EFFECT OF CULTURE ON
FIRM PERFORMANCE:
1
EVIDENCE FROM CEO SUCCESSIONS
John Klappe
2Faculty of Economics and Business, University of Amsterdam
September 11, 2014
Abstract
Does a CEO’s own cultural traits affect firm performance? By looking at CEO (chief executive officer) successions at 207 non-utility, non-financial, publicly traded U.S. firms I find no evidence that the cultural traits of a CEO affects firm performance. I measure culture by four indicators of individual beliefs: trust, confidence in self-determination, respect for others and obedience and authority. Of the previously mentioned indicators of culture I construct two principal components. In addition, I test for a possible non-linear relationship between the indicators for culture and firm performance. I measure firm performance by taking operating return on assets (OROA) and the Market to book (M-B) ratio. The OLS and GMM results clearly show there is no effect of the CEO’s cultural traits on firm performance. In sum, the cultural traits of the CEO have no effect on firm performance. The evidence shows caution is required in understanding culture and the way how it affects economic outcomes.
Keywords: Culture; Cultural, Executiveyy JEL classification: Z1, M14
1
I would like to thank for the useful comments: Carmine Guerriero and Jeroen Wismans. And I would like to thank for the useful comments during the thesis seminars Jeroen van de Ven, Joeri Sol, and my fellow class mates. University of Amsterdam, Faculty of Economics and Business, Roetersstraat 11, 1018 WB Amsterdam, the Netherlands, email: john.klappe@student.uva.nl.
2
2
“Individuals have less control over their culture than over other social
capital […], culture is largely a “given” to individuals throughout their lifetimes” (Gary Stanley Becker, Preferences and values, 1996).
1
Introduction
Culture affects fundamental economic decision making. In experimental settings, there is a large variation in offered stakes between various different small-scale societies at played ultimatum-games (the offered stakes vary from 0.26 to 0.58 per stake)34 (Henrich et al.,2001). Economist also find evidence that culture is relevant in non-experimental settings. By making use of better measurement techniques, more available data to measure culture and by defining culture in a measurable way it becomes possible to determine the causal relationship of culture in many areas of interest (Sapienza et al., 2006). For example, regarding to the number of mergers between companies founded in different countries, Ahern et al. (2012) observe that the volume of mergers between companies lowers when countries are culturally more distant (Ahern et al., 2012).
In general, a culture which promotes cooperation (which is revealed by norms like the degree of trust, and respect for others) is strongly correlated with economic regional development (Tabellini, 2010). Perrson and Tabellini (2009) also mention the importance of having inclusive political institutions to promote economic development. However, Guerriero (2013) shows that the effectiveness of these inclusive political institutions also depend to a great extent of the degree of culture. If citizens are morally not willing to punish politicians for misbehavior and are not interested in the selection of public spirited politicians, the advantage of having inclusive political
3 In an ultimatum game two players divide a sum of money that is allocated to them. One player proposes how to divide the allocated sum of
money, the second player can either reject of accept the offer. If the offer is accepted the money gets divided in order with the proposed offer, if the offer is refused both players get nothing (Cameron, 1999)
4 “The vast majority of the Ache (94 percent) made offers above 40 percent of the stake size. This coincides neatly with ethnographic descriptions
indicating widespread meat-sharing and cooperation in community projects despite the absence of a fear of punishment in Ache society” (Henrich et al., 2001)
3
institutions dilutes. In other words, even if the political system makes it possible to punish misbehavior within political institutions, it all depends of the degree of culture if people make use of this opportunity. This is showed by Guerriero (2013) who observe that the degree of criminal prosecutions of politicians in Italian in a region with a higher degree of culture is considerable lower.
Though, despite the growing evidence of culture as a valid determinant for economic development and as a relevant influencer on the functioning of political institutes, there is little known if the level of culture also affect firm performance. To be more specific the topical question is, does the election of a new CEO with different cultural traits affect firm performance?
In this paper, I provide evidence that culture shifts caused by CEO successions have no impact on the performance of publicly traded U.S. firms. I use data from 207 CEO successions and observe if firm performance, measured by the Market to book ratio (M-B) and operating return on assets (OROA), is influenced by shifts in cultural characteristics due to CEO successions. I correct firm performance for the average firm and industry performance, and I take into account other characteristics that influence performance like the CEO’s amount of ownership. I also control for the entrenchment of the CEO by looking at the degree of shareholder rights. More shareholder rights leads to a higher firm value, profit, and sales level (Gompers et al., 2003). Thus, for 71 organizations for which I have information on the degree of shareholders rights, I use the governance index of Gompers et al. (2003) to control for the degree of quality of firm governance.
In order to measure culture I use four indicators for culture which are: trust, confidence in self-determination, respect for others, and obedience and authority. I also construct two principal components of culture, PC_Culture 1 and PC_Culture 2, which are constructed out of the above named indicators of individual beliefs. PC_Culture 1, consists out; trust, confidence in individual self-determination and respect for others and PC_Culture 2, consists out: trust, confidence in
4
individual self-determination, respect for others, and authority and obedience. Regarding to the possibility that the relationship between culture and firm performance is non-linear and has an inverted u shape (Guerriero and Boranbay, 2012; Butler ea. , 2009) I also test if the squared indicators for culture and components of culture have a significant impact on firm performance.
OLS estimates provides no empirical support that changes in cultural values due to CEO successions influence firm performance. Second, if I estimate the values by treating culture as endogenous and estimate the relationship by employing GMM, again I find no significant effect of culture on firm performance.
The key contribution of this paper is that culture shifts at the top of the firm due to CEO successions have no effect on firm performance. This is in line with some of the existing theory (Gomez-Mejia and Palich, 1997; Oosterbeek et al., 2004) which also find no evidence of culture as a relevant influencer for economic outcomes. An alternative explanation which could explain the insignificancy of the results is that the effect of the cultural characteristics of the CEO are diluted by the composition of the remainder of the board (Gomez ea., 1997). A second alternative explanation could be that the cultural values of the remainder of the employees within the firm is more important than the cultural values of the CEO. This is in line with Guerriero (2013) who observes that the average culture of the citizens is important with respect to the degree of economic development. A last explanation is the results are biased by hidden reasons of the successions of CEO’s which leads to muddy results (Beatty, 1987).
The rest of the paper is organized as follows. Section 2 describes an overview of the literature on the definition of culture, the indicators of values and beliefs to measure culture and findings on their expected influence on the CEO and firm performance. Section 3 describes my hypothesis based on the previously mentioned theory, the used data to measure firm performance
5
and culture, and my empirical strategy. Section 4 presents some summary statistics, results and my analysis. Finally, section 5 presents some concluding remarks.
2
Culture
In this section I first describe the definition of culture. Second, I discuss which values and believes are suitable to measure culture. Third, I discuss how these values an believes affect my topic of interest.
2.1 Definition of culture
It is not so straightforward to find a causal relationship between culture and firm performance. Due to the broad notion of culture and the various ways culture can interact with performance it becomes hard to measure (Sapienza et al., 2007). The following definition, which is also used by others, overcomes this difficulty to measure culture (Sapienza et al., 2007).
“Those customary beliefs and values that ethnic, religious, and social groups transmit fairly unchanged from generation to generation (Sapienza et al., 2007)”.
The above definition of culture minimises the risk of being influenced by other unobserved factors and thus the possibility of the elaboration of constructing refutable hypothesis rises (Sapienza et al., 2006). However, there is some ambiguity whether these values can be considered as approximately exogenous. Boranbay and Guerriero (2012) show that culture is likely to be endogenous. In their analysis of political institutions and monasticism between the 11th and the 16th centuries Boranbay and Guerriero (2012) reveal that in times of uncertainty cultural values of citizens alter due to factors that aggravate consumption risk. A possible threat to democracy leads
6
to an overinvestment in a culture of cooperation (Boranbay and Guerriero, 2012). In sum, although these values are transmitted fairly unchanged from generation to generation these values can be affected by unobservable factors.
