Printer Friendly

The financial performance of diversified hospital subsidiaries.

The service mix of acute inpatient care hospitals in the United States is continually changing in response to technological changes, community needs and, more recently, perceived environmental threats such as increased competition. Prior to the last decade, new services were simply added to the hospital in its preexisting organizational structure.

During the 1980s, however, many hospitals restructured. By restructuring, we mean that these hospitals have formed separately incorporated subsidiaries to house some of the new services (Gerber 1983; Alexander, Morlock, and Gifford 1988). These restructured firms can take on several organizational forms. One structure consists of a relatively small parent firm with hospital, fund-raising, and service-producing subsidiaries. This form of restructuring is typically termed the parent holding company model. Another frequent structure uses the hospital as the parent firm for subsidiaries. These kinds of restructuring have occurred both among members of multihospital systems and among freestanding hospitals.

A study by Alexander, Morlock, and Gifford (1988) indicates that restructuring is a common change in hospital organization that, at least until the mid-1980s, was increasing in prevalence. They analyzed data from 3,189 acute care community hospitals responding to an American Hospital Association survey in May 1985. The results showed that 34 percent of the hospitals had restructured in the past five years (1980-1985) and that restructuring had increased steadily during the five-year period. The most common form of restructuring by far (80 percent) was the parent holding company model.

The many motivations for restructuring in the last decade have included protecting tax-exempt status, increasing market share, developing new sources of revenue through diversification, and avoiding certificate-of-need regulations (Gerber 1983; Ernst and Whinney 1982). Many of these motivations can be summarized as attempts to improve a hospital firm's financial viability and its ability to deliver services.

Despite the proliferation of hospital restructuring, relatively little research has been done to examine its impact. The purpose of this exploratory study is to develop an understanding of factors that affect the financial performance of hospital subsidiaries. More specifically, the article examines the relationship between product and market diversification and financial performance in separately incorporated business units (subsidiaries) of acute care inpatient hospital firms.


As noted, although there has been some documentation of changes in services (Alexander 1990), factors associated with offering new services (Shortell, Morrison, Hughes, et al. 1987), and changes in organizational structure of hospitals (Alexander, Morlock, and Gifford 1988), there have been few empirical studies of how service mix or restructuring affects financial performance. No empirical studies have specifically addressed the financial performance of subsidiaries of a cross section of restructured hospital firms. Instead, the performance of services offered by multihospital systems, through a variety of organizational units, and by hospitals themselves have been analyzed.

Shortell, Morrison, and Hughes (1989) reported that only one-third of the diversified services offered by eight multihospital systems were profitable, or operating revenues exceeded direct and indirect operating costs. Although Shortell, Morrison, Hughes, et al. (1987) evaluated variables associated with offering unprofitable alternative services, they did not investigate the association of such variables with the degree of profitability. In neither report did the authors distinguish between services offered in subsidiaries or those included as part of the hospital.

Clement (1987) studied the financial performance of diversified nonprofit hospitals in the late 1970s and early 1980s. She found no relationship between diversification and profitability or risk, but was unable to study business units.

Results from firm-level studies of industries other than health care have shown that diversification both is (Rumelt 1974; Palepu 1985; Christensen and Montgomery 1981, for example) and is not (Keats and Hitt 1988) associated with better financial performance of the firm. Early studies, which measured diversification with counts of businesses or variations of business counts, reported little evidence of any relationship between corporate diversification and financial performance measured with accounting ratios (Gort 1962; Rhoades 1973, 1974; Carter 1977; Bass, Cattin, and Wittink 1978). Studies defining diversification as the conglomerate firm produced inconsistent results using accounting performance measures (Melicher and Rush 1973; Holzmann, Copeland, and Hayya 1975).

However, when different types of diversification have been evaluated, a positive relationship between related diversification and financial performance has often been reported. Rumelt's (1974) pioneering work showed not only that related diversification was associated with higher profitability than unrelated diversification, but also that more narrowly focused (related-constrained) diversification was more profitable than the looser related-linked diversification. Subsequent work has supported Rumelt's work with regard to diversification of products (e.g., Christensen and Montgomery 1981; Rumelt 1982; Nathanson and Cassano 1982; Grant, Jammine, and Thomas 1988; Lubatkin and Rogers 1989). In addition, Capon et al. (1988) recently found that more market specialization by diversifying firms is related to better financial performance.

