Division of Research Graduate School of Business Administration OUTSIDER EXECUTIVE SUCCESSION WITHIN THE FORTUNE 1000: A FAVORABLE IMPACT ON SHAREHOLDER WEALTH Working Paper #511 Margarethe F. Wiersema University of Michigan FOR DISCUSSION PURPOSES ONLY None of this material is to be quoted or reproduced without the expressed permission of the Division of Research. Copyright 1987 University of Michigan School of Business Administration Ann Arbor Michigan 48109 April 1987

OUTSIDER EXECUTIVE SUCCESSION WITHIN THE FORTUNE 1000: A FAVORABLE IMPACT ON SHAREHOLDER WEALTH By Margarethe F. Wiersema Graduate School of Business The University of Michigan Ann Arbor, Michigan 48109 (313)763-0599

-2 - OUTSIDER EXECUTIVE SUCCESSION WITHIN THE FORTUNE 1000: A FAVORABLE IMPACT ON SHAREHOLDER WEALTH Top management's influence on corporate strategy and thus firm performance has been a central premise within the field of business policy. Yet empirical research has focused on market and economic determinants of performance. The succession literature, on the other hand, has researched the link between leadership and performance with mixed results. Yet there is evidence from the divestiture and turnaround literature that executive successions that lead to changes in managerial perspectives may lead to important strategic events. Given this link, it is hypothesized that the stock market may reflect anticipated strategic change due to changes in the executive constellation. This study explores the effect of succession within top management on shareholder wealth for a random sample of Fortune 1000 firms during the 1968 -1982 time period. The utilization of stockmarket data enables the isolation of the succession effect while holding industry and organizational factors constant. This study differentiates succession events according to the top management position: CEO or President; and the organizational background of the executive: insider or outsider. Since the top management structure varies within firms, not all executive selections may be regarded as significant events impacting the organization's strategic course. In addition the executive succession literature provides support that organizational bakcground may be important in determining the degree of turbulence experienced by the organization. The empirical results indicate CEO and President succession events are not treated uniformly by the stock market. A significant, positive shareholder effect occurs when firms announce president successions from outside the organization. On the other hand, CEO succession regardless of organization background did not result in a significant shareholder effect.

-3 - OUTSIDER EXECUTIVE SUCCESSION WITHIN THE FORTUNE 1000: A FAVORABLE IMPACT ON SHAREHOLDER WEALTH Introduction Within the strategy field, top management has been viewed as being primarily responsible for the development and execution of corporate strategy which determines long-run organizational performance. Corporate strategy can be viewed as the "stream of decisions over time that reveal management's goal for the corporation and the means they choose to achieve them" (Donaldson and Lorsch, 1984). Exactly how this responsibility is managed can be viewed quite differently: through direct intervention in the substance of strategy (Barnard, 1938; Andrews, 1980; Ansoff, 1965); by designing the appropriate administrative context (Bower, 1970;) or by assimilating and influencing dominant coalitions within the firm (Quinn, 1980). Regardless of the means used, it is the CEO who ultimately influences the strategic direction for the organization. Anecdotal evidence abounds describing the importance of chief executives on the direction of strategy within their firms. In addition Levinson and Rosenthal in their study of six firms (1984) have provided a detailed account of the role of the CEO in the direction of corporate strategy. Their research provides evidence that the CEO has a variety of tools at his disposal to alter the strategic course of an organization and thus its long-run performance. In light of the strategic importance conferred on top management by both academic literature and the popular press, empirical research exploring the management-strategy-performance linkage may provide some insight into the exact importance of top management. If firm performance can be attributed to the strategic course chosen by top management, then such a link might be detected empirically. Hambrick and Mason (1984) have argued for just such an avenue of research.

-4 - Previous Research Previous research on the top management-strategy-performance linkage has come from two different perspectives. Research addressing the strategyperformance link has examined differences in firm performance in light of diversification strategy pursued (Rumelt, 1974; Montgomery, 1979; Bettis, 1981); industry characteristics (Porter, 1980); firm specific characteristics (Hatten, Schendel, and Cooper, 1978) as well as specific business level strategies (Miles, 1982; Miles and Snow, 1978; Harrigan, 1980; Hall, 1980). In addition, there have been numerous studies addressing the strategyperformance linkage at the business level using the PIMS database. Much of this research takes into account industry and firm specific factors but neglects to examine the role that succession and management composition may have on strategic choice and firm performance. In addition, the performance measures used to evaluate strategic choice have been largely accounting measures. Within the sociology literature, researchers have focused on the linkage between managerial succession and organizational performance. In examining this linkage, researchers have found there is little (Brown, 1982) or no (Smith et al., 1984) succession effect on organizational performance. Grusky (1963) found that "rates of administrative succession and degree of organizational effectiveness are negatively correlated." Allen, Panian, and Lotz (1979) found that "there is a negative relationship between past team performance and the frequency of managerial succession." Thus, poor team performance led to increased turnover. However, Eitzen and Yetman (1972) found that once previous levels of performance were controlled for, managerial changes made no differences. Gamson and Scotch (1964) found that changes in leadership have little bearing on organizational performance. The replacement

