Chief Executives from 1960 - 1989: A Trend toward Professionalization

There has been a longstanding debate whether or not business executives, most notably Chief Executive Officers (CEOs), populate an occupation worthy of the status of professions, the apex of the occupational hierarchy. On one side, the demands and importance of the position, as well as the prestige and remuneration of CEOs justify their place in society's highest strata, not just for the individuals, but also for their occupation. On the other side is a branch of sociology, oftentimes referred to as occupational theory, that more systematically defines professions, and to several key theorists, (Moore, 1970; Goode, 1969; Abbott, 1988) chief executives fail to qualify as a profession under their definitions. They cite there is not a prescribed educational requirement, or a self-governing association that credentials and monitors its members, much like the American Bar Association does for attorneys and the American Medical Association does for doctors.

The fact that there is no self-governing association is important since such organizations typically oversee codes of ethical conduct, thus hoping to ensure professionalism of its members. In recent years, a rash of corporate scandals such as those at Enron, WorldCom, and ImClone, has made ethical conduct especially relevant, as the CEOs of these companies have been implicated for mismanagement, thus questioning the competence and trustworthiness (in essence the professionalization) of CEOs (Byrne, 2002).

Despite the arguments by organizational theorists that chief executives do not merit the designation of profession, the fields of business and business education have argued otherwise. As long as 40 years ago, Bernard Barber (1963) performed a sociological analysis of business to conclude "American business ought to be, and is in fact becoming, more professionalized."

This paper attempts to analyze the debate whether Chief Executive Officers deserve being members of a profession. It has been an interesting question for occupational theorists, yet business and management scholars have largely ignored it, perhaps assuming executives are indisputably entrenched in the professional realm. One exception to the management literature is Trank's and Rynes's (2003) recent essay that argues that business education has become less professional, thus suggesting the professionalism of business school graduates has diminished as well.

To analyze the question of the professionalization of CEOs, this paper borrows from the work of occupational theorists who have identified criteria for an occupation to be established as a legitimate profession, and compares these criteria with the demographic traits of the CEOs of 247 companies over the course of 30 years, beginning in 1960, the approximaate time of Barber's sociological analysis, to 1989. If the field of CEOs has become more professionalized during this period, the demographic traits of CEOs should change over time to reflect the increased professionalization.

This paper has five sections. The first section reviews the literature pertaining to professions including the criteria that define professions, with arguments debating why, and why not, chief executive management deserves to be called a profession. The second section provides a brief overview of research on chief executives and the labor markets executives pursue to reach CEO positions. Section three presents the study's hypotheses, followed by section four which outlines the data, methodology and results of this research. A discussion with summery in section five concludes the paper.

Professions

Distinguishing professions from other occupations is typically the domain of sociologists, who, from a Weberian perspective, try to identify ways in which society organizes itself into tiers of social strata. Careers are but one metric to stratify society with professions occupying the highest tier of occupations. These are the jobs that require the most education, receive the greatest public prestige, and generally pay higher salaries. Medicine, law and the clergy are the quintessential professions (Carr-Saunders & Wilson 1962), but other fields such as accounting have entered into the domain. In a modern economic society, Moore (1970) notes, professionals represent the fastest growing occupational sector.

Interestingly, different countries consider professionalization differently. In the United States, professionals are more associated with the private sector, but in much of Europe, professionalization is more associated with the public sector (Torstendahl, 1990). This paper will focus on the American interpretation of professionalization.

The 1960s and 1970s were an especially fruitful era for sociologists studying occupations and professions (Collins, 1990), and in their research, several have proffered various definitions or criteria distinguishing professions from other types of jobs. Ultimately, there is no consensus definition that the field has accepted. Some may be as broad as, "occupations that claim to join theory and practice" (Broman, 1995, p. 836), or as specific as Cullen's (1978) list of eleven discrete criteria constituting professions. A review of some of the criteria is helpful to identify the more common themes in this type of research.

Two themes with which virtually all scholars agree refer to the intellectual difficulty of the job and the extensive training required. Cullen (1978) sites professions as being "complex occupations," while others have recognized the "expert" (Freidson, 1994) or "systematic" knowledge (Barber, 1963) or "theory" (Greenwood, 1962) required to perform the professional tasks. Such knowledge does not happen on the job, nor is it learned quickly. Instead, it requires university training (Wilensky, 1964) or some other form of professional education (Abbott, 1988).