2.2 Indicators of individual beliefs and values
The indicators of culture are the following values and believes: 1) trust, 2) respect for others, 3) confidence in individual self-determination, and 4) obedience and hierarchy. Trust is necessary for coordination within communities and is widely studied by others within the areas of finance and economics (Adler, 2001: Ahern et al., 2012). The other values and believes are named in various models in sociology in attempt to define culture. Hofstede (1984) mention in his five cultural dimension model, the indicators confidence in self-determination (individualism) and obedience and authority (power distance) as two of the five relevant dimensions to define culture. The authors Trompenaars and Hampden-Turner (1998) mention these two dimensions in their seven-dimension culture model. All dimensions except for trust can be allocated in one of the three dimensions of culture in the model of Fiske (1992). All named dimensions are also used by Tabellini (2010) to examine the effect of culture on regional economic development.
1) Trust: is the dependence on someone else to fulfill an implicit or explicit obligation (Sapienza et al., 2007) and is studied in various context. In experimental settings, played games between two more trusting subjects, are more likely to lead to an efficient outcome (Tabellini, 2010). While when there is a shortage in trust between two subjects, it will be hard to come to an efficient outcome. From another experimental game: ‘the trust game’ played with CEO’s and played with students there is a surprising outcome that CEO’s show in general more trust for others (Fehr and List, 2004). At the trust game (one game with and without the opportunity to punish) both players are endowed with money and have the opportunity to transfer their obtained money to
7
another and have the opportunity to triple their stack for each transferred token (Fehr and List, 2004). The evidence that CEO’s in general perform more trust conflicts with the assumption that culture characteristics of the CEO alters economic outcomes. Although, one could argue if the results would differ between students and CEO’s if the size of the stakes change. The amount of money in the game is relative larger for the average student compared to the average CEO.
In non-experimental settings, trust can be a complement or even a substitute to formal contracting and can reduce transaction cost. For instance, Sako and Helper (1998) find differences between trust and opportunism levels between the U.S. and Japanese automotive industry. And as a consequence Sako and Helper (1998) observe the length of written contracts in the U.S. is larger compared to the Japanese who considered these contracts as irrelevant.
Also within the firm a lack of trust can influence performance. Falk and Kosfeld (2006) reveal how control negatively affects motivation, and how it can lead to a reduction of effort. People who perceive control as negative see it as a sign of distrust (Falk and Kosfeld, 2006). The authors Zak and Knack (2001) confirm the consequences of these negative perceptions of trust by observing that a lack of trust leads to costly monitoring.
In sum, in general a higher trust level reduces transaction cost and monitoring cost within the company. In a high-trusting environment, the need for external enforcement is lower. While in a low trusting environment the suspicion of fraud will be larger, and this will raise transaction costs and the cost of monitoring which reduces potential benefits (Tabellini, 2010).
Notice an increase in the average trust level of the CEO does not automatically leads to a better firm performance. Two sources of suboptimal behavior cause a hump-shaped relationship between trust and performance, too little trust worsen performance through cautious decision making that leads to missed profit opportunities (Butler, ea., 2009). Too much trust undermines performance by the chance to be cheated (Butler, ea., 2009).
8
2) Respect for others: people who practise in general more morality are more reluctant to free ride on others (Tabellini, 2010). This not only holds for economic decision making but also holds for reducing effort in group activities (also known as the free riding effect) (Tabellini, 2010). A CEO who exhibits a lower morality and is self-interested could harm the organization by taking decisions or actions which are not in the primary interest of the firm (Jensen, 1993). This could lead to activities like empire building, shirking, or delivering incomplete information. Or it can lead to high-compensation packages or lead to attempts to strengthen the own position (entrenchment) (Bebchuk and Cohen, 2005). A lower morality could also lead to illegal activities of the CEO like theft, fraud, insider trading or bribing. In sum all these activities harm firm performance. Otherwise, if the morality of the new CEO rises this will lower the probability the firm encounters higher agency costs, because the CEO is more reluctant to engage in these kind of activities (Tabellini, 2010).
3) Confidence in individual self-determination: is an important drive under economic progress, individual effort has to be likely to pay off. More individualism leads to more innovation and higher growth; this is due to the higher social rewards which are associated with innovation in an individualist culture (Tabellini, 2010). If individuals are highly motivated to succeed and view economic success as related to their deliberate choices, people are more likely to work hard in order to invest for the future, and to innovate and undertake newly economic initiatives (Tabellini, 2010). The author Lazaer (1996) show the positive effect of individual incentives by the evidence that switching from a fixed wage to piece rate improves productivity. However, one could argue if this increase in productivity is due to the piece-rate scheme or that there might be other reasons that cause this rise in productivity like for example the change of the management (Lazaer, 1996).
Although people always react on incentives, this does not mean that responses to incentives are necessarily beneficial (Prendergast, 1999). A firm can set incentives such that it could harm the
9
organization, Kerr (1975) gives various examples in his article5 that demonstrate how the outcome of initial set incentives can lead to a removal from the initial set goal. Introducing incentives can even backfire an persons confidence in own abilities or decrease the value of the rewarded task (Bernabou and Tirole, 2003).
The consequence of the interaction between individualistic and non-individualistic employees is ambiguous. Concerning Tabellini (2010) it is plausible that combining individualist and collectivist employees within the firm this would inhibit the operations of the firm more than combining employees who share the same cultural values. Although there is no information on the effect of interaction between individualistic and non-individualistic employees the U.S. can be considered as a relative individualistic society compared to other societies and thus this may matter in the degree how sensitive people are to incentives (Triandis et al. 1988). Though, (although the U.S. is relative individualistic) there are still large differences in the degree of individuality between the various regions within the U.S. (Vandello and Cohen, 1999).
4) Obedience and Authority: Hierarchical cultures delineate members into multiple vertical ranks of power (Ahern et al., 2012; Tabellini, 2010). Members from lower ranks defer to higher ranked members, who in turn have an obligation to ensure that the needs of lower ranked members are satisfied (Ahern et al., 2012). A shortage of trust and a shortage for respect for others are
characteristics of hierarchical societies. In such societies, the type of behaviour that takes advantage of trust or takes advantage at cost of others is being punished (Ahern et al., 2012). For example, individualism is mistrusted and suppressed. The goal of hierarchical societies is to force everyone in good behaviour and also the role of parental education in hierarchical environments is targeted to control the negative instincts of children, often through recourse to violence (Tabellini, 2010).
This punishing can stifle individual initiative and cooperation within a group and a 5
10
hierarchical society can paralyse, individual initiative and group cooperation and can hurt growth and development (Tabellini, 2010). The consequence for a more hierarchical CEO could be that he is not able to understand that egalitarian workers are unlikely to follow orders without justification (Ahern et al., 2012). Likewise, an egalitarian CEO could not be respected by hierarchical workers if the CEO treats workers as her equal (Ahern et al., 2012). Thus, a CEO with a higher hierarchic level could inhibit coordination and other forms of interpersonal frictions could harm cooperation (Ahern et al., 2012).
11
3
Methodology
The main idea of the theoretical framework is that culture influences the relationship between the CEO and his stakeholders. In this section I first discuss my predictions based on my theoretical framework. Second, I discuss my data on culture and on firm performance. Finally, I discuss my empirical strategy.