Still, there are important exceptions. Michel and Shaked (1984) and Dubofsky and Varadarajan (1987) have reported that unrelated diversifiers earned higher risk-adjusted market returns than related diversifiers. Lubatkin and Rogers (1989) have noted that the results may be due to employing product count type measures of diversification. Keats and Hitt (1988), on the other hand, found no relationship between a measure of diversification based upon Rumelt's typology and accounting performance, but found higher financial market performance to be related to higher (unrelated) diversification.

There is less evidence concerning the relationship of diversification to financial performance for divisions of firms. Ravenscraft (1983) has reported that no statistically significant relationship exists between diversification and operating margin. Biggadike (1979) found that business units that could draw upon market similarities performed better than those that could not. His sample consisted of only 40 business units, however.

In sum, the relationship between diversification and financial performance is still not clear (Ramanujam and Varadarajan 1989). Previous research is largely limited to the study of large firms (Grant, Jammine, and Thomas 1988), manufacturing firms participating in multiple industries (Palepu 1985; Bass, Cattin, and Wittink 1978; Dubofsky and Varadarajan 1987), and for-profit firms. Studies of health care delivery firms have been few. It has not been clear that the unit of analysis for these studies has been the business unit, and the work has often been restricted to multihospital system members.


Drawing upon previous research on organizational structure and performance, especially research on diversification, we propose that three sets of factors will influence a business unit's financial performance: (1) the extent to which the activity of the business unit allows it to make useful resource exchanges (e.g., information, skills, technology) with the firm's main business, its hospital(s); (2) the processes involved in resource exchanges between business units and the hospital; and (3) market conditions faced by the business units.(1)


Exchanges of skills and knowledge that enhance financial performance between business units and the hospital (the main business) are more likely to occur to the extent that hospitals and business units produce similar products or services and have similar customers. When this occurs, corporatewide distinctive competencies can be transferred across the restructured hospital's separate business units (Govindarajan and Fisher 1990; Bower 1981; Yavitz and Newman 1982; Barney 1986). Recent research focusing on the strategic business units (SBUs) of diversified firms indicates that the more resources they share, the higher their levels of effectiveness (Govindarajan and Fisher 1990). An example of resource sharing is the use of a common management staff or physical facility for several business units in a firm. Rumelt (1974) used the concept of the "relatedness" of business units that occurs when "a common skill, resource, market, or purpose applies to each" (Rumelt 1974, 29).

As discussed previously, empirical evidence exists that suggests that financial performance is likely to be better when product and market diversification are related (Rumelt 1974; Capon et al. 1988). Consequently, we expect that the financial performance of business units whose products, services, and markets directly involve inpatient acute care will be better than that of business units whose business is not related to acute inpatient care. Similarly, we expect that the financial performance of business units whose products and services or markets are both related to health care will be better than that of business units whose products and services are not related to acute inpatient care.


Other relevant organizational research has focused on the rules and norms that regulate the transfer of resources between two or more organizations. The transaction cost perspective has been particularly useful for focusing attention on exchange processes (Williamson 1975, 1981). Mick and Conrad (1988), for example, have used this perspective to analyze vertical integration by health care firms. They note that a key assumption of the transaction cost perspective is that firms seek to maximize efficiency in transactions or resource transfers to improve financial performance. The ability of firms to maximize efficiency in transactions will depend on (1) the length of time that subsidiaries are associated with the hospital and parent firm, and (2) the extent to which the units share goals.

The longer the hospital and parent firm exchange resources with the subsidiary, the more likely it is that resource transfers will become routine, and thus more efficient. Transaction costs related to coordinating exchanges between two or more units in a firm will decrease over time as they learn to work with each other. Increased efficiency may also facilitate higher revenues because more time can be devoted to developing external rather than internal exchanges.

