, states, governments), V (= venture capital firms) and Y (= hedge funds).
I assume investment funds to be under the reign of banks and consequently replace investment funds by banks. In general the definition must hold for companies that are more likely to monitor and those who doesn’t. In my research the alternative investors clearly have the power to monitor. For example the whole business model of private equity firms is to explicitly take stakes in firms to control and improve their financial performance, to sell them later. The traditional investor, e.g. a bank, often has close contact with companies by providing them loans and cater financial services.
Other types are e.g. F (= financial company), C (= industrial company). Financial and industrial companies are institutions, but are ambiguous for my definition. They do not particularly belong to the pressure-resistant or pressure-sensitive side of the spectrum; therefore they do not make a case for controlling effects.
Others to exclude are I (= one or more known individuals or families), M (= employees/managers/directors), H (= self-ownership), Q (= branch) and W (= marine vessel). These shareholders do not match the definition of an institution that manages other people’s money. Moreover, they represent insider ownership, which is clearly not the intent of this
study (please visit for more information:
https://help.bvdinfo.com/mergedProjects/WHDOTNETOWNERSHIP_EN/Home.htm). According to Orbis, Z (= public quoted companies), D (= unnamed private shareholders aggregated), L (= other unnamed shareholders aggregated) are unable to exert control over a company, so these are excluded as well.
Lastly, I want to emphasize that I measure only the direct ownership. Indirect ownership brings unnecessary complexity to measuring institutional ownership. Moreover, in this way we examine the ultimate control of the company.
24 4.3.2 International Board Diversity
For the measurement of demographic diversity I use the Blau-index. The Dutch Monitoring Commission of Corporate Governance Code also uses this measure. It can be applied to non-numeric variables.
The Blau-indicator (BLAU) is used in a report of Ees, Hooghiemstra, van der Laan, & Veltrop (2007). In their study they compare diversity in the supervisory board of Dutch listed firms for the years 2001 and 2006. The Dutch Monitoring Commission of Corporate Governance Code has commissioned their research. They conclude that diversity has risen, but the consequences for firm performance are not researched yet and set it out as a
recommendation for future research.
(http://www.commissiecorporategovernance.nl/download/?id=804)
The formula is as follows:
Blau𝑑𝑗 = 1 − ∑ (𝑥𝑛𝑖𝑗 𝑗) 2 𝑘 𝑖 =1 (1)
In this formula xij is the number of board or supervisory members from a certain category i in board j. The total number of members in board j is shown as nj. K resembles the total collection of different categories from a certain characteristic d. The proportion of members from category i in board j is calculated by dividing the total number of members of category i in council j by the total members in board j. This proportion is then squared and summed over all categories in set K. The sum is then subtracted from 1 to come to the indicator.
25 There are exceptional cases that need attention. For example, in latter board of Arcadis NV one U.S. national, namely Mr. Perez (16 January 2016 †) passed away. This research considers contributions of larger than one day accountable for a year. Therefore, Mr. Perez is included in the annual nationality board diversity. In cases of double nationalities the first-named nationality has been chosen as the country of origin of the board member. Moreover, notion worthy is the firm Oranjewoud who maintained listed without having a supervisory board and only one management board member throughout 2008-2010.
4.3.3 Control variables
Ferreira & Matos (2008: 508) provide me with a list on determinants of institutional ownership, which in turn I use to control for the relationship between institutional ownership and firm performance. Consequently, the only thing that is not covered is the nationality board diversity. Most research on the effect of boards on firm performance controls for the board size. However, after Variance Inflation Factor-analysis in Eviews I find this variable in conflict with the size of the firm, which I argue to be more important to be included in the models.
Ferreira & Matos (2008) identify two major drivers of institutional ownership: market capitalisation (SIZE) and closely held shares. Size is a major driver, because institutional money managers prefer largely sized firms to invest in. Moreover, firm size is important in international investment because of liquidity and transaction cost concerns (Ferreira & Matos, 2008).
Secondly, they identify closely held shares to be a major driver of institutional ownership as a large amount of closely held shares implies concentrated control rights. Institutional investors tend to not invest in poorly governed firms i.e. firms with a large amount of closely held shares. After Variance Inflation Factor-analysis I find closely held shares to correlate with other variables in my model.
26 from Orbis I do not expect to have any more insider ownership in my sample. In line with Chirinko (1999) I find 30% of the sample to have this anti-takeover construction.
Other drivers for institutional ownership are investment opportunities and firms respresented in the MSCI World-index, but also these variables seems to conflict with other variables by displaying multicollinearity issues.
