Barriers to market exit.
Porter (1976) discusses three major types of barriers to exit, namely, economic, strategic, and managerial. Economic barriers refer to the costs associated with selling a business as a unit that will be run by someone else, or dismantling it and then selling. Strategic barriers are created due to the strategic posture of the firm. For instance, a firm following a single business strategy would cease to exist if liquidated. Personal barriers are also a potent force. Managers may perceive severe damage to their image of competence if their business is being divested. This may translate into losing their jobs, or having to take a reduction in pay.
Which of these factors do decision makers really consider while making the exit decision? What is the relative importance of the factors considered? How do these differ across industries? Some researchers have attempted to answer several of these questions, notably Porter (1976) and Harrigan (1981, 1985a). However, the studies conducted so far have relied on some combination of historical databases and interviews. Databases restrict the scope of the research, simply because they may not include all the factors relevant to the issue to be addressed. Using interviews and questionnaires, on the other hand, has been criticized for potential biases in the data (see Arnold and Feldman, 1981; Hyman, 1954; Stahl, 1986; Stahl and Harrell, 1982). The research to date has not been able to include enough of the relevant factors that deter market exit, especially the personal barriers.
This study attempts to identify the determinants of market exit behavior from the literature and find their relative importance to decision makers, using a decision-making exercise (Slovic et al., 1977; Stahl and Zimmerer, 1984). Furthermore, this study examines the differences in the relative importance of barriers to exit in declining versus mature industries.
Market Exit as a Strategy
Up until the 1980s, market exit was not regarded as a strategy, but a helpless response to a failure to compete. This view, however, has changed over the past decade or two, which saw a trend towards leaner businesses that focused their attention on related industries (Lynch, 1980). Since Rumelt's (1974) classification of firms according to their diversification strategy, several studies have shown that related diversified firms, especially those in the related-constrained (closely related business units owned by a firm) category, are more profitable and less risky. Subsequent research also supported the view of researchers who proposed the concept of strategic fit and the resulting synergy as the way to increased profitability and reduced risk. Christensen and Montgomery (1981) found that unrelated diversified firms performed poorly when compared to related-constrained firms.
In the last decade corporate market exit behavior seems to confirm the findings of Rumelt (1974), and Christensen and Montgomery (1981). The increasing role of divestment of unrelated business units as a strategy for making the firm profitable and competitive (Sicherman and Pettway, 1987) further support the view of market exit as a formal strategy. A study conducted by Taylor (1992) showed that in the four years from 1982 through 1985, there was a forty percent increase in divestment activity among corporations. Furthermore, in 1984 and 1985, 2,000 subsidiaries and divisions were spun off by their parent companies. Marriott, for example, divested its restaurant business division after a strategic analysis showed a weak competitive position in that area (Muller, 1990).
There are a variety of reasons for divestment. The decision to divest does not have to be connected to a decline in the industry, it may simply represent a change in the corporate goals concerning which industry to compete in. A business unit may be far too unrelated to the main operations of the firm, so that the top management does not have enough knowledge concerning its operations, and hence gets divested (Dubin, 1986). Pepsico, for example, sold its unrelated freight business and followed a strategy of related diversification in the food business, acquiring restaurant chains like Kentucky Fried Chicken. Westinghouse also expanded in the 1960s and 70s into areas as diverse as watchmaking, low income housing and uranium contracts. It has had as many as 135 divisions at one time. A one billion dollar loss in 1991 forced the company to liquidate assets and sell several industrial business divisions (Schroeder, 1992). Lynch (1980) noted that many firms are selling off their unrelated acquisitions to lift their stock price by improving their image as firms with a clear sense of their future direction. Size of business is another possible contributor to the exit decision (Miles and Rosenfeld 1983; Mitchell, 1994).
Financial Benefits of Market Exit
Several studies have focused on the financial impact of market exit (Alexander et al., 1984; Klein, 1986; Miles and Rosenfeld, 1983; Rosenfeld, 1984; Zaima and Hearth, 1985). When a firm voluntarily announces a divestment, it is a reasonable assumption that the decision was reached to create wealth for the firm. This does not imply, however, that divestment always benefits a firm financially. A study performed by Miles and Rosenfeld (1983) found a statistically significant increase in the stock prices around the announcement date of the divestment. Stewart and Glassman (1988) argue that selling unprofitable or unsuitable business units creates value for both the divestor and the acquirer. Feinberg (1988) noted that many small firms sometimes divest the entire firm as a financing strategy. Other reasons that may persuade a firm to divest businesses include legal, political (Alexander et al., 1984; Flath, 1994; Hite and Owers, 1983), social, or moral (Ennis and Parkhill, 1986; Hall, 1986).
Barriers to Exit
Divestment can be an appropriate and profitable strategy, but the decision to divest is complicated by the presence of several barriers. Firms often find themselves involved with running businesses earning well below the expected rate of return, and are unable or unwilling to divest them because of exit barriers. Porter (1976) classified the barriers into three broad categories - structural or economic, strategic, and managerial. Table 1 summarizes the exit barriers literature using the three broad categories and their components originally proposed by Porter.
