“Executive turnover and nationality
diversity: Evidence from the 2007-2009
financial crisis”
By: Rick Nieters
Student Number: s1640658
Email:
r.b.nieters@student.rug.nl
MSc International Business & Management
Groningen, 30 August 2012
University of Groningen
Faculty of Economics and Business
P.O. Box 800, 9700 AV Groningen, the Netherlands
First supervisor:
Dr. K. van Veen
ABSTRACT
This study checks whether the trend toward more multinational boards has continued during the recent financial crisis in a sample of the 50 largest financial TNCs in the world. In addition, it examines the effect of three crisis-specific variables (government intervention, write-downs and capital raisings) on executive turnover. In order to capture these effects, data on the firms’ top managers was collected for the years 2006 and 2009. Due to data availability, write-downs and capital raisings were measured from the first quarter of 2007 until the third quarter of 2008.
The results suggest that government intervention is positively related to executive turnover. Hence, governments were not hesitant to replace poorly performing executives, when they had the chance and rights to do so. Further, the hypothesized, positive effects of write-downs and capital raisings on turnover were not found.
In addition, government intervention has a significant, negative effect on the nationality diversity in TMTs. This supports the notion that governments have a home-bias in their activities. But, despite the negative effect of government intervention, the overall trend shows an increase in nationality diversity. The proportion of non-nationals on TMTs increased from 23% in 2006 to 26% in 2009. So the TMTs of the world’s largest financial TNCs have become more multinational during the crisis. Moreover, there is no reason to believe that the social forces spurring this trend toward more multinational boards are likely to disappear soon.
Keywords: Top Management Teams (TMTs), executive turnover, nationality diversity,
ACKOWLEDGEMETS
With organizational and strategic leadership theories as my prime interest during my studies, I wanted to combine a study of executive turnover with an examination of nationality diversity in the boards of large multinational corporations. Such research is an appropriate response to the globalizing trend in the modern economy in which multinational corporations play an ever more significant role. Moreover, studying such corporations and their management teams has been a great, informative way to finish my studies in the form of this thesis.
TABLE OF COTETS
Introduction 5
Literature Review 9
The 2007-2009 financial crisis 9
Write-downs and capital raisings 10
Systemic failure 10
Government intervention 11
The consequences for nationality diversity 12
Research Methodology 15
Sample 15
Time frame 15
The top management team 15
Data collection 16
Variables 16
Control variables – executive turnover 18
Industry effects 20
Control variables – nationality diversity 21
Method of analysis 22
Results 24
Sample description 24
Executive turnover 24
Nationality diversity 28
Discussion & Conclusion 31
Executive turnover 31
Nationality diversity 32
Limitations of the study 34
Directions for future research 35
References 36
ITRODUCTIO
An unprecedented large number of financial institutions have collapsed or were bailed out by governments worldwide since the beginning of the global financial crisis in 2007. Lehman Brothers went bankrupt, just like 25 smaller US banks, and Merrill Lynch, AIG, Freddie Mac, Fannie Mae, HBOS, Royal Bank of Scotland, Bradford & Bingley, Fortis, Hypo and Alliance & Leicester all came within a whisker of doing so and had to be rescued1.
Many observers attribute these events to failures in corporate governance, such as lax board oversight and flawed executive compensation practices that encouraged aggressive risk taking. Kashyap, Rajan & Stein (2008), for example, argue that the root cause of the crisis lies in the breakdown of shareholder monitoring and ill-conceived managerial incentives. In this way, corporate governance did not prevent the losses incurred during the crisis, but instead intensified them by encouraging executives to focus on short-term performance. Erkens, Hung & Matos (2009) argue that this short-term focus of boards and investors has not only encouraged risk taking, but has also led to the replacement of poorly performing CEOs. This can be seen in figure 1: While financial firms exhibited CEO turnover rates below those of non-financial firms in the 2004-2006 period, this pattern reversed in the 2007-2008 crisis period.
Figure 1: CEO turnover rates for financial versus non-financial firms (Erkens et al., 2009)
1
In their study, Erkens et al. (2009) find that approximately 24 percent of the financial services companies in their sample experienced CEO turnover from 2007 to 2008, an increase of more than 10 percent as compared to the years before. Also the turnover of other corporate board directors was relatively high. According to the authors, this proportion increased to 24 percent in 2007 and 2008. Karlsson & Neilson (2009) find similar results in their study of 578 financial services companies. They reveal that 18 percent of these companies lost their CEO in 2008. Of these successions, more than half were dismissals – a rate of forced succession 158.8 percent higher than the historical average.
Overall, these studies show a significant increase in executive turnover in boards of financial firms during the crisis period. However, they hardly try to explain why and how the differences in turnover rates in boards have developed. This is remarkable because the decision to replace executives is arguably among the most important and difficult decisions made by a board of directors. It has long-term implications for a firm’s investment, operating, and financing decisions, and can often lead to distraction from the company’s core strategies.
So why do executives lose their jobs?
The majority of earlier studies on executive turnover have focused on the accountability of leaders for organizational outcomes. In these studies, performance has been the most frequently studied determinant of turnover, and a negative relationship between performance and turnover has been found consistently (Finkelstein, Hambrick & Cannella, 2009). Other causes of executive turnover have been found in individual characteristics such as age (Weisbach, 1988; Murphy & Zimmerman, 1993; Goyal & Park, 2002) and tenure (Wagner, Pfeffer & O’Reilly, 1984; Fredrickson, Hambrick & Baumrin, 1988), in board characteristics such as board size (Fredrickson et al., 1988; Helmich, 1980) and board composition (Salancik & Pfeffer, 1980; Weisbach, 1988), in organizational characteristics such as firm size (Grusky, 1961; James & Soref, 1981) and ownership structure (McConnell & Servaes, 1990; Denis, Denis & Sarin, 1997; Shleifer & Vishny, 1986), and in environmental characteristics such as industry performance and competition (Meindl, Ehrlich & Dukerich, 1985; DeFond & Park, 1999). As the studies above illustrates, executive turnover has been the subject of a huge volume of research and many antecedents have already been investigated.
and overall bank performance from July 2007 to December 2008 was the worst since the Great Depression (Beltratti & Stulz, 2010). As a result, governments in many countries had to adopt measures to combat the looming meltdown of the financial system. Among the most frequently adopted measures around the world were deposit guarantees, capital injections and bad bank schemes (Dietrich & Hauck, 2012). Some of these measures provided state aid conditional on some future events, such as a bank failure. In some cases, governments even insisted on changes in top management as a condition for a bank to receive a capital injection (Erkens et al., 2009). Royal Bank of Scotland, for instance, cleared out its boardroom in 2008 as part of a shake-up, demanded by the British government, for a state-assisted recapitalization of the bank. Such events have undoubtedly affected executive turnover rates and can help to explain the increased turnover observed during the crisis. Moreover, investigation of the crisis period can help to determine the extent to which executives are held responsible for the economic turmoil. Therefore, the first part of this paper will study the determinants of executive turnover, by focusing, in particular, on crisis-specific factors.