-5 - of a leader in a poorly performing organization is often a kind of "ritual scapegoating" (Brown, 1982). Given the constraints on strategy imposed by economic and market forces (industrial organization literature) as well as the organizational impediments (Cyert and March, 1963; Thompson, 1967), leadership may not be the primary factor influencing organizational performance. The need to control for organizational context has been highlighted by Carroll (1984) and Pfeffer and Davis-Blake (1986). In their study of single industry firms, Lieberson and O'Connor (1972) found that leadership accounted for very little of the performance variance once year, industry, and company effects were accounted for. The actual effect of leadership was also constrained by the nature of the industry. These findings confirm the importance of the economic and competitive environment in determining firm profitability (Porter, 1980). In addition they highlight the problem of using accounting measures by which to isolate leadership effects. Several recent studies have used stock market data as the dependent variable in measuring a succession effect (Dalton and Kesner, 1984; Reinganum, 1985; Worrell, et al., 1986; Beatty and Zajac, 1987). By focusing on the stock market's reaction to the actual succession effect, Dalton and Kesner found that succession yielded a positive effect on stock prices, whereas Worrell, et al and Reinganum found no effects with the exception of a positive effect where outsiders where brought into small firms (Reinganum, 1985). Beatty and Zajac found a significant negative effect associated with the succession event. While the results are mixed, stock market data can isolate the market's perception of leadership changes within the firm in light of organizational and environmental factors.

-6 - Much of the research on executive succession explores the premise that leaders, by their decisions and actions, should be able to influence organizational performance. It can be argued that a succession effect will only occur when new managers differ from previous managers. This change in perspective may be unlikely to occur since according to March and March (1977) the "filtering process in organizations leads to selection from a fairly homogeneous group." This filtering process ensures that new managers are similar to old ones and as a result, little effect is realized. In many companies, managers are groomed from within and carefully observed prior to being chosen to join the ranks of top management. In addition to the selection process, organizations can also constrain managerial autonomy. Executives are embedded in a social system which constrains behavior (Pfeffer, 1977). As a result, organizational expectations may reinforce previous modes of operation. To understand the effect of executive succession on the strategy of an organization, one needs to examine the exact nature of the transition within the top management team. While managers can be identified by a wide variety of differentiating characteristics, research on executive succession has come to focus on managerial origin in terms of the organization from which that individual came. Grusky (1960), in his studies of administrative succession has defined insider/outsider roles as follows: insider: individual who emerged from among the membership of the executive's staff. outsider: an individual transferred into the organization from another organization. This typology is similar to Carlson's (1961) study which identified managers as "place-bound" versus "career-bound." Similar terms are used to describe executives as "institutionally oriented" or "maintenance oriented." The

-7 - underlying difference is whether the manager's career was within or outside of the organization. Based on differences in the origin of the successor, one might predict rather different organizational outcomes. According to Carlson (1962), insider school superintendents would be more inclined to maintain the organizational status quo than to bring about dramatic change. On the other hand, outsiders are brought in to promote change and creativity. Thus, insider succession should be less disruptive than outsider succession. Helmich and Brown (1972) found confirming evidence in their study of 208 corporations that "organizations experiencing inside succession in the office of the president exhibit less organizational change than firms undergoing outside succession." In studies relating managerial origin to organizational performance, both Gouldner (1954) and Grusky (1964) found that team performance deteriorates when outsiders became managers. In a more recent study using stock prices as a performance measure, Reinganum (1985) found that external successors in smaller organizations yielded a positive effect on performance. Conceptual Framework In light of the literature on successor type, we would expect the market to react differently to the succession event based on the organizational background of the manager. Executive succession from within would be viewed as a "maintenance strategy" and should therefore not have financial significance for the firm. As a result we would hypothesize that insider succession announcements will lack any significant impact on stockholder wealth. H1: Insider succession within the top management ranks will not result in abnormal stock market returns.