In addition to the costs associated with exit, the decision to divest a large business unit is also dependent on the effect of the firm being divested to the other businesses owned by the [TABULAR DATA FOR TABLE 1 OMITTED] same company, rather than on the cost of solving the internal problems of the business unit (Duhaime and Baird, 1987; Von Krogh and Roos, 1994). Also, from a buyer's perspective, the value of the business to be purchased is a critical question. Therefore, when a firm considers a market exit decision, the degree of operating fit, marketing fit, forward integration, and backward integration becomes important to both sellers and buyers.
The portions of the capital investment that cannot be recovered, sunk costs, also create a barrier to exit (MacLeod, 1987; Kleindorfer and Knieps, 1982; Dixit, 1980; Bernheim, 1984). Balakrishnan and Wernerfelt (1986) argued that capital invested alone does not form a barrier to entry. Rather, entry is deterred by the extent to which the costs are sunk upon entry. Therefore, entry and exit barriers are closely related.
Some multibusiness firms stay in a declining business for another strategic reason. A firm may have a business unit that is a poor competitor, but its competitors also compete with the firm in other businesses. The firm can use such a business unit to play the role of a "competitive harasser" who pulls the competition down with itself (MacMillan, 1982).
One of the most important strategic barriers to divesting a business is its relationship with the other businesses in the firm. If different business units share production facilities or marketing channels, the economies so achieved could be lost when any of the business units are divested. Further, one business may build up an image or reputation for the company that helps the sales of the other businesses. Divesting such a business, if it shows subnormal earnings, may seem economically justified, but one must consider the damage that it might do to the sales of the other business units (Porter, 1976). Harrigan (1986) studied the effect of vertical integration on exit barriers, and concluded that the greater the degree of vertical integration, the greater the exit barriers. A similar argument was made by Porter (1980).
The length of time a business unit has been with a company is expected to influence the magnitude of the above factors such as the relatedness of a business unit including vertical and forward integration, as well as marketing and operating fit. Before considering exit, managers are more likely to expend additional resources to rebuild the business. The greater such an investment is, the harder it becomes to divest. In addition, personal attachments to products or businesses as well as the psychological effects (i.e., failure) of market exit serve as barriers to exit (Duhaime and Baird, 1987; Hilton, 1972; Porter, 1976).
The factors discussed above are essentially transaction costs associated with exit decisions. These factors give rise to the development of the first hypothesis as stated below:
[H.sub.1.]: Each of the six transaction costs, including sunk costs, marketing fit, operating fit, vertical integration, forward integration, and the number of years a business has been a part of the firm, is inversely related to a market exit decision.
Harrigan (1980) argues that the impact of the stage of the product life cycle is an important consideration in a firm's decision to exit a market. It is obvious that a firm would be more likely to exit a market in the decline stage than when the industry is mature. Hall (1980) studied several firms that achieved success in mature (e.g. Inland Steel) and even in declining (e.g., Phillip Morris - cigarettes) industries by following better strategies. Does the relative importance of the factors change in making the market exit decision for a mature industry as opposed to a declining industry? In other words, does the model used by the decision maker depend on the stage of the life cycle? This was examined by testing the following hypothesis.
[H.sub.2]: The decision to divest a business unit is affected by the stage of the life cycle (maturity or decline) of the industry to which it belongs. Specifically, vertical integration is a more important barrier during the decline stage than during the maturity stage.
Based on many studies of human information processing in a variety of settings, it was expected that decision makers would not interactively process combinations of the six variables. Previous research also shows that an additive linear model of the six variables would be the most powerful and stable model of how the decision makers processed the information interactively. Slovic et al.'s (1977) literature review demonstrated that the occasional interactive model did not substantially increase the amount of explained variation. Stahl (1989) and Stahl and Zimmerer (1984) showed that strategic decision makers process information additively, not interactively, in several different strategic decision-making situations involving several different samples of executives. Thus, this study used an additive linear model without interaction terms among the variables to test the above hypothesis.
Having the literature review in mind, a study of exit barriers was conducted using a decision-making exercise. This approach has been widely used in behavioral decision-making research (see for example, Baker et al., 1989; Harrell and Stahl, 1984; Stahl, 1986).
The major advantage of such a study was that all of Porter's proposed major exit barriers could be included in the study, without the possible biases involved with the use of databases, interviews, and questionnaires (Brenner et al., 1985; Hyman, 1954; Phillips, 1973; Ramanujam and Venkatraman, 1984). The relative importance of factors is inferred from the decisions made. This has an advantage over asking the decision maker to indicate the importance of the factors directly, since the decision makers may have poor insight into their own decision-making process (Stahl and Zimmerer, 1984). Also, decision-making exercises seem to be immune from a social desirability response bias often found in interviews and Likert scale based questionnaires in which the respondent indicates an answer that is socially acceptable (Arnold and Feldman, 1981). A possible disadvantage of using decision-making exercises is that the researcher may specify an inappropriate set of variables for the decision task. Given the literature review conducted for this study and Porter's (1976) model, that risk has been minimized.