Lastly, I exclude all country-level regressors. I choose to exclude those for the reason that institutional ownership in the Netherlands is merely affected by domestic country-level variables, which are for all institutional owners exactly the same. As I regard foreign countries to be different but attach the same weights to them I cannot weight them.
Moreover, they prefer recent positive stock performance e.g. annual stock return (RETURN). On the contrary, domestic institutional investors prefer high-dividend (DIV) paying stocks, so I need to control for that. Foreign investors seem to avoid these, because of tax-withholding concerns. Other firm-characteristic control variables are: leverage (LEV) and cash (CASH).
27
Table 4
Variables and Definitions Type of Variable Acronym Variable Definition and Measurement
Panel A: Institutional ownership variables
Total institutions INSTI Direct institutional ownership by all researched institutions as a percentage of common equity Domestic institutions DOM_INSTI Direct institutional ownership by Dutch institutions as a percentage of common equity Foreign institutions FOR_INSTI Direct institutional ownership by non-Dutch institutions as a percentage of common equity
Traditional institutions TRAD_INSTI Direct institutional ownership by traditional institutions (insurance companies, banks and mutual and
pension funds/nominees/trusts/trustees) as a percentage of common equity
Alternative institutions ALT_INSTI Direct institutional ownership by alternative institutions (foundations/research institutes, private equity firms, public authorities, states, governments, and venture capital firms) as a percentage of common equity
Panel B: Valuation, operating performance and investment variables
Tobin’s Q TOBINSQ Sum of total assets plus market capitalisation minus common equity divided by total assets
Return on Assets ROA Net income divided by total assets
Capital Expenditures CAPEX Capital expenditures divided by total assets
Panel C: Moderator variable
International Board Diversity BLAU One minus the fractions of nationalities in management and supervisory board separately squared and aggregated.
Panel D: Firm level control variables
Market capitalization (natural log) SIZE Log annual market capitalization in EUR€
Foreign sales FOREIGN Foreign sales divided by total sales
Leverage LEV Ratio of total debt to total assets