Just as doctors in the U.S. have the American Medical Association and attorneys have the American Bar Association, several scholars note that other occupations seeking professional recognition tend to have selfgoverning organizations or associations (Abbott,1988; Cullen, 1978; Moore, 1970; Wilensky, 1964). The role of these associations is to provide self-monitoring for the professions' members through licensing (Abbott, 1988; Cullen, 1978) or competence testing (Cullen, 1978); establish codes of conduct (Barber, 1963) or ethics (Cullen, 1978; Greenwood, 1962; Wilensky, 1964); and sanction members who fail to comply with the profession's standards (Greenwood, 1962).

Several other characteristics of professionalism refer to the relationship between the practitioner and his or her client. Moore (1970) describes the role of professionals as having a service orientation and being autonomous, while Cullen (1978) adds that professionals are self-employed. Together these characteristics suggest a client relationship (Freidson, 1994) more like a consultant than an employee. Clients hire professionals to perform services with no certainty of a long-term relationship. As such, the nature of a professional's work is not limited to a single client or organization; instead the professional can practice in multiple settings.

Other attributes of professions worthy of mentioning, (if not mentioned as frequently) include that professionals perform altruistic service (Carr-Saunders & Wilson, 1962; Goode, 1969) as if it were a "calling" (Moore, 1970), and receive high prestige (Barber, 1963; Goode, 1969) and income (Carr-Saunders & Wilson, 1962) for their services.1

Even though these dimensions read like a checklist of discrete criteria, scholars point out that identifying professions is more complicated. Professions develop gradually, becoming institutionalized in a normative (DiMaggio & Powell, 1983) manner, as the fields adopt criteria over time. Moreover, the absence of a singularly accepted definition of a profession leaves existing ones open to interpretation. To accommodate the subjectivity of these concepts, there must be a way to acknowledge differences in interpretation of the professionalization process. Barber (1963) and Carr-Saunders & Wilson (1962) liken the professional process to a "scale," while Greenwood (1962) states that professions "fall into a continuum." These expressions suggest that some professions may be relatively weak in certain criteria, yet still deserving of professional status. While established professions may score high on the above-mentioned criteria, others fall elsewhere on the continuum. For occupations posessing some evidence of professionalism, but not enough to justify full professional status, some recognize a category of "semi-professions" among which Etzioni (1969) includes teachers, nurses and social workers.

Because of the prestige awarded to professional occupations, it is not surprising there have been efforts to elevate certain occupations to professional status. In the recent literature, Stephen Banning (1998) contends that the field of journalism has been a profession since the 1860s, with the advent of the first press associations. Charlotte Borst's (1995) historical study comparing the fields of medicine and midwifery documents the professionalization of midwives.

So too has business attempted to ensure its status as a profession; as Leicht and Fennell (2001) write: ". . .few people who have studied the history of business management would deny that management (and their academic representatives in business schools) have engaged in attempts to become more like a classic profession." Christopher McKenna (1995) argues that management consulting became a profession during the great depression as a result of changes in banking laws. Advances in organizational theory including the increased popularity of flatter and simpler organizational structures is what Bresnen and Fowler (1996) posit is advancing the professionalism of management in Great Britain. They argue flatter structures require more theoretical approaches to management, rather than technical specialties. Joseph Raelin (1997) argues the increase in CEOs being hired from outside of the organization is evidence of an executive profession in that professional CEOs can utilize their skill sets across organizational and industry boundaries.

Do CEOs Belong to a Profession?

If professionalization is reflected by career status, it would be difficult to argue that CEOs (especially of large corporations) do not belong in this elite tier. CEOs of large corporations are responsible for hundreds of thousands of workers and answer to millions of shareholders. Their companies conduct billions of dollars in business annually, and in return, CEOs earn handsome salaries and compensation packages. Due to the importance of the organizations they lead, CEO decisions often have repercussions far beyond their organizations' boundaries. Moreover, as a reviewer suggested, it is illogical that a CEO who oversees accountants and attorneys would be less of a professional than some of his or her own employees.