Specifically, the study used six factors; namely, cost of divestment relative to the firm's asset base, operating fit, marketing fit, degree of forward vertical integration, degree of backward vertical integration, and number of years the unit has been with the firm. The selection of these factors was based on Porter's (1976) classification of exit barriers as economic, strategic, and managerial. The first variable, "cost of divestment," is obviously the economic barrier faced by the firm. Absolute cost is not a meaningful variable, since different firms are of different sizes. Hence, the variable used represents a cost relative to the firm's asset base.
Fifty-five Fortune 500 corporations were mailed a decision-making exercise designed to model executives' decisions related to the six market exit barriers mentioned earlier. Since single and dominant firms were not likely to have participated in divestment, they were excluded from the sample selected. The categorization into single and dominant groups was made based on Rumelt's (1974) scheme. Data on divisional sales for the categorization were obtained from the Compustat tapes. Forty-two of the companies responded, providing market exit decisions for 82 executives. Fifty-seven percent of the responding firms had two or more executives that completed the deci-sion-making exercise. The list of the executives involved in strategic planning and development was obtained from the Standard and Poor's Register of Corporations, Directors and Executives. The executives targeted were most likely to be involved in making market exit decisions, typically the Vice Presidents of Corporate Planning, Senior Vice Presidents, Group Vice Presidents, and Presidents.
A package containing the decision-making exercise, a personalized cover letter, and a postage paid return envelope was mailed. The only incentive offered for completing the exercise was the summary results of the study and the decision-making profile of the participating executive.
The instrument was designed around a one-fourth replicate of a full factorial design with six factors at two levels each (1/4 2x2x2x2x2x2). This yielded 16 scenarios with two market exit decisions or a total of 32 decisions per respondent. An example of a market exit scenario is presented in Figure I. Since a 1/4 replicate of a [2.sup.6] factorial experiment was used, the treatments to be used were decided by confounding the higher order interactions. Four blocks of 16 treatments each were constructed by confounding on three higher order interactions. The six way interaction and the two three way interactions were confounded to obtain 16 treatments in each of 4 blocks (Hicks, 1982). One of these blocks was chosen randomly as the one to be used. The 16 treatments in this block, each representing a different business unit, were then presented in a random order to construct the instrument.
The analysis of the data involved performing separate regressions for individual respondents. Multiple regression analysis was used and standardized regression coefficients were obtained to derive the relative weights of the various factors for the orthogonal design. The relative importance placed by each of the decision makers on the 6 factors was computed as follows (Ward, 1962):
[Rw.sub.i] = [[B.sub.i]sup.2]/[R.sup.2] for i = 1 ,......, 6
[Rw.sub.i.] = relative weight of factor i
[B.sub.i.] = standardized Beta weight of factor i
[R.sup.2] = coefficient of determination
These relative weights are objective in that they are determined by the regression analysis (Darlington, 1968) and sum to 1. This allows comparison of the relative importance of the decision cues (i.e., market exit barriers) used in the decision-making exercise.
Hypothesis two was tested in two steps. First, an equivalent of the Chow's F-test (Chow, 1960) was used to examine the differences in the regression equations for exit decisions in declining and mature industries. Second, once the Chow's F-test indicated the presence of significant differences, six paired t-tests were performed to establish which exit barriers were different. In implementing the Chow's F-test, a dummy variable was defined with a value 0 for business units in a declining industry, and a value 1 for units in a mature industry. The product of the dummy variable and each of the six variables were defined as six additional variables. A multiple regression with all of the data was performed using the following model:
y = [B.sub.0] + [summation of] [B.sub.i][X.sub.i] where i = 1 to 13 + [[Epsilon].sub.i]
Where, y = Market Exit Decision
Bo = Constant
[B.sub.i] = Beta Weights associated with decision cues [X.sub.1] to [X.sub.13]
[X.sub.1] to [X.sub.6] = Barriers to Exit
[X.sub.7] = Dummy Variable
[X.sub.8] to [X.sub.13] = Interaction Variables
[[Epsilon].sub.i] = Error Term
The interaction variables [X.sub.8] through [X.sub.13] are obtained by multiplying [X.sub.1] through [X.sub.6] with the dummy variable. Testing the hypothesis that the regression models for declining and mature industries are the same is equivalent to testing if the regression coefficients of the dummy variable and variables [X.sub.8] through [X.sub.13] are all zero in the above model.
Hence, the hypothesis tested was as follows:
[H.sub.0]: [B.sub.7] = [B.sub.8] = [B.sub.9] = [B.sub.10] = [B.sub.11] = [B.sub.12] = [B.sub.13] = 0.
Operationalization of Variables
The operationalization of the variables is based on the methodology used by Stahl and Zimmerer (1984). Each variable of interest has two levels. By changing the level of one or more variables, several scenarios can be simulated. In fact, exactly [2.sup.f] different scenarios can be constructed, where f is the number of factors used as the independent variables. However, as f gets larger, fractional replication can be used for practical reasons with a minimum loss of information (Hicks, 1982).
Each of the six factors was considered at two levels, high or low. For cost, "high" means greater than 10 percent of the firm's assets and "low" means less than 10 percent. For production/operating fit and marketing fit, "high" means the presence of shared resources, and "low" means the absence of such sharing. For forward vertical integration, "high" means a unit that buys a large fraction of the total sales of the unit or units upstream. For backward vertical integration, "high" means a unit whose sales contribute to a large fraction of the total purchases of the unit or units downstream. If one considers an automobile manufacturer owning an iron ore mining company, a steel company and automobile dealerships, then the iron ore mining business would be upstream and the dealerships would be downstream businesses. For number of years, "high" means more than five years.
