Business leaders’ characteristics and decisions - a synthesis of research in business administration and behavioral economics
During the sabbatical, the plan is to develop a synthesis based on a broader interdisciplinary project studying how business leaders make decisions in strategic games and incentivized elicitation tasks, and how such decisions can be associated with how firms are run. In the synthesis, the aim is to elaborately and critically merge the research from experimental and behavioral economics, primarily on business leaders, with the existing empirical and theoretical work in the management and finance literature.
The research of my co-authors and me that will be one part of the synthesis. This is based on a unique random stratified sample of 700 Chinese CEOs who have participated in three waves from 2009 to 2016, where they have provided information about their background, and their firms, and also made incentivized decisions in games and elicitation tasks. The project has so far studied decisions under uncertainty, in strategic games, and reciprocity situations resulting in six publications in relevant journals (Acad Man Jnl (1), Exp Econ (1), Man Sci (1), Org Sci (3)).
The plan is to write a book manuscript on the synthesis. The sabbatical involves two four-week-long visits to Cornell University and plans for deepened collaboration.
The research of my co-authors and me that will be one part of the synthesis. This is based on a unique random stratified sample of 700 Chinese CEOs who have participated in three waves from 2009 to 2016, where they have provided information about their background, and their firms, and also made incentivized decisions in games and elicitation tasks. The project has so far studied decisions under uncertainty, in strategic games, and reciprocity situations resulting in six publications in relevant journals (Acad Man Jnl (1), Exp Econ (1), Man Sci (1), Org Sci (3)).
The plan is to write a book manuscript on the synthesis. The sabbatical involves two four-week-long visits to Cornell University and plans for deepened collaboration.
Final report
Much of the time of the sabbatical has been spent working on a manuscript about the characteristics of business leaders and observations about their decisions and outcomes for their businesses. This is an attempt at synthesizing research on this topic in behavioral economics and business administration. I will describe here the most important parts of this work.
While an important part of this work is about summarizing the results of my co-authors' and my studies on CEOs in China, a fairly significant part has been about structuring the data, especially the data captured in the last "wave" of our study of CEOs in China. As for the work on the manuscript, this consists of a couple of main parts which are briefly described below.
In the introduction, a brief historical review of the research on the characteristics of business leaders is given based on UET (upper echelons theory) and a behavioral economics perspective. Among other things, it is noted that while the formation of theory within UET can be seen as a natural development within business administration, behavioral economics has its roots in a dissatisfaction with the neoclassical economics’ assumption of "homo economicus". Based on these different starting points, the various research fields have gradually approached each other in such a way that economic actors (and thus CEOs) are seen as boundedly rational and have preferences that are not entirely self-centered.
The characteristics studied are cognitive abilities, risk aversion, cooperativeness, trust, reciprocity, altruism, and narcissism/self-esteem. For each characteristic, how to use theoretical reasoning in the management and organizational literature as well as in behavioral economics is discussed. The ambition is to link the characteristics to expected observations about CEOs' decisions and company outcomes. For example, a starting point within UET is that business leaders act under bounded rationality and that this leads to different interpretations depending on the degree of rationality and what experiences the business leader has had. A consequence of this is that the cognitive abilities of business leaders can be expected to play a role in which decisions are made. Within behavioral economics, there are also theories about the degree of rationality. Within the so-called level-k theory, the ability of individuals to create expectations about the opponents' strategy and degree of sophistication is decisive for the choice of strategies in certain situations. Based on both theories, it can be argued that there is heterogeneity concerning cognitive abilities and that these can have an impact on how business leaders interpret their decision-making situation and thus which decisions are made. This primarily applies to strategic decisions that involve other actors. A concrete result that is analyzed in this context is a strong statistical relationship between our Chinese CEOs' cognitive reflection score and their choice of reciprocity strategy in a strategic game.