-8 - On the other hand, outsider succession would be viewed as turbulent, i.e. a departure from a maintenance strategy and should therefore elicit an abnormal return in the stock market. H2 Outsider succession within the top management ranks will result in abnormal stock market returns. Given the strategic posture of the organization in terms of the mix of businesses within which it competes, changes in top management might lead to strategic changes within the firm. If strategy is a reflection of managerial choice made within an organization, then how one views the firm's current strategic posture should influence the direction and extent of strategic change pursued. Managerial perceptions of reality are not necessarily objective, but are biased by one's previous experiences and encounters as well as by one's organizational commitments. Donaldson and Lorsch (1984) have described how psychological constraints act to influence managerial perceptions. Managers recruited from outside the organization are less likely to share the previous strategic perceptions and thus might be viewed as more inclined to chart a new strategic direction for the firm. Evidence that outsiders can have an impact on the strategy of the business has been found in several studies. Hammermesh (1977) found that in order to respond effectively to a divisional profit crisis within a multibusiness firm, it may be necessary to break commitments through personnel changes. Existing top management can stand in the way of the strategic steps necessary to resolve the profit crisis because of prior commitments to individuals, structural relationships, and divisional strategic plans. In his study on turnarounds, Charles Hofer (1980) found that "a precondition for almost all successful turnarounds is the replacement of the current top management of the business in question".

-9 - Stuart Gilmour (1973) came to similar conclusions on the role of top management in divestiture situations. According to Gilmour, the "crucial block was that recognition of the divestment solution as a viable and acceptable option required some admission of failure or placing of blame on the part of the key decision maker." Gilmour found that actual divestment decision-makers were all new to the companies and situations they faced. While these studies focused at the business level, divestments and turnarounds have strategic consequence for the firms involved. Hammermesh, Hofer, Gilmour, and Duhaime(1984) have shown that the likelihood that decisions of this nature will be made is influenced by the choice of top management. In addition, from a previous study of portfolio shifts (Wiersema, 1986) the author found that executive background was highly significant in explaining the magnitude of portfolio shifts experienced by the firm. Firms managed by CEOs and Presidents recruited from outside the organization exhibited, on average, twice the extent of strategic change in their core business than firms managed by executives from within the firm. The stock market's perception of managerial succession is in part a perception of changes in the organization's strategic course as the result of managers with different perspectives. Executive succession from within the organization would not convey strategic turbulence since managerial perspectives are unlikely to change. Their perspective is influenced by the same factors that were at work before them: coalitions within the firm, past decision making, and commitments within the firm. The current officeholders can influence the selection of their successors and undoubtedly will chose managers that concur with their strategic vision. Given the average size of the firms in the sample, there should be an adequate supply of capable and experienced managers within the firm who could be chosen for the top

-10 - management positions. When the board of directors chooses to go outside the organization for its top executives, it is sending a clear signal that the prospective managers within the organization are not adequate for the job at hand. While their experiences are undoubtedly valuable, insiders bring the biases of their career within the organization into the executive suite whereas outsiders lack this excess baggage. If outsiders are hired to change the strategic course of the firm, then this type of managerial selection must be viewed in light of the long-run best interest of the firm and its shareholders. Given this information, we would expect the stock market to react favorably to outsider succession appointments. H3 Outsider succession within the top management ranks will result in positive abnormal stock market returns. Measuring The Succession Effect The principal assumption in the succession research stream has been that managerial succession is an important organizational event and therefore should have an impact of performance. In analyzing the succession effect on organizational outcome, most studies have relied on win/loss records for sport teams and financial measures (ROI, ROS) for business organizations. Accounting measures, however, not only reflect general economic trends, they are also somewhat industry- and company-specific, making it difficult to isolate the succession effect. As a result of economic and market determinants, performance measures tend to exhibit little variance over relatively short periods of time. The changes in strategic direction stimulated by the arrival of new executives may never be accurately detected due to intervening factors which impact firm performance as well as to the problem of ascertaining the correct time frame. Exactly when does the new

-11 - executive begin to influence organizational performance? Even without these research design problems, causality would be difficult to ascertain. Is the firm's long-run performance the result of the new manager, or the financial result of a strategy put in place by his/her predecessor? Given accounting measures' inherent limitations in capturing succession effects, the lack of conclusive results found in the succession literature are expected. If the firm's strategic course and its specific business policies are a reflection of its top management (as the policy literature affirms), then succession may provide us with information as to likely changes in the strategic course of the firm. Within large business organizations (Fortune 1000 firms) replacements within the top management team are usually the domain of the board of directors and are therefore given significant public importance. The purpose of this study is to examine the capital market's reaction to succession within the top management ranks. If top management succession is an important event for a corporation, then its ramifications should be discernible in stock price data on and closely around the announcement date. Given this narrow event window, it is highly likely that other events impacting the firm's future earnings flow are being held constant. This is the key advantage to using market returns versus accounting returns in evaluating the management-performance linkage. In an informationally efficient market, succession announcements would be evaluated for their possible impact on the firm's future income stream. At some date close to the succession announcement, efficiently determined stock prices should adjust to that information. This assumption is based on the efficient markets hypotheses, that capital markets are informationally efficient. Based on this theory, items of information about the prospects of the firm should be reflected in changes in the price of a security when this information reaches

