Are competitors advantageous or disadvantageous in consolidated versus fragmented industries?

Wright, Peter ; Ferris, Stephen P. ; Vaughan, Mary Jo 等

ABSTRACT

We contend that competitors may be mutually disadvantageous in fragmented industries. Consequently, we expect that announcements of firm distress will be associated with positive implications for non-distressed competitors in fragmented industries. Alternatively we speculate that, in consolidated industries, rivals may be advantageous because they may offer net mutual benefits to each other. Thus, we predict that the announcement of distress by a firm in a consolidated industry will be received as negative news by its rivals since the contribution of that firm to the industry may cease. We utilize the event-study methodology to empirically test our hypotheses.

INTRODUCTION

While some senior executives may view the existence of rivals as advantageous, others tend to perceive competitor firms as detrimental to the interests of their own enterprises. We contend that whether competitors are advantageous or detrimental may be situational. Our conjecture is that in fragmented industries rivals may ordinarily be a threat; whereas, in consolidated industries they may be beneficial. That is because in fragmented industries competitors tend to be confrontational but in many consolidated industries enterprises can be non-confrontational and mutually advantageous in their rivalry. Whether the presence of rivals is advantageous or disadvantageous to a firm may be related to a variety of theories across a number of disciplines. We discuss the implications of these theories in the context of two settings--fragmented versus consolidated industry settings.

For reasons that we will subsequently provide, our premise is that in fragmented industries competing enterprises may be reciprocally detrimental. In consolidated industries, however, we presume that rivals can be mutually beneficial. Yet whether firms in distinct industries offer net mutual advantages or disadvantages to each other remains an empirical question. If a firm is a threat to its rivals, its distress should be good news for these rival firms. Alternatively, if the firm can make a positive contribution to the industry, then its potential demise should be seen as an unfavorable event.

The focus of our research is on exploring whether market values of firms respond negatively or positively to announcements of distress by a competitor, contingent on their industry affiliation. More specifically, we examine how announcements of bankruptcy impact the equity values of non-bankrupt competitors in consolidated and fragmented industries. An event-study methodology, more completely described later in this study, is used to accomplish the empirical analysis. If an announcement of a firm's bankruptcy within an industry has adverse economic implications for its non-bankrupt competitors, then their stocks should suffer negative returns. Alternatively, if the announcement of a bankruptcy benefits rivals, their stocks should experience positive returns.

We organize this study into several sections. In the following section we provide a literature review and present our hypotheses. We then present our empirical analysis, describing both the sample construction procedure and our application of the event-study methodology. Finally, we report our findings and offer a discussion of their interpretation.

RELATED LITERATURE AND HYPOTHESES

Traditionally, advocates of determinism have assigned to organizations limited discretion within their environments. For instance, industrial organization theorists have explained firm conduct and performance as reflections of the structure of industry environment (Bain, 1956; Mason, 1939). Conversely, in the past, proponents of strategic choice have credited organizations with significant proactivity within their environments (Andrews, 1971; Chandler, 1962). More recently, however, advocates of determinism as well as strategic choice have begun to move closer together. Strategic choice theorists have recognized that the external environment may play a critical role in a firm's quest for survival (Hambrick, 1983; Hrebiniak & Joyce 1985). Alternatively, industrial organization theory has given increased credence to firm strategy which may not only affect the strategy of rivals but may also modify the structure of an industry (Fudenberg & Tirole, 1984; Porter, 1980, 1985; Tirole, 1988).

The preceding discussion, however, provides different implications for prospects of firms operating in fragmented industries versus those in consolidated industries. In our view, firms competing in fragmented industries may be externally constrained in their capability to structure non-adversarial interrelationships. Note that due to low entry (Baumol, Panzar, & Willig, 1982; Scherer & Ross, 1990) and mobility barriers (Caves & Porter, 1977; Waring, 1996), many enterprises with smaller market shares operate in fragmented industries. In effect, numerous firms in such industries are forced to compete toe-to-toe for resources and customers in overlapping industry niches. Consequently, the proactive struggle for gains by any one firm, in this setting, will likely be made at the expense of other enterprises (Stigler, 1957). The ramification of these contentions is that as the number of firms increase in an industry, so should the probability that they may become detrimentally interrelated.

The notion that with a higher number of enterprises adversarial interactions may be promoted is consistent with the implications of other works, such as Cournot's equilibrium pricing model (Cournot, 1971; Dixon, 1986; Novshek, 1980), research in game theory (Axelrod, 1984; Brandenburger & Nalebuff, 1995, 1996; Hill, 1990; Rappoport & Chammah, 1965), as well as institutional (DiMaggio & Powell, 1983; Meyer & Rowen, 1977) and population ecology works (Baum, 1996; Hannan & Carroll, 1992; Hannan & Freeman, 1977). According to Cournot's model, numerous firms in the industry--characteristic of fragmented industries--are related to a more intense competition and reduced profitability. That is because as the number of rivals increases in an industry, the equilibrium price declines until it approaches the product's marginal cost (Cournot, 1971; Dixon, 1986; Novshek, 1980).