Biggadike's (1979) study of a sample of 40 surviving new businesses begun by Fortune 200 companies showed that, on average, the businesses incurred large losses during their first four years. When he supplemented his sample with data from older surviving business units, he concluded that an average of eight years was required to reach profitability, and 10 to 12 years to achieve the same profitability as the mature businesses of the firm. Thus, we expect that the financial performance of business units will be better the longer they are affiliated with their hospital.

Transaction costs between business units and hospitals are also likely to be lower if the units and hospitals share goals and objectives (Ouchi 1980). When exchange partners share values and goals, they are likely to trust each other and engage in more efficient transactions. They may also exhibit commonalities in management methods and style and thus share the heuristic principles required to deal with unfamiliar situations (Prahalad and Bettis 1986; Grant 1988).

For the hospital business units studied, a proxy for possible goal differences is reflected in the match between the parent company and business unit tax status. Although tax laws and other environmental conditions may have influenced the initial choice of legal form, subsequent goal differences may affect subsequent financial performance.

Not-for-profit firms are typically believed to pursue community and service goals subject to financial limitations. In contrast, the goal of a for-profit firm is to maximize the wealth of its owners. Thus, an aspect of goal congruence that can reduce transaction costs between the main hospital and the other business units is the extent to which their orientation to health care as a profit-making business is similar. Consequently, we expect that the financial performance of business units whose goals are congruent with those of the hospital or parent firm (i.e., they share profit orientation and tax status) will be better than that of business units whose goals are not congruent with the goals of their main businesses. However, it should also be noted that incongruence in ownership between the parent and subsidiary is a partial and indirect measure of goal incongruence.


Previous research indicates that efficient exchanges and exchange processes are not sufficient for effective financial performance in diversified firms (Biggadike 1979; Montgomery 1985). Market factors are also important to financial performance. Such variables are also becoming increasingly important for health care providers as more competition is introduced into markets (Zwanziger and Melnick 1988).

More specifically, where competition among providers is low, consumers will be more likely to enter into exchanges with the firm because there are fewer alternative exchange partners. Further, service firms located in markets where demand is high will have more potential exchange partners. Thus, we expect that the financial performance of business units will be better in markets with higher demand and less competition.


The data for the study consist of product, ownership, and financial performance data from separately incorporated subsidiaries of 35 acute inpatient hospital firms operating in Virginia.(2) The data were reported to the Virginia Health Services Cost Review Council (VHSCRC) for the 1987 fiscal year in response to a new legislative mandate. Because no similar national or state reporting requirements exist, the data set is unique.(3) It should be noted, however, that data are available only for surviving subsidiaries. In addition, no consolidated firm financial data are available.

Firms operating specialty hospitals, including rehabilitation, psychiatric, children's, heart, eye, and university teaching hospitals, are excluded from the analyses. The parent, hospital, and fund-raising subsidiaries of the remaining firms are also excluded from the study. Only the nonhospital product- or service-producing subsidiaries are examined. Finally, the 162 total number of subsidiaries studied excludes the seven product subsidiaries of the two for-profit hospital firms operating in the state. These are not available for study because of more lenient reporting requirements. Thus, the only subsidiaries studied are those of not-for-profit parent firms. Although the reporting requirement was limited to subsidiaries of hospital firms in which the hospital firm had at least 50 percent ownership, several firms reported information for subsidiaries in which their ownership stake was less.

These data are supplemented by health care market and socioeconomic data for the same time period. The variables are summarized in Table 1 and are described briefly below.


Dependent Measure: Financial Performance

Financial performance of the business unit is measured with two common measures, return on assets (ROA) and total margin (MARGIN). MARGIN, an indicator of the firm's short-term performance, is defined as the business unit's net income -- the difference between total revenues and expenses -- divided by total revenues. Return on assets compares the net income for the fiscal year to the total assets used to generate it and is a better measure of long-term viability. Both measures are stated in percentage terms. The availability of only a single year of data does not permit adjustment of the return measures for risk.