-12 - the market. This type of analysis is based on isolating abnormal returns associated with particular events -given what the normal or expected return should be, based on a market model (Fama, 1976). If managerial succession is viewed as a nonevent; i.e., the new manager is not viewed as being significantly different from the preceding manager, the market will interpret this information as not impacting the firm's future earnings flow. As a result, this type of succession announcement would not result in significant changes in the stock price of the firm on or around the announcement date. On the other hand, if managerial succession is viewed as likely to lead to significant changes in the firm's strategic decisions, then one would expect a positive impact on shareholder wealth where these changes are viewed favorably and a negative impact where unfavorably. As Reinganum (1985) pointed out, these short-run changes in the market value of the firm can be used to measure changes in management policies and practices designed to improve the long-run prospects of the firm. One of the assumptions in event studies examining succession is that investors act on information regarding succession events. In other words, investors understand that top management changes can signal changes in an organization's future strategy and performance. Upon decoding the message implied in executive changes, investors can chose whether to react to this piece of information. This event study methodology has been used extensively in the finance literature as well as more recently within the succession literature. Brown and Warner (1985) have examined in detail a number of potential problems associated with the use of daily stock return data in event studies. Despite the particular characteristics of daily stock data, they conclude that "methodologies based on the OLS market model are well-specified under a variety of conditions."

-13 - By using event study methodology, researchers can isolate the impact of succession on shareholder wealth. Environmental, competitor, and organizational realities can be held constant since the market already reflects the influence of these factors on future firm performance. The capital market's reaction to the succession announcement reflects its expectation about future events, given the new manager. The succession effect on shareholder wealth thereby provides one way of examining the managementstrategy-performance linkage empirically. Method Sample The population of firms chosen for this study was the 1000 largest manufacturing firms as listed by Fortune magazine in 1982. Of these firms, a random sample of 200 firms was chosen representing a cross section of the economy with a variety of industries represented - from energy to electronics, forest products to food processors, metals to motor vehicles. From this initial sample of 200 firms, ten firms were eliminated due to their acquisition. Of the remaining sample of 190 firms, several firms were eliminated due to missing cusips (37); insufficient stock market data points on the CRSP tape (8); and because they lacked specific announcement dates regarding succession events in the Wall Street Journal Index (34). This left a sample of 111 firms for this study. Within the original sample of 190 firms, Presidents promoted from outside the organization represented 32 percent of the sample. Presidents hired from outside the organization increased to 41 percent of the final sample of 111 firms. The ratio of outsiders to insiders is significantly higher than previous succession studies using stock market data (the Zajac and Beatty sample had 12 percent outsider

-14 - succession; Reinganum's sample had 8 percent outsiders). The outsider/insider ratio in this study is not an anomaly due to sample selection since it was chosen at random from the Fortune 1000. The outsider/insider ratio for this study is consistent with recent studies of executive organizational backgrounds. In the early 1950s, Newcomer (1955) in his study of 9000 CEOs found that the percentage of outsiders was about 17 percent. However two decades later Helmich (1977) observed that 33 percent were outsiders. The trend towards increasing outsiders was confirmed in the 1980s by Brady, Fulmer, and Helmich (1982) who in a sample of 1100 firms, found that 37.8 percent of the CEOs were outsiders. Based on these succession studies, the ratio of outsiders for our sample does not seem to be out of line. In addition, this study collected data on the top management composition for the sample over an extended period of time. From this historical perspective outsiders predominantly replaced managers from inside the organization. And in addition, over time firms exhibited an increasing tendency to recruit from outside their organizations for all four top management positions. Succession Measures Data were collected on top management changes for the period 1968 to 1982. The top management team examined included the chairman, chief executive officer, president, and chief operating officer. These four positions comprised the top management team in most business organizations. The event study however will focus on the two predominant positions: CEO and President. Data on the composition and background of these executives were gathered from Dun and Bradstreet Reference Book of Top Management(1983) as well as from financial reports and 10Ks filed annually by the firms for the period 1970 -1982. Where the specified executive's career was within the firm, the