3.1 Predictions
The reasoning behind my first two predictions are based on the theoretical framework discussed before and leads to the following testable predictions:
Prediction 1: a positive change in at least one of the CEO’s indicators of individual beliefs for
culture ( Trust, Confidence in individual self-determination, Respect for others) due to a CEO succession will have a positive effect on firm performance (OROA, M-B)
Prediction 2: a change in the level of Authority and Obedience of the CEO due to a succession will
have a negative effect on firm performance (OROA, M-B)
To capture the underlying variation in culture, I construct two principal components of culture. By constructing a principal component of culture I observe how the vector of the variations of the dimensions of culture affect performance (Jolliffe, 2005; Wold ea.,1987). I construct two different principal components of culture: PC_Culture 1, which consist out; trust, confidence in individual self-determination and respect for others and PC_Culture 2: which consists out: trust, confidence in individual self-determination, respect for others, and authority and obedience. I
12
expect PC_Culture 1 has a positive effect on firm performance. Since I expect obedience and authority has a negative effect on firm performance I expect PC_Culture 2 has a lower but still positive effect on firm performance.
Prediction 3: an increase in the average culture PC_Culture 1 (trust, confidence in individual
determination ,respect for others) and PC_Culture 2 (trust, confidence in individual self-determination, respect for othersand authority and obedience) of the CEO due to a succession will have a positive effect on firm performance (OROA.M-B)
Guerriero and Boranbay (2012) find that culture has an inverted u shape relationship with forces who aggravate consumption risk. Butler ea. (2009) also observe an inverted U-shape relationship between trust and performance. Considering the relationship between culture and firm performance could be non-linear and is inverted u shaped I also test for the following predictions.
Prediction 4: there is an inverted U-shaped relationship between firm performance (OROA, M-B)
and one of the indicators for culture (trust, confidence in individual self-determination, respect for others) because of a CEO succession
Prediction 5: there is an inverted U-shaped relationship between firm performance (OROA, M-B)
and one of the indicators for culture (PC_Culture 1 (trust, confidence in individual determination ,respect for others) and PC_Culture 2 (trust, confidence in individual self-determination, respect for othersand authority and obedience) because of a CEO succession
13
3.2 Measuring firm performance
To measure firm performance, I use the database constructed by Pérez-González (2006)6. which includes all U.S. (United States) nonfinancial, nonutility firms registered in 1994 in COMPUSTAT7. There are three restrictions made by the author: 1) firms have to be founded prior 1971; 2) in the proxy statement information have to be available on 2a) a relative with at five percent ownership 2b) two or more individuals related as directors, officers, shareholders, 2c) a founder as an executive or director; 3) a management change needs to have occurred as identified by news searching using the Dow Jones INC (Pérez-González, 2006). In addition to the origin database, I make one additional restriction. Information on the origin region of birth of both CEO's of every succession should be obtainable and should lie within the U.S.. Due to the scope of the GSS (General Social Survey) it is only possible to measure culture for nine U.S. regions. I analyze firm performance three years before and three years after each CEO succession. If the financial information (three year before and after a succession) is not available I take firm performance of a five year period (I take the firm performance of the current year, two years before and two years after the succession). I end up with an unbalanced panel data of 207 CEO-successions and 1237 data points within a period of 1994 and 2003 (see table 1 for a list with all firm observations).
Validity- The requirement that firms must have been active since 1970, introduces a survivor bias (Pérez-Gonzalez, 2006). This means that firms who went bankrupt are not considered in the analysis and this could positively bias the image of the results of the firms who manage to survive. The second requirement that the information on CEO should lie within the U.S. and should be available causes a selection bias. Cultural differences of American regions are smaller compared
6
The author constructed this database to determine the effect of inherited control on firm performance. 7
14
to cultural differences worldwide (Ahern et al. ,2012). Due to this selection bias it may become difficult in finding differences in firm performance due to the smaller differences in cultural traits (Ahern et al., 2012). The other restrictions; ownership of at least five percent, the inclusion of only firms that are nonfinancial, and non-utility sector, make the sample of firms non-random and thus this harms the extern validity of the outcomes (Pérez-González, 2006). In sum, consider the manner how the database is constructed one should be careful of interpreting the results as a valid predictor for firms all over the world (Pérez-González, 2006).
3.3 Measuring the CEO’s culture
The origin region of birth of every CEO is collected by the use of; 1) direct email contact with companies (investor relations, corporate relations departments), 2) contact with CEO's through social media (LinkedIn, Facebook), 3) contact with journalists, 4) contact with other organizations that are linked to the CEO (foundations, companies, etc.), and 5) information obtained by search of ‘In Memorial’ notifications. The proxy of the cultural traits of the CEO is obtained from the General Social Survey (GSS)8. The most detailed level the data can be
aggregated in which of the nine U.S. regions the CEO is born. Building on previous literature on measuring culture the following indicators of individual beliefs and values are used: 1) trust, 2) confidence in individual self-determination, 3) respect for others, and 4) obedience and hierarchy. To measure my first variable trust I consider the share of the answer “can trust” on the question “can people be trusted”9
. Turning to confidence in individual self-determination I use the share of the answers between “some choice and control and “a great deal of choice and control” on the
8
The GSS is a study on the structure and development of American Society and contains various demographic and attitudinal questions. The GSS started in 1972 and completed its 28th round in 2010 and many of the core questions have been unchanged since 1972 I obtained this information on the website of http://www3.norc.org/GSS+Website/, visited on 30-06-13 I use the cumulative dataset of 1972-2012
9 Generally speaking, would you say that most people can be trusted or that you can’t be too careful in dealing with people An equivalent question of
15
question “I feel choice and control over own life”. In order to measure my third variable respect for others I use the share of the answers “strongly agree” and “agree” on the question “ people are treated with respect”. To determine my last variable obedience and authority I use the share “strongly agree” and “agree” to the question “teach children respect for authority and obedience”.
Validity- A first concern is that the measured individual cultural straits are poor proxies for the real cultural values of the CEO’s. This is contradicted by Pérez-González (2006) which
concludes that the predicted response levels of trust are a good predictor of the trusting behavior in experimental settings. The second concern is that the origin region of birth of the CEO does not represent a relevant image on the real origins of the CEO. Although this concern cannot be
excluded completely I do observe that information on early childhoods are often in the same region as the CEO’s origin region of birth. Even the largest removals of CEO’s in their childhood are often within the same region, thus I consider It safe to assume the origin region of birth reflect the CEO’s individual cultural straits
3.4 Identification strategy
I estimate my predications by running panel regressions of the following model:
(1) Where is the firm performance in the operating return on assets, OROA, or the market to book ratio, M-B in year t for firm f and lies between a time period between 1994-2003. accounts for an equal constant for all CEO’s denotes for proxies of the CEO’s culture which is measured by the variables trust, or, confidence in individual self-determination, or, respect for others, or, obedience and authority, or by a principal component of culture, PC_Culture 1, or, PC_Culture 2.whereas
displays a set of control variables like Ln Sales, Industry Adjusted OROA, Industry Adjusted M-B, Mean pre transition industry and performance adjusted OROA and Mean pre transition
16
industry and performance adjusted M-.B a to pick up industry and average return year effects. And I control for other characteristics that affect firm performance like the degree of ownership, board ownership, and the quality of the governance, governance index (see table 2 for all summary variables).
A first naïve attempt to measure a relationship between firm performance and culture is to estimate the equation (1) by OLS (ordinary least squares). However, there are some issues with performing this strategy. Considering the construction of the sample is not-random and due I analyze panel data and thus with a regular cross section analysis I lack to incorporate the effect of lagged values I may overestimate the effect of culture on firm performance. Second, knowing culture is endogenous and can be influenced by omitted characteristics or could even be triggered by reverse causality this also makes the OLS estimates biased. To account for these problems and to elaborate the OLS bias I estimate equation (1) also by GMM (Roodman, 2006). I compare the OLS-estimates with the GMM-estimates and I treat the proxies for culture: PC_Culture 1 and PC_Culture 2 as endogenous. To avoid too many instruments which can lead to an inconsistent estimator of the treated endogenous variables I limit the lag depth by only use lagged two and three instruments. Lagged instruments who are correlated with the dependent variable are not orthogonal thus I use only lagged variables which are uncorrelated witch the error term (Roodman, 2009)
17
4
Results
First, I present some summary statistics. Second, I discuss the outcomes of the OLS and GMM estimates. Finally, I comment on the results and I provide possible explanations why I find no significant results.