Game theory also provides implications for enterprises in fragmented industries. This theory, initially developed by von Neumann and Mortenstern (1944), has been applied to the study of negative-sum and zero-sum-game circumstances (e.g., Axelrod, 1984; Brandenburger & Nalebuff, 1995, 1996; Hill, 1990; Rappoport & Chammah, 1965). In a zero-sum-game situation, competitors are constrained from being non-adversarial. Instead, they tend to be confrontational because the gain of one rival is only possible at a cost to another. In a negative-sum-game context, rivals are also forced to be mutually detrimental because for one competitor just to keep what it has requires that another lose. Given our deliberation so far, both negative-sum and zero-sum-game situations are characteristic of fragmented industry circumstances, where firms proactively compete head-on for resources and customers.

The conjecture that an increase in the number of enterprises may promote adversarial interactions is also addressed in institutional theory. For instance, DiMaggio and Powell (1983) suggest that as the number of organizations that are competitively interrelated expands, they are more likely to be mutually detrimental. Meyer and Rowan (1977) similarly propose that as more organizations become competitively interconnected, they will be increasingly confrontational. These arguments are likewise compatible with the premise of population ecology theory. Several population ecologists have suggested that as organizations grow in number within sub-populations, they assume correspondingly more adversarial roles, presumably because the environment's carrying capacity limit is approached (Baum, 1996; Hannan & Carroll, 1992; Hannan & Freeman, 1989).

Consistent with the previous discussion as well as the arguments of Miles, Snow, and Sharfman (1993), we contend that toe-to-toe competition, characteristic of a fragmented industry, not only may be harmful for each enterprise but also such rivalry may undermine the health of the industry. That is because intense rivalry tends to lower profitability, inhibiting investments that could enhance product performance or industry efficiencies. Under these circumstances, since the existence of rivals tends to be disadvantageous, the potential exit of a firm from a fragmented industry may be beneficial to other enterprises because a reduction in the number of rivals may lessen the intensity of competition. More specifically, a firm's distress and its potential bankruptcy may be fortunate for survivors since they may subsequently face fewer rivals in their jockeying for resources and customers. Thus, we offer the following hypothesis:

H1: In fragmented industries, announcements of bankruptcy will be associated with positive abnormal returns for non-bankrupt competitors because the existence of rivals is disadvantageous.

Alternatively, a limited number of firms, but with larger market shares, are contained in consolidated industries due to high entry (Baumol, et al., 1982; Scherer & Ross, 1990) and mobility barriers (Caves & Porter, 1977; Waring, 1996). In these industries, since there are fewer rivals, enterprises tend to have the discretion to compete in non-overlapping environmental niches, potentially avoiding head-on competition for resources and customers (Baumol, et al., 1982; Scherer & Ross, 1990; Tirole, 1988). In consolidated industries, therefore, firms need not establish mutually detrimental interrelationships. Indeed, enterprises may deliberately avoid confrontational strategies in order to offset the possibility of retaliation by others. The reason is that even though enterprises in such industries may operate entirely independently, they recognize their interdependence since any one rival's move has considerable effect on others (Chamberlin, 1929; Machlup, 1952). Each firm, consequently, may be hesitant to implement an adversarial strategy which, when countered, would ultimately leave all industry members worse off.

These contentions are in conformance with other implications inherent in Cournot's equilibrium pricing model (Cournot, 1971; Dixon, 1986; Novshek, 1980), research in game theory (Axelrod, 1984; Brandenburger & Nalebuff, 1995, 1996; Hill, 1990; Rappoport & Chammah, 1965), as well as select strategy arguments (Buzzell & Gale, 1987; Miles & Snow, 1986). According to Cournot's model, fewer firms in an industry--characteristic of consolidated industries--are associated with reduced confrontations and consequently enhanced profitability, presumably because limited adversarial interfirm behavior allows for the equilibrium price to be higher in an industry. Alternatively, according to game theory, in positive-sum-game situations, all players can win without resorting to destructive interfirm behavior. Positive-sum-game situations are more applicable to consolidated industry circumstances, where fewer rivals may operate in non-overlapping industry niches; consequently, they need not proactively compete head-on for resources and customers.

The speculation that fewer organizations in a sub-population of enterprises may be non-confrontational is also recognized in institutional and population ecology theories (Baum, 1996; Carroll, 1984; DiMaggio & Powell, 1983; Hannan & Freeman, 1977). Accordingly, given that fewer firms may not impose on the environment's carrying capacity limit, they need not resort to adversarial inter-firm behavior. In these theories it is further proposed that organizations could be advantageously interconnected (Carroll, 1984; Gould, 1977; Hannan & Carroll, 1992; Hawley, 1968; Meyer & Rowan, 1977). Consider, for instance, that the R & D efforts of some firms may spill-over and benefit the other firms in the industry. Alternatively, advertising by some enterprises may increase the demand for the outputs of all organizations in the industry. Consequently, the legitimacy of select firms (and their industry) may in the preceding ways be enhanced vis-a-vis firms in substitute industries. In a number of strategy related works, it is also argued that firms may be beneficially interconnected as they adopt various strategies within select industries that may require divergent resources. Moreover, clusters of firms may address the particular needs of various groups of customers, while obviating confrontational interfirm behavior (Buzzell & Gale, 1987; Miles & Snow, 1986). Because the needs of various customer groups may be more effectively met by different clusters of firms addressing their unique needs, the long-term viability of an entire industry may be enhanced, implying beneficial outcomes for all firms operating in the industry. Note that fewer firms, potentially adopting non-confrontational strategies, are more applicable to circumstances prevailing in consolidated industries.