But not only the causes of executive turnover are worthwhile to study; its consequences are equally interesting. Information on this topic is useful for organizations and their stakeholders because turnover at the top changes the composition of the top management team (TMT), which is assumed to affect the future direction of the organization. This assumption is based on Hambrick & Mason’s (1984) upper echelon theory, which argues that individual attributes influence the values and ideas of TMT members, as well as the resulting team dynamics. In turn, these affect the strategic choices managers make, and therefore, organizational outcomes (Smith, Smith, Olian, Sims, O’Bannon & Scully, 1994; Finkelstein & Hambrick, 1996; Tsui & Gutek, 1999).
Are boards becoming more ‘global’?
diversity in executive boards between these countries and companies. A situation also observed by Greve, Nielsen & Ruigrok (2009) who added that over time, diversity levels not only steadily rise at all places but can also locally drop after a while. The authors suggest that these variations are due to differences in the degree of internationalization of firms2.
From a sociological perspective, a further understanding of this process is interesting because nationality diversity can be seen as an important indicator of the formation of a transnational business class (Carroll & Fennema, 2002; Carroll & Carson, 2003). Recent studies provide convincing evidence that such elite transnational networks have formed or are forming in Europe (Van Veen & Kratzer, 2011; Heemskerk, 2011). An increase in board globalization might stimulate this process worldwide by increasing the number of cross-border interlocking directorates (Staples, 2006). In order to inform this discussion, the second part of this study will focus on the consequences of the financial crisis for board globalization and the resulting changes in nationality diversity.
The analyses in this study are based on a large dataset of the 50 largest financial TNCs in the world, which contains information on TMT members, company characteristics and financial performance during the period 2006-2009. I decided to focus on TMTs because these teams are seen as the most important group of decision-makers in a company (Murray, 1989; Pettigrew, 1992). No other small group has nearly as much effect on the form and fate of an enterprise (Finkelstein et al., 2009). Moreover, evidence suggests that studying TMTs, rather than CEOs alone, provides better predictions of organizational outcomes (Finkelstein, 1988; Ancona, 1990; O’Reilly, Snyder & Boothe, 1993; Trushman & Rosenkopf, 1996).
2
LITERATURE REVIEW
The present study examines the effect of the 2007-2009 financial crisis on executive turnover, and the subsequent change in nationality diversity. In order to get a better understanding of this period, it is relevant to start with a brief introduction.
The 2007-2009 financial crisis
The financial crisis was triggered in the first quarter of 2006 when the U.S. housing market turned. A number of the mortgages designed for a subset of the market, namely subprime mortgages, were designed with a balloon interest payment, implying that the mortgage would be refinanced within a short period to avoid the jump in the mortgage rate. The mortgage refinancing presupposed that home prices would continue to appreciate. Thus, the collapse in the housing market necessarily meant a wave of future defaults in the subprime area, which led to unprecedented losses throughout the financial industry, particularly for banking institutions with significant exposure to the U.S. real estate market (Acharya, Philippon, Richardson & Roubini, 2009).
As the number of mortgage delinquencies soared, the crisis produced large write-downs on risky mortgage-related positions, including loans, mortgage- and asset-backed securities, and related derivatives. One particularly noteworthy example was the announcement of write-downs totaling $19 billion by Citigroup in the final quarter of 2008. This led to the largest ever loss in the bank’s 197-year history (Beltratti & Stulz, 2010). Moreover, between mid-2007 and mid-2009, financial institutions worldwide reported many hundreds of economically significant write-downs totaling approximately $1.3 trillion (Acharya et al., 2009).
Write-downs and capital raisings
This introduces the first set of factors that could have had an impact on executive turnover: write-downs and capital raisings. Where write-downs capture the losses associated with the risky investments in mortgage-backed securities, capital raisings are a good proxy for the extent of losses, in that the firm had a need to raise distressed capital (Erkens et al., 2009). Many studies attribute these losses to the risk taking behavior of banks (e.g. Dooley, Folkerts-Landau & Garber, 2009; Stiglitz, 2010). The idea is that riskier banks were more engaged in excessive risk taking causing them to make larger losses. Following this line of reasoning, write-downs and capital raisings (which capture the crisis-related losses) can serve as a proxy for the level of risk incurred by financial firms.
In turn, the level of risk incurred by firms has been found to be negatively associated with the tenure of executives (Pfeffer & Leblebici, 1973; Gilson, 1990). Or, to put it differently, positively associated with the turnover of executives. This makes sense, because taking large risks may result in large losses, for which executives are held responsible. Therefore, I expect higher turnover in banks that had to undertake larger write-downs and capital raisings, as these factors indicate excessive risk taking. This translates into the following hypotheses:
H1: Write-downs are positively related to executive turnover.
H2: Capital raisings are positively related to executive turnover.
Systemic failure
As a result of the large devaluations of assets, many banks suffered from a debt overhang. The biggest concern for the financial industry, then, was that the failure of single banks would spread through the entire financial system causing substantial disintermediation3 and a credit crisis (Dietrich & Hauck, 2012).
The most identifiable event that eventually led to such systemic failure was the collapse on June 20, 2007, of two highly levered Bear Stearns-managed hedge funds that invested in subprime asset-backed securities (Acharya et al., 2009). The subsequent bankruptcy of Leman Brother in September 2008 triggered a further confidence crisis in the financial sector. The market realized that if Lehman Brothers was not too big to fail, then that might be true for
3
other investment banks as well. From then on, not just the liquidity, but also the solvency of financial institutions was suddenly questioned. This led to classic bank runs on several financial institutions, irrespective of the fact that they were most likely more solvent than Lehman Brothers. In hindsight, Lehman Brothers contained considerable systemic risk and led to the near collapse of the financial system (Acharya et al., 2009). Arguably, this stopped only when the U.S. government announced its first bailout plan, the Troubled Asset Relief Program (TARP).
Government intervention
This introduces the next feature of the financial crisis that could have had an impact on executive turnover: government intervention. In order to stabilize the financial system during late 2007 and early 2008, governments not only intervened in the United States, but also in the United Kingdom and several other Western European countries, including Belgium, France, Germany, Ireland, Luxembourg and the Netherlands. They purchased large amounts of illiquid, risky mortgage-backed securities from financial institutions, and provided deposit guarantees and bad bank schemes (Dietrich & Hauck, 2012). As mentioned before, in some of these cases, governments insisted on changes in top management as a condition for a company to receive a government bailout (Erkens et al., 2009). In other cases, governments simply replaced executives after nationalizing a bank. For example, after obtaining full control of ABN AMRO in 2008, the Dutch government appointed former Dutch finance minister Gerrit Zalm as CEO to restructure and stabilize the bank. In this way, governments have triggered resignations at some banks and placed new board members at others.