4.1 Summary statistics
Table 2 reports information on all the variables of equation (1). The OROA has an average of 14,4 percent with a standard deviation of 0.09. The M-B has an average 1.49 ratio which means that the market on average undervalues the collection of firms, however the standard deviation of 0,83 accommodate for a large variation in possible M-B ratios. All proxies for culture lay between 0 and 1 which is in line with the expectations since a proportion of answers to multiple choice questions is measured. Figure 1 presents the number of changes in cultural traits after CEO-successions. The number of no changes at the top is considerable larger than the number of changes in cultural traits. Approximately the ratio of 1:2 clearly show that shifts in cultural traits of the CEO is less likely to happen. Figure 2 displays the distribution of the origins of the new and old CEO’s divided on the nine U.S. regions. The regions Middle Atlantic and East North Central are overrepresented and cover approximately 40 percent of the total sample. The distribution of the new CEO’s compared to the old CEO’s do not differ from each other. Figure 3 reflects the individual proxies of regional values and beliefs to measure culture. The regions New England, East North Central and West North Central perform in general more trust compared to East South Central and West South Central. Regarding to all other regions the regions new England and Middle Atlantic in general believe less in confidence in self-determination. In the region New England, West North Central and the Pacific people in general believe more in respect for others is
18
important. On the question if children should be learned respect for authority and obedience the regions South Atlantic, West South Central, East South Central score high.
4.2 Main results
Table 3 reports the OLS estimates of equation (1) with the dependent variable OROA in columns (1) to (7). All indicators and principal components of culture are not significant and thus have no effect on firm performance. Table 4 reports the OLS estimates of equation (1) with the dependent variable M-B in columns (1) to (7). Only the variable confidence in individual self-determination is significant at 10 percent level and has a positive effect on firm performance with a coefficient of 0.516. All other values of culture are insignificant. The results from table 3 and 4 clearly show that a shift in culture due to a CEO succession has no impact on firm performance and the first two predictions are not supported by the data. Tables 5 and 6 report the values for squared variables of culture to test if there is maybe a nonlinear relationship of culture. Again, there are no significant results for any of the squared indicators of components of culture.
Table 7 reports the GMM estimates compared to the OLS estimates of the principal components of culture. I use as excluded instruments PC-Culture 1 and PC-Culture 2 lagged two and three years. This choice reflects the fact that the error term in equation 1 is autoregressive of order 1. Therefore, all instruments correlated to the dependent variable and lagged less than two years will be endogenous. The Hansen test indicates that the over- identifying restrictions cannot be rejected, and the AR estimates show there is not a higher order of an autoregressive structure of the errors. The GMM results confirm the results of the OLS estimates, again there are no significant results of culture on firm performance.
In sum, the estimates of table 3,4,5 and 6 show (besides for one indicator in table 3) no significant impact of culture on firm performance. If culture is treated as endogenous the
19
components of culture still have no significant effect on firm performance. The evidence triggers to conclude that culture shifts due to successions at the top have no effect on firm performance.
4.3 Analysis
The basic estimates of the OLS/GMM results suggest there is no significant effect of culture
on firm performance. The possibility that financial markets already anticipate the arrival of a new CEO and thus stock prices already incorporate the expectations of the new CEO could bias the
results. Therefore, the possibility to find a significant effect with an event study is inhibited by the
financial market.
A possible conclusion is that culture is not so critical as thought. This is in line with
research of Gomez-Mejia and Palich (1997) who, in an attempt to define how cultural diversity affects firm performance, did not find any evidence of culture influences firm performance. This is also in line with previous research of Oosterbeek ea. (2004) who analyzed ultimatum games
(meta-analysis) and conclude that the differences in outcomes cannot be attributed to culture traits. It
could also be the elusive goal in general to find a relationship between firm performance and CEO characteristics. Due the nature of a succession remains unknown it is hard to investigate the relationship between firm performance and CEO-successions (Beatty, 1987).
Another possible reason that could explain the insignificant relationship of the CEO’s
cultural traits on firm performance is that the effect of the culture shifts due to the succession is
mitigated by the composition of the remainder of the board. Firms are able by employee selection
to minimize the negative effect of cultural differences (Gomez ea., 1997).
I end with my final explanation that could cause the insignificant results. The cultural values of the remainder of the employees within the firm could be more important than the cultural values of the CEO. This is in line with Guerriero (2013) who observes that the average culture of
20
the citizens is important with respect to the degree of economic development and the degree of control on key persons like politicians.
5
Concluding comments
If a CEO’s cultural traits impact the cooperation between him/her and his/her employees or affect his/her contracting activities between him/her and his/her contracting partners, culture becomes a relevant dimension to take into account. Here, I have tested if culture shifts due to CEO successions affect firm performance and I have found no significant relationship between a shift in cultural traits of the CEO and firm performance. The insignificant results support theory that question the effect of culture as a relevant determinant on economic outcomes (Gomez-Mejia and Palich, 1997; Oosterbeek et al., 2004) and it becomes attractive to conclude culture has no impact on firm performance. Yet, due to the mixed evidence within economic theory on culture and performance one should be cautious in drawing conclusions that culture has no effect on firm performance.
Thus, I close with some opportunities for further research. The first one is to expand this current research and build on a larger sample of companies that also include non-U.S. companies. My expectation is that a larger variance in cultural traits of CEO’s increases the probability to find results. An alternative option is not to focus on key persons like CEO’s but focus on the employee’s culture. One could build further on research of Guerriero (2013) and look if organizations with a higher degree of culture possess a lower degree of company boards with criminal prosecutions. Or one could build further on research of Perez-Gonzalez (2006) and observe if differences in culture not only affect the number of mergers but also inhibit the performance and/or number of cooperation’s of joint ventures between companies who are culturally more distant.
21
References
Adler, P. S. (2001). Market, hierarchy, and trust: The knowledge economy and the future of capitalism. Organization Science, 12, 215-234
Ahern, K. R., Daminelli, D., & Fracassi, C. (2012). Lost in translation? the effect of cultural values on mergers around the world. Journal of Financial Economics, Vol. 101-110,1-23
Beatty, R. P., & Zajac, E. J. (1987). CEO change and firm performance in large corporations: Succession effects and manager effects. Strategic Management Journal, 8, 305-317
Bebchuk, L. A., & Cohen, A. (2005). The costs of entrenched boards. Journal of Financial Economics, 78, 409-433
Becker, G. (1996). Preferences and values. Accounting for Taste, Harvard University Press: Cambridge, 1-156
Benabou, R., & Tirole, J. (2003). Intrinsic and extrinsic motivation. The Review of Economic Studies, 70, 489-520
Boranbay and Guerriero (2012). Endogenous (in) formal institutions. Amsterdam Law School Research Paper, (2012-79)
Butler, J., Giuliano, P., & Guiso, L. (2009). The Right Amount of Trust, National Bureau of Economic Research, w15344, 1-43
Cameron, L. A. (1999). Raising the stakes in the ultimatum game: Experimental evidence from indonesia. Economic Inquiry, 37, 47-59
Falk, A., & Kosfeld, M. (2006). The hidden costs of control. The American Economic Review, 96, 1611-1630
Fehr, E., Herz, H., & Wilkening, T. (2013). The lure of authority: Motivation and incentive effects of power. The American Economic Review, 103, 1325-1359
Fehr, E., & List, J. A. (2004). The hidden costs and returns of incentives—trust and trustworthiness among CEOs. Journal of the European Economic Association, 2, 743-771
Fiske, A. P. (1992). The four elementary forms of sociality: Framework for a unified theory of social relations. Psychological Review, 99, 689
Gomez-Mejia, L. R., & Palich, L. E. (1997). Cultural diversity and the performance of multinational firms. Journal of International Business Studies,28 , 309-335
Gompers et al. (2003). The role of lockups in initial public offerings. Review of Financial Studies, 16, 1-29
22
Guerriero, C. (2013). Endogenous institutions and economic outcomes. Amsterdam Law School Research Paper, 2013-2039
Guiso, L., Sapienza, P., & Zingales, L. (2009). Cultural biases in economic exchange? The Quarterly Journal of Economics, 124, 1095-1131
Henrich, J., Boyd, R., Bowles, S., Camerer, C., Fehr, E., Gintis, H., et al. (2001). In search of homo economicus: Behavioral experiments in 15 small-scale societies. American Economic Review,91 , 73-78
Hofstede, G. (1984). Cultural dimensions in management and planning. Asia Pacific Journal of Management, 1, 81-99.