Further, the existence of different firm strategies, in the context of consolidated industries, not only is advantageous because it obviates competition for the same resources and customers, but can also provide industrywide benefits because healthy industries may require a diversity of complementary strategies. On this subject, Miles and Snow (1978, 1986) as well as Miles, Snow, and Sharfman (1993) argue that in some industries there is an implicit complementary interdependence among firms and that each enterprise with a different strategy may have a synergistic role to play with its rivals for the industry to maintain a long-run viability. Thus, the performance of each firm as well as the industry's aggregate performance may suffer if any of the current competitors potentially exit (Miles, Snow, & Sharfman, 1993). Under these circumstances, a firm may create value for its rivals as it creates value for itself. Thus, with mutually beneficial interrelationships, the potential exit of a firm will eliminate the benefit provided to competitors. Brandenburger and Nalebuff explain as follows:

 "In business, what does a particular player bring to the game? To
 find the answer, look at the value created when everyone is in the
 game, and then pluck that player out and see how much value the
 remaining players can create. The difference is the removed
 player's added value" (1995: 58).

Based upon the preceding discussion, we speculate that the potential exit of a firm may be detrimental to rivals in consolidated industries. Consequently, we propose the following hypothesis:

H2: In consolidated industries, announcements of bankruptcy will be associated with negative abnormal returns for non-bankrupt rivals because the existence of competitors is advantageous.

SAMPLE CONSTRUCTION AND METHODOLOGY

Over the period of one decade--January 1980 through December 1989--we identify 841 bankruptcy announcements through a search utilizing the Dow Jones News Retrieval Service. We eliminate from this sample any firms that report other potentially contaminating events (e.g., earnings reports, management turnover, strikes, capital expenditures, restructuring) in the period surrounding the bankruptcy announcement. We also exclude firms from further consideration if they are not listed on the New York (NYSE), American (AMEX), or Over the Counter (OTC) stock exchanges, if they have announcements of bankruptcies following their delisting from an exchange, if they lack common stock returns on the day of the announcement (day 0), or have less than 50 daily stock returns over the estimation period (day -265 to day -16). Our final sample consists of bankruptcy announcements for 274 firms, with 113 listed on either the NYSE or AMEX and the remaining 161 listed as OTC issues. The non-bankrupt competitor sample is formed by selecting firms in the same four-digit Standard Industry Classification (SIC) code industries as those firms announcing bankruptcies. Each bankrupt firm averages over 9 competitors and results in a total sample of 2,563 competitor firms.

To assess the impact of a bankruptcy announcement, we separately examine the abnormal returns to the NYSE/AMEX and OTC-listed competitors of the bankrupt firms. The reason for a separate analysis, based on exchange listing, is to test for possible industry concentration effects. NYSE/AMEX bankruptcies typically represent consolidated industries with fewer firms that have larger market shares, while those on the OTC generally reflect bankruptcies in fragmented indus-tries with more numerous companies that have smaller market shares (Scherer & Ross, 1990; Tirole, 1988). This is indicated by the values for Herfindahl-Hirschman (HH) Index, calculated as the sum of the squared market shares of companies with the same four-digit SIC codes as the firm filing for Chapter 11. Thus the HH Index can be interpreted as a measure of industry concentration, with higher values indicating greater consolidation, and by implication, fewer but larger individual firms. The mean HH index value for the NYSE/AMEX-listed companies is 0.32 while the corresponding value for OTC-listed firms is 0.06.

We emphasize that some competing firms within an industry may be listed on different exchanges. Yet, this does not distort our results. By separately calculating CARs accruing to non-bankrupt competitors listed on the exchanges over the period surrounding an announcement of bankruptcy (Table 1), we assure that the abnormal returns are attributable to the impact of the bankruptcy rather than any idiosyncratic characteristic of the exchange. Specifically, we employ the market model event methodology which analyzes a daily series of mean excess returns, calculated by equally weighting returns across the samples of non-bankrupt competitor firms (Brown & Warner, 1985). Additionally, we sum these daily excess returns to obtain a cumulative abnormal return (CAR), which provides a more comprehensive measure of the event's unanticipated impact on equity values. To enhance robustness, we provide our analysis for several CAR reporting periods or windows.

Although there are other empirical models available to test abnormal stock performance, we elect to use the market model for two reasons. First, the market model has been widely used in empirical studies to estimate measures of excess return beginning with the first event study by Fama, Fisher, Jensen and Roll (1969) and continuing to the present. Indeed, the Brown and Warner (1985) review of event methodology reports that the market model is more powerful in terms of its ability to identify abnormal performance than other models that are available. This suggests the second reason for our use of the market model. Unlike the CAPM or the empirical market line, the market model is not vulnerable to Roll's (1977) criticism. Hence, the abnormal return analysis provided by the market model estimates in this study are not subject to the mathematical criticisms originally noted by Roll and now generally recognized as compelling in the related literature.