Additionally, stock markets tend to respond favorably to executive changes in financially distressed firms, because the succession is seen as a step to lead the company out of financial distress (Bonnier & Bruner, 1989; Davidson, Worrell & Dutia, 1993). Hence, governments may replace executives of bailed out firms in order to help to restore investor confidence.
All in all, this leads to the expectation that:
H3: Government interventions are positively related to executive turnover.
Next to these crisis-specific determinants of executive turnover, I will include several control variables in my analyses to account for the effects of well-known predictors of turnover, such as performance, firm size, ownership concentration, and the average tenure of executives. These variables and their relations to turnover will be covered in the section ‘Research Methodology’. For now, I will turn my attention to nationality diversity.
The consequences for nationality diversity
Inherent to executive turnover are changes in the composition of the TMT. If executive turnover increases, more executives are replaced, and as a result, more changes in TMT composition may come about. The relevant question then remains as to whether or not the team will become more diverse as a result of the change.
Diversity advocates would assume that the boards of the world’s largest corporations would take on more foreign executives. This assumption comes along with the recommendation that the boards of top corporations should be more globally diverse in order to deal with the increasing international and complex business environment. The underlying rationale here is that diverse groups make better decisions in a complex environment, because diversity enhances the breadth of perspectives, cognitive resources and overall problem-solving capacity (Hambrick, Cho & Chen, 1996; Barsade, Ward, Turner & Sonnenfeld, 2000).
that people consciously and unconsciously prefer others who are similar to them is one of the most robust and reliable social psychological findings (Barsade et al., 2000).
During such a turbulent period as the recent financial crisis, I expect firms to follow this last, ‘save’ route and favor domestic managers over foreign managers. These managers are familiar with the company’s home country, and do not need extra time to adapt to a new country, culture and language. Instead, they can have a direct impact on the organization and try to improve the situation from the moment they are appointed. Therefore, it is hypothesized that:
H4: Executive turnover during the crisis has a negative impact on the nationality diversity in TMTs.
In addition, the earlier discussion on government intervention suggests that governments have initiated exits and entries of executives as well. As a result, they might have influenced the diversity levels in the TMTs. However, the way in which this might have occurred is unknown and at least debatable.
On the one hand, governments may be more ethnocentric oriented in times of crises due to negative sentiments toward other countries. A study by Ang, Jung, Kau, Leong, Pornpitakpan & Tan (2004) showed this effect to hold for consumers in the context of the Asian economic crisis in the late 1990s. Their results indicated that the more severely hit a country was, the more ethnocentric respondents were. If governments would act in the same way, their interventions would lead to fewer appointments of foreign managers. On the other hand, governments may acknowledge that the presence of intricate knowledge on foreign markets in top management teams is beneficial for TNCs (Athanassiou & Nigh, 2000; Tan & Mahoney, 2006), and therefore take on more foreign directors to improve performance.
managers, thereby lowering the nationality diversity in TMTs. This results in the following hypothesis:
H5: Government interventions have a negative impact on the nationality diversity in TMTs.
Last but not least, I expect that the trend toward multinational boards has been halted in my sample of financial TNCs. These companies experienced substantial turnover during the crisis and some of them needed government assistance to survive. Following the directions of H4 and H5, this leads to the expectation that:
H6: ,ationality diversity has decreased during the 2007-2009 financial crisis.
RESEARCH METHODOLOGY Sample
Since the late 1980s, the United Nations Conference on Trade and Development (UNCTAD) has been conducting research on direct foreign investment, trade flows and TNCs. As part of this research, the UNCTAD published a list of the world’s 50 largest financial TNCs in 2008, which is presented in Appendix 1.
My initial sample consisted of all the companies from this list. This choice was made for two reasons. First, firms from the financial services industry were at the center of the economic downturn, so they are expected to be mostly affected by the financial crisis. By including these companies in my sample, the effects of the crisis can be examined. Secondly, firms from a single industry face similar pressures in their business environment, whereas business conditions across industries may differ substantially. Thus, by focusing on one specific industry, industry effects are isolated and comparisons can be better drawn (Tihanyi, Ellstrand, Daily & Dalton, 2000).
Only those firms for which complete information on nationalities and personal information was available for all TMT members were included in the final sample. This resulted in a list of 48 firms, spread across 15 countries. While this sample might seem small, it is only slightly smaller than in previous single industry studies in this field (e.g. Reuber & Fischer, 1997; Greve et al., 2009).
Time frame
The start of 2007 is generally regarded as the period when the market first realized the severity of the losses related to sub-prime mortgages (Ryan, 2008). The massive government bailouts were initiated from October 2008 and onwards. In 2009, the financial markets stabilized and most analysts marked the official end of the financial crisis (Sherman, 2011). Because I want to draw a comparison between the situation before the financial crisis and the situation after the crisis, the investigation period starts in 2006 and ends in 2009.
The top management team
two-tier board structure the members of both the executive board and the supervisory board are included. This approach makes the board size for companies with a two-tier board structure comparable to companies with a single board structure.
Managers in lower positions are equally interesting, but these data are very difficult to collect on such a large scale. Therefore, I decided to focus exclusively on the highest tiers of management. This choice resulted in a dataset of exactly 1000 managers in 2006 and 957 managers in 2009.
Data collection
Data on the firms’ top managers was primarily collected from company websites and annual reports. In addition, articles from the Financial Times, Business Week and other media sources were consulted for additional information on executives. Most financial information was gathered from DATASTREAM4 and the Bloomberg WDCI menu, which covers banks, brokers, insurance companies, and government-sponsored entities (such as Freddie Mac and Fannie Mae).
Variables
In this section, the independent and dependent variables that were used in the analyses are discussed and operationalized.
Dependent variable I: executive turnover
Executive departures come in several forms, including death, illness, mandatory retirement, early retirement for personal reasons, leaving for an executive position in another company, and dismissal. These various forms could be modeled separately, but this introduces further methodological problems, such as determining whether job changes or retirements were voluntary or involuntary. Furthermore, firms may not reveal the true reasons of a departure. A dismissal, for example, may be said to be an early retirement to save the reputation of the manager or to avoid concerns about the company’s future. Therefore, I studied executive turnover without regard for the form of turnover.
4
Executive turnover was calculated by comparing the groups of executives in 2006 and 2009 and taking the proportion of executives that were replaced. By using relative instead of absolute values, I account for differences in TMT sizes between companies as well. For example, 10 replacements are more substantial for a TMT consisting of 15 members than for a TMT consisting of 30 members.