Jensen, M. C., & Meckling, W. H. (1979). Theory of the firm: Managerial behavior, agency costs, and ownership structure Springer
Jolliffe, I. (2005). Principal component analysis Wiley Online Library
Kerr, S. (1975). On the folly of rewarding A, while hoping for B. Academy of Management Journal, 18, 769-783
Lazear, E. P. (1996). Performance Pay and Productivity, The American Economic Review, Vol. 90, 1346-1361
Oosterbeek, H., Sloof, R., & Van De Kuilen, G. (2004). Cultural differences in ultimatum game experiments: Evidence from a meta-analysis. Experimental Economics, 7, 171-188
McGillivray, F., & Smith, A. (2000). Trust and cooperation through agent-specific punishments. International Organization, 54, 809-824
Pérez-González, F. (2006). Inherited control and firm performance. The American Economic Review, Vol 96.No 5, 1559-1588
Persson, T., & Tabellini, G. (2006). Democratic Capital: The Nexus of Political and Economic Change, American Economic Journal:, 1:2, 88–126
Prendergast, C. (1999). The provision of incentives in firms. Journal of Economic Literature, Vol. XXXVII, 7-63
Roodman, D. (2009). How to do xtabond2: An introduction to difference and system GMM in stata. Stata Journal, 9, 86
Sako, M., & Helper, S. (1998). Determinants of trust in supplier relations: Evidence from the automotive industry in japan and the united states. Journal of Economic Behavior &
23
Sapienza, P., Toldra, A., & Zingales, L. (2007). Understanding Trust,National Bureau of Economic Research No. w13387., 1-31
Sapienza, P., Zingales, L., & Guiso, L. (2006). Does Culture Affect Economic Outcomes? Journal of Economic Perspectives, Vol. 20, 1-50
Tabellini, G. (2010). Culture and institutions: Economic development in the regions of europe. Journal of the European Economic Association, 8, 677-716
Triandis, H. C., Bontempo, R., Villareal, M. J., Asai, M., & Lucca, N. (1988). Individualism and collectivism: Cross-cultural perspectives on self-ingroup relationships. Journal of Personality and Social Psychology, 54, 323
Trompenaars, F., & Hampden-Turner, C. (1998). Riding the waves of culture McGraw-Hill New York
Vandello, J. A., & Cohen, D. (1999). Patterns of individualism and collectivism across the united states. Journal of Personality and Social Psychology, 77, 279
Wold, S., Esbensen, K., & Geladi, P. (1987). Principal component analysis. Chemometrics and Intelligent Laboratory Systems, 2, 37-52
24
Tables and figures
Table 1: Observations CEO’s of all Non-utility, Non-financial US firms
ABM INDUSTRIES INC: William W. Steele, Sydney J. Rosenberg; ACETO CORP: Leonard S. Schwartz, Arnold J. Frankel; ADAPTIVE BROADBAND CORP: Philip F. Otto, David B. Leeson; AEROSONIC CORP, J. Mervyn Nabors, David S. Goldman; ALBERTSONS INC: Gary G Michael, Warren E McCain; ALICO INC: Ben Hill Griffin III: Ben Hill Griffin Jr. : ALL AMERICAN SEMICONDUCTOR: Bruce M. Goldberg, Paul Goldberg, ALLERGAN INC: William C Shepherd, Gavin S. Herbert; ALPHA INDUSTRIES INC: Martin J. Reid, George S. Kariotis; AMERADA HESS CORP: John Hess, Leon Hess: AMERICAN BILTRITE INC: Roger S Marcus, George E Safiol: AMERICAN VANGUARD CORP: Eric G. Wintemute, Herbert A. Kraft; AMREP CORP: Anthony B. Gliedman, Howard W. Friedman, ANALOG DEVICES: Jerald G. Fishman, Ray Stata, ANDERSEN GROUP INC: Oliver R. Grace Jr., Francis E. Baker: APOGEE ENTERPRISES INC: Donald W. Goldfus, Russell H. Baumgardner, APPLICA INC: David M. Friedson, Belvin Friedson; ARCHER-DANIELS-MIDLAND COG: Allen Andreas, Dwayne O. Andreas; BADGER METER INC: James L. Forbes, James O. Wright; BAIRNCO CORP: Luke E. Fichthorn III : Glenn W. Bailey, BARR LABORATORIES INC: Bruce L. Downey, Edwin A. Cohen: BASSETT FURNITURE INDS: Robert H. Spilman, Jr., Ed Bassett; BELO (AH) CORP -SER A COM: Robert W. Decherd, James M. Moroney Jr.; BEMIS CO: John H. Roe, Howard Curler; BLAIR CORP: Murray K. McComas, John L. Blair; BOB EVANS FARMS: Stewart Owens, Daniel E. Evans; BONTEX INC: James C. Kostelni, Hugo N. Surmonte; BON-TON STORES INC: Heywood Wilansky, Tim Grumbacher; BOWNE & CO INC: Robert M. Johnson, Richard H. Koontz: BRADY CORP: Katherine M. Hudson, Paul G. Gengler; BROWN (TOM) INC: Donald L. Evans, Thomas Brown: BUCKLE INC: Dennis H. Nelson, Daniel J. Hirschfeld; CABOT CORP: Samuel W. Bodman, Robert A. Charpie; CALLON PETROLEUM CO/DE: Fred L. Callon, John S. Callon, CASCADE CORP: Robert C. Warren, Jr. , Joseph J. Barclay; CASEYS GENERAL STORES INC: Ronald M. Lamb, Donald F. Lamberti; CATO CORP -CL A: John Cato, Wayland Cato, Jr: CELESTIAL SEASONINGS INC: Steve Hughes, Mo Siegel: CENTRAL NEWSPAPERS -CL A: Louis A. Weil III, Frank E. Russell, CHARMING SHOPPES: Dorrit J Bern, David V. Wachs, CLAYTON HOMES INC: Kevin Clayton, James L. Clayton, COACHMEN INDUSTRIES INC: Claire C. Skinner, Thomas H. Corson, COBRA ELECTRS CORP: Jerry Kalov, Carl Korn; COMPUTER DATA SYSTEMS INC: Gordon S. Glenn, Clifford M. Kendall; CONTINENTAL MATERIALS CORP: James G Gidwitz, Gerald S Gidwitz , COORS (ADOLPH) -CL B: Peter Coors, William Coors; CRANE CO: Robert S. Evans, Thomas M. Evans; DIGITAL EQUIPMENT: Robert B. Palmer, Kenneth Olsen; DILLARDS INC -CL A:William Dillard II , William Dillard, DIODES INC: David M. Lloyd, Wilbert Lloyd; DOW JONES & CO INC: Peter R. Kann, Warren H. Phillips; DRS TECHNOLOGIES INC: Mark S. Newman, Leonard Newman, EMC CORP/MA: Michael C. Ruettgers, Richard Egan; EQUITY OIL CO: Paul M. Dougan, Fred H. Evans, FAMILY DOLLAR STORES: Howard R. Levine, Leon Levine, FEDDERS CORP: Salvatore Giordano Jr., Salvatore Giordano; FIRST YEARS INC: Ronald J. Sidman, Marshall