One of the possible biases that may be present in an event study results from the thin trading of the firm's securities. Specifically, thin or reduced trading of a firm's equities may result in serial correlation between observed security and index returns and a consequent bias in ordinary least square (OLS) estimates of the systematic risk (beta) coefficient. An econometric approach to solving this potential problem is the use of the Scholes-Williams (1977) technique for beta estimation. The Scholes-Williams beta coefficient results from weighting a series of beta estimates for a security that have been calculated against both synchronous and non-synchronous market return data. We find that the reestimation of our event residuals using the Scholes-Williams betas do not qualitatively change our results and leave our conclusions the same. Hence, we conclude that thin trading volume is not present in our sample of firms and that our excess return estimates are statistically unbiased.

It is important to note that the event methodology controls for macroeconomic activity that may influence the level of stock returns and thereby distort our estimate of bankruptcy's impact upon equity returns (Brown & Warner, 1985). Thus, if one observes a negative abnormal rate of return for a non-bankrupt competitor's stock surrounding an announcement of bankruptcy, the event methodology allows us to attribute it to the bankruptcy event rather than to other factors (e.g., changes in gross domestic product, shifts in the structure of industries, expansion or contraction in specific industries, technological innovations in substitute industries).

Moreover, the event methodology is based on an assumption of capital market efficiency that requires investors to revise their expectations about a firm's prospects only upon the announcement of new, economically relevant information. If an announcement does not affect a firm's economic prospects, one should not observe significant abnormal returns at the time of its release. Likewise, there should be no significant abnormal returns at the time of the announcement if the details of the bankruptcy have been anticipated or leaked in advance.

In this context, we should also emphasize that some observers suggest that a bankruptcy is a downward spiral that may be predictable several years in advance (Hambrick & D'Aveni, 1988). For such bankruptcies, we do not anticipate significant abnormal returns on the announcement date. Not all downward spirals, however, will ultimately result in bankruptcy. For instance, some downward spirals may be reversed through a turnover in top management (Tushman, Newman, & Romanelli, 1986; Tushman & Romanelli, 1985; Warner & Watts, 1988) or through retrenchment (Blackwell, Marr, & Spivey, 1990) or through voluntary, internal restructuring (Brickley & Van Drunen, 1990; Donaldson, 1990; John, Lang, & Netter, 1992). We apply the event methodology to our sample, presuming bankruptcies were unanticipated by the market. Specifically, such bankruptcy announcements would not have been anticipated by investors, either because downward spirals were not evident or because projected results were previously expected to be offset by such measures as senior manager turnover, retrenchment, or voluntary restructuring. To the extent that bankruptcies are anticipated, CARs of competitor firms will necessarily be less significant. That is because an efficient capital market will have already capitalized its response to the news of the bankruptcy announcement in the share prices of competitors, resulting in only an insignificant price change for rivals at the time of the actual announcement.

Finally, our event methodology results are not significantly effected by the presence of more asset diverse corporations. That is, one may argue that a bankruptcy announcement in a given industry may have a lesser impact on the equity value of conglomerate-like rivals. Such a situation would bias our methodology against finding significant returns, as opposed to observing significant negative abnormal returns accruing to non-bankrupt competitors.

In addition to the loss of contributions provided by a competitor, a bankruptcy may negatively affect the returns of a non-bankrupt rival for an alternative reason, provided by contagion theory. The contagion theoretical reason is based on the presumption that if one firm in an industry is distressed, then others in the industry may confront similar distresses. Known as the contagion effect (Altman, 1984; Bernanke, 1983; Lang & Stulz, 1992), accordingly, non-bankrupt firms may be negatively impacted by a bankruptcy announcement since it may signal that an entire industry is threatened, with consequent negative implications for the asset values of the remaining firms.

The contagion theory appears to most directly apply to entropic firms lacking growth opportunities (Altman, 1984; Lindenberg & Ross, 1981). Firms lack growth opportunities because their internal resources are not valuable or because of their position in declining industries (Barney, 1991; Lado, Boyd, & Wright, 1992; Lindenberg & Ross, 1981; Wright, Ferris, Sarin, & Awasthi, 1996). The contagion impact is based on the assumption that the distress of one firm and its steady deterioration implies that its competitors may have similar difficulties because of internal vulnerabilities or external threats associated with the lack of growth opportunities (Altman, 1984; Barney, 1991; Lado, et. al., 1992; Lindenberg & Ross, 1981). In Schumpetarian (1934) terms, the contagion effect suggests that the distress of a single firm may imply an industrywide threat as enterprises external to the industry develop new technologies which render the outputs of existing firms less desirable or possibly obsolete, suggesting lack of growth prospects for these firms.