Dependent variable II: nationality diversity
In this research, nationality was considered to be the legal status of the executives as reported by the companies. When this data was not available, I searched for a person’s country of birth on other sources like Business Week or Reuters. In case of any more doubts, additional indicators were taken into account, such as the country where education was taken, and the country-of-origin of previous employers. Consistent with previous research, I defined a ‘non-national’ or ‘foreign’ TMT member as an individual whose citizenship was different from the company’s legal domicile (Alexander & Esser, 1999). In the few cases where individuals had a dual citizenship, I coded them as a ‘national’ instead of a ‘non-national’ director in order to produce a conservative estimate of TMT globalization.
The proportion of foreigners in the TMT was used as a proxy for nationality diversity. Subsequently, the change in nationality diversity from 2006 (t1) to 2009 (t2) was used as
dependent variable in the analyses (t1 – t2).
Independent variable: government intervention
Government intervention is defined as the situation in which a government offers money to a failing organization in order to prevent the consequences that arise from an organization’s downfall. Such ‘bailouts’ can take the form of loans, bonds, stocks or cash, and they may or may not require reimbursement. The variable government intervention takes a value of 1 if the bank received a government bailout and 0 otherwise. A list of the banks that received government assistance is presented in Appendix 2.
Independent variables: write-downs and capital raisings
articles, and company press releases (such as quarterly earnings announcements). Write-downs and capital raisings were measured from the first quarter of 2007 until the third quarter of 2008 due to data availability. An overview of these numbers can be found in Appendix 3.
Control variables – executive turnover
In order to control for other factors that may influence executive turnover, the following control variables have been used: firm performance, firm size, ownership concentration, TMT size, TMT composition, executives’ average tenure and age.
Firm performance
Empirical evidence generally suggests that poor corporate performance precedes CEO departure and the departure of lower-level executives. Some studies have documented the effects of poor stock returns on executive departure (Benston, 1985; Warner, Watts & Wruck, 1988), where others have relied on profitability measures as predictors of executive departure (James & Soref, 1981; Wagner, Pfeffer & O’Reilly, 1984; Harrison et al., 1988). Essentially, all the studies have sufficiently arranged the chronology of their data to allow the conclusion that the poor performance preceded the departures. This issue of temporal order becomes more problematic when considering research on the effects of executive departures on performance, but that is beyond the scope of this research.
A widely used indicator of firm performance in succession research is return on assets (ROA). This variable is measured as a firm’s net income divided by total assets (Murphy & Zimmerman, 1993; Shen & Cannella, 2002; Arthaud-Day, Certo, Dalton & Dalton, 2006). The ROAs of 2007, 2008, 2009 were added together to measure the firms’ performance during the crisis years.
Firm size
Firm size has been previously measured using total sales, assets, and the number of employees (Harrison et al., 1988). The choice for one of these measures is probably not critical, since the three measures of size are highly correlated. Due to data availability, I have used the logarithm of total assets to control for firm size. By doing so, I follow the standard prescription of using the log of size (Blau, 1970).
Ownership concentration
Denis et al. (1997) examined ownership structure and its impact on the firm performance– departure relation. They concluded that executive turnover was positively associated with the presence of at least one large outside blockholder. Blockholders were said to be “the most direct way to align cash flow and control rights of outside investors” and have the incentives and the cost-efficiency means of monitoring a company’s board. Therefore, I examined whether my baseline results are robust to controlling for the firm’s ownership concentration. This variable was measured as the fraction of shares owned by blockholders and institutions owning 5 per cent or more at the end of 2006.
TMT size
While additional directors can improve monitoring, they may also affect a firm’s mode of operations. Clendenin (1972), for example, observed that large boards were unmanageable, because as the size of the board increased, it became less cohesive. This lack of cohesion typically manifests itself in factions within the board. The CEO’s preferred strategies will be more difficult to sell to a factionalized board, his or her performance may be subject to diverse and potentially conflicting criteria, and the support of one group of board members may alienate another group (Fredrickson et al., 1988). As a result, board allegiances and values will be divided, making it more likely that a CEO will become a victim of the board’s internal disagreements. These disagreements may also lead to the subsequent replacement of other executives. Therefore, TMT size is used as a control variable and is measured by the number of individuals in each TMT.
TMT composition
First of all, inside directors have stronger loyalty to the firm’s leadership than do outside directors, since insiders are employees of the firm (Harrison et al., 1988). Moreover, outsiders are not likely to be encumbered in the operational activities of the firm because they do not have the required firm-specific knowledge. This suggests that outside directors must focus on the firm’s bottom-line performance and respond to any dissatisfaction by taking one of the few actions available to them – dismissing executives. In order to control for outside board members, TMT composition is used as a control variable and is measured as the number of independent, outside directors on a board divided by the board’s size (Daily & Dalton, 1994; Ocasio, 1994; Shen & Cannella, 2002).
Average tenure
Just as a TMT’s size affects its cohesion, so does its average tenure. Boards whose members have shared a reasonably long service to the firm are likely to exhibit strong ties (Fredrickson et al., 1988). Such boards are expected to be more cohesive than those in which members vary widely in their tenure, or those in which members have short average tenures. Boards in which members have wide variations or short average tenures are more likely to be factionalized, and their members will tend to evaluate management on a variety of potentially conflicting criteria (Fredrickson et al., 1988). These factors are expected to influence executive turnover rates, so I took the average of the number of years each TMT member served on the TMT to account for tenure.
Average age
Another variable that is highly correlated with the probability of a resignation is the age of an executive. Weisbach (1988), Murphy & Zimmerman (1993), and Goyal & Park (2002) all found a strong positive relation between executive turnover and age. Since these resignations are likely to be actual retirements, unrelated to performance, average age is included as a control variable in the analyses.
Industry effects
management (Meindl et al., 1985). Others argue that in competitive industries, it is relatively easy for boards to identify unfit CEOs, resulting in a higher frequency of CEO turnovers (DeFond & Park, 1999). Higher levels of competitive uncertainty also put more pressure on top managers and make their jobs more difficult, thereby leading to higher turnover rates (Harrison et al., 1988). These characteristics may thus explain a great deal of variance in executive turnover across industries.
This study, however, is carried out in a single industry setting. In this respect, the financial services sector provides a relatively homogeneous platform where all companies face similar pressures from the external environment. Hence, there is no need to account for differences across industries.
Control variables – nationality diversity
Also in the second analysis I have included several control variables to control for other factors that may influence nationality diversity. These are: firm size, the number of host countries a company is active in, and involvement in cross-border mergers and acquisitions (M&As).
Firm size
According to Van Veen & Marsman (2008), firm size can have an independent effect on nationality diversity. Large companies are often (cross)listed at several stock markets, which is associated with a higher firm visibility and, hence, investor awareness. As a result, larger companies may attract personnel from a larger pool of candidates, both domestic and foreign, which will eventually lead to a higher diversity of their personnel. In order to account for this effect, I have included firm size as a control variable in the analyses. Again, the logarithm of total assets was used to measure this variable.