B. Sidman; FLEETWOOD ENTERPRISES: Glenn F. Kummer, John C. Crean; FLOWERS INDUSTRIES INC: Amos R. McMullian, Langdon S. Flowers; FOSTER (LB) CO: Lee Foster II, Joseph H. Dugan; FREQUENCY ELECTRONICS INC: Joseph P. Franklin, Martin Bloch; FRISCH'S RESTAURANTS INC: Craig F. Maier, Jack C. Maier; FROZEN FOOD EXPRESS IND: Stoney M. Stubbs Jr., Edgar O. Weller; GALOOB TOYS INC: Mark Goldman, David Galoob; GENCORP INC: A. William Reynolds, Gerald O'Neil; GENESEE CORP -CL B: John L. Wehle, John (Jack) L. Wehle; GEORESOURCES INC: Jeffrey P. Vickers, Rollin C. Vickers GLATFELTER (P H) CO: George H. Glatfelter II, Thomas C. Norris; GOLDEN ENTERPRISES: John S. Stein, Sloan Y. Bashinsky, HARCOURT GENERAL INC: Robert J. Tarr Jr., Richard A. Smith; HAROLDS STORES INC: Rebecca Powell Casey, Harold G. Powell; HELMERICH & PAYN: Hans Helmerich, Walter H. Helmerich III; HILLENBRAND INDUSTRIES: W. August Hillenbrand, Daniel A. Hillenbrand, HILTON HOTELS CORP: Stephen F. Bollenbach, Barron Hilton, HOWELL CORP: Donald W. Clayton, Paul N. Howell ; HUGHES SUPPLY INC: David H. Hughes, Harry C. Hughes; IMPERIAL SUGAR CO: James C. Kempner, Robert C. Hanna; INSTEEL INDUSTRIES: H.O. Woltz III, Howard O. Woltz Jr.; JACLYN INC: Robert Chestnov, Abe Ginsburg; JACOBSON STORES: P. Gerald Mills, Mark K. Rosenfeld; JO-ANN STORES INC -CL A: Alan Rosskamm, Martin Rosskamm; KAMAN CORP -CL A: Paul R. Kuhn, Charles H. Kaman KEITHLEY INSTR INC: Joseph P. Keithley, Thomas G. Brick KENAN TRANSPORT CO: Lee P. Shaffer, Frank H. Kenan; KIMBALL INTERNATIONAL -CL B: Thomas L. Habig, Arnold F. Habig; KNIGHT-RIDDER INC: P. Anthony Ridder, James K. Batten; KOSS CORP: Michael J. Koss, John C. Koss, KRUG INTERNATIONAL CORP: C.L. Haslam, Maurice F. Krug; LA BARGE INC: Craig E. LaBarge, Pierre L. LaBarge; LANCASTER COLONY CORP: John B. "Jay" Gerlach Jr., John B. "Bernie" Gerlach, LANDS END INC: Richard C. Anderson, Gary Comer; LEE ENTERPRISES: Richard D. Gottlieb, Lloyd G. Schermer; LENNAR CORP: Stuart A. Miller, Leonard Miller; MACDERMID INC: Daniel H. Leever, Harold Leever; MARTEN TRANSPORT LTD: Randolph L. Marten, Roger R. Marten; MCDONNELL DOUGLAS CORP: John McDonnell, Sanford McDonnell; MCI COMMUNICATIONS: Bert C. Roberts Jr., William G. McGowan; MEDIA GENERAL -CL A: J. Stewart Bryan III, D. Tennant Bryan; MEREDITH CORP: Jack D. Rehm, Robert A. Burnett; MIKASA INC: Raymond B. Dingman, Alfred J. Blake; MINE SAFETY APPLIANCES CO: John T. Ryan III, Lee N. Short Jr; MISONIX INC: Joseph Librizzi, Michael Juliano; MOCON INC: William N. Mayer, H.L. Demorest; MOLEX INC: Fred Krehbiel, John Krehbiel Sr; MOOG INC -CL A: Robert T. Brady, William C. Moog; MOORE PRODUCTS CO: William B. Moore, Edwin G. Rorke; MUELLER (PAUL) CO: Daniel C Manna, Donald E. Golik; MURPHY OIL CORP: Claiborne P. Deming, Jack W. McNutt, NASH FINCH CO: Alfred N. Flaten, Harold B. Finch Jr.; NATIONAL PRESTO INDS INC: Maryjo Cohen, Melvin S. Cohen; NEW BRUNSWICK SCIENTIFIC INC: Ezra Weisman, David Freedman; NEW YORK TIMES CO -CL A: Russell T. Lewis, Arthur Sulzberger; NEWELL RUBBERMAID INC: William P. Sovey, Daniel C. Ferguson; NOLAND CO: Lloyd U. Noland III, Lloyd Noland Jr; NOODLE KIDOODLE INC: Stanley Greenman, Bernard Greenman; NORDSON CORP: William P. Madar, Eric T. Nord; OHIO ART CO: William C. Killgallon, W. C. Killgallon; OIL DRI CORP AMERICA: Daniel S. Jaffee, Richard M. Jaffee; OLSTEN CORP: Frank N. Liguori, William Olsten; OMNI USA INC: Jeffrey K. Daniel, Edward L. Daniel; OPTICAL COATING LAB INC: Herbert M. Dwight, Rolf F. Illsley, ORLEANS HOMEBUILDERS INC: Jeffrey P. Orleans, MARVIN ORLEANS, OSHKOSH B'GOSH INC -CL A: Douglas W. Hyde, Charles F. Hyde; OXFORD INDUSTRIES INC: J. Hicks Lanier, Sartain Lanier; P & F INDUSTRIES -CL A: Richard A. Horowitz, Sidney Horowitz; PARK ELECTROCHEMICAL CORP: Brian E. Shore, Jerry Shore; PARKER DRILLING CO: Robert L. Parker Jr., Robert L. Parker, PARKER-HANNIFIN CORP: Paul G Schloemer Patrick S. Parker, PAYCO AMERICAN CORP: Neal R. Sparby, William A. Inglehart; PEERLESS MFG CO: Sherrill Stone, Donald A. Sillers Jr.;PENOBSCOT SHOE: Paul Hansen, Irving Kagan; PILGRIMS PRIDE CORP: David Van Hoose, Lonnie Pilgrim; PINKERTONS INC: Denis R. Brown, Thomas W. Wathen; PLYMOUTH RUBBER -CL A: Maurice J. Hamilburg, Daniel M. Hamilburg; POPE & TALBOT INC: Michael Flannery, Peter T. Pope POWELL INDUSTRIES INC: Thomas W. Powell, William E. Powell; PRECISION CASTPARTS CORP: William C. McCormick, Edward H. Cooley, PRIDE INTERNATIONAL INC: Paul A. Bragg, Ray H. Tolson; PULASKI FURNITURE CORP: John Wampler, Bernard Wampler; QUAKER OATS CO: Willieam D. Smithburg, Robert D. Stuart; QUALITY FOOD CENTERS INC; Stuart Sloan, Jack Croco; QUIXOTE CORP; Leslie J. Jezuit, Philip E. Rollhaus, Jr; RAVEN INDUSTRIES INC: Ronald M. Moquist, David A. Christensen; RAYCHEM CORP: Robert J. Saldich, Paul M. Cook; REYNOLDS METALS CO: Richard G Holder, David . Reynolds, Bristol, Tenn; RICHTON INTERNATIONAL CORP: Fred R. Sullivan, Franc M. Ricciardi; RITE AID CORP: Martin Grass, Alex Grass; ROBINSON NUGENT INC: Larry W. Burke, Samuel C. Robinson, ROLLINS INC; R. Randall Rollins, O. Wayne Rollins; RUDDICK CORP: Thomas W. Dickson, John W. Copeland; RUSSELL CORP: John C. Adams, Eugene C. Gwaltney; SAGE LABORATORIES INC: Carl A. Marguerite, Theodore S. Saad; SANDERSON FARMS INC:Joe Frank Sanderson Jr., Joe Frank Sanderson; SCHEIB (EARL) INC: Donald R. Scheib, Earl Scheib; SCI SYSTEMS INC: A.E. Sapp Jr., Olin B. King; SCOTT'S LIQUID GOLD: Mark E. Goldstein, Jerome J. Goldstein; SEAWAY FOOD TOWN INC: Richard B. Iott, Wallace D. Iott; SHOREWOOD