-15 - succession was classified as an insider promotion. In firms where the new executive came directly from outside the organization or had less than three years of previous experience within the organization, the succession was classified as an outsider promotion. The exact announcement dates for managerial successions within the sample of firms were gathered from the Wall Street Journal Index. This is in line with previous event study methodology in the succession literature as well as in the financial merger and acquisition literature. To the extent that the Wall Street Journal does not always report succession events in a timely fashion, we may fail to identify the correct announcement dates and thus make abnormal returns more difficult to detect. In addition, the lack of announcement dates for many firms reduced the size of the sample. Data Analysis To research the impact of top management on firm strategy, this study examines the stock market's reaction to executive succession events within a sample of firms. It is hypothesized that when a firm announces an executive succession, the stock market will evaluate the long-run strategic ramification of this event on future earnings and incorporate this information in the market value of the firm. The market's valuation of executive succession information was analyzed by examining stockholders' returns. A return is the difference in the price of a stock during a specified period divided by its original price. t = (Pt - Pl)/ t-l where R = return of a security for day t Pt = stock price for the firm on day t

-16 - The following market model was used to predict the normal return for a period of 40 days prior to the announcement date and 40 days afterwards. jt = a + B. Rmt jt where Rj = return on security j for period t; a. = intercept term for security j; B. = historical market relationship for security j for period t; R = market return; mt ej = unexplained residual Regressing the return on each security against the return on a market index provided a predictive model. The actual returns on the stock were compared to the predicted returns, and the difference measured as the abnormal return (AR). Ajt jt(actual) Rjt(expected) The hypothesis of whether or not this abnormal return could be equal to zero was tested giving T-statistics. In addition a cumulative abnormal return (CAR) was calculated for the seven days immediately surrounding the event date to capture more fully the information effect on the market. The stock market data necessary for this analysis was gathered from the CRSP data base. Results Tables 1 and 2 and Figures 1, 2, and 3 present results for the stock market valuation of insider versus outsider succession in the office of the President. As can be seen, while mean abnormal return on the event date was not significant, the cumulative abnormal return for seven days surrounding the succession event was significant for the sample as a whole. The source of the significant succession effect can be seen when the sample is broken down

-17 - between insider and outsider successions. Both the mean abnormal return and the cumulative abnormal returns for outsider succession have a highly significant positive effect on abnormal returns. Insider succession, on the other hand, did not exhibit abnormal returns. Since outsider succession represents 41 percent of the total sample, the sample as a whole also exhibits a significant succession effect. The results from this study confirm the two hypotheses discussed: one, that outsider succession would be viewed favorably by the market, and two, that insider succession would result in the absence of abnormal returns. Insert Tables 1 and 2 About Here Insert Figures 1,2, and 3 About Here When we examine the extent of abnormal return, we see that, on average, firms with outsider succession have a 2.0 percent cumulative abnormal return from the seven trading days around the announcement date. These abnormal returns indicate a significant increase in the wealth for these firms' shareholders. Succession in the office of the CEO however, was not statistically significant. As can be seen in Tables 3 and 4 and Figures 4, 5, and 6, outsider CEO succession did not result in a significant abnormal return in the stock market. Insert Tables 3 and 4 About Here ____________________________________

-18 - Insert Figures 4, 5, and 6 About Here Why should outsider succession in the office of the president have an abnormal stock market return when outsider CEO successions have no effect? The results can be explained by the composition and succession process within the firm's top management team. Succession within the offices of the CEO and President are not necessarily two separate and discrete events with two players. Vancil (1987) in his study of CEO succession noted three modes of succession: solo; duo; and team. In the solo model, one individual is selected as both CEO and President simultaneously. In the duo mode, the firm has two individuals within the top management ranks: typically a Chairman-CEO and President. When the current CEO goes into semi-retirement he keeps the title of chairman, but passes on the title of CEO to the current President. This elevation of duties from President to CEO-President can be viewed as a nonevent since it only validates the previous appointment. Only when a new president is appointed is there an actual selection event. Vancil found this duo mode of executive selection as the most common form in his sample of firms. Within this study's sample, almost half of the firms have only one individual in the offices of President and CEO. If the firm has a solo model of top management it would result in one announcement date. Under a duo model of top management, however it would result in two announcement dates: first the selection of the executive as President and then the promotion to CEOPresident. As a result the CEO succession event is not as distinct an event in terms of impacting the composition of the top management team. Since the position of President comes first, the announcement regarding President succession indicates to the market the executive's entry into the very top