,umber of host countries
have a positive effect on nationality diversity, the number of host countries is used to control for the extent to which a company is operating internationally.
Cross-border M&As
Cross-border M&As occur when a corporation from one country acquires or merges with a corporation from another country. As one firm is absorbed by or merged with another, the two boards are merged in various ways too, of which the outcomes can affect the nationality diversity of the future company. In his study on the 148 largest TNCs and commercial banks, Staples (2008) found evidence for this relationship. He concluded that a large cross-border M&A almost always results in a more multinational board of directors. In order to account for this effect, the involvement in large cross-border M&As is used as a control variable. This variable takes a value of 1 if the company was involved in a cross-border M&A with a value of over $1 billion in the period 2006-2009 and 0 otherwise. Following Staples (2008), I excluded M&As valued at under $1 billion because such deals are not expected to have much of an impact, if any, on board composition.
Method of analysis
In order to identify possible relationships between the independent and dependent variables, a multiple regression analysis is carried out in SPSS5. This analysis tests whether and how the independent variables are related to the dependent variable. Hence, not only the existence and significance of the relationships are identified, but also the direction is shown (Hair, Black, Babin, Anderson & Tatham, 2006).
Furthermore, this type of regression analysis allows for known predictors (i.e. the control variables) to be entered into the regression model first (Model 1). These variables are kept constant to prevent their influence on the effect of the independent variables on the dependent variable. Subsequently, write-downs are added to the equation to test the first hypothesis (Model 2). The third model includes capital raisings in order to test the second hypothesis. In the fourth model government intervention is added to check whether H3 holds. Lastly, the fifth model contains the control variables, government intervention, write-downs, and capital raisings to assess the simultaneous influence of all variables on the dependent variable.
5
Testing these relationships may provide useful information about the determinants of executive turnover.
In the second step of this study, I conduct another 4 regression analyses to examine the drivers of nationality diversity. The first model contains the control variables firm size, number of host countries, and involvement in cross-border M&As. The second model tests H4 by adding executive turnover to the equation. Model 3 tests the effect of government intervention on nationality diversity (H5), while Model 4 includes all variables in order to assess their effects on nationality diversity simultaneously.
RESULTS
This section will start with some descriptive statistics of the sample and the variables. Subsequently, the specific tests related to the hypotheses are presented and discussed.
Sample description
From the initial sample of 50 financial institutions, 2 were excluded from the analyses. The first one, Merrill Lynch, ceased to exist as a separate entity after its sale to Bank of America on September 14, 2008. The second one, Bank of New York Mellon, was not formed yet in 20066. As a result, 48 firms from 15 countries were included in the final sample. 20 of those firms received government aid during the crisis.
The United States and United Kingdom contribute the most firms to the sample. Both countries are represented by 6 companies (12.5%). Ireland, Norway and Denmark have the smallest number of companies in the sample, namely 1 (2.1%). An overview of the other countries involved in the sample can be found in Appendix 4.
The TMTs in the sample vary from 36 members to 11 members, resulting in an average TMT size of approximately 20 people. The maximum number of foreigners in a team is 16, and the average number is 4.83. This means that on average almost 5 foreigners can be found in a TMT. Moreover, the average TMT member is 57.6 years old and stays almost 5 years in his or her position. A summary of the descriptive statistics per variable is presented in Appendix 5.
Executive turnover
In the first series of tests, the determinants of executive turnover are examined. Correlation and multiple regression analyses have been used to test the hypotheses.
Correlation matrix
Table 1 below presents the Pearson correlation coefficients among the key variables used in the first series of tests. These coefficients vary from +1 to -1. If the value is 0, the variables are not related. If the value is positive, the variables are positively correlated: one increases
6
when the other increases. If the value is negative, the variables are negatively correlated: one increases when the other decreases and vice versa.
Table 1: Correlation matrix I (Pearson correlation)
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 1. Executive turnover 2. TMT size 0.11 3. TMT composition -0.18 -0.24 4. Average tenure -0.33* -0.25 0.34* 5. Average age -0.16 -0.16 0.31* 0.47** 6. Firm size 0.05 0.37* 0.06 -0.17 0.01 7. Firm performance -0.60** -0.22 0.32* 0.27 0.18 -0.03 8. Ownership concentr. -0.21 0.04 -0.18 -0.06 0.05 -0.25 0.13 9. Write-downs 0.41** 0.33* -0.14 -0.09 0.00 0.30* -0.48** -0.21 10. Capital raisings 0.30** 0.35* 0.07 0.11 0.01 0.37* -0.66** -0.22 0.42** 11. Government interv. 0.38** 0.13 -0.20 -0.10 -0.08 0.22 -0.44** 0.17 0.37** 0.40** *. Correlation is significant at the 0.05 level (2-tailed)
**. Correlation is significant at the 0.01 level (2-tailed)
The coefficients indicate that the variable ‘executive turnover’ is negatively and significantly related to average tenure and firm performance. The same variable is positively related to write-downs, capital raisings and government intervention. Furthermore, there seems to be a positive relationship between TMT size and firm size, write-downs and capital raisings. Write-downs and capital raisings are also positively related to the size of the organization. Firm performance has a significant, negative effect on write-downs, capital raisings and the occurrence of a government bailout. Last but not least, write-downs, capital raisings and government intervention are positively related to each other.
Empirical results
In addition, it must be checked that the independent variables are not too strongly correlated, because in such a situation the coefficient may change inconsistently in response to small changes in the model or data (Hair et al., 2006). The collinearity analysis in SPSS produces a Tolerance and VIF value. The Tolerance can be calculated by extracting the R² from 1 and explains the extent to which the variance of that independent variable cannot be explained by the other independent variables. This value should not be too low. The VIF value is calculated as 1 / Tolerance, and should not exceed 10. As can be seen in Appendix 6, this is not the case for any of the independent variables. Hence, there are no multicollinearity issues.
As mentioned before, I have used 5 models to test the hypotheses concerning executive turnover. Model 1 includes the control variables: firm performance, firm size, ownership concentration, TMT size, TMT composition, average tenure and average age. Model 2 adds write-downs to the equation. Model 3 includes capital raisings, and Model 4 accounts for government intervention. Model 5 contains all variables. The results are presented in Table 2.
Table 2: Results of the regression analyses I
Dependent variable: Executive turnover
Independent variables Model 1 Model 2 Model 3 Model 4 Model 5
Write-downs 0.198 (1.031) 0.439 (1.129) Capital raisings 0.124 (0.616) 0.317 (0.776) Government intervention 0.254* (1.745) 0.257* (1.738) R² 0.418 0.433 0.423 0.460 0.480
Note: The standardized coefficients are displayed (t-values between brackets) *. Correlation is significant at the 0.05 level (2-tailed)
The standardized coefficients (betas) listed above indicate the relative importance of each independent variable. A greater difference from zero means a stronger influence on the dependent variable. Furthermore, the p-values provided by SPSS are based on a two-tailed test. These should be divided by 2, because the hypothesized relationships are directional (i.e. they go one specific way). A p-value lower than 0.05 indicates a significant relationship between the independent and dependent variable, based on a 95% confidence level.
a negative relationship between firm performance and turnover, and a negative relationship between average tenure and turnover.