Cash CASH Ratio of cash to total assets
Dividend yield DIV Dividend yield of firm
Annual stock return RET Annual (end-of-year) geometric stock rate of return, new minus old return index divided by old Firm age (natural log) AGE Natural logarithm of firm age
Time dummy DT Year dummy, equals one in respective year, zero otherwise
Industry dummy DM Industry dummy, equals one if a firm belongs to SIC group classification, zero otherwise
28 4.3.4 Method
For my empirical analysis I use Pooled OLS to conclude on the relationship between institutional ownership and firm performance. Pooled regression assumes that the intercepts are the same for each firm and for each year. This may be an inappropriate assumption. I do control for heterogeneity across industries and time by adding industry and time dummies. When estimating the Pooled OLS fixed effects model the adjusted R2 increases, but dramatically decreases the significance of my regressors. Thus, I choose for Pooled OLS.
Model 1: 𝑇𝑂𝐵𝐼𝑁𝑆𝑄, 𝑅𝑂𝐴, 𝐶𝐴𝑃𝐸𝑋 = 𝛼 + 𝛽𝐼𝑁𝑆𝑇𝐼𝑖𝑡+ 𝛾1𝑆𝐼𝑍𝐸𝑖𝑡+ 𝛾2𝐹𝑂𝑅𝐸𝐼𝐺𝑁𝑖𝑡+ 𝛾3𝐿𝐸𝑉𝑖𝑡+ 𝛾4𝐶𝐴𝑆𝐻𝑖𝑡+ 𝛾5𝐷𝐼𝑉𝑖𝑡+ 𝛾6𝑅𝐸𝑇𝑈𝑅𝑁𝑖𝑡+ 𝛾7𝐴𝐺𝐸𝑖𝑡+ 𝛾𝐷𝑡+ 𝛾𝐷𝑚+ 𝜀𝑖𝑡 (2) Model 2: 𝑇𝑂𝐵𝐼𝑁𝑆𝑄, 𝑅𝑂𝐴, 𝐶𝐴𝑃𝐸𝑋 = 𝛼 + 𝛽𝐷𝑂𝑀_𝐼𝑁𝑆𝑇𝐼𝑖𝑡+ 𝛾1𝐷𝑂𝑀_𝐼𝑁𝑆𝑇𝐼∗𝐵𝐿𝐴𝑈 + 𝛾2𝐵𝐿𝐴𝑈𝑖𝑡+ 𝛾3𝑆𝐼𝑍𝐸𝑖𝑡+ 𝛾4𝐹𝑂𝑅𝐸𝐼𝐺𝑁𝑖𝑡+ 𝛾5𝐿𝐸𝑉𝑖𝑡+ 𝛾6𝐶𝐴𝑆𝐻𝑖𝑡 + 𝛾7𝐷𝐼𝑉𝑖𝑡+ 𝛾8𝑅𝐸𝑇𝑈𝑅𝑁𝑖𝑡+ 𝛾9𝐴𝐺𝐸𝑖𝑡+ 𝛾𝐷𝑡+ 𝛾𝐷𝑚+ 𝜀𝑖𝑡 (3) Model 3: 𝑇𝑂𝐵𝐼𝑁𝑆𝑄, 𝑅𝑂𝐴, 𝐶𝐴𝑃𝐸𝑋 = 𝛼 + 𝛽𝐹𝑂𝑅_𝐼𝑁𝑆𝑇𝐼𝑖𝑡+ 𝛾1𝐹𝑂𝑅_𝐼𝑁𝑆𝑇𝐼∗𝐵𝐿𝐴𝑈 + 𝛾2𝐵𝐿𝐴𝑈𝑖𝑡+ 𝛾3𝑆𝐼𝑍𝐸𝑖𝑡+ 𝛾4𝐹𝑂𝑅𝐸𝐼𝐺𝑁𝑖𝑡+ 𝛾5𝐿𝐸𝑉𝑖𝑡+ 𝛾6𝐶𝐴𝑆𝐻𝑖𝑡 + 𝛾7𝐷𝐼𝑉𝑖𝑡+ 𝛾8𝑅𝐸𝑇𝑈𝑅𝑁𝑖𝑡+ 𝛾9𝐴𝐺𝐸𝑖𝑡+ 𝛾𝐷𝑡+ 𝛾𝐷𝑚+ 𝜀𝑖𝑡 (4) Model 4: 𝑇𝑂𝐵𝐼𝑁𝑆𝑄, 𝑅𝑂𝐴, 𝐶𝐴𝑃𝐸𝑋 = 𝛼 + 𝛽𝑇𝑅𝐴𝐷_𝐼𝑁𝑆𝑇𝐼𝑖𝑡+ 𝛾1𝑆𝐼𝑍𝐸𝑖𝑡+ 𝛾2𝐹𝑂𝑅𝐸𝐼𝐺𝑁𝑖𝑡+ 𝛾3𝐿𝐸𝑉𝑖𝑡+ 𝛾4𝐶𝐴𝑆𝐻𝑖𝑡+ 𝛾5𝐷𝐼𝑉𝑖𝑡+ 𝛾6𝑅𝐸𝑇𝑈𝑅𝑁𝑖𝑡+ 𝛾7𝐴𝐺𝐸𝑖𝑡+ 𝛾𝐷𝑡+ 𝛾𝐷𝑚+ 𝜀𝑖𝑡 (5) Model 5: 𝑇𝑂𝐵𝐼𝑁𝑆𝑄, 𝑅𝑂𝐴, 𝐶𝐴𝑃𝐸𝑋 = 𝛼 + 𝛽𝐴𝐿𝑇_𝐼𝑁𝑆𝑇𝐼𝑖𝑡+ 𝛾1𝑆𝐼𝑍𝐸𝑖𝑡+ 𝛾2𝐹𝑂𝑅𝐸𝐼𝐺𝑁𝑖𝑡+ 𝛾3𝐿𝐸𝑉𝑖𝑡+ 𝛾4𝐶𝐴𝑆𝐻𝑖𝑡+ 𝛾5𝐷𝐼𝑉𝑖𝑡+ 𝛾6𝑅𝐸𝑇𝑈𝑅𝑁𝑖𝑡+ 𝛾7𝐴𝐺𝐸𝑖𝑡+ 𝛾𝐷𝑡+ 𝛾𝐷𝑚+ 𝜀𝑖𝑡 (6)
29 variables. Equation (3) and (4) segregates domestic and foreign institutional ownership and examines their influence on the dependent variables, moreover it includes all control variables and the moderator variable. The same holds for equation (5) and (6), however these equations separate the impact of traditional and alternative investors. Here the moderator variable is excluded. The i stands for the entity i.e. the firm, and t stands for the time period i.e. a year. The 𝛼 is the intercept and the 𝜀 is the error term. The 𝐷 stands for the dummy and respectively the 𝑡 for the time (in years) and the 𝑚 stands for the industry-types.