To support the argument that executive leadership is a profession, the field of business education is well established, with the University of Pennsylvania's Wharton School being the first in 1881. In the subsequent 120 years, hundreds of colleges and universities have added business curricula. The Association to Advance Collegiate Schools of Business (AACSB) currently recognizes 454 accredited programs (http://www.aacsb.edu). This number does not include the majority of business programs not accredited by the AACSB.

Associations or professional organizations also exist for the field of management although membership is voluntary and not required as in the medical and bar associations. Examples of business associations include The American Management Association with over 700,000 members (http://www.amanet.org), and the Business Roundtable, which includes CEOs as its members (http://www.brtable.org).

However, many occupational theorists are not yet persuaded that being a CEO qualifies as being a professional. They sight no formal organization with licensing or certification responsibilities, and there exists no code of ethics to which all CEOs must adhere, (Abbott, 1988). One of the most repeated reasons why CEOs are undeserving of professional status is that there is no prescribed educational requirement, so that CEOs can have considerably different levels of education, and vastly different educational specialties, (Goode, 1969, & Moore, 1971). As Abbott (1988, p. 132) writes, "...indeed, perhaps the reason that business administration has never successfully managed to 'professionalize' is that it commonly recruits from other professions, particularly law, accounting, and engineering."

Professionals or not, CEOs have been the subject of an enormous amount of research, and the following section will briefly outline some types of CEO scholarship.

CEO Research

No organizational position has been probed and prodded by researchers more than that of the Chief Executive Officer (CEO). Studying CEOs and their succession offers a window into organizational phenomena, and due to their high-profile positions, data on CEOs is becoming increasingly obtainable. "It comes as little surprise, then, that executive succession has been the subject of a huge volume of research, which has grown exponentially along with other research on top executives over the past fifteen years" (Finkelstein & Hambrick 1996, p. 163).2 Research topics using CEOs as subjects include compensation (Kerr & Bettis, 1987, O'Reilly, Main, & Crystal, 1988; Finkelstein & Hambrick, 1988; Carpenter, Sanders, & Gregersen, 2001; Sanders, 2001), organizational performance (Zajac, 1990; Osborn, Jauch, Maratin, & Glueck, 1981; Dalton & Kesner, 1985; Salancik & Pfeffer, 1980; Warner, Watts, & Wruck, 1988), career patterns (Useem, 1984; Useem & Karabel, 1986; Daily, Certo, & Dalton, 2000; Ocasio & Kim, 1999), industry structure (Datta & Rajagopalan, 1998), succession events (Boeker, 1992; Dalton & Kesner, 1983; Bommer & Ellstrand, 1996; Brown, 1989; Canella & Shen, 2001; Chung, Rogers, Lubatkin, & Owers, 1987; Khurana, 2001; Ocasio, 1994; Ocasio, 1999; Pitcher, Chreim, & Kisfalvi, 2000), power (Fligstein & Brantley, 1992; Ocasio, 1994; Pfeffer, 1981), personality (Miller, & Droge, 1986, Miller & Toulouse, 1986), and a myriad of other topics and combinations. Yet, for all the field knows about CEOs, there are regrettably few examples of systematic research about executive traits, and whether the demographic traits of individuals in this top position have changed over time.

Mabel Newcomer was among the first to systematically study executives in her 1955 book, The Big Business Executive. Among her findings were that chief executives were better educated and came from more privileged backgrounds than the general population. Twenty years later, Burck (1976) largely supported Newcomer's findings, although he found educational levels increased slightly, and more chief executives had come from middleclass backgrounds. In a study of executives (not just chief executives) of 1967 and 1976, Swinyard and Bond (1980) found profiles similar to Newcomer's and Burk's, and that executive profiles had not changed significantly during the time of their study. The one notable exception is that subjects with Masters of Business Administration (MBA) degrees, got their executive positions at a younger age (44 years old) than those without MBAs (47).

Neil Fligstein (1990) studied the careers of corporate presidents and CEOs between 1919 and 1979 of the 100 largest companies to document the increase of MBAs over time, with twenty percent having MBAs by the end of his study. He also studied the functional backgrounds of these CEOs. In the early years of his study, CEOs were more likely to have backgrounds in manufacturing. By the 1940s and 1950s, the number of CEOs with sales and marketing backgrounds increased, while CEOs at the end of his study were most likely to have finance backgrounds.