A central part of the work has been to describe potential links between the CEOs' characteristics and observations of their decisions. This connection is not trivial as described, among other things, in a well-known follow-up article by Donald Hambrick in 2007 which will be analyzed in detail in a section on methodological issues to be shortly described below. Purely theoretically, the connection between characteristics and decisions can be straightforward, while empirical investigations of the connections are more complicated, not least because certain data are difficult to capture. For example, one can expect that CEOs with a low degree of risk aversion are more likely to make investments in markets where the variance in returns is high compared to more risk-averse CEOs. By extension, such differences in risk preferences should be reflected in observations of firms' investments and returns. The decisions and outcomes studied include the choice of strategy in different types of games, their leadership style in their firms (regarding acquisition activity, investments, debt ratio, CSR activities) as well as outcomes for firms (development of profits, market value, etc.) However, it is not easy to find unambiguous empirical measures of CEOs' risk aversion that can be linked to clear measures of the degree of risk exposure in decisions. An important contribution to the literature on the characteristics and decisions of business leaders is an analysis and categorization of the different methods used. Within the research on CEOs' characteristics and decisions, a large number of methods have been used, which is why such an analysis is important to evaluate the results. The methods used to collect data include interviews, questionnaires, collected register data, and decisions in experiments. Concerning indications of CEOs' characteristics, a distinction is made between direct measures that are observed in connection with CEOs' answers to questions regarding a characteristic (e.g. risk aversion) and indirect measures such as e.g. design of an option contract which is interpreted as indicating the same characteristic. Similarly, more or less direct observations of decisions and outcomes are discussed. In cases where the observations are not directly related to a characteristic, decision, or outcome, possible moderating factors and whether these affect the validity of the study are discussed. In this context, a number of known methodological problems are analyzed, such as the fact that certain results tend to be linked to the same method being used, that statistical significance levels are not corrected when testing many hypotheses at the same time, and the risk of so-called "data mining". Part of the sabbatical was spent completing a paper on reciprocity behavior among CEOs. We ask several questions and can report several results in this study. The first question is whether reciprocity strategies are common among CEOs. To answer this question, we study the frequency of different strategies according to the typology proposed in the literature. We find that reciprocity strategies, also called conditional cooperation strategies, are remarkably common among our sample of CEOs. We also examined the relationship between reciprocity and CEOs' individual characteristics and their leadership style. In the theoretical part, we discuss potential individual background factors that may influence the CEO's propensity to choose a reciprocity strategy. In the same section, inspired by UET, we discuss the possibility that the CEO's reciprocity strategy can be traced to different outcomes for the company or to how the CEO leads the company. We observe that CEOs with the strictest reciprocity strategy called perfect conditional cooperators are more likely to have high cognitive reflection test scores and Buddhist beliefs than those with other strategies. When we study the leadership style of CEOs, we find that perfect conditional partners are more likely to be the largest shareholder in their companies and less likely to be involved in business disputes with their suppliers and customers.
Collaboration and dissemination:
In the spring of 2023, I visited Professor Victor Nee at Cornell University in Ithaca. We discussed several issues about our work and the data we collected. In particular, this concerned questions concerning reciprocity behavior as this is a new area in our collaboration. Collaboration has continued throughout the sabbatical and has resulted, among other things, in my being invited to give a talk at a workshop at Cornell Center in New York in February 2024.
During the fall of 2023, I visited the University of Amsterdam for a couple of weeks, where Professor Randolph Sloof was my host. He is one of the researchers in Europe who has conducted research on entrepreneurs and business leaders that is closest to my research on CEOs. It was a very fruitful couple of weeks with long discussions with him and his colleagues. I also had the opportunity to present my latest research in the field at a larger seminar. I have already been asked if I can participate in a grading committee at a university in Amsterdam later this spring. As for the work on the synthesis, the work on the manuscript and the data connected to it has been extensive and is still ongoing. The plan is that when I finally decide how to structure the various parts, I will send in a book proposal to an international publisher. Part of this script will build on the research on reciprocity that I recently conducted with Victor Nee. This part has resulted in an essay entitled "Reciprocity among CEOs". The paper has not yet been published but has been submitted to the Academy of Management Annual Meeting, which will be held in Chicago, in August 2024. If accepted, I expect it to receive significant exposure.
While an important part of this work is about summarizing the results of my co-authors' and my studies on CEOs in China, a fairly significant part has been about structuring the data, especially the data captured in the last "wave" of our study of CEOs in China. As for the work on the manuscript, this consists of a couple of main parts which are briefly described below.
In the introduction, a brief historical review of the research on the characteristics of business leaders is given based on UET (upper echelons theory) and a behavioral economics perspective. Among other things, it is noted that while the formation of theory within UET can be seen as a natural development within business administration, behavioral economics has its roots in a dissatisfaction with the neoclassical economics’ assumption of "homo economicus". Based on these different starting points, the various research fields have gradually approached each other in such a way that economic actors (and thus CEOs) are seen as boundedly rational and have preferences that are not entirely self-centered.
The characteristics studied are cognitive abilities, risk aversion, cooperativeness, trust, reciprocity, altruism, and narcissism/self-esteem. For each characteristic, how to use theoretical reasoning in the management and organizational literature as well as in behavioral economics is discussed. The ambition is to link the characteristics to expected observations about CEOs' decisions and company outcomes. For example, a starting point within UET is that business leaders act under bounded rationality and that this leads to different interpretations depending on the degree of rationality and what experiences the business leader has had. A consequence of this is that the cognitive abilities of business leaders can be expected to play a role in which decisions are made. Within behavioral economics, there are also theories about the degree of rationality. Within the so-called level-k theory, the ability of individuals to create expectations about the opponents' strategy and degree of sophistication is decisive for the choice of strategies in certain situations. Based on both theories, it can be argued that there is heterogeneity concerning cognitive abilities and that these can have an impact on how business leaders interpret their decision-making situation and thus which decisions are made. This primarily applies to strategic decisions that involve other actors. A concrete result that is analyzed in this context is a strong statistical relationship between our Chinese CEOs' cognitive reflection score and their choice of reciprocity strategy in a strategic game.