-19 - management of the firm. The market does not view the President's assuming CEO responsibility as additional information since it serves as a validation of the previous succession event. Since 30 percent of executives promoted to the office of CEO already hold the office of the president, the lack of significant abnormal stock market returns associated with CEO succession events might be expected. The significant abnormal returns indicated by the total sample for President succession announcements has not been found by previous succession studies. Reinganum (1985) and Beatty and Zajac (1987) did not find significant positive results for their respective population of firms. Reinganum did find a positive abnormal return effect on outsider succession but only in smaller firms (mean capitalization of $39.2 million). The significant impact that outsiders had on shareholder wealth in this study was unrelated to firm size. In addition, the ratio of outsider to insider executive succession was also unrelated to firm size. The difference in this study's findings from previous studies can be explained by the nature of the sample as well as by the managerial positions examined. Previous succession studies using stock market data have used samples of convenience, i.e. all firms having succession announcements as recorded by the WSJ Index within a specific time frame. This study, on the other hand, has focused on first selecting a random sample of firms, and then tracking down the appropriate event date. This study took the initial announcement date for the President or CEO who held the office during the 1973-1980 period. If there were more than one individual holding the office during this time period, the initial appointment was used. As a result the event dates are not concentrated in a two or three year time span. In addition, the sample of 200 firms was drawn from the 1982 Fortune 1000 firms

-20 - which by definition are very large and predominantly diversified organizations. Previous succession studies drew samples from the universe of all New York and American Stock Exchange firms (about 2500 firms). These population differences lead to samples of firms varying in their nature and size which may explain the differences in the results found. In addition, researchers have focused on different executive positions. Reinganum tracked succession within the office of the President and Chairman, while Beatty and Zajac researched CEO succession exclusively. As previously mentioned, the office of the CEO, and in all likelihood, the office of the Chairman as well are not necessarily entry level positions within the top management ranks. From data gathered on executive succession within a sample of firms, the office of the President is clearly the predominant first step in becoming CEO and eventually Chairman. As a result, information provided to the market from succession events is clearly influenced by the office in question. When a CEO succession is announced the added value of the information is dependent on whether the executive previously held the office of the President or is indeed a newcomer to the executive suite. Similarly, announcements regarding the office of the Chairman are unlikely to provide investors with additional information regarding the strategic course of the firm since the majority of these positions are filled by the previous CEO. Conclusions and Implications This research provides evidence that the stock market's reaction to succession within the top management ranks of very large firms is dependent on the organizational background of the new executives as well as on the position in question. Outsider succession in the office of the President is viewed favorably, resulting in positive abnormal returns to shareholders. These

-21 - results imply that the stock market perceives the announcement of outsiders into the executive ranks as leading to favorable expectations about the firm's future earnings. Insider succession on the other hand yielded no impact on shareholder wealth indicating that insiders were perceived as not leading to changes in the firm's future earnings. These results support Helmich and Brown's findings that inside succession leads to less organization change while outsiders were found to be more turbulent. In addition, the turbulence that outsiders are perceived as bringing to their respective organizations is viewed as being financially favorable by the capital market. While the organizational background of the President led to significantly different stock market reactions, this was not the case for CEO succession events. Neither insider or outsider CEO succession resulted in any significant impact on shareholder wealth. These results can be explained in light of the structure and succession process within top management. The position of CEO is typically not an entry level to the top management ranks. Many of the CEO announcements involved the semi-retirement of the Chairman-CEO to Chairman and the promotion of the President to CEO-President. In these cases, the CEO succession event is a reaffirmation of the previous selection event and thus may be viewed by the market as a nonevent. The top management structure and the multiple succession events may explain the differences found between President and CEO succession events. By examining the impact of succession events on shareholder wealth, we can isolate the managerial effect and hold constant industry structure and organizational factors. Using this research design, this study provides empirical evidence that the nature of top management succession can lead to differences in the perceived value of the firm. In light of the stockmarket's reaction, succession is perceived as an important strategic event. Changes in

-22 - leadership which result in outsider succession within the office of the President is judged favorably by the market. Whereas insider succession is viewed as a continuation of the status quo. These findings affirm the notion that top management can make a difference in the strategic course for the firm. In addition, changes in the expectations of the firm are more likely to occur when executives are not promoted from inside the organization. The notion that insiders and outsiders will lead to different expectations about the firm is supported by previous succession research. Future research should examine top management composition and succession and its impact on firm strategy. Investors are perhaps making assumption about changes in the strategic course for the firm based on top management composition. Additional research may confirm that outsiders are more likely to redirect an organization's strategy.