More recent studies including high technology industries have determined that CEOs in higher technology industries are more inclined to have backgrounds in research and development, and tend to be younger than CEOs in lower technology industries (Hambrick, Black, & Fredrickson, 1992). Meanwhile, within the high technology industries, organizations going through growth phases were more apt to have CEOs with research and development backgrounds, than organizations in more mature phases, which tended to have CEOs with backgrounds in administration, finance, and marketing (Drazin & Kazanjian, 1993).

From a sociological perspective, Michael Useem's research (1984, Useem & Karabel, 1986) identifies an elite class of CEO he labels, "The Inner Circle." These CEOs are connected by board directorships and membership on influential non-corporate entities such as the Business Roundtable. As a group, these CEOs have backgrounds of social privilege as well as educational backgrounds from the country's most elite academic institutions, suggesting social capital partly determines who becomes CEOs. Similarly, Peter Temin, (1999) analyzed why, throughout the 20th century, the business elite has remained predominantly white males from Protestant families. He attributes this to this group's access to superior education over time, rather than blatant discrimination.

Two theories explain career trajectories for chief executives, Firm Internal Labor Market (FILM) Theory and Occupational Internal Labor Markets (OILM) Theory, (Althauser & Kalleberg, 1981). Internal labor markets exist in a single organization where executives are groomed by their organization through a, "system of 'promotions' according to seniority, or to achievement, or both" (Weber, 1947, p. 334). This theory predicts individuals get hired at entry-level positions, and work their way to executive positions. The prestige and financial rewards associated with promotions are the motivation for employees to excel at their jobs. According to FILM theory, only the most talented and loyal will reach the apex to become CEO.

The alternative perspective, occupational internal labor markets suggests individuals are still motivated by promotions (as in internal labor markets), but they may have to cross organizational boundaries to achieve their advancement. This can happen only when the individual's occupation requires skills that are transferable to other organizations, a distinct characteristic of a profession (Raelin, 1997). Therefore, increased use of occupational internal labor markets would be a signal of greater professionalization.

Hypotheses

If an occupation has elevated its stature to that of a profession, occupational sociologists previously mentioned in this paper (Abbott, 1988; Carr-Saunders & Wilson, 1962; Wilenski, 1964) claim the field would require certain criteria such as high levels of required education or a professional organization monitoring the membership and activities of its participants. Rather than observe the changes in the field of chief executive, this paper argues one can document the professionalization of a field by documenting changes in its participants. The following hypotheses pertaining to CEO traits will determine whether CEOs have become more professionalized. They are organized into two categories: traits reflecting professional knowledge and human capital (versus social capital), and traits reflecting professional career patterns.

Occupational theorists' most repeated criterion of professions refers to the knowledge and educational requirements of practitioners. The following four hypotheses refer to the educational levels of CEOs, the existence of a specialized degree (MBA), and increasing reliance on human capital, rather than social capital to earn their positions.

1) Education level: Since knowledge is a key component of virtually all definitions of a profession, and that knowledge is gained through professional or university educations, one should find the educational levels of CEOs to have increased over the course of the study.

2) Existence of MBA: As the Masters of Business Administration becomes more recognized as the necessary training for business professionals, the likelihood of CEOs having an MBA should increase over the course of the study.

More reliance on human capital rather than social capital: As the skill sets of CEOs become more established, executive knowledge, or human capital (Becker, 1975), will matter more than social capital such as family connections and friendship networks.

3) Since Ivy League colleges are associated with social capital (Judge, Cable, Boudreau & Bretz, 1995; Useem & Karabel, 1986), the percentage of undergraduate alumni from the eight Ivy League colleges and universities will decrease during the course of the study.

4) Like elite educational institutions, memberships in social clubs are another metric of social capital, so one should find the average number of social club memberships to decrease over time.