Lindenberg and Ross (1981) and Wright and colleagues (1996) propose that individual firm values of Tobin's q ratio can proxy for the existence of growth opportunities. Tobin's q is defined as the market value of a firm standardized by the replacement cost of its assets. Consequently, firms with q's of unity or less can be judged as overinvested and lacking growth prospects. Lindenberg and Ross (1981) and Wright and colleagues (1996) explain that the absence of growth opportunities for firms may be due to inefficiency, technological inferiority, locational disadvantages or declining industries. Alternatively, firms with q's in excess of unity may be viewed as under-invested, whose value is largely driven by the existence of growth opportunities. Such firms may possess valuable resources and operate in profitable industries. Thus, to control for a possible contagion effect, we separately examine the abnormal returns of firms based on the value of their Tobin's q ratio. If contagion does exist, then firms which lack growth opportunities (i.e., low-q firms) should be most negatively impacted by an announcement of a competitor's bankruptcy since they are less capable of exploiting subsequent business opportunities. Conversely, high-growth firms (i.e., high-q firms) should be less adversely affected by such bankruptcy announcements since they are likely to possess the resources and competencies to exploit their competitors' misfortunes.

Through the use of Tobin's q, we can test for the existence and relative dominance of the contagion versus the advantageous competition effects. The subsample of low-q firms are those firms with an unfavorable market valuation of the future earnings capability of their assets within their industries. These firms lack attractive investment options in their industry and are unlikely to generate growth in their corporate cashflows. Because of this, these firms are less capable of attracting new investment capital or exploring new technologies and projects within their industries. The high-q subsample, however, represents a set of firms with a favorable market valuation of their future growth opportunities within their industries. These firms possess a set of profitable investment opportunities. As such, they are financially resilient and have the greater access to external capital markets. They have more resources, both actual and potential, with which to develop new technologies or extend operations into new markets.

Consequently, we contend that on a relative basis, high-q enterprises are less likely to suffer from an industry contagion than the low-q companies. This is an important, yet subtle point in our argument. Although high-q firms may suffer negative returns from an industry contagion, they will suffer less than the low-q firms. The reason is that such firms have greater resources with which to address adverse developments. Thus, the potential existence of the contagion effect requires that we compare the magnitudes of the CARs for the two q-based subsamples. As discussed subsequently, we find that based upon an economic interpretation of Tobin's q, the high-q firms are more negatively impacted by the news of a rival's bankruptcy than the low-q firms. We conclude that this is inconsistent with the impact of an industry contagion on share prices in a rational capital market.

RESULTS AND A FURTHER TEST

In Table 1, we provide CARs for impacts of bankruptcies listed on the various exchanges. Non-bankrupt competitors of larger NYSE/AMEX-listed firms that announce bankruptcies experience negative CARs. Note that these negative CARs are statistically significant for the standard two-day event window of (-1, 0) whether non-bankrupt competitors are listed on either the NYSE/AMEX or the OTC exchange. The negative CARs are not only indicative that non-bankrupt firms are worse off with a potential demise of a member firm, but also that these firms are not beneficiaries of a lesser competitive intensity. That is because the loss to the firm announcing bankruptcy is not a gain to its non-bankrupt rivals. If the competitive effect were present, we would expect to witness positive CARs in response to a bankruptcy announcement instead of negative CARs.

Regarding the possibility of competitive effect, Altman (1984) as well as Lang and Stulz (1992) suggest that the potential demise of a firm may favorably impact its rivals if a redistribution of wealth (or resources) from the bankrupt firm to its competitors can be anticipated. For instance, suppliers and customers may be reluctant to do business with a firm announcing bankruptcy. Thus, their business activity will be switched to various non-bankrupt competitors. In this way, the value of the firm announcing bankruptcy would be lowered while the value of its rivals may be enhanced, reflecting the anticipated redistribution of wealth from the bankrupt firm to its rivals.

Note that the CARs for competitors are insignificant, however, when the bankruptcy announcements are made by a smaller OTC-listed firm. These results are consistent with hypothesis H2, but not H1. We emphasize that the CARs for the smaller and larger firms in our sample are unbiased. That is because we employ the widely cited market model (Fama, 1976) to estimate the abnormal returns reported in our study. The market model contains a market risk adjustment as well as an adjustment for non-market risk factors. We empirically accomplish this through the estimation of a series of linear regression models between the market's returns and those of the sample firms over a 250-day estimation period that precedes the days of the event periods. This provides us with estimates of an intercept term (alpha) and a slope (beta) coefficient. These parameters are then used to net an expected return from the realized returns to generate the abnormal return. These abnormal returns are then cross-sectionally averaged and summed to obtain the CARs. Thus, whatever impact average firm size may exert on a firm's returns is incorporated in the alpha term, leaving the abnormal returns unbiased. This approach represents a standard application of the event-study methodology to examine the valuation impact of an unanticipated event.

In Table 2 we present the CARs of firms separated into low- and high-q portfolios. The high-q firms are significantly and negatively affected (while the low-q firms are insignificantly impacted) by the news of a rival's bankruptcy. Moreover, the results are fairly dramatic as high-q firms experience negative CARs over two different event windows. Alternatively, the low-q firms experience CARs that are insignificantly different from zero. This is inconsistent with the impact of a contagion effect on share prices in an efficient capital market. Thus, the equity price reaction by firms to a rival's bankruptcy announcement does not indicate that rivals face similar troubles due to common external threats or internal vulnerabilities. These results are further supportive of hypothesis H2.