The values of the dependent variables, the recommendations for divestment of units in the mature and declining industries, were the decisions made by the respondents on an eleven point scale (from 0 to 10, with 0 representing a decision strongly against divestment, and 10 representing a decision strongly in favor of divestment). For each simulated business unit, the respondent made two decisions, one assuming the industry is mature and the other assuming that it is declining.
Two multiple regressions were performed for each respondent, one each for the maturity and declining stages to test hypothesis one. First, using the divestment decision in a declining industry as the dependent variable, the relative weights were calculated for each of the six independent variables. This was accomplished by performing multiple regression analysis for each executive and then transforming the standardized regression coefficients. Five of the 82 regression models were not significant at the 5% level of significance. The same procedure was followed for the divestment decisions in the mature industry. Six of the 82 models were found not significant at the 5% level of significance. These six included the same five respondents with insignificant regression results for the declining industry decisions. The responses of the executives with insignificant regression models were excluded from analyses since they would not be significantly different from zero and indicate random decision (Stahl, 1989). The ability to detect random decisions makers is a strength of this approach and permits further analysis of stable variation.
Table 2. Group Regression Results
Exit Relative Weight for Relative Weight for Barriers(a) Declining Industry Mature Industry
FORWARD-VI 32.12(*) 28.56(*) BACKWARD-VI 21.52(*) 24.63(*) COST 21.39(*) 20.33(*) OPERFIT 14.12(*) 13.02(*) MKTGFIT 10.65(*) 12.38(*) N-YEARS 00.17 01.11(*)
n = 1228 for declining industry.
n = 1212 for mature industry.
[R.sup.2] = 0.3604 for declining industry, p [less than] 0.01.
[R.sup.2.] = 0.3652 for mature industry, p [less than] 0.01.
* indicates significantly different from zero, p [less than] 0.01.
a. COST = Cost of divestment-including dismantling, selling, and sunk costs.
OPERFIT = Operating fit between divestment candidate unit and other units of the firm.
MKTGFIT = Marketing fit between the divestment candidate unit and other units of the firm.
FORWARD-VI = Forward vertical integration of the unit (degree to which it buys from other units of the firm).
BACKWARD-VI = Backward vertical integration of the unit (degree to which it sells to other units of the firm).
N-YEARS = The number of years the unit has been with the firm.
The average [R.sup.2] value for the individual regressions after excluding the ones that were not significant was 0.85, indicating a high degree of internal consistency. As expected, the regression coefficients were negative showing an inverse relationship between the presence of the barriers and decisions to divest.
To test whether each of the six factors can be considered significant overall, a cross sectional regression analysis was performed with the data from all the respondents. Those with insignificant individual regressions were excluded. The combined regressions involved 77 respondents for the declining industry and 76 for the mature industry. The results of the two regressions with the combined data are summarized in Table 2. The table shows the relative weights of the six factors with the divestment decision in declining industry and mature industry as the dependent variables.
The above results indicate that all variables except the number of years in business significantly affect the decision to exit from the market in a declining industry. In a mature industry, each of the six variables is significant. The sharp decrease in the value of [R.sup.2] (from 0.85, the average [R.sup.2] for the individual regressions, to 0.36 for the group regression) suggests that decision-making models are specific to the individuals. In other words, this suggests that there is a suggestion that there is no common model that most executives agree upon.
Another way of looking at the overall results is to compute the average of the relative weights of the six factors [TABULAR DATA FOR TABLE 3 OMITTED] obtained from the individual regressions. To find the relative importance of the six factors, an Analysis of Variance was performed, followed by the Duncan's Multiple Range Test. The results for the declining industry are shown in Table 3.
Forward vertical integration was thus the most important deterrent to divestment of a business in a declining industry, while the number of years the business unit has been with the firm was found to be the least important. No significant difference was found between the mean relative weights of backward vertical integration and cost of divestment, or between operating fit and marketing fit. Similarly, no significant difference was observed between backward vertical integration and operating fit or cost and operating fit.
Table 4 shows the mean relative weights and Duncan's Multiple Range Tests results for the mature industry. The results in Table 4 show that both types of vertical integration and the cost of divestment are not significantly different from each other. These three factors are the major barriers to exit. The average relative weight for vertical integration is higher than operating fit, marketing fit, and the number of years.
The importance of forward vertical integration is denoted in Tables 3 and 4. The Duncan's test shows that for a mature industry, forward and backward integration and the cost of divestment are not significantly different from each other in terms of the barriers they pose to market exit. However, in the declining industry, forward integration is significantly more important. This is not surprising, since a captive customer is more difficult to let go of during decline, when other buyers are more difficult to find for the products of upstream units.