In order to determine the goodness of fit of the regression analyses, the R² value can be used. This coefficient provides a measure of how well future outcomes are likely to be predicted by the model (Hair et al., 2006). The R² of the first regression analysis is 0.418. In other words, the independent variables included in this analysis explain almost 42% of the observed executive turnover. The other 58 percent can be explained by variables not included in this model. After adding write-downs to the model, the R² increases slightly to 0.433. The variable capital raisings contributes less to the model, resulting in an R² value of 0.423. The biggest R² improvements are found for the models that include government intervention. This underlines the significant impact of government intervention on executive turnover.
ationality diversity
In the second series of tests, the determinants of nationality diversity are examined. Again, correlation and multiple regression analyses have been used to test the hypotheses.
Correlation matrix
Table 3 below presents the Pearson correlation coefficients among the key variables used in the second series of tests.
Table 3: Correlation matrix II (Pearson correlation)
1. 2. 3. 4. 5. 6.
1. ∆ Nationality diversity
2. Firm size 0.05
3. Number of host countries 0.11 0.52**
4. Involvement in M&As 0.24 0.16 0.20
5. Executive turnover -0.04 0.05 0.09 0.04
6. Government intervention -0.19 0.22 0.17 0.19 0.38** **. Correlation is significant at the 0.01 level (2-tailed)
a positive relationship between government intervention and executive turnover. The other variables seem to be uncorrelated, at least in this correlation matrix. The positive coefficients between the number of host countries and involvement in cross-border M&As on the one hand, and the change in nationality diversity on the other hand, were expected but these are not significant. Further analysis is required to test the hypotheses regarding nationality diversity.
Empirical results
I have conducted 4 regression analyses to test the second series of hypotheses. Similar to the earlier-explained procedure, the first model comprises the control variables: firm size, number of host countries, and involvement in cross-border M&As. In the second model, executive turnover is added to the equation to test H4. Subsequently, the effect of government intervention is examined in Model 3, which may provide evidence for H5. Finally, the fourth model includes all variables to assess their effects on nationality diversity simultaneously. The results are presented in table 4 below.
Table 4: Results of the regression analyses II
Dependent variable: ∆ Nationality diversity
Independent variables Model 1 Model 2 Model 3 Model 4
Firm size (t-value) -0.031 (-0.178) -0.030 (-0.175) 0.014 (0.080) 0.016 (0.095) Number of host countries 0.083
(0.481) 0.088 (0.501) 0.097 (0.573) 0.094 (0.549) Involvement in M&As 0.225 (1.507) 0.227 (1.501) 0.266* (1.800) 0.267* (1.790) Executive turnover -0.055 (-0.369) 0.047 (0.298) Government intervention -0.263* (-1.770) -0.281* (-1.736) R² 0.061 0.064 0.125 0.127
Surprisingly, most of the control variables are not significantly related to the change in nationality diversity. The only significant control variable is ‘involvement in cross-border M&As’. This variable is positively related to the change in nationality diversity in Model 3 and 4, as was expected from existent literature. The inclusion of executive turnover as a predictor of nationality diversity in Model 2 seems to have no added value. The correlation coefficients are rather low and insignificant. Hence, more turnover does not necessarily lead to more (or less) diverse TMTs, which rejects H4. The fifth hypothesis, on the other hand, can be accepted: government intervention has a significant, negative effect on the change in nationality diversity. This is shown by the negative coefficients in Model 3 and 4, which are significant on a 95% confidence level (p-value < 0.05).
The R² values of these analyses are lower than those of the first series of analyses (on executive turnover). As a result, the extent to which future outcomes are likely to be predicted by the model is lower as well. In the first model, the independent variables explain only 6.1% of the observed change in nationality diversity. After adding government intervention to the model, this percentage increases to 12.5%. The other independent variable, executive turnover, contributes less to the model, resulting in an R² value of 0.064. In the final model, which includes all the variables, the R² increases to 0.127.
Last but not least, H6 was tested by comparing the nationality data of 2006 and 2009. This comparison resulted in several interesting findings. In 2006, 85.4% of the sampled companies had at least one non-national TMT member, while 7 firms had no ‘foreign’ TMT member at all. The same proportions were found in 2009. However, the proportion of companies with more than 3 non-national TMT members increased from 54.2% in 2006 to 58,3% in 2009. In addition, almost 26% of all TMT members of the sampled companies were non-nationals by 2009, up from 23% three years earlier. Switzerland is by far the most “globalized” country, with 62% of all its TMT members being non-nationals, followed by the Netherlands and Sweden (39%), the United Kingdom (37%), and Belgium (33%). Lastly, the proportion of TMTs that were dominated by non-nationals ( > 50% non-nationals) increased from 10.4% in 2006 to 12.5% in 2009.
DISCUSSIO & COCLUSIO Executive turnover
This study extends previous research on executive turnover by examining the impact of the 2007-2009 financial crisis and its distinct characteristics. More specifically, the role of government intervention was evaluated, and additional factors were included to capture the unprecedented losses caused by large exposures to subprime mortgage-backed assets and other complex structured credit instruments. In order to test the effects of these variables on executive turnover, three hypotheses were formulated.
The first and second hypothesis suggested that write-downs and capital raisings would be positively related to executive turnover. Both variables capture the losses associated with the risky investments done in the period before the crisis. These losses may be attributed to the excessive risk-taking of the firms’ executives, who in turn may be held responsible for the losses and be dismissed. However, the analyses (Model 2 and Model 3) did not provide empirical evidence for this line of reasoning. There were no signs of significant relationships between write-downs and capital raisings on the one hand, and executive turnover on the other. This suggests that the executives of the firms included in this study were not disciplined by their stakeholders for the large losses incurred during the crisis. It might be that these losses were seen as hard to prevent, out of the control of the executives, and as a failure of the global financial system as a whole.
The third hypothesis dealt with the impact of government interventions on executive turnover. It was hypothesized that government interventions would lead to higher turnover rates, because the need to rely on government assistance could be seen as a result of poor managerial decision making, and eventually lead to executive dismissals. The positive impact of government interventions on executive turnover was confirmed by Model 4 and Model 5. These models identified a strong, positive and significant relationship between the two variables. This suggests that governments were not hesitant to replace poorly performing executives, when they had the chance and rights to do so.
with what was expected from previous studies. Another significant control variable was average tenure. This variable had a significant, negative effect on executive turnover in Model 2, 4 and 5. Following the argumentation of Fredrickson et al. (1988), this can be explained by the strong ties that exist in boards in which board members have shared a long service to the firm. Such boards are expected to be more cohesive, and tend to experience lower turnover rates than those in which members have shared a short service to the firm.