Why is hypothesis H1 not supported? That is, why are the CARs insignificant when a smaller firm in a fragmented industry announces a bankruptcy? Two reasons may explain this lack of significance. First, one may argue that the potential demise of a troubled firm will not impact its rivals if other organizations can easily replace it. We speculate that the probability of other enterprises replacing the activities of a small firm located in a fragmented industry that may fail is higher. That is because fragmented industries have lower entry (Baumol, et al., 1982; Scherer & Ross, 1990) and mobility barriers (Caves & Porter, 1977; Waring, 1996). Low barriers facilitate the entry of new firms into an industry, thus making it possible for new competitors to replace the activities of the bankrupt firm.

Second, bankruptcies of smaller firms may be due to the inherent liability of small-scale operations or their newness (Hambrick & D'Aveni, 1988, 1992; Stinchcombe, 1965). These fac-tors are idiosyncratic to the circumstances of smaller troubled firms and hence are likely to have little impact on rivals. Alternatively, however, the probable impact on surviving firms due to the failure of a larger troubled competitor in a consolidated industry is likely to be significant. Consolidated industries have higher entry (Baumol, et al., 1982; Scherer & Ross, 1990) and mobility barriers (Caves & Porter, 1977; Waring, 1996). These barriers preclude the easy entrance of new rivals that can substitute for the activities of the larger failed firm.

Indeed, in aggregate, we expect that the significant impact on share prices of rivals due to larger firms announcing bankruptcies will dominate the insignificant impact on stock prices of competitors in response to smaller enterprises announcing bankruptcies. To test this, we analyze the effect of all bankruptcy announcements in our sample on share prices of non-bankrupt rivals. As shown in Table 3, we present a time series of daily abnormal returns to the portfolio of competitors of all firms announcing a Chapter 11 filing. Although these daily abnormal returns vary, most observations are negative. The significant finding occurs in the standard two-day window of (-1, 0) where the cumulative abnormal return is -0.39% and statistically significant at the 0.05 level. This finding is consistent with our expectation.

DISCUSSION AND IMPLICATIONS

The contentions of our study are consistent with the theoretical implications of the arguments of diverse scholars (e.g., Axelrod, 1984; Baumol, et al., 1982; Brandenburger & Nalebuff, 1995, 1996; Caves & Porter, 1977; Chamberlin, 1929; Cournot, 1971; DiMaggio & Powell, 1983; Hannan & Carroll, 1992; Machlup, 1952; Miles & Snaw, 1978, 1986; Miles, Snow, & Sharfman, 1993; Scherer & Ross, 1990; Stigler, 1957; Tirole, 1988; Waring, 1996). Based on the related literature, we have speculated that competing enterprises in fragmented industries may be reciprocally disadvantageous. Although our speculation may be intuitively appealing, the empirical findings do not unambiguously support such a speculation potentially because of easy entry of new firms into frag-mented industries (Caves & Porter, 1977; Scherer & Ross, 1990; Waring, 1996) as well as factors which may be idiosyncratically relevant to the circumstances of smaller enterprises (Hambrick & D'Aveni, 1988, 1992; Stinchcombe, 1965). Nevertheless, we surmise that managers in fragmented industries may be justified to view their rival firms as a threat. A paradoxical implication of frag-mentation, however, is that such industries can represent a unique strategic opportunity. While fragmented industries with low barriers are unattractive because their firms are mutually disadvantageous, it is possible for a firm to advantageously consolidate some of these industries. Indeed, "the payoff to consolidating a fragmented industry can be high because the costs of entry into it are by definition low, as there tend to be small and relatively weak competitors who offer little threat of retaliation" (Porter, 1980: 200).

Our results, however, tend to be in contrast to the view of those senior executives who may perceive their rivals as primarily a threat in consolidated industries. Such managers typically do not assume that competitors can contribute to the long-term viability of either the industry or their own firm. Our findings are consistent with the notion that competitors may beneficially contribute to each other in consolidated industries. Consequently, the performance of firms may suffer in such industries if the contributions of their rivals are temporarily or permanently withdrawn through an announcement of bankruptcy (Brandenburger & Nalebuff, 1995, 1996; Miles, Snow, & Sharfman, 1993).

An implication of the preceding discussion is that in consolidated industries a constructive approach to rivalry may create opportunities for mutually beneficial strategies. Additionally, an implication of our work may be that viewing rivals as advantageous in consolidated industries may become even more important in the future. As the pressure for enhanced efficiency and innovation increases, more firms may find it necessary to view their rivals as a constructive force and a poten-tial ally. Indeed, recent trends suggest that even the most vigorous competitors can form beneficial strategic interconnectedness (Templin, 1995).

REFERENCES

Altman, E. (1984). A further investigation of the bankruptcy cost question. Journal of Finance, 39: 1067-1089.

Andrews, K.R. (1971). The concept of corporate strategy. Homewood: Dow Jones Irwin.

Axelrod, R. (1984). The evolution of cooperation. New York: Basic Books.

Bain, J. (1956). Barriers to new competition. Cambridge: Harvard University Press.

Barney, J.B. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17: 99-120.

Baum, J.A.C. (1996). Organizational ecology. In Clegg, S.R., Hardy, C., & Nord, W.R. Handbook of Organization

Studies. London: SAGE Publications, 77-114.

Baumol, W.J., Panzar, J.C. & Willig, R.D. (1982). Contestable markets and the theory of industry structure. New York: Harcourt Brace Jovanovich.