To test whether executives use different models in making divestment decisions for the declining and mature [TABULAR DATA FOR TABLE 4 OMITTED] stages, a Chow's F-test was performed for each respondent. A significant difference in the two models (at [varies] - 0.05) was found for 36 of the 82 respondents. To test if there is a difference overall, a Chow's F-test was performed with the combined regressions for declining and mature industry (with 77 and 76 respondents, respectively). The results indicate that there is a statistically significant difference in the market exit decision models used for the two stages of the product life cycle (Chow's F = 27.41, p [less than] 0.01, n = 2,424). In testing the differences in importance of exit barriers with exit decisions in declining and mature industries, six paired t-tests were performed. The results indicate that four of the six barriers compared differ significantly (Table 5).
Conclusions and Discussion
The individual and group regressions show that all barriers but the number of years a business unit has been with the firm are significant in the case of a declining industry. This does not mean, however, that the number of years a business has been with the firm is not important. It only implies that it is not important relative to the other barriers tested in this study. For a mature industry, all six factors were found significant in the overall regression. The regression coefficients in each case were negative, indicating that each factor deterred the decision to divest and thus is an exit barrier.
The relative importance of the factors is not contrary to expectations. Forward vertical integration of the business unit was the most important deterrent to its divestment, followed by backward vertical integration, cost of divestment, and operating and marketing fit. The results are consistent with Harrigan's (1985a) argument that vertical integration is an important exit barrier, and that a downstream unit faces a greater exit barrier than an upstream one. Forward vertical integration is found to [TABULAR DATA FOR TABLE 5 OMITTED] raise a greater barrier than backward vertical integration in a declining industry. This seems logical, since divesting a captive customer, especially during decline, leaves the upstream units with no customers. Also, the presence of a captive buyer can lead to the upstream units being less efficient, and consequently the divestment of the downstream unit can seriously affect the sales of the upstream unit. Divestment of the upstream unit, however, is perhaps not as great a threat to the downstream unit, since the likelihood of finding another supplier during decline is greater than that of finding a buyer.
As Harrigan (1985b) pointed out, however, the phase of the product life cycle is important, and can change the relative importance of forward and backward vertical integration. While forward integration raised a greater barrier during decline, there was no significant difference between forward and backward integration during the mature phase. This can be explained by the fact that during maturity there are more options available to the upstream units than during decline. Hence the divestment of a downstream unit is a greater problem during decline than during the mature phase. In general, vertical integration creates greater interdependence between business units of a firm than horizontal relatedness. While relatedness due to shared production or marketing resources is important, the divestment of a related unit does not have as much of a direct impact on sales as in the case of a vertically integrated unit.
The lack of significance of the number of years a business unit has been with the firm in a declining industry suggests that the managerial barrier, as measured by the number of years of association with the firm, does not play as important a role in divestment decision making as proposed by Porter (1976). The lack of significance of this factor for declining industry and its significance for mature industry suggests that the economic considerations for the declining industry would be too strong to permit personal attachments to come in the way of divestment. In a mature industry, however, there is a greater chance of revitalization being used as the corporate strategy instead of divestment, and personal attachment may be a deciding factor in the decision.
The results of the Chow's F-tests with the individual reggessions show that 36 respondents (about 45% of the total number of respondents) used different models for declining and mature industries. Thus, on an individual basis, executives do seem to make decisions differently depending on the stage of the life cycle of the business unit. The overall regressions, when compared using the Chow's F-test, confirm this difference in the models for the two stages of the life cycle. This is also evidenced by the six paired t-test results. As mentioned before, there are two notable differences in the models. First, forward vertical integration is a more important factor during decline than during maturity. Secondly, the variable N-YEARS changes from being insignificant for the declining industry to being significant for the mature industry. The relative weight of the factor is still very low (1.11), indicating that while managerial barriers may be present, their influence on the decision is very small compared to other factors.
This was an exploratory study to discover whether the factors mentioned by researchers as important exit barriers were really considered important by corporate planners. The study makes two major contributions. First, it confirms the presence of the major barriers to exit and found their relative importance to executives in making a divestment decision. Secondly, it shows the difference in the importance of these factors to exit decisions in two different stages of the product life cycle, maturity and decline. The results support earlier literature which proposed that the stages of the product life cycle determine, at least in part, corporate strategies.
An objection to the product life-cycle theory in the literature is that executives cannot accurately tell when an industry is in a certain stage, hence the theory cannot be used to formulate strategy. This study shows that when the stage of the life cycle is known, executives use that information for strategy formulation. An implication for managers, therefore, is that the ability to identify where the industry is in the life cycle is important in formulating competitive strategy.
The relative importance of the exit barriers also has some implications for strategic management. The finding that vertical integration is the most important barrier to exit raises a question about the pros and cons of following that strategy. Opinion is divided about the benefits of vertical integration (see Rumelt, 1974; Harrigan, 1980, 1986; Buzzell, 1983; Balakrishnan and Wernerfelt, 1986). While the strategy can provide better control over raw materials and distribution networks and raise entry barriers for potential competitors (Harrigan, 1980), there are disadvantages, too. When the capital requirements are high, the firm is less able to respond to changing market conditions and exit barriers are raised. Ford Motor Company is the last vertically integrated of the U.S. auto makers, but GM is 40% less efficient in production time than Ford (Taylor, 1992).