-23 - Table 1 Cumulative Abnormal Returns for Insider versus Outsider Successions in the Office of the President Organizational CARs Test Background t= -3 to +3 Statistics Insider -1.29 0.19 n = 65 59% Outsider 2.00 2.40* n = 46 41% Total.132 2.01* n = 111 Table 2 Mean Abnormal Returns for Insider versus Outsider Successions in the Office of the President Organizational ARs Test Background t= 0 Statistics Insider 0.06 0.242 n = 65 59% Outsider 0.64 1.742** n = 46 41% Total 0.29 1.395 n = 111 * significant at p 4.05 ** significant at p <.10

-24 - Table 3 Cumulative Abnormal Returns for Insider versus Outsider Successions in the Office of the CEO Organizational CARs Test Background t= -3 to +3 Statistics Insider -1.09 n = 63 0.65 66% Outsider 4.01 1.12 n = 33 34% 0.01 Total n = 96 1.24 Table 4 Mean Abnormal Returns for Insider versus Outsider Successions in the Office of the CEO Organizational ARs Test Background t= 0 Statistics Insider -0.34 n = 63 -1.25 66% Outsider 0.10 n = 33 0.22 34% Total -0.19 n = 96 -0.80

-25 - Figure 1 Cumulative Average Residuals President Succession - Total Sample CAR 1.4 -...6.~ -*1. - -1..8 --1 4....................................................................... _.............................. -1.8 = = = -= ---- — ~ — < --- —-1Se --- —-* --- —- = =- = -25 -20 -1 -10 -5 0 5 10 15 20 25 DAY Figure 2 Cumulative Average Residuals President Succession - Outsiders Figure 3 Cumulative Average Residuals President Succession - Insiders CAR CAR 1 3.2 2.8............................................................................................................ 1 h................................................................ _,~ ' ' -;4 /,...,,....,_.... _......_.... It,- -,,I 2.............................................. _._........... _._._............................................................................................ IL.+~~~~~~~~~~~~~~~ r 1.2.8.............................................................. -1.............-. — - - - - - - -. -....-......./ -12. --- - 0 -1 5 - — 2 5 - 2 0 1 1 5 0 5 2 2 -25 -20 -15 -10 -5 10 15 20 25 DAY -25 -20 -15 -10 -5 10 15 20 25 DAY

-26 - Figure 4 Cumulative Average Residuals CEO Succession - Total Sample CAR DAY Figure 5 Cumulative Average Residuals CEO Succession - Outsiders CAR Figure 6 Cumulative Average Residuals CEO Succession - Insiders CAR;,:,,...................... -1.2.. ~~-....... -2 - -25 -20 -t5 -10 -5 5 10 15 20 25

-27 - Bibliography Allen, M. P., Sharon K. Panian, and Roy C. Lotz. "Managerial Succession and Organizational Performance: A Recalcitrant Problem Revisited." Administrative Science Quarterly, 24 (1979):167-180. Andrews, K. R. The Concept of Corporate Strategy. 4th edition. Homewood, Illinois: Dow Jones-Irwin, 1980. Ansoff, I. H. Corporate Strategy. New York: McGraw-Hill, 1965. Beatty, Randolph, and Edward Zajac. "CEO Change and Firm Performance in Large Corporations: Succession Effects and Manager Effects." Strategic Management Journal, forthcoming, 1987. Bettis, Richard. "Performance Differences in Related and Unrelated Diversified Firm." Strategic Management Journal, Vol. 2: 379-393 (1981). Bower, J. L. Managing the Resource Allocation Process. Homewood, Illinois: Irwin, 1970. Brady, Gene F., Robert M. Fulmer, and Donald L. Helmich. "Planning Executive Succession: The Effect of Recruitment Source and Organizational Problems on Anticipated Tenure," Strategic Management Journal, Vol.3, No.3 (1982):269-275. Brown, M. C. "Administrative Succession and Organizational Performance: The Succession Effect." Administrative Science Quarterly, 27 (1982):1-16. Brown, Stephen and Jerold Warner. "Using Daily Stock Return: The Case of Event Studies." Journal of Financial Economics. Vol. 14 (1985): 3-31. Carlson, R. O. "Succession and Performance Among School Superintendents." Administrative Science Quarterly, 6 (1961):210-277., R. 0. Executive Succession and Organizational Change. Danville, Illinois: Interstate Printer & Publishers, 1962. Carroll, Glenn R. "Dynamics of Publisher Succession in Newspaper Organizations." Administrative Science Quarterly, Vol. 29 (1984): 93 -113. Cyert, R., and March, J., A Behavioral Theory of the Firm, Prentice-Hall,1963. Dalton, D., and Kesner, I. "Inside/Outside Succession and Organizational Size: The Pragmatics of Executive Replacement. Academy of Management Journal, Vol. 26, (1984): 736-742. Donaldson, and Lorsch. Decision Making at the Top. New York: Basic Books, 1984. Duhaime, Irene M. Influences on the Divestment Decisions of Large Diversified Firms. Ann Arbor, Mi.: University Microfilms, 1981.