Bernanke, B.S. (1983). Nonmonetary effects of the financial crisis. American Economic Review, 73: 257-276.

Blackwell, D.W., Marr, M.W. & Spivey, M.F. (1990). Plant-closing decisions and the market value of the firm. Journal of Financial Economics, 26: 277-288.

Brandenburger, A.M., & Nalebuff, B.J. (1995). The right game: Use game theory to shape strategy. Harvard Business Review, 73(4): 57-71.

Brandenburger, A.M. & Nalebuff, B.J. (1996). Co-opetition. New York: Currency/Doubleday.

Brickley, J.A. & Van Drunen, L.D. (1990). Internal corporate restructuring: An empirical analysis. Journal of Accounting and Economics, 12: 251-280.

Brown, S.J. & Warner, J.B. (1985). Using daily stock returns: The case of event studies. Journal of Financial Economics, 14: 3-31.

Buzzell, R.D. & Gale, B.T. (1987). The PIMS principles: Linking strategy to performance. New York: The Free Press

Carroll, G.R. (1984). The specialist strategy, in G. Carroll & D. Vogel (Eds.), Strategy and or-ganization: A west coast perspective. Boston: Pitman: 117-128.

Caves, R.E. & Porter, M.E. (1977). From entry barriers to mobility barriers: Conjectural decisions and contrived deterrence to new competition. Quarterly Journal of Economics, 91: 241-262.

Chamberlin, E.H. (1929). Duopoly: Value where sellers are few. Quarterly Journal of Economics, 43: 63-100.

Chandler, A.D. (1962). Strategy and structure: Chapters in the history of American industrial enterprise. Cambridge: MIT Press.

Cournot, A.A. (1971). Research into the mathematical principles of the theory of wealth. Translated by N.T. Bacon. New York: Augustus M. KeRlley, Publisher.

Dixon, H. (1986). The Cournot and Bertrand outcomes as equilibria in a strategic metagame. Economic Journal, 96: 59-70.

DiMaggio, P.J. & Powell, W.W. (1983). The iron cage revisited: Institutional isomorphism and col-lective rationality in organizational fields, American Sociological Review, 48: 147-160.

Donaldson, G. (1990). Voluntary restructuring: The case of General Mills. Journal of Financial Economics, 27: 117-141.

Fama, E.F. (1976). Foundations of Finance. New York: Basic Books, Inc.

Fama, E.F., Fisher, L., Jensen, M.C. & Roll, R. (1969). The adjustment of stock prices to new in-formation. International Economic Review, 10: 1-21.

Fudenberg, D. & Tirole, J. (1984). The fat cat effect, the puppy dog ploy and the lean hungry look. American Economic Review, 74 (May): 361-368.

Gould, S.J. (1977). Ever since Darwin. New York: Norton Press.

Hambrick, D.C. & D'Aveni, R.A. (1988). Large corporate failures as downward spirals. Administrative Science Quarterly, 33: 1-23.

Hambrick, D.C. (1983). High profit strategies in mature capital goods industries: A contingency approach. Academy of Management Journal, 26: 687-707.

Hannan, M.T. & Carroll, G.R. (1992). Dynamics of organizational populations: Density, legitima-tion and competition. New York: Oxford University Press.

Hannan, M.T. & Freeman, J.H. (1977). The population ecology of organizations, American Journal of Sociology, 82: 929-964.

Hawley, A. (1968). Human ecology, in Sills, D.L. (Ed.), International Encyclopedia of the Social Sciences. New York: Macmillan.

Hill, C.W.L. (1990). Cooperation, opportunism, and the invisible hand: Implications for transaction cost theory. Academy of Management Review, 15: 500-513.

Hrebiniak, L.G. & Joyce, W.F. (1985). Organizational adaptation: Strategic choice and environ-mental determinism. Administrative Science Quarterly, 30: 336-349.

John, K., Lang, L.H.P. & Netter, J. (1992). The voluntary restructuring of large firms in response to performance decline. Journal of Finance, 47: 891-917.

Lado, A.A., Boyd, N.G. & Wright, P. (1992). A competency-based model of sustained competitive advantage: Toward a conceptual integration. Journal of Management, 18: 77-91.

Lang, L.H.P. & Stulz, R.M. (1992). Contagion and competitive intra-industry effects of bankruptcy announcements: An empirical analysis. Journal of Financial Economics, 32: 45-60.

Lindenberg, E. & Ross, S. (1981). Tobin's q ratio and industrial organization. Journal of Business, 54: 1-32.

Machlup, F. (1952). The political economy of monopoly. Baltimore: Johns Hopkins Press.

Mason, E.S. (1939). Price and production policies of large-scale enterprises. American Economic Review, 29: 61-74.

Meyer, J.W. & Rowan, B. (1977). Institutional organizations: Formal structure as myth and ceremony, American Journal of Sociology, 83, 340-363.

Miles, G., Snow, C.C. & Sharfman, M.P. (1993). Industry variety and performance. Strategic Management Journal, 14: 163-177.

Miles, R.E. & Snow, C.C. (1978). Organizational strategy, structure, and process. New York: McGraw-Hill Book Company.