Empirical studies relating vertical integration to performance do not show much agreement. Rumelt (1974) and Buzzell (1983) found that highly vertically integrated firms were less profitable than other kinds of diversified firms. Harrigan (1986), however, found the reverse to be true. Balakrishnan and Wernerfelt (1986) proposed that vertical integration could be profitable if certain market conditions were present.
The results of this study confirm the contention that vertical integration has a major disadvantage in that it raises exit barriers. No case can be made, however, about the overall impact of the strategy on the performance of a firm. It obviously depends on competition, time period, industry, and other factors. One cannot agree that there is one "best" strategy. The outcomes of the most profitable strategy are time period and situation specific. Simply put, a strategy that is successful today may be a failure tomorrow.
An important area of study that is closely related to market exit barriers is that of market entry barriers. Many of the barriers that prevent timely exit are usually caused by strategies implemented during the growth phase to raise entry barriers against potential competitors (Porter, 1980). Vertical integration and capital investment are both deterrents to market entry (Harrigan, 1981; Karakaya and Stahl, 1989). Thus, there is a tradeoff that firms must consider between the added advantage during the growth and maturity phases and the potential cost of running an unprofitable business during decline.
The second major finding of this study was the difference in the models used for divestment decision making in the mature and declining stages of the product life cycle. This brings up the question of cause and effect that has been discussed in the literature (Wind and Claycamp, 1976). Do strategies differ based on the stage of the life cycle, or do the strategies followed determine the stage of the life cycle and the longevity of the product? A distinction between the life cycle for the industry and for an individual firm is required to answer the question. If by the product life cycle one refers to the industry as a whole, then it is more likely that strategy is a response to the stage of the life cycle. Environmental factors like the economic conditions and consumer tastes can affect the life cycle, but it is unlikely that an individual firm's strategy would change the characteristics of the life cycle, except in the case of a monopoly or perhaps an oligopoly. However, an individual firm's sales may grow, level off and decline independent of the stage of the industry life cycle. This is where the strategy of the firm determines the stage of the life cycle that the firm is in.
The hypothesis tested in this study ([H.sub.2]) is about the difference in models for divestment for mature and declining stages of the industry as a whole. In both cases, the individual firm's sales were assumed to be declining. Hence, the responses in this case indicate that strategy depends on the stage of the life cycle.
Hall's (1980) study of firms that were very successful even in mature and declining industries shows that they follow one of two major strategies - a low cost leadership approach with acceptable quality, or a product differentiation approach with an acceptable cost structure.
In the declining cigarette industry, American Brands increased sales by introducing cheaper brands (Misty, Montclair, and Bill Durham) that have sold very well and now account for 36% of the firm's business (Nulty, 1992). It has been pointed out that since cigarettes are not fashionable anymore, perhaps people smoke cheaper brands because they have to go outside to light one anyway and cannot flaunt the brand name to anyone.
Cooper Tire Co., which ranks 9th in market share in the mature tire market, has reaped profits by targeting the replacement tire market through independent distributors. This strategy has produced more profits for the company than selling to auto makers (Erdman, 1992).
In each case, it seems that the strategy was a successful response to the stage of the life cycle. The stage of the life cycle itself seems to be a result not of the strategies of the firms, but other factors (e.g., sociocultural factors in the case of the cigarette industry).
Limitations and Future Research
Despite some advantages of the decision modeling methodology over questionnaires and studies that rely on historical data, there are limitations to using such methodology. Since decision making is an exercise that relies on varying the levels of certain cues, it is not possible to completely recreate a situation that corporate planners may face. Several factors must be left out. The respondents are presented with a set of cues to base their decisions on, whereas in reality one of the tasks of the decision makers would be to sift through much information to determine which factors to consider.
Emotional involvement with decisions is also a limiting factor. While emotions can play a role in a real decision, the responses on an exercise are more likely to be based more on reason. A specific example in this study is the factor number of years a business unit has been with a firm. This was used as a proxy for emotional involvement, and the low relative importance of the factor could be due to the fact that executives do not actually feel the emotional involvement while responding to the exercise. Also, the sample of executives surveyed were mostly the strategic planners, whose responsibilities probably included making a recommendation only, and not the final decision to divest. Emotional barriers would be more likely to affect the final decision maker. Thus, there is a possibility of error regarding factors dealing with emotions.
While this study found the importance of exit barriers by studying divestment decision making, no attempt was made to study the strategic alternatives to divestment. This is a promising area for research. While researchers have identified several strategies that can be followed during decline, such as revitalization and harvesting, such studies have been essentially prescriptive. An attempt needs to be made to find out what corporate planners actually do. Under what conditions is one of these strategies preferred by the executives over another? What factors do they consider important in formulating the strategies?
Alexander, G. J., P. G. Benson, and J. M. Kampmeyer. 1984. "Investigating the Valuation Effects of Announcements of Voluntary Corporate Selloffs." The Journal of Finance 19 (June): 503-517.
Ambrosio, J. 1993. "ICC Stalked by Leasing Woes?" Computerworld 27 (October 11): 1-2.
Arnold, H. J., and D. C. Feldman. 1981. "Social Desirability Response Bias in Self-report Choice Situations." Academy of Management Journal 24: 377-385.