Looking at the results of the analyses, it seems that stronger monitoring by boards and investors is not associated with stronger disciplining of executives after the crisis began. Executives were not held responsible for poor organizational performance, even when organizations were subject to large write-downs and capital raisings. Such failures in corporate governance (lax board oversight and shareholder monitoring) may have encouraged excessive risk-taking. However, when things got worse and governments had to intervene, action was taken against the sitting executives. A government bailout is a clear sign of serious problems, so it might be that organizations and governments had to undertake such measures in order to restore investor confidence and sustain their credibility and trustworthiness.
ationality diversity
In the second part of this study, I focused on the topic of TMT globalization. More specifically, I examined how the crisis had affected this phenomenon. Again, the role of governments was assessed, and I checked whether executive turnover had affected the diversity levels. The final analysis dealt with the question whether the trend toward multinational boards had continued during the crisis. To guide the analyses, three hypotheses were formulated.
Secondly, it was hypothesized that government intervention would have a negative impact on the nationality diversity in TMTs (H5). This hypothesis was confirmed after including government intervention in the regression analyses. Thus, the results indicate that when governments step in, the nationality diversity in TMTs goes down. This may be due to a more ethnocentric mindset in times of crisis, as observed by Ang et al. (2004) in the context of the Asian economic crisis. In addition, the findings support the notion that governments have a home-bias in their activities (Ganelli, 2005; Fujiwara & Van Long, 2012; Brülhart & Trionfetti, 2004), assuming that they have replaced executives after intervening.
Lastly, it was hypothesized that the nationality diversity in TMTs had decreased during the financial crisis (H6). A substantial number of firms needed government assistance to survive, so the negative impact of government intervention on nationality diversity was expected to be strong enough to lower the overall nationality diversity in my sample. However, the proportion of non-nationals on TMTs increased from 23% in 2006 to 26% in 2009. Furthermore, the proportion of TMTs that were dominated by non-nationals increased from 10.4% in 2006 to 12.5% in 2009. Hence, I had to reject H6, and conclude that the TMTs of the world’s largest financial TNCs had become more multinational during the crisis. Moreover, there is no reason to believe that the social forces spurring this trend are likely to disappear anytime soon. As companies increasingly seek investment opportunities worldwide, they will continue to look for directors with the kinds of international experience and expertise that will allow them to capitalize on these investments (Staples, 2007).
In conclusion, I have examined the question whether the 2007-2009 financial crisis has affected executive turnover and the nationality diversity in TMTs. A significant, positive relationship was found between government intervention and executive turnover. The hypothesized, positive effects of write-downs and capital raisings on turnover were not confirmed. In addition, government interventions were found to be negatively related to nationality diversity. Despite this negative effect, the TMTs of the world’s largest financial TNCs became more multinational during the crisis. These findings contribute to organizational and strategic leadership theories with a specific focus on the TMT.
insurance, for example, is a heavily regulated industry, where strict restrictions on licensing and the type of services provided are imposed on foreign companies (Greve et al., 2009). Furthermore, whereas banks and financial services firms do not face tariff barriers to foreign expansion (since they do not export material goods), they are more subject to non-tariff barriers such as government regulations and the domestic environment, including business, cultural and language differences (Rugman, 2005). Therefore, banks and insurance companies may be experiencing a stronger liability of foreignness than, for example, manufacturing industries. As a result, foreign top managers with specific knowledge of key countries, regions and useful local contacts play a crucial role in overcoming the liability of foreignness that a bank is facing. This may lead to relatively high levels of nationality diversity in the financial services industry compared to other industries. However, it is beyond the scope of this research to make such a claim as it would require more data from multiple industries.
Limitations of the study
Several limitations of this study could provide additional impetus for future research. First of all, the generalizability of my findings is low because they are based on a limited dataset from a single industry. Consequently, the results of this research might differ from outcomes found in other studies. Future studies could include companies from other industries and draw a comparison between them. This could bring new insights into the differences in turnover rates and nationality diversity across industries.
Secondly, the negative relationship between government interventions and nationality diversity might be a result of the actions of the bailed-out firms themselves, who might have replaced TMT members irrespective of government pressure. In that case, the conclusions about the governments’ influence on recruitment practices may not hold. Further investigation into this matter would be an interesting direction for future research. This, however, may be difficult because the variables are correlated anyway. Also, it would require information on the motives of each individual dismissal, which may be hard to acquire since firms are not obligated to communicate the genuine reasons behind a replacement.
or depart, from their jobs (Sonnenfeld, 1988). Such factors may explain a considerable part of executive turnover as well.
Lastly, the number of government interventions in my sample may be out of proportion. It is likely that only very large banks receive such assistance, since they are “too big to fail”. My sample consists of the 50 largest financial TNCs, so it presumably includes many of such institutions. A larger dataset with both smaller and larger firms may result in a more realistic number of government interventions. This will result in a more valuable measure of firm size as well, because differences between firms become larger.
Directions for future research
Research on executive succession shows no signs of slowing, and many interesting and challenging questions remain. Therefore, I note here several issues that are in need of further investigation.
In this study, write-downs and capital raisings were used to capture the losses related to subprime mortgage-backed assets and other complex structured credit instruments. These variables showed no significant relationship towards executive turnover. Future research could extend the measurement of losses and financial performance by using market-based indicators, such as cumulative stock returns or market-to-book ratios. These measures may yield better results as they capture the full extent to which the market believes the crisis has impacted the firm.