5. Results
5.1 Descriptive Statistics
Table 5 shows the descriptive analysis results. The mean Tobin’s Q of the firms in this sample is 1.38. The average return on assets is 0.02 and the capital expenditures have a mean of 3.87. The average institutional ownership is 24.64, where domestic institutional ownership has a mean of 16.40 and foreign institutional ownership has a mean of 8.23. Moreover, traditional institutional owners hold a mean of 19.02, and alternative institutional owners hold an average 6.82 percent of shares in the sample firms.
The diversity of nationalities in boards has an average of 0.31. In addition, the average age natural logarithm of the firms is 3.91. The leverage mean is 0.25 and the mean of the natural logarithm of the assets is 13.06. Moreover, the average cash holdings are 0.11 and the average foreignness is 0.50. In terms of shares the average dividend yield was 3.34 and annual stock return 0.16.
Table 5 Summary Statistics
Type of Variable Mean Median Std
Dev
Min Max N
Panel A: Institutional ownership variables Total institutions 24.64 21.99 19.79 0.00 100.93 463 Domestic institutions 16.40 12.00 17.29 0.00 91.59 463 Foreign institutions 8.23 5.00 11.60 0.00 77.29 463 Traditional institutions 19.02 16.31 15.76 0.00 75.94 463 Alternative institutions 6.82 0.00 13.05 0.00 66.30 463
Panel B: Valuation, operating
performance and investment variable
Tobin’s Q 1.38 1.24 0.55 0.45 4.57 419
30
Capital Expenditures 3.87 2.54 4.08 0.00 27.06 450
Panel C: Moderator variable
International Board Diversity 0.31 0.28 0.29 0.00 0.85 464
Panel D: Firm level control variables
Market capitalization (natural log) 13.06 12.89 2.35 7.65 17.95 460
Foreign sales 0.50 0.59 0.36 0.00 1.20 464
Leverage 0.25 0.22 0.22 0.00 2.72 441
Cash 0.11 0.07 0.13 0.00 1.53 460
Dividend yield 3.34 2.58 6.85 0.00 118.75 412
Annual stock return 0.16 0.10 0.44 -1.00 2.72 362
Firm age (natural log) 3.91 3.83 0.91 0.69 5.81 464
Time dummy 0.13 0.00 0.33 0.00 1.00 464
Industry dummy 0.45 0.00 0.50 0.00 1.00 464
Source: Vliet (2017)
In order to check whether there is multicollinearity in the sample I use Variance Inflation Factor (VIF) in Eviews. I exclude domestic and foreign institutional ownership in the analysis for the reason that they are supposed to correlate. Moreover, I exclude traditional and alternative ownership for the same reason. The results range from -0.15 (between firm age and cash holdings) to 0.57 (between firm size and foreignness). The correlation coefficients are shown in the following Table 6. I assume no multicollinearity issues.
Table 6
Correlation Matrix of Variables
TOBINSQ ROA CAPEX BLAU SIZE FOREIGN LEV CASH DIV RETURN AGE TOBINSQ 1.00 -0.01 0.17 0.22 0.26 0.15 -0.09 0.22 -0.09 -0.07 -0.11 ROA -0.01 1.00 0.18 -0.06 0.27 0.25 0.00 -0.06 0.05 0.13 0.16 CAPEX 0.17 0.18 1.00 -0.02 0.12 0.07 0.25 -0.12 0.03 0.09 -0.09 BLAU 0.22 -0.06 -0.02 1.00 0.53 0.44 0.07 0.14 -0.10 0.06 0.02 SIZE 0.26 0.27 0.12 0.53 1.00 0.57 0.12 -0.03 0.03 0.02 0.24 FOREIGN 0.15 0.25 0.07 0.44 0.57 1.00 -0.01 0.03 -0.01 0.12 0.12 LEV -0.09 0.00 0.25 0.07 0.12 -0.01 1.00 0.04 0.01 -0.06 -0.05 CASH 0.22 -0.06 -0.12 0.14 -0.03 0.03 0.04 1.00 -0.08 0.09 -0.15 DIV -0.09 0.05 0.03 -0.10 0.03 -0.01 0.01 -0.01 1.00 0.06 0.15 RETURN -0.07 0.13 0.09 0.06 0.02 0.12 -0.06 0.09 0.06 1.00 -0.03 AGE -0.11 0.15 -0.09 0.02 0.24 0.12 -0.05 -0.15 0.15 -0.03 1.00 Source: Vliet (2017) 5.2 Accounting Performance