Miles, R.E. & Snow, C.C. (1986). Organizations: New concepts for new forms. California Management Review, 28: 62-73

Novshek, W. (1980). Cournot equilibrium with free entry, Review of Economic Studies, 47: 473-486.

Porter, M.E. (1980). Competitive strategy: Techniques for analyzing industry and competitors. New York: The Free Press.

Porter, M.E. (1985). Competitive advantage: Creating and sustaining superior performance. New York: The Free Press.

Rappoport, A. & Chammah, A.M. (1965). Prisoner's dilemma. Ann Arbor: The University of Michigan Press.

Roll, R. (1977). A critique of the asset pricing theory's tests; Part I: On past and potential testability of theory. Journal of Financial Economics, 4: 129-176.

Scherer, F.M. & Ross, D. (1990). Industrial market structure and economic performance. Boston: Houghton Mifflin Company.

Scholes, M. & Williams, J. (1977). Estimating betas from non-synchronous data. Journal of Financial Economics, 5: 309-327.

Schumpeter, J.A. (1934). The theory of economic development. Cambridge: Harvard University Press.

Stigler, G.J. (1957). Perfect competition historically contemplated. Journal of Political Economy, 65: 1-17.

Stinchcombe, A.L. (1965). Social structure and organizations. In March, J.G. (Ed.). Handbook of organizations. Chicago: Rand McNally.

Templin, N. (1995). Strange bedfellows: More and more firms enter joint ventures with big competitors. Wall Street Journal, November 1, A1 and A12.

Tushman, M., Newman, W.H. & Romanelli, E. (1986). Convergence and upheaval: Managing the unsteady pace of organizational evolution. California Management Review, 29: 29-44.

Tushman, M. & Romanelli, E. (1985). Organizational evolution: A metamorphosis model of convergence and reorientation. In Cummings, L.L., & Staw, B.M., (Eds.), Research in Organizational Behavior, Greenwich: JAI Press.

von Neumann, J. & Mortenstern, O. (1944). Theory of games and economic behavior. Princeton, N.J.: Princeton University Press.

Waring, G.F. (1996). Industry differences in the persistence of firm-specific returns. American Economic Review, 86: 1253-1265.

Warner, J.B. & Watts, R. (1988). Stock prices and top management changes. Journal of Financial Economics, 20: 461-492.

Wright, P., Ferris, S.P., Sarin, A. & Awasthi, V. (1996). Impact of corporate insider, blockholder, and institutional equity ownership on firm risk taking. Academy of Management Journal, 39: 441-463.

Peter Wright, University of Memphis

Stephen P. Ferris, University of Missouri at Columbia

Mary Jo Vaughan, Mercer University

William T. Jackson, University of Texas of the Permian Basin

Table 1: Cumulative Abnormal Returns of Competitors Listed By
Their Exchanges for Select Windows

 Bankrupt Firm Listing

 NYSE/AMEX OTC

Competitor Firm
Listing

 Window CAR T-Stat CAR T-Stat

NYSE/AMEX -1, 0 -0.0059 -2.00 * -0.028 -0.99
 -1, +1 -0.0082 -2.28 * -0.0018 -0.52

OTC -1, 0 -0.0047 -1.88 ([dagger]) -0.0020 -0.90
 -1, +1 -0.0049 -1.61 -0.0019 -0.70

([dagger]) p < .10

* p < .05

Table 2: Cumulative Abnormal Returns to Competitors Sorted by
Tobin's q Ratio

Windows Firms with q > 1 Firms with q =1

-1, 0 -0.0028 * 0.0011
 -2.150 1.201
-1, +1 -0.0030 * 0.0017
 -1.992 0.967

* p< .05

Table 3: Daily Abnormal Returns (ABRET for Competitors of Firms

Announcing Chapter 11 Filings (1979-1989)

 Daily Stock Price
Day Reaction (ABRET) T-Statistic

-15 -0.0018 -0.789
-14 -0.0003 -0.172
-13 -0.0003 -0.560
-12 -0.0010 -2.004 *
-11 -0.0000 -0.561
-10 -0.0010 -0.152
-9 0.0003 0.512
-8 -0.0016 -1.041
-7 -0.0005 -0.340
-6 -0.0012 -1.824 ([dagger])
-5 -0.0009 -0.069
-4 -0.0006 -0.513
-3 -0.0010 -0.098
-2 -0.0038 -3.434 **
-1 -0.0015 -0.714
0 -0.0025 -1.232
1 0.0005 0.848
2 0.0005 0.831
3 0.0002 0.871
4 -0.0005 -0.382
5 -0.0016 -0.336
6 -0.0016 -0.638
7 -0.0007 -0.126
8 0.0008 1.323
9 -0.0010 -0.252
10 0.0002 0.663
11 0.0011 1.535
12 -0.0014 -0.181
13 -0.0016 -1.871 ([dagger])
14 0.0004 1.060
15 -0.0012 -0.073

Cumulative Abnormal Returns (CARS of Competitors for Select Windows

Windows CAR Z-Statistic

(-1, 0) -0.0039 -2.05 *
(-1, +1) -0.0034 -1.47
(-5, +5) -0.0034 -0.76
(-10, +10) -0.0074 -1.20

([dagger]) p < .10

* p < .05

** p < .01