Professional career patterns should reflect the autonomous (Moore, 1970) nature of professionals. As such, CEOs should be able to pursue their careers through Occupational Internal Labor Markets (Althauser & Kalleberg, 1981) rather than through a single organization. This should be reflected in the number of previous employers, the organizational status in which CEOs joined their organizations, length of organizational tenure before becoming CEOs, and the percentage of CEOs hired from outside the organization.

5) Number of previous employers: Greater professionalization predicts CEOs will take advantage of the transferability of their professional skill sets (Raelin, 1997) and increasingly change employers within the occupational internal labor market Therefore, over the course of the study, the number of previous employers for CEOs should increase.

6) Status of position upon entering organization: Higher reliance on occupational internal labor markets, and less reliance on firm internal labor markets suggests that CEOs will join their organization at increasingly high levels during the course of the study.

7) Organizational tenure before becoming CEOs: Greater professionalization promotes less reliance on firm internal labor markets, so that, over time, individuals will have less tenure in their organizations prior to reaching CEO.

8) Percentage of outside CEOs: Relying on external labor markets and transferable skill sets, the percentage of CEOs hired from outside their organizations will increase over the course of the study.

Data, Method and Results

Sample of Organizations

All of the organizations in this sample were listed in Fortune magazine's 1975 listing of the largest organizations in industry, commercialbanking, insurance, diversified financial, retailing, transportation or utilities. The magazine publishes its annual "Fortune 500" of the largest 500 industrial companies, and lists the largest fifty firms in six other industries. The sample came from the 1975 because it is the mid-point of the time period of observation and therefore the firms would have the greatest probability of having collectable data during all thirty years of the collection period. From the different lists in Fortune magazine, the possible population of firms was 800 organizations. Because personal data is more available for CEOs of larger firms, there is predictably a bias towards large organizations. However, the thoroughness of the data collection justifies these possible biases in sampling.

In order to ensure that the firms in the sample were representative of industry distributions in the population, the firms were categorized by the first two digits of their 1975 Standard Industrial Classification (SIC) code. Standard and Poor's Register of Corporations, Directors and Executives publishes SIC codes for companies. In cases where more than one SIC code was listed, the firm was categorized by its first code as that represents the primary industry of the organization. After eliminating firms that did not exist through the entirety of the study, a combination of sequential and stratified sampling resulted in a sample of 247 companies, an arbitrary number large enough to be statistically viable, while manageable for data collection.

Sample of CEOs

CEO names came from the annual listings in Standard and Poor's Register of Corporations, Directors and Executives. Interestingly, "Chief Executive Officer" was not a title used during the first years of the study. It was first used in 1964 by seven of the firms and was quickly adopted by others. By, 1989 fully 95% of the firms in the sample listed someone with "CEO" as a title (see Figure 1).

For organizations not specifically mentioning a CEO, the first name listed was included into the sample (unless that person had, "emeritus" in his or her title). Normally, the title of the first person listed represented the organization's top position, often, "Chairman," "Chairman of the Board," or "President." Sampling CEOs (or other top executives, whom I'll refer to as "CEO" for simplicity) in this manner gave the name of the individual as well as the year in which they were first listed as CEO. During the thirty-year period of the study, the 247 companies in the sample had 1,070 different CEOs.

Biographical Data of CEOs

The individual data on CEOs came predominantly from a biographical directory, Who's Who in Commerce and Industry up until 1970, after which it changed its name to Who's Who in Finance and Industry. In cases where CEOs' biographies were not included in this source, biographical information was sometimes available in Who's Who in America, or Forbes magazine's annual listing of the 800 highest paid executives in America. Missing data is a persistent problem in archival research (Fligstein, 1990, p. 282) especially when being listed in directories such as Who's Who is voluntary. Nevertheless, of the 1,070 CEOs who worked for the companies during the period of the study, 867 had some biographical information available from the sources.

For this study, data collected from the biographical sources were: educational history including degrees earned, and alma maters; the number of previous employers; the year in which they joined their current employer; organizational status upon joining their current employer; the year in which they took over as CEO (or other chief executive position); and the number of social club memberships.

Education level was coded "0" if the individual did not have a college degree; "1" if they had only an undergraduate degree, "2" if they had two degrees such as a masters, "3" (or higher) for every degree, if they earned at least three degrees. Masters of Business Administration was coded separately as a dummy variable, "1" if they had an MBA, and "0" if they did not.