Among the well-documented limitations of such measures based upon accounting data (Benston 1985; Fisher and McGowan 1983) are these: (1) accounting values are not the same as economic (market) values and (2) adjustments for the use of different accounting methods that lead to different reported values are difficult if not impossible. However, because the firms and business units studied are small and not-for-profit, and because their stock is not publicly traded, market performance data are not available. In addition, although not all of the problems with accounting data can be alleviated, the fact that the main businesses are confined to a single industry means that the differences in accounting methods should be relatively less than in studies of firms involved in multiple industries. Further, in combination with other data, accounting data are important for internal control decisions in the organization. These decisions include pricing, output, and product and managerial performance evaluation. Finally, as Grant, Jammine, and Thomas (1988) note, "The impact of corporate strategy on a firm's performance is more directly reflected in accounting profit than in stock price, which measures investors' expectations about future profits."

As is shown in Table 1, the means of both ROA and MARGIN were negative for the study subsidiaries in 1987. The mean of the total margin was -24.22 percent and the mean of the return on assets was -8.05 percent.(4) The larger size of most subsidiaries' assets than revenues yields smaller absolute ROAs than MARGINs.


Independent Measures: Internal Exchanges

Diversification: Definitions. Consistent with previous research, product and market diversification are defined relative to the main business of the firm (Alexander 1990; Pitts and Hopkins 1982). The first step in measuring diversification is to determine whether product or market diversification has occurred. Product diversification exists when the firm must use a different technology, including equipment, labor skills and the like, to produce the output. Since the great majority of hospitals provide acute inpatient care and, to a lesser extent, outpatient care, product diversification exists if the business unit does not deliver acute inpatient or outpatient care.(5) Examples of common product diversification activities of the subsidiaries are provided in Figure 1. They include physician billing, real estate development, management consulting, health plans (preferred provider or health maintenance organization plans), and wellness programs.

Product diversification is further classified as related or unrelated to the firm's main business. Products are related if they are not health care delivery services but are products or services with which employees are likely to have some experience. For example, management consulting, data processing, and hospital laundry services are related to health care delivery because they are typically a part of delivering acute inpatient care but are not the traditional final outputs of hospitals. Unrelated activities involve producing totally different outputs such as development or management of health plans, wellness programs, retirement communities, and real estate. Health plans, for instance, involve producing insurance products rather than acute care inpatient hospital services.

Firms diversify not only with respect to products but also with respect to markets (Capon et al. 1988). More specifically, customer characteristics differ when a new type of market is entered (Abell 1980). Previous research has shown that consumer and industrial markets differ (Day, MacMillan, and Hambrick 1983; Hambrick and Lei 1985) and that firms may have difficulty responding to the needs of various markets. Thus, as Capon et al. have discussed, it is useful to make additional distinctions within consumer markets because various groups of consumers probably differ greatly in their behavior, presenting different challenges for managers. For example, Capon et al. (1988) argue that customers for durable goods differ from customers for nondurable goods in the amount of information they need to make a purchase.

The customers for acute inpatient and outpatient care services have consisted of physicians and patients. Most of the latter are either beneficiaries of governmental payment programs or subscribers to private insurance. Both physicians and third party payers act as agents for the patient, the ultimate consumer. Therefore, the market that a business unit in this study faces is considered to be a new type when the consumers for the output differ from physicians or insured patients (Clement 1988). When this occurs, the firm cannot draw upon commonalities between a new and an old market (Ansoff 1965).

New markets are involved when, for example, the following services are offered: real estate, wellness programs, management consulting, and health plans. The customers for wellness programs and health plans are typically a diverse set of well people rather than ill patients. The clients for management consulting and real estate services are usually a variety of other firms or individuals, many of whom are not patients or physicians. Diversification has occurred since these customer groups have different needs and wants than individuals who are ill or their physician agents.(6)

Diversification: Measurement. Diversification is measured with both continuous and binary variables. Because of the way the data were collected, the business unit financial data cannot be consolidated to show the firm's overall performance or the proportion of the firm's resources derived from a particular unit. As a result, the importance of the unit's activities to the firm is determined relative to the assets of the main hospital business. Assets reflect the commitment of resources to the business unit.

As indicated in Table 1, the size of the business units relative to the hospital business is small. The mean of PERURA, the ratio of the assets of subsidiaries pursuing unrelated activities to hospital assets, is only 0.86 percent. Similarly, the means of PERRELA, PERNOA, and PERUMKTA, which indicate the investment in subsidiaries pursuing related product, no diversification, and unrelated market diversification, respectively, are near 1 percent.