PACKAGING CORP: Marc P. Shore, Paul B. Shore; SHOWBOAT INC: J. Kell Houssels III; J.K. Houssels; SKYLINE CORP: Ronald F. Kloska, Arthur J. Decio; SMITH (A O) CORP: Thomas I Dolan, L.B. Smith; SMUCKER (JM) CO -CL A: Timothy P. Smucker, Paul H. Smucker; STANDARD CP: Stephen J. Scarborough, Arthur E. Svendsen; STANDARD REGISTER CO: Peter S. Redding, John K. Darragh; STEPAN CO: F. Quinn Stepan, Alfred C. Stepan; STEWART ENTERPRISES -CL A: Joseph P. Henican III, Frank B. Stewart, Jr.; SWANK INC John Tulin, Marshall Tulin: SWISS ARMY BRANDS INC: J. Merrick Taggart, James W. Kennedy; TAYLOR DEVICES INC: Douglas P. Taylor, Paul H. Taylor; TELEPHONE & DATA:Le Roy T. Carlson Jr., Le Roy T. Carlson Sr.:TENET HEALTHCARE CORP: Jeffrey C. Barbakow, Richard K. Eamer; TENNANT CO: Janet M. Dolan, Roger L. Hale; TEREX CORP: Ronald DeFeo, Randolph Lenz; TESORO PETROLEUM CORP: Michael D. Burke, Robert V. West Jr; TIMES MIRROR COMPANY -SER A: Mark Willes, Robert F. Erburu; TOROTEL INC:Dale H. Sizemore Jr., Alfred F. Marsh; TRANSNET CORP: Steven J. Wilk, John J. Wilk; TRIDEX CORP: Seth M. Lukash, Alvin Lukash, TRINITY INDUSTRIES: Timothy R. Wallace, W. Ray Wallace; TWIN DISC INC: Michael E. Baten, John Batten; TYSON FOODS INC -CL A: Leland E. Tollett, Don Tyson; UNIFIRST CORP: Ronald D. Croatti, Aldo A. Croatti ;UNITED FOODS -CL A: James I. Tankersley, J. O. Tankersley; VALMONT INDUSTRIES: William F. Welsh II, Robert B. Daugherty; VALSPAR CORP: Richard M. Rompala, C. Angus Wurtele; WALBRO CORP: Laabert E Althaver, Walter E. Walpole; WALGREEN CO: L. Daniel Jorndt, Charles R. Walgreen III; WAL-MART STORES: David D. Glass, Sam M. Walton; WASHINGTON POST -CL B: Donald E. Graham, Katharine Graham; WATKINS-JOHNSON: Keith Kennedy, H. Richard Johnson; WAUSAU-MOSINEE PAPER CORP: Daniel R. Olvey, Arnold Nemirow; WAVERLY INC: William M Passano Jr, Edward M. Passano;WELLS-GARDNER ELECTRONICS: Francis J. Myers, Albert S. Wells Jr.; WENDY'S INTERNATIONAL INC: Gordon F. Teter, James W. Near; ;WESTON (ROY F) -SER A: William J. Marrazzo, Roy F. Weston; WESTVACO CORP: John A. Luke Jr., John A. Luke; WEYCO GROUP INC: Thomas W. Florsheim Jr., Thomas W. Florsheim; WEYERHAEUSER CO: John W. Creighton Jr., George H. Weyerhaeuser; WILLAMETTE INDUSTRIES: Steven R. Rogel, William Swindels; WINN-DIXIE STORES INC: Allen Rowland, Dano Davis; WOLOHAN LUMBER CO: David F. Wallace, Richard V. Wolohan; WOODHEAD INDUSTRIES INC: C. Mark DeWinter, Alan Reed; WORKSAFE INDUSTRIES INC: Lawrence Densen, Alan Densen; WORTHINGTON INDUSTRIES: John P. McConnell, John H. McConnell; WRIGLEY (WM) JR CO: William Wrigley,jr, William Wrigley; YELLOW CORP: George E. Powell III, George Powell Jr
Note: 1. The names of the firms are in capital letters and the names of the CEO’s are in small letters. The order is as follows: NAME FIRM: name new CEO, Name old CEO. The information on the CEO’s and firms are constructed by Perez-Gonzalez (2006) and obtained from COMPUSTAT
25
Table 2: Summary of Variables
Variables Name
Definition and Sources Mean
(Standard deviation)
Dependent Variables
OROA: The industry and performance- adjusted OROA (operating return on assets) around
CEO transitions 0 . 0 4 9 0.144 (0.090)
M-B: The industry and performance- adjusted M-B (market to book) around CEO transitions 0
. 0 2 1 1.494 (0.833) Individual Beliefs
Trust: The level of (in percentage) of trust between new CEO and old CEO 0
. 0 5 4 0.383 (0.058) Indicators of individual
Confidence in self-determination: The level of (in percentage) of confidence in self-determination between new CEO and old CEO
0 . 0 1 7 0.793 (0.006)
beliefs for culture Respect for other: The level of (in percentage) of respect for others between new CEO and old CEO 1 . 3 7
0.816 (0.065)
Obedience and authority: The level of (in percentage) of obedience and authority between new CEO and old CEO 1 . 4 9 0.905 (0.021) Summary Variables Culture
PC_Culture 1: Principal component of indicators: trust, confidence in self-determination and respect
for others 1 . 7 2 2.21e-08 (1.337)
PC_Culture 2: Principal component of indicators: trust, confidence in self-determination, respect for
others and obedience and authority
0 . 2 0 1.47e-08 (1.5158)
Ln Sales: The natural logarithm of sales at year t=-1 0
. 0 5
6.009 (1.712)
Board ownership: Fraction of ownership held by officers and directors 0
. 9 3
0.265 (0.197)
Governance Index: From Gompers et al. (2003) 0
. 0 5
8.818 (3.032)
Other Controls Industry Adjusted OROA: Difference between OROA and Median of its relevant industry 0 . 9 3
0.039 (.0938)
Industry Adjusted M-B: Difference between M-B and Median of its relevant industry 0
. 9 3
0.103 (.905)
Mean pretransition industry and performance adjusted OROA Index:
Three year pre-transitions average of the industry and performance adjusted OROA 0 . 0 5
0.009 (0.046)
Mean pretransition industry and performance adjusted M-B Index
Three year pre-transitions average of the industry and performance adjusted M-B 0.097 (0.764)
Note: 1. All the statistics are obtained from the database of (Pérez-González, 2006) except for the variables: trust, confidence in self-determination, respect for others an obedience and authority, PC_Culture 1 and PC_Culture 2, which is hand collected additional information on the origins of CEO’s
26
Figure 1: Number of changes in cultural traits after CEO-successions
Figure 2: Origin region of each CEO 72 135 0 20 40 60 80 100 120 140 160
Changes in CEO's cultural traits Change No change 0 10 20 30 40 50 60 New CEO Old CEO
27
Figure 3: Proxies of culture in each region in the U.S.