Organizational status was determined by a group of seven organizational scholars (five Ph.D. candidates and two faculty) who rated a sampling of job titles on a five-point scale. For job titles that were believed to be non-exempt from the Fair Labor Standards Act, or paid on an hourly basis, the group members assigned a value of "1." Jobs that had titles suggesting exemption from wage and hour laws, but were non-managerial in nature were assigned a value of "2." Titles assigned a "3," were those suggesting entry-level management. Titles that were executive in nature, received a "4." And finally, "5s" were awarded to jobs signifying the very most senior positions.

Social capital was determined by the CEOs' undergraduate alma mater and memberships in social organizations. CEOs who graduated from the eight Ivy-League colleges and universities were coded as "1" while graduates from other schools were coded as "0." Because the ivyleague institutions are widely regarded as some of the most elite and expensive schools in the country, those who matriculate into these colleges would likely have more social capital or social advantages (e.g., wealth and family connections) than those who graduate from other colleges (Judge, Cable, Boudreau & Bretz, 1995; Useem, 1984; Useem & Karabel, 1986). Additionally, social capital was measured by the number of different social organizations listed in the CEOs' biographies. Since memberships in social clubs are a means for networking, it follows that the number of memberships would be a measure for social capital.

Methods and Operationalization

To determine whether CEO traits have changed over time, each of the individual variables was tested independently using analysis of variance (ANOVA) techniques across the time of the study to determine if these traits had changed for new CEOs. Traits of all new CEOs for each year were collected and clustered into six five-year periods. Combining new CEOs into five-year clusters allows for greater sample size in each of the categories, thus improving the power of the analysis of variance. Clustering the data into six halfdecade clusters was an arbitrary decision based on statistical power of the ANOVA analysis and simplicity in identifying trends. Scatter plots using individual years rather than five-year clusters reflected the same patterns as the six separate clusters.

Results

Table 1 contains the results of the analysis of variance for CEO traits, showing significant changes across all traits. With the exception of one variable, Number of Previous Employers, all the changes were in the direction predicted by the hypotheses, such that scores for the earliest cohort, 1960-1964 was significantly different than the 1985-1989 cohort. Charts for each of the characteristics offer visual evidence of changes and trends.

The level of education of new CEOs increased every time period, except for the early 1980s when there was a slight dip. As illustrated in Figure 2, the typical CEO of the early 1960s had an undergraduate degree, and about a forty percent chance of having a second degree. By the late 1980s, new CEOs were nearly eighty percent likely to have the second degree. This trend supports Burck's (1976) findings that CEOs were better educated than those in Newcomer's 1955 study.

The increase in educational levels and second degrees appears to be partially explained by the increasing presence of MBAs during the thirty years of the study. In the earliest cohort, less than seven percent of new CEOs had MBAs, but that increased nearly five-fold for the latest cohort. Figure 3 presents the percentage of new CEOs with MBAs showing consistently higher representation except for the early 1980s when there was a slight drop. The percentage of CEOs with MBAs is quite consistent with Fligstein's (1990) findings.

Supporting the assertion that CEOs will be decreasingly reliant on social and cultural capital and more reliant on human capital, Figure 4 shows that each cohort of new CEOs was less likely to have graduated with an undergraduate degree from an elite Ivy League university than its predecessor. Over the observed time period, the likelihood of a new CEO with an Ivy League bachelor's degree halved from 34.5% in the early 1960s to 17.2% in the late 1980s.

Figure 5 continues the analysis of the trend from social capital to human capital by displaying the average number of memberships in social organizations. Using social club memberships as a metric for social capital, we find CEOs were members of fewer social organizations throughout the years of study. In the 1960s, the average new CEO was a member of nearly five social organizations (e.g., city clubs, university clubs, golf clubs, fraternal organizations, etc.), whereas by the end of the study, the average CEO was a member of threeto-four clubs. The downward trend was mostly continuous with one exception in the early 1980s, which found a slight rise over CEOs in the late 1970s.