Diversification is also measured with binary variables because of the small size of many of the business units. Since each binary variable measures the effect of the variable with respect to a reference, two different measures of product diversification are evaluated. The reference category for the first, DIV, which takes a value of one for related or unrelated diversification, is "not diversification." The reference category for the second, CLOSE, is unrelated diversification. CLOSE takes a value of one when the business unit's activity is part of the main business or is related to it. Market diversification is represented with a single binary variable (MKT) which is one when a new market type is involved.

In addition to being measured separately, product and market diversification are measured when the business unit both faces a new market and produces a new product. BOTHDIV is one when the unit engages in both product and market diversification. Results from continuous measures of this type of diversification did not differ and are not reported in this article. When the business unit pursues more than one activity and is classified as engaging in both related and unrelated diversification, it is eliminated from the study.(7) Only 12 units are eliminated for this reason.

The data in Table 1 show that the majority of subsidiaries studied did not offer products or services that differed much from inpatient acute care or outpatient care. Sixty-two percent of the reported activities were either part of the main business or closely related to it (CLOSE). Further, only 24 percent of the study units reported an activity involving both an unrelated product and unrelated market type (BOTHDIV).

Independent Measure: Exchange Processes

Duration of Exchange Relationship. The duration of the relationships between parent firms and business units is measured as length of time in months (LAM) that a business unit has been affiliated with a firm. The mean of LAM for the 1987 fiscal year was 43.41 months.(8)

Goal Incongruence. The tax status of the business unit may differ from that of the parent firm and main business. As discussed previously, the inconsistency in tax status is a proxy for possible goal incongruence of one type. Since subsidiaries of for-profit firms, all of which were for-profit, were not required to report all information for all variables in the study, this kind of goal incongruence occurs only between for-profit subsidiaries and not-for-profit parent firms. Goal incongruence is thus measured with a binary variable (NPFP) that reflects a for-profit subsidiary and a tax-exempt main business. As can be seen in Table 1, 61 percent of the subsidiaries of tax-exempt firms were taxable entities in 1987.

Independent Measures: External Exchanges

Competition and demand variables reflect the likelihood that the business unit will enter into exchanges with customers. Competition among health care delivery firms and demand for services are primarily local phenomena because services, unlike tangible products, must be custom-made and produced on-the-spot for consumers. Therefore, available data for the smallest geographic region representing the market area, either the county or the Metropolitan Statistical Area (MSA), are used to construct each of the variables. Data available at this level are relevant primarily for health care services. Additional data for the non-health care services are not readily available.(9)

The unemployment rate (URATE) in the MSA or county in which the firm is located is used to measure demand. In addition to reflecting the purchasing power of potential customers, the unemployment rate typically affects the demand for many health care and related services because most medical insurance is provided through employers (Friedman and Shortell 1988).

A Herfindahl index (HERFBEDS) measures competition among hospitals in the MSA or county in which the business unit is located (Zwanziger and Melnick 1988). It is defined as

|Mathematical Expression Omitted~

where i is the hospital; n is the number of hospitals in the county or MSA; and |p.sub.i~ is the hospital's proportion of total beds in the MSA or county. HERFBEDS increases as competition decreases. When HERFBEDS is one, there is only one hospital in the geographic region.(10)

A correlation matrix of the variables is presented in Table 2. Except for the alternative definitions of the diversification variables, the correlation coefficients are generally low.


The results from estimates of multiple regression models are presented in Table 3 and Table 4. As can be seen in Table 3, unrelated product (PERURA), unrelated market (PERUMKTA and MKT) and both unrelated product and unrelated market (BOTHDIV) diversification are associated with lower business unit total margins.(11) Similarly, product diversification (DIV) is associated with lower margins and the related, or not diversification variable (CLOSE) is associated with higher margins. As expected, then, subsidiary MARGIN is better where the unit can benefit most from resource exchanges with the main business. The diversification variables, however, are not significantly associated with ROA. Thus, there is some support for our expectations that unrelated diversification is associated with poorer financial performance with the short-term financial performance measure, MARGIN, but no support with the indicator of longer-term financial viability, ROA.