REFERECES
Acharya, V.V., Philippon, T., Richardson, M. & Roubini, N. (2009) “Prologue: A Bird’s-Eye View – The Financial Crisis of 2007-2009: Causes and Remedies”, in: Restoring Financial Stability: How to Repair a Failed System, Hoboken, New Jersey: John Wiley and Sons
Alexander, L. & Esser, S.U. (1999) “Globalizing the Board of Directors”, New York: The Conference Board
Allen, M.P. (1981) “Power and privilege in the large corporation: Corporate control and managerial compensation”, American Journal of Sociology, 86, 1112-1123
Ancona, D.G. (1990) “Outward bound: Strategies for team survival in the organization”, Academy of Management Journal, 33(3), 334-365
Ang, S.H., Jung, K., Kau, A.K., Leong, S.M., Pornpitakpan, C. & Tan, S.J. (2004) “Animosity towards economic giants: what the little guys think”, Journal of Consumer Marketing, 21(3), 190-207
Arthaud-Day, M., Certo, S.T., Dalton, C.M. & Dalton, D.R. (2006) “A changing of the guard: Executive and director turnover following corporate financial restatements”, Academy of Management Journal, 49(6), 1119-1136
Athanassiou, N. & Nigh, D. (2000) “Internationalization, Tacit Knowledge and the Top Management Teams of MNCs”, Journal of International Business Studies, 31(3), 471-488
Barsade, S.G., Ward, A.J., Turner, J.D.F., Sonnenfeld, J.A. (2000) “To your heart’s content: a model of affective diversity in top management teams”, Administrative Science Quarterly, 45, 802-836
Benston, G.J. (1985) “The self-serving management hypothesis: Some evidence”, Journal of Accounting and Economics, 7, 67-84
Blau, P.M. (1970) “The Formal Theory of Differentiation in Organizations”, American Sociological Review, 35, 201-218
Boeker, W. (1992) “Power and managerial dismissal: Scapegoating at the top”, Administrative Quarterly, 37, 400-421
Bonnier, K.A. & Bruner, R.F. (1989) “An Analysis of Stock Price Reaction to Management Change in Distressed Firms”, Journal of Accounting and Economics, 11, 95-106
Brülhart, M. & Trionfetti, F. (2004) “Public Expenditure, International Specialisation and Agglomeration”, European Economic Review, 48(4), 851-881
Carpenter, M.A., Geletkanycz, M.A. & Sanders, W.G. (2004) “Upper echelons research revisited: Antecedents, elements and consequences of top management team composition”, Journal of Management, 30(6), 749-778
Carroll, W.K. & Carson, C. (2003) “The Network of Global Corporations and Elite Policy Groups: A Structure For Transnational Capitalist Class Formation?”, Global ,etworks, 3(1), 29-57
Carroll, W.K. & Fennema, M. (2002) “Is There a Transnational Business Community?”, International Sociology, 19(3), 369-378
Clendenin, W.D. (1972) “Company presidents look at the board of directors”, California Management Review, 14(3), 60-66
Daily, C.M. & Dalton, D.R. (1994) “Bankruptcy and corporate governance: The impact of board composition and structure”, Academy of Management Journal, 37(6), 1603-1617
Davidson, W.N., Worrell, D.L. & Dutia, D. (1993) “The Stock Market Effects of CEO Succession in Bankrupt Firms”, Journal of Management, 19(3), 517-533
DeFond, M. & Park, C. (1999) “The Effect of Competition on CEO Turnover”, Journal of Accounting and Economics, 27, 35-56
Denis, D.J., Denis, D.K. & Sarin, A. (1997) “Ownership Structure and Top Executive Turnover”, Journal of Financial Economics, 45, 193-221
Dietrich, D. & Hauck, A. (2012) “Government interventions in banking crises: Effects of alternative schemes on bank lending and risk taking”, Scottish Journal of Political Economy, 59(2), 133-161
Dooley, M.P., Folkerts-Landau, D., Garber, P.M. (2009) “Bretton Woods II still defines the international monetary system”, Pacific Economic Review, 14, 297-311
Erkens, D., Hung, M. & Matos, P.P. (2009) “Corporate Governance in the 2007-2008 Financial Crisis: Evidence from Financial Institutions Worldwide”, ECGI Finance Working Paper, 249
Finkelstein, S. (1988) “Managerial orientations and organizational outcomes: The moderating roles of managerial discretion and power”, unpublished doctoral dissertation, Columbia University
Finkelstein, S. & Hambrick D.C. (1996) “Strategic leadership: Top executives and their effects on organizations”, Mineapolis/St. Paul: West Educational Publishing
Finkelstein, S., Hambrick, D.C. & Cannella, A.A. (2009) “Strategic leadership: Theory and research on executives, top management teams, and boards”, Oxford University Press
Fujiwara, K. & Van Long, N. (2012) “Welfare Effects of Reducing Home Bias in Government Procurements: A Dynamic Contest Model”, Review of Development Economics, 16(1), 137-147
Ganelli, G. (2005) “Home Bias In Government Spending And Quasi Neutrality Of Fiscal Shocks”, Macroeconomic Dynamics, Cambridge University Press, 9(2), 288-294
Gilson, S.C. (1990) “Bankruptcy, boards, and blockholders: Evidence on changes in corporate ownership and control when firms default”, Journal of Financial Economics, 27, 355-387
Goyal, V.K. & Park, C.W. (2002) “Board leadership structure and CEO turnover”, Journal of Corporate Finance, 8, 49-66
Greve, P., Nielsen, S. & Ruigrok, W. (2009) “Transcending borders with international top management teams: A study of Europeans Financial MNCs”, European Management Journal, 27, 213-224
Grusky, O. (1961) “Corporate size, bureaucratization, and managerial succession”, American Journal of Sociology, 67, 263-269
Hair, J.F., Black, W.C., Babin, B.J., Anderson, R.E. & Tatham, R.L. (2006) “Multivariate Data Analysis”, 6th edition, Pearson Education Inc., Upper Saddle River, New Jersey
Hambrick, D.C., Cho, T. & Chen, M. (1996) “The influence of top management team heterogeneity on firms’ competitive moves”, Administrative Science Quarterly, 41, 659-684
Hambrick, D.C. & Mason, P.A. (1984) “Upper Echelons: The Organization as a Reflection of Its Top Managers”, Academy of Management Review, 9(2), 193-206
Heemskerk, E.M. (2011) “The social field of the European corporate elite: A network analysis of interlocking directorates among Europe’s largest corporate boards”, Gobal ,etworks, 11(4), 440-460
Helmich, D. (1980) “Board size, variation, and rates of succession in the corporate presidency”, Journal of Business Research, 8, 51-63
James, D.R. & Soref, M. (1981) “Profit constraints on managerial autonomy: Managerial theory and the unmaking of the corporate president”, American Sociological Review, 46, 1-18
Karlsson, P. & Neilson, G.L. (2009) “CEO Succession 2008: Stability in the Storm”, Strategy & Business, Issue 55, published by the global management consulting firm Booz & Company:
http://www.booz.com/media/uploads/CEO_Succession_2008.pdf
Kashyap, A., Rajan, R. & Stein, J. (2008) “Rethinking Capital Regulation”, Working Paper, University of Chicago
Masselink, M. & Van den Noord, P. (2009) “The global financial crisis and its effects on the Netherlands”, Country Focus, 6(10), 1-7
McConnell, J.J. & Servaes, H. (1990) “Additional evidence on equity ownership and corporate value”, Journal of Financial Economics, 27, 595-612
Meindl, J.R., Ehrlich, S.B. & Dukerich, J.M. (1985) “The romance of leadership”, Administrative Science Quarterly, 30(1), 78-102
Murphy, K.J. & Zimmerman, J.L. (1993) “Financial performance surrounding CEO turnover”, Journal of Accounting and Economics, 16, 273-315
Murray, A.I. (1989) “Top management group heterogeneity and firm performance”, Strategic Management Journal, 10, 125-141