Testing the transition from internal to external labor markets, the number of previous employers of new CEOs was predicted to increase over the course of the study. With the exception of the first cohort, this appears to be the case. Interestingly, the first cohort had, on average, 2.16 previous employers an amount slightly less than the last cohort's average of 2.27. However, average number of previous employers in the second cohort dropped to 1.56 and increased steadily throughout the remainder of the study. Post hoc analysis finds the Number of Previous Employers is the only variable where the 1960-64 group of CEOs was not significantly different from the 1985-89 group. One possible explanation for the earliest group's high number of previous jobs is that these individuals would have been more likely than other cohorts to have had their careers disrupted by World War II.

Not surprisingly, executives with broader work experience entered their organizations at higher levels later in the study. The 1960-64 group averaged an entry score of just over 3.70 signifying management-level positions. By the end of the study, CEOs had started in their organizations at nearly 4.26, or within the executive ranks. Figure 7 shows the overall trend with slight dips in the second and fifth cohorts that kept this trend continuing every time period. Consistent with Temin's (1999) research, this data show that relatively few CEOs experienced Horatio Alger-like career progressions where they started at the bottom of the organization and worked their way to the top. In fact, most CEOs begin working for their companies well positioned within their organizational hierarchies.

That eventual CEOs were joining their organizations at higher levels, it follows their tenures before coming CEOs would decrease. As Figure 8 shows, new CEOs in the early 1960s had to wait in their organizations over 27 years before becoming CEOs. By the end of the study, the waiting period was nearly a decade less at 17.4 years. The downward trend would have been continuous except for a increase in the early 1980s.

One other way of analyzing this is looking at the number of CEOs who joined their organizations as CEOs, or in positions preparing them to be the next CEO. Figure 9 contains a graph of the percentage of CEOs who had two or fewer years of tenure, suggesting they became CEOs immediately, or had a brief transition before their promotion. Of the earliest cohort, 3.3 percent joined their organizations within two years of becoming CEO. Throughout the time of the study, the percentage increased slowly and steadily until the 1985-89 period, when the percentage of new CEOs who came from outside their organizations leapt to over 28 percent.

Discussion

As predicted, each of the traits of CEOs changed significantly during the thirty-year period, 1960-1989 in a manner suggesting greater professionalization of the field. New CEOs through this period increasingly relied more heavily on human capital (Becker, 1975) as evidenced by increasing educational levels and greater reliance on a specialized graduate degree, the Masters of Business Administration. As human capital increased, it appears that new CEOs relied less on their social capital. Ivy League affiliation, a measure of social capital (Useem, 1984; Useem & Karabel, 1986; Judge et al., 1995) decreased every period, as did the number of memberships in social organizations. This decreased reliance on social capital, with increased educational backgrounds suggest new CEOs bring more intellectual strength to their position, and this human capital matters more, over time, than do the affiliations and connections of membership organizations and elite educational institutions.

In comparison to their predecessors, CEOs later in the study were less likely to pursue internal labor markets to become CEOs, increasingly relying on occupational labor markets, suggesting their skills and talents were transferable across organizational boundaries, which Raelin (1997) cites as evidence of a profession. Newer CEOs had more experience with other employers, and they joined their organizations later than their predecessors. Over the course of the study, the wait within an organization decreased by nearly ten years from 27 to 17. Similarly, the number of CEOs hired from outside the organization within two years jumped markedly, especially at the end of the study when over 28% of new CEOs had been with their employer for two years or fewer.

However, none of these results addresses the most fervent critics' (Abbott, 1988; Goode, 1969; Moore, 1970) concerns of why management does not deserve to become a profession. There still exists no mandatory educational requirement, nor is there a professional association credentialing members and monitoring ethical codes of conduct. Yet these criteria are those of the most conservative occupational theorists, and other sociologists (Freidson 1983, 1986; Leicht & Fennel, 2001) have questioned whether these definitions are the last word in what defines a profession. Freidson (1986) recognizes confusion and lack of consensus of what constitutes a profession, arguing that the perceptions of other groups such as practitioners and the public at large have their own validity (1983). Leicht and Fennell (2001) see greater overlap between management and the established professions. They believe managers are becoming more professional, while professionals are becoming more managerial. They cite the rise of HMOs where doctors have to concern themselves with administrative and cost-cutting responsibilities in addition to practicing medicine.