-28 -Dun and Bradstreet. Reference Book of Corporate Managements. Volume I, II, 1984. Parsippany, New Jersey, 1983. Eitzen, D. S. and N. Yetman. "Managerial Change, Longevity, and Organizational Effectiveness." Administrative Science Quarterly, 17 (1972):110-116. Fama, E.F. Foundations of Finance, New York: Basic Books, 1976. Fortune, The 500 Largest Industrial Corporations. May 3, 1982, pp. 260-278. Fortune, The Second 500 Largest Industrial Corporations. June 14, 1982, pp. 180-198. Gamson, W. A., and N. Scotch. "Scapegoating in Baseball." American Journal of Sociology, 70 (1964):69-72. Gilmour, Stuart. The Divestment Decision Process. Unpublished Doctoral Dissertation, Harvard Graduate School of Business, 1973. Gouldner, A. Patterns of Industrial Bureaucracy. Glencoe, Illinois: Free Press, 1954. Grusky, O. "Administrative Succession in Formal Organization." Social Forces, 39 (1960):105-115. __, O. "Managerial Succession and Organizational Effectiveness." American Journal of Sociology, 69 (1963):21-31., O. "The Effects of Succession: A Comparative Study of Military and Business Organizations." in M. Janowitz (ed.) The New Military. New York: Russel Sage Foundation, 1964. Hall, William K. "Survival Strategies in a Hostile Environment." Harvard Business Review, (1980):75-85. Hambrick, D. E. and P. A. Mason. "Upper Echelons: The Organization as a Reflection of its Top Managers." Academy of Management Review, 9 (1984):193-206. Hammermesh, Richard G. "Responding to Divisional Profit Crises," Harvard Business Review, Vol.55, No.2 (March/April 1977):124-130. Harrigan, K. R. "Strategies for Declining Industries." Journal of Business Strategy, 1 (Fall, 1980):20-34. Hatten, Kenneth, Dan Schendel, and Arnold Cooper. "A Strategic Model of the U.S. Brewing Industry, 1952-1971," The Academy of Management Journal. Vol.58, No.l (1978):590-610. Helmich, D. L. "Executive Succession in the Corporate Organization: A Current Integration." Academy of Management Review, (1977):252-264.

-29-, D.L., and Warren B. Brown. "Succession Type and Organizational Change in the Corporate Enterprise," Administrative Science Quarterly. (1972):371-382. Hofer, Charles W. "Turnaround Strategies," Journal of Business Strategy, Vol.1, No.1 (1980):19-31. Levinson, H. and S. Rosenthal. CEO: Corporate Leadership in Action. New York: Basic Books, 1984. Lieberson, S., and J. F. O'Connor. "Leadership and Organization Performance: A Study of Large Corporations." American Sociological Review, 37 (1972):117-130. Miles, Raymond E., and Charles E. Snow. Organizational Strategy, Structure, and Process. McGraw-Hill, 1978. Miles, Robert H. Coffin Nails and Corporate Strategies. Englewood Cliffs, New Jersey: Prentice-Hall, Inc., 1982. Montgomery, Cynthia A. Diversification, Market Structure, and Firm Performance. Ann Arbor, Michigan: University Microfilms, 1979. Newcomer, M. The Big Business Executive. New York: Columbia University Press, 1955. Pfeffer, Jeffrey. "The ambiguity of leadership." Academy of Management Review, 2 (1977):104-112. Porter, M. E. Competitive Strategy. New York: Free Press, 1980. Quinn, J. B. Strategies for Change: Logical Incrementalism. Homewood, Illinois: Richard D. Irwin, Inc., 1980. Reinganum, M. R. "The Effect of Executive Succession on Stockholder Wealth." Administrative Science Quarterly, 30 (1985):46-60. Rumelt, R. P. Strategy, Structure, and Economic Performance. Boston, Mass.: Division of Research, Harvard Graduate School of Business, 1974. Selznick, P. Leadership in Administration. New York: Harper and Row, 1957. Smith, J.E.; Carson, K.P.; and R.A. Alexander. "Leadership: It Can Make A Difference." Academy of Management Journal, 27 (1984):765-776. Thompson, James D., Organizations in Action, McGraw-Hill, 1967. Vancil, Richard F. "A Look at CEO Succession." Harvard Business Review, (March/April, 1987):107-117. Wiersema, Margarethe F. "Executive Succession, Top Management Team Consolidation, and Corporate Strategic Redirection." Working Paper #480, Division of Research, University of Michigan, Ann Arbor, Michigan.

-30 -Worrell, D. L., S. Davidson, P.Chandly, and S. Garrison, "Management Turnover Through Deaths of Key Executives: Effects on Investor Wealth." Academy of Management Journal. 1986, Vol.29, No.4: 675-694. Wrigley, L. Divisional Autonomy and Diversification. Unpublished Doctoral Dissertation, Harvard Graduate School of Business, 1970.