Baker, D. D., R. Ravichandran, and D. M. Randall. 1989. "Exploring Contrasting Formulations of Expectancy Theory." Decision Sciences 20 (1): 1-13.
Balakrishnan, S, and B. Wernerfelt. 1986. "Technical Change, Competition, and Vertical Integration." Strategic Management Journal 7: 347-359.
Baumol, W., J. C. Panzar, and R. Willig. 1982. Contestable Markets and the Theory of Industry Structure. San Diego: Harcourt Brace Jovanovich.
Bernheim, B. D. 1984. "Strategic Deterrence of Sequential Entry into Industry." Rand Journal of Economics 15:1-11.
Brenner, M., J. Brown, and D. Canter. 1985. The Research Interview Uses and Approaches. Orlando, FL: Academic Press.
Buzzell, R. D. 1983. "Is Vertical Integration Profitable?" Harvard Business Review (Jan.-Feb.): 92-102.
Chow, Gregory. 1960. "Tests of Equality Between Sets of Coefficients in Two Linear Regressions." Econometrica 28 (Fall): 591-605.
Christensen, K. H., and C. A. Montgomery. 1981. "Corporate Economic Performance: Diversification Strategy vs. Market Structure." Strategic Management Journal 2: 327-343.
Darlington, R. B. 1968. "Multiple Regression in Psychological Research and Practice." Psychological Bulletin 69 (3): 161-182.
Dixit, A. 1980. "The Role of Investment in Entry Deterrence." Economic Journal 90: 95-106.
Dubin, R. N. 1986. "Divestments, Some Practical Thoughts on How to Sell a Division." Financial Executive's Manual 2 (November): 37-39.
Duhaime, I. M., and I. S. Baird. 1987. "Divestment Decision-making: The Role of Business Unit Size." Journal of Management 13: 483-498.
-----, and J. H. Grant. 1984. "Factors Influencing Divestment Decision-making: Evidence From a Field Study." Strategic Management Journal 5: 301-318.
The Economist. 1995. "The End of Good Times: Personal Computer." 337 (December 9): 61-62.
Ennis, R. M., and R. L. Parkhill. 1986. "South Africa Divestment: Social Responsibility or Fiduciary Folly?" Financial Analysts Journal (July-Aug.): 30-38.
Erdman, A. 1992. "Now Hear This Jack Welch." Fortune 125 (7): 94-96.
Feinberg, P. 1988. "The Ultimate Financing: Putting the Company on the Block." Corporate Cashflow Magazine (Sept.): 60-61.
Flath, D. 1994. "Keiretsu Shareholding Ties: Antitrust Issues." Contemporary Economic Policy 12 (January): 24-37.
Hall, J.P. 1986. "Ethics in Investment: Divestment." Financial Analysts Journal (July-Aug.): 7-10.
Hall, W. K. 1980. "Survival Strategies in a Hostile Environment." Harvard Business Review (Sept.-Oct.): 75-85.
Harrell, A., and M. J. Stahl. 1984. "Modeling Managers' Efforts Level Decisions for a Within Persons Examination of Expectancy Theory in a Budget Setting." Decision Sciences 15: 52-73
Harrigan, K. R. 1980. Strategies for Declining Businesses. Lexington Books, D.C.: Heath & Co.
-----. 1981. "The Effect of Exit Barriers Upon Strategic Flexibility." Strategic Management Journal 1:165-176.
-----. 1985a. "Exit Barriers and Vertical Integration." Academy of Management Journal 28: 686-697.
-----. 1985b. "Vertical Integration and Corporate Strategy." Academy of Management Journal 28: 397-425.
-----. 1986. "Matching Vertical Integration Strategies to Competitive Conditions." Strategic Management Journal 7: 535-555.
Hicks, C. R. 1982. Fundamental Concepts in the Design of Experiments. New York: CBS College Publishing.
Hilton, P. 1972. "Divestiture: The Strategic Move on the Corporate Chessboard." Management Review (March): 16-19.
Hite, G., and J. Owers. 1983. "Security Price Reactions Around Corporate Spinoff Announcements." Journal of Financial Economics 12 (December): 409-436.
Hyman, H. 1954. Interviewing in Social Research. Chicago: University of Chicago Press.
Holcomb, J. H., and D. A. Evans. 1987. "The Effect of Sunk Costs on Certain Decisions in Experimental Markets." Journal of Behavioral Economics (Fall): 59-64.
Karakaya, F., and M.J. Stahl. 1989. "Barriers to Entry and Market Entry Decisions in Consumer and Industrial Goods Markets." Journal of Marketing 53 (April): 80-91.
Klein, A. 1986. "The Timing and Substance of Divestiture Announcements: Individual, Simultaneous and Cumulative Effects." Journal of Finance 41 (July): 685-696.
Kleindorfer, P., and G. Knieps. 1982. "Vertical Integration and Transaction Specific Sunk Costs." European Economic Review 19: 71-87.
Lynch, M. C. 1980. "Many Firms are Selling Off Acquisitions to Clarify Their Images, Lift Their Stocks." Wall Street Journal (Dec. 4): 48.
MacLeod, W. B. 1987. "Entry, Sunk Costs, and Market Structure." Canadian Journal of Economics (February): 140-152.