Independent of this research, recent corporate scandals and bankruptcies have raised questions regarding the ethics of several high-profile CEOs. In the wake of these events, the Security and Exchange Commissions has recently required CEOs to sign financial statements swearing to the accuracy of the documents (SEC File No. 4-460), thereby making the SEC the organization that monitors the ethics of CEOs. Even though this is a coercive, rather than normative (DiMaggio & Powell, 1983), institutional process of professionalization, it, to some degree, addresses the concerns of some occupational theorists (Abbott, 1988; Moore, 1970) that CEOs had no code of ethics, or an organization to police them. It's important to note that codes of ethical conduct do not guarantee ethical behavior for established professions. In the Enron case, Arthur Andersen was found to have acted unethically, even though the accounting profession has its own code of ethics.

In the context of the findings reported here, it is important to remember that the transition from an occupation to a profession is not instantaneous, rather it progresses along a scale or continuum (Barber, 1963; Carr-Saunders & Wilson, 1962; Greenwood, 1962). An unanticipated finding of this research was discovering how recently the title Chief Executive Officer had been adopted in major organizations. Prior to 1963, the year Barber suggested that business be considered a profession, no companies in this sample of 247 Fortune companies listed a CEO among their executives or directors. Yet, within the next 26 years, all but a very few of the companies in the sample had adopted a CEO position in their hierarchy. That the title became institutionalized during the period in this study offers evidence that organizational structures changed along with the demographic traits of individuals in the position.

If Barber was premature to suggest management was a profession in 1963, this study presents evidence that while the position of CEO grew from infancy in the early 1960s through the 1980s, the traits of CEOs changed in a manner, such that the field of chief executive rightly deserves to move considerably higher on the continuum toward professionalization.

Implications for Further Research

This paper identifies changes in general demographic characteristics consistent with an occupation that is becoming more professionalized, but these findings do little to explain the skills and abilities these individuals had that promoted them to the highest echelon of business. Studying functional backgrounds offers some insight, but leading a large organization requires much more than a strong foundation in a single functional area. Functional backgrounds fail to tell a complete story since CEOs essentially forsake some their functional expertise to take on broader administrative responsibilities that position them to lead an entire organization, (Freidson, 2001).

It would behoove leadership scholars to identify the management or organizational skills that enabled CEOs to progress through the labor markets (internal and external) to reach their positions. Yet these skills and attributes are difficult to identify and collect from traditional data sources. Similar to skills and attributes is the concept of social capital. This paper used Ivy League affiliation and social club memberships, but these are only two forms of social capital. Identifying different types of social capital from existing databases would offer scholars more opportunity to identify the networks in which CEOs operate.

To complicate matters further, the 21st century brings new challenges that were not as evident during the time of this study. Global competition, new technologies, and changes in public policy (Hitt, 2000; Nadler and Heilpern, 1998) are the key drivers of the new hyper-competitive business environment, and the leadership traits of CEOs will need to accommodate the changing environment. Whether new leadership requirements will equate to changed demographic characteristics remains to be seen with additional research.

With the changing competitive environment, another interesting pursuit would be to study, not just how CEOs have changed, but also how the nature of the position has changed. Hitt (2000) argues that being a CEO is more difficult, and he cites decreasing tenure for those in the position as evidence. Integrating changing demands of the position with qualifications of CEOs would shed considerable light on what types of individuals successfully become CEOs.

Summary

Due to their positions of prestige and importance, CEOs garner a lot of attention from scholars, not limited to the field of business. The purposes of this paper were largely twofold, 1) to present the criteria and the debate why some occupational theorists have concluded that CEOs do not belong to a profession, and 2) to analyze changes in CEO traits over time to determine whether CEOs have changed in a manner consistent with an evolving profession.

During the period of this analysis, CEOs have seen increases in their human capital by being more education and having greater likelihood of earning a business-specific degree, the MBA. They are less reliant on social capital as evidenced by fewer Ivy League graduates and fewer memberships in social organizations. CEOs have also seen their skills become more transferable, a characteristic of professions, and they increasingly capitalized on occupational labor markets rather than firm internal labor markets. These changes signify greater professionalization of the individuals who became CEOs, and despite the criticisms of conservative occupational theorists, these changes reflect professionalization of the field.