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Confidence in self determination
a great deal of choice and control
some choice and control no choice and control 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Level of Trust cannot trust depends can trust
28 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Respect for others
strongly disagree disagree agree strongly agree 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Authority and obedience
strongly disagree disagree
agree
29
Table 3 : P e r f o r m a n c e c h a n g e s a r o u n d s u c c e s s i o n s : O L S
Operating return on assets (OROA)
(1) (2) (3) (4) (5) (6) (7) Trust 0.0135 (0.0269) Confidence in self-determination -.03246 -0.018 -0.012 -0.044 -0.029 -0.332 -0.428 -0.0064 -0.339 -0.220 (0.010) (0.015) (0.014) (0.019) (0.014) (0.10) ) (0.182) (0.134) (0.09 ) (0.119) ∆ │Trust│ -0.0076 (0.0203)
Respect for other: 0.0133
(0.0250)
Obedience and authority: -.03246 -0.018 -0.012 -0.044 -0.029 -0.332 -0.428 -0.0064 -0.339 -0.220
(0.010) (0.015) (0.014) (0.019) (0.014) (0.10) ) (0.182) (0.134) (0.09 ) (0.119) ∆ │Trust│ -0.0067 (0.0844) PC_Culture 1 -1.07e-05 (0.0012) PC_Culture 2 -.03246 -0.018 -0.012 -0.044 -0.029 -0.332 -0.428 -0.0064 -0.339 -0.220 (0.010) (0.015) (0.014) (0.019) (0.014) (0.10) ) (0.182) (0.134) (0.09 ) (0.119) ∆ │Trust│ -3.29e-05 0.00013 (0.00101) (0.0023) Ln sales -.03246 -0.018 -0.012 -0.044 -0.029 -0.332 -0.428 -0.0064 -0.339 -0.220 (0.010) (0.015) (0.014) (0.019) (0.014) (0.10) ) (0.182) (0.134) (0.09 ) (0.119) ∆ │Trust│
3.25e-05 2.54e-05 2.95e-05 3.25e-05 3.49e-05 3.46e-05 -0.0025 (0.00111) (0.00111) (0.00111) (0.00111) (0.00111) (0.00111) (0.00359)
Industry adjusted OROA 0.884*** 0.884*** 0.882*** 0.883*** 0.883*** 0.883*** 0.799***
(0.0286) (0.0289) (0.0290) (0.0288) (0.0288) (0.0288) (0.0704) Industry adjusted M-B -.03246 -0.018 -0.012 -0.044 -0.029 -0.332 -0.428 -0.0064 -0.339 -0.220 (0.010) (0.015) (0.014) (0.019) (0.014) (0.10) ) (0.182) (0.134) (0.09 ) (0.119) ∆ │Trust│ 0.0101*** 0.0101*** 0.0101*** 0.0101*** 0.0101*** 0.0101*** 0.0152** (0.00325) (0.00327) (0.00329) (0.00326) (0.00328) (0.0033) (0.00600) Board ownership 0.00230 0.00268 0.00253 0.00250 0.00247 0.00248 0.00200 (0.00978) (0.00988) (0.00983) (0.00980) (0.00983) (0.00981) (0.0181)
Mean pre transition -0.0654 -0.0650 -0.0667 -0.0654 -0.0655 -0.0655 -0.0568
industry and performance adjusted OROA (0.0445) (0.0445) (0.0441) (0.0445) (0.0443) (0.0445) (0.0760) Governance Index -0.00077 (0.00098) Number of observations 207 207 207 207 207 207 71 R-squared -.03246 -0.018 -0.012 -0.044 -0.029 -0.332 -0.428 -0.0064 -0.339 -0.220 (0.010) (0.015) (0.014) (0.019) (0.014) (0.10) ) (0.182) (0.134) (0.09 ) (0.119) ∆ │Trust│ 0.933 0.933 0.933 0.933 0.933 0.933 0.906
Notes: 1. The unit of observation is CEO change per firm
2. *** denotes significant at the 1% confidence level;**, 5%;*10%. 3. The parentheses are reported the robust standard errors.
30
Table 4 : P e r f o r m a n c e c h a n g e s a r o u n d s u c c e s s i o n s : O L S
Market to Book Ratio (M-B)
(1) (2) (3) (4) (5) (6) (7) Trust -0.258 (0.307) Confidence in self-determination -.03246 -0.018 -0.012 -0.044 -0.029 -0.332 -0.428 -0.0064 -0.339 -0.220 (0.010) (0.015) (0.014) (0.019) (0.014) (0.10) ) (0.182) (0.134) (0.09 ) (0.119) ∆ │Trust│ 0.516* (0.265)
Respect for other 0.351
(0.287)
Obedience and authority -.03246 -0.018 -0.012 -0.044 -0.029 -0.332 -0.428 -0.0064 -0.339 -0.220
(0.010) (0.015) (0.014) (0.019) (0.014) (0.10) ) (0.182) (0.134) (0.09 ) (0.119) ∆ │Trust│ 0.292 (0.795) PC_Culture 1 0.019 (0.013) PC_Culture 2 -.03246 -0.018 -0.012 -0.044 -0.029 -0.332 -0.428 -0.0064 -0.339 -0.220 (0.010) (0.015) (0.014) (0.019) (0.014) (0.10) ) (0.182) (0.134) (0.09 ) (0.119) ∆ │Trust│ -0.009 -0.016 (0.012) (0.024) Ln sales -.03246 -0.018 -0.012 -0.044 -0.029 -0.332 -0.428 -0.0064 -0.339 -0.220 (0.010) (0.015) (0.014) (0.019) (0.014) (0.10) ) (0.182) (0.134) (0.09 ) (0.119) ∆ │Trust│ 0.0235* 0.0241** 0.0233* 0.0236* 0.0235* 0.0234* 0.0432 (0.0120) (0.0120) (0.0120) (0.0120) (0.0120) (0.0120) (0.0363)
Industry adjusted OROA 0.512* 0.511* 0.504* 0.529* 0.497* 0.517* 1.750**
(0.287) (0.279) (0.285) (0.286) (0.285) (0.288) (0.671) Industry adjusted M-B -.03246 -0.018 -0.012 -0.044 -0.029 -0.332 -0.428 -0.0064 -0.339 -0.220 (0.010) (0.015) (0.014) (0.019) (0.014) (0.10) ) (0.182) (0.134) (0.09 ) (0.119) ∆ │Trust│ 0.781*** 0.775*** 0.777*** 0.786*** 0.775*** 0.781*** 0.742*** (0.0725) (0.0734) (0.0724) (0.0738) (0.0720) (0.0727) (0.122) Board ownership 0.0965 0.0776 0.0939 0.0921 0.0938 0.0956 -0.287 (0.0991) (0.0982) (0.0983) (0.0992) (0.0984) (0.0990) (0.221)
Mean pre transition 0.0399 0.0434 0.0430 0.0339 0.0449 0.0401 0.0710
industry and performance adjusted M-B (0.0655) (0.0647) (0.0651) (0.0666) (0.0647) (0.0657) (0.140) Governance Index -0.0012 (0.013) Number of observations 207 207 207 207 207 207 71 R-squared -.03246 -0.018 -0.012 -0.044 -0.029 -0.332 -0.428 -0.0064 -0.339 -0.220 (0.010) (0.015) (0.014) (0.019) (0.014) (0.10) ) (0.182) (0.134) (0.09 ) (0.119) ∆ │Trust│ 0.899 0.900 0.900 0.899 0.900 0.899 0.923
Notes: 1. The unit of observation is CEO change per firm
2. *** denotes significant at the 1% confidence level;**, 5%;*10%. 3. The parentheses are reported the robust standard errors.