A central part of the work has been to describe potential links between the CEOs' characteristics and observations of their decisions. This connection is not trivial as described, among other things, in a well-known follow-up article by Donald Hambrick in 2007 which will be analyzed in detail in a section on methodological issues to be shortly described below. Purely theoretically, the connection between characteristics and decisions can be straightforward, while empirical investigations of the connections are more complicated, not least because certain data are difficult to capture. For example, one can expect that CEOs with a low degree of risk aversion are more likely to make investments in markets where the variance in returns is high compared to more risk-averse CEOs. By extension, such differences in risk preferences should be reflected in observations of firms' investments and returns. The decisions and outcomes studied include the choice of strategy in different types of games, their leadership style in their firms (regarding acquisition activity, investments, debt ratio, CSR activities) as well as outcomes for firms (development of profits, market value, etc.) However, it is not easy to find unambiguous empirical measures of CEOs' risk aversion that can be linked to clear measures of the degree of risk exposure in decisions. An important contribution to the literature on the characteristics and decisions of business leaders is an analysis and categorization of the different methods used. Within the research on CEOs' characteristics and decisions, a large number of methods have been used, which is why such an analysis is important to evaluate the results. The methods used to collect data include interviews, questionnaires, collected register data, and decisions in experiments. Concerning indications of CEOs' characteristics, a distinction is made between direct measures that are observed in connection with CEOs' answers to questions regarding a characteristic (e.g. risk aversion) and indirect measures such as e.g. design of an option contract which is interpreted as indicating the same characteristic. Similarly, more or less direct observations of decisions and outcomes are discussed. In cases where the observations are not directly related to a characteristic, decision, or outcome, possible moderating factors and whether these affect the validity of the study are discussed. In this context, a number of known methodological problems are analyzed, such as the fact that certain results tend to be linked to the same method being used, that statistical significance levels are not corrected when testing many hypotheses at the same time, and the risk of so-called "data mining". Part of the sabbatical was spent completing a paper on reciprocity behavior among CEOs. We ask several questions and can report several results in this study. The first question is whether reciprocity strategies are common among CEOs. To answer this question, we study the frequency of different strategies according to the typology proposed in the literature. We find that reciprocity strategies, also called conditional cooperation strategies, are remarkably common among our sample of CEOs. We also examined the relationship between reciprocity and CEOs' individual characteristics and their leadership style. In the theoretical part, we discuss potential individual background factors that may influence the CEO's propensity to choose a reciprocity strategy. In the same section, inspired by UET, we discuss the possibility that the CEO's reciprocity strategy can be traced to different outcomes for the company or to how the CEO leads the company. We observe that CEOs with the strictest reciprocity strategy called perfect conditional cooperators are more likely to have high cognitive reflection test scores and Buddhist beliefs than those with other strategies. When we study the leadership style of CEOs, we find that perfect conditional partners are more likely to be the largest shareholder in their companies and less likely to be involved in business disputes with their suppliers and customers.
Collaboration and dissemination:
In the spring of 2023, I visited Professor Victor Nee at Cornell University in Ithaca. We discussed several issues about our work and the data we collected. In particular, this concerned questions concerning reciprocity behavior as this is a new area in our collaboration. Collaboration has continued throughout the sabbatical and has resulted, among other things, in my being invited to give a talk at a workshop at Cornell Center in New York in February 2024.
During the fall of 2023, I visited the University of Amsterdam for a couple of weeks, where Professor Randolph Sloof was my host. He is one of the researchers in Europe who has conducted research on entrepreneurs and business leaders that is closest to my research on CEOs. It was a very fruitful couple of weeks with long discussions with him and his colleagues. I also had the opportunity to present my latest research in the field at a larger seminar. I have already been asked if I can participate in a grading committee at a university in Amsterdam later this spring. As for the work on the synthesis, the work on the manuscript and the data connected to it has been extensive and is still ongoing. The plan is that when I finally decide how to structure the various parts, I will send in a book proposal to an international publisher. Part of this script will build on the research on reciprocity that I recently conducted with Victor Nee. This part has resulted in an essay entitled "Reciprocity among CEOs". The paper has not yet been published but has been submitted to the Academy of Management Annual Meeting, which will be held in Chicago, in August 2024. If accepted, I expect it to receive significant exposure.