The results also provide some support for the expectations regarding exchange processes between the main business and subsidiaries. A longer length of affiliation is associated with both a higher total margin and return on assets. When more efficient exchanges of resources between the subsidiary and the hospital or parent firm occur, expenses are lower and revenues may increase -- the latter because the subsidiary can focus more on external rather than internal exchanges.

This study provides only limited evidence that goal incongruence between the parent firm and the subsidiary influences financial performance. NPFP is significantly negatively associated only with lower ROA and not with MARGIN. An alternative explanation of the negative relationship cannot be entirely ruled out, however. For-profit subsidiaries of nonprofit firms may have an incentive to minimize net income to avoid paying income taxes. Our data do not permit us to examine this possibility.

Finally, only one market variable is weakly significant. HERFBEDS, the competition measure is positively associated with ROA. Thus, when there are fewer competitors, ROA is higher. URATE, which is a broader measure than HERFBEDS of market conditions for all industries, is not statistically significant in any of the regression results.





The results of multivariate linear regression analyses reported here indicate that several variables are associated with financial performance of the subsidiaries of nonprofit acute care hospital firms operating in Virginia. More specifically, the strongest results indicate that business units that had existed longer and produced health care or related products were more profitable in 1987. These results are consistent with previous studies of businesses in other industries (Rumelt 1974; Palepu 1985; Biggadike 1979). In addition, better performance by nonprofit subsidiaries of nonprofit parents and those located in less competitive markets are also weakly supported by the results.

Although the results for length of affiliation of the subsidiary with the firm are consistent for both the short-term (margin) and long-term (return on assets) financial performance variables, the results concerning type of diversification and ownership type are not. Some of the difference in the diversification results may be attributable to (1) larger assets than revenues for most subsidiaries and (2) the larger amount of assets invested in diversified and unrelated units than in nondiversified or related units. A larger asset base improves the ROA performance relative to MARGIN for units reporting a negative net income (numerator), which is the case for the majority of these units. The effect would be larger for the diversified and unrelated units since they reported more assets than the other subsidiary types.(12) Thus, it is possible that the asset size difference led to a statistically insignificant result for ROA in contrast to the MARGIN results.

Nonprofit subsidiaries reported both more assets and a larger difference between assets and revenues than the for-profit subsidiaries.(13) The larger asset base may have improved the ROA performance of the nonprofit units relative to the MARGIN performance enough to produce a statistically significant negative association between ROA and NPFP even though the relationship between MARGIN and NPFP is not significant. Recall, however, that these are relative performance differences and that, in general, the financial performance of the units studied was poor.

Although this study was exploratory in nature and limited to hospital subsidiaries operating in one state and by the availability of a single year of performance data, the results are important because of the extent of restructuring by hospital firms. Alexander, Morlock, and Gifford (1988) reported that 34 percent of hospitals responding to a national survey had restructured by 1985 and that most had formed subsidiaries. Our data from Virginia indicate that 62.5 percent of all hospitals in the state had active nonparent, nonhospital subsidiaries that engaged in activities other than fund-raising.

However, further work is necessary to clarify the diversification-financial performance relationship. First, since the data for this study are cross-sectional and cover only surviving businesses, we cannot assess how much or how quickly performance improves over time. We also cannot assess the reliability and stability of the performance measures. Second, further work is needed to determine the effect of the subsidiaries on the entire firm. Although no information was available concerning the consolidated firm's performance in 1987, the very small size of the subsidiaries relative to the hospitals would indicate limited ability for meaningful financial subsidization of the main hospital business. The attention required by these relatively small and new businesses, however, may be disproportionate to their size and contribution to the firm's overall performance. Diversification, whether through subsidiary formation or otherwise, has been criticized for diverting management's attention away from the main business. On the other hand, some of the subsidiaries may improve the performance of the main hospital business, not by providing direct cash subsidies but by increasing patient flows.

In addition, flexibility in the corporate office-business unit relationship should be a focus of study. Shortell and his colleagues in their study of eight multihospital systems (Shortell, Morrison, and Friedman 1990), and Govindarajan and Fisher's (1990) study of large midwestern firms suggest that the relationship between corporate offices and business units should