MacMillan, I. C. 1982. "Seizing Competitive Initiative." The Journal of Business Strateg) 2 (Spring): 43-57.
McDowell, E. 1996. "The Cruise Industry is Moving Toward More Campaigns in the Face of Declining Business." The New York Times 145 (February 22): 4.
Miles, J. A., and J. D. Rosenfeld. 1983. "The Effect of Voluntary Spin-off Announcements on Shareholder Wealth." Journal of Finance 38 (December): 1597-1606.
Mitchell, W. 1994. "The Dynamics of Evolving Markets: The Effects of Business Sales and Age on Dissolutions and Divestitures." Administrative Science Quarterly 39 (December): 575-603.
Muller, C. 1990. "The Marriott Divestment: Leaving the Past Behind." Cornell Hotel and Restaurant Administration Quarterly 30: 7-13.
Nulty, P. 1992. "How to Win With a Value Strategy." Fortune 126 (2): 94-96.
Phillips, D. L. 1973. Abandoning Method. San Francisco, CA: Jossey-Bass Co.
Porter, M. E. 1976. "Please Note Location of Nearest Exit: Exit Barriers and Strategic and Organizational Planning." California Management Review 19 (2): 21-33.
-----. 1980. Competitive Strategy: Techniques for Analyzing Industries and Competitors. New York: Free Press.
Ramanujam, V., and N. Venkatraman. 1984. "An Inventory and Critique of Strategy Research Using the PIMS Database." Academy of Management Review 9: 138-151.
Rosenfeld, J. D. 1984. "Additional Evidence on the Relation Between Divestiture Announcements and Shareholder Wealth." Journal of Finance 39 (December): 1437-1448.
Rumelt, R. P. 1974. Strategy, Structure, and Economic Performance. Boston: Division of Research, Graduate School of Business Administration, Harvard University.
Schroeder, M. 1992. "Westinghouse Gets a Big Dose of Reahty." Business Week February 17: 110-15.
Sicherman, N. W., and R. H. Pettway. 1987. "Acquisition of Divested Assets and Shareholder's Wealth." The Journal of Finance 42 (5): 1261-1273.
Slovic, P., B. Fischhoff, and S. Lichtenstein. 1977. "Behavioral Decision Theory." Annual Review of Psychology 28: 1-39.
Stahl, M. J. 1986. Managerial and Technical Motivation: Assessing Needs for Achievement, Power, and Affiliation. New York: Praeger Publishers.
-----. 1989. Strategic Executive Decisions. Westport, CT: Quorum Books.
-----, and A. M. Harrell. 1982. "Evolution and Validation of a Behavioral Decision Theory Measurement Approach to Achievement, Power, and Affiliation." Journal of Applied Psychology 67: 744-750.
-----, and T. W. Zimmerer. 1984. "Modeling Strategic Acquisition Policies: A Simulation of Executives' Acquisition Decisions." Academy of Management Journal 27: 369-386.
Stewart, G. B., and D. M. Glassman. 1988. "Why Restructuring Adds Value: When Subtracting is a Goal." Cashflow Magazine (February): 48-50.
Taylor, A. 1992. "Can GM Remodel Itself?." Fortune 125 (1): 26-34.
Von Krogh, G., and J. Roos. 1994. "Corporate Divestiture and the Phantom Effect. European Management Journal 12 (June): 171-178.
Ward, J. 1962. "Comments on the Paramorphic Representation of Clinical Judgment." Psychological Bulletin 59 (1): 74-76.
Wind, Y., and H. Claycamp. 1976. "Planning Product Line Strategy: A Matrix Approach." Journal of Marketing 40 (January): 2-9.
Zaima, J. K., and D. Hearth. 1985. "The Wealth Effects of Voluntary Spinoffs: Implications for Divesting and Acquiring Firms." Journal of Financial Research 8 (Fall): 227-236.
Figure I. An Example of A Market Exit Decision Making Exercise
MARKET EXIT CANDIDATE UNIT # 1
The characteristics of this business unit are as follows.
* Number of years unit is with firm is more than five
* Forward vertical integration (selling to other units) is low
* Backward vertical integration (buying from other units) is high
* Production/operations technology is not similar to other units of the firm.
* Cost of divestment is less than 10% of firm's assets.
* Marketing resources are shared with other units of the firm.
DECISION A: Given the levels of the factors above, for a unit with low returns in a declining industry, indicate how strongly you would recommend its divestment.
Strongly Against Strongly For Divestment 0 1 2 3 4 5 6 7 8 9 10 Divestment
DECISION B: Given the levels of the factors above, for a unit with low returns in a mature industry, indicate how strongly you would recommend its divestment.
Strongly Against Strongly For Divestment 0 1 2 3 4 5 6 7 8 9 10 Divestment
|Printer friendly Cite/link Email Feedback|
|Author:||Nargundkar, Satish V.; Karakaya, Fahri; Stahl, Michael J.|
|Publication:||Journal of Managerial Issues|
|Date:||Jun 22, 1996|
|Previous Article:||Family responsibilities and absenteeism: employees caring for parents versus employees caring for children.|
|Next Article:||An examination of job tension and coping in the relationship between stressors and outcomes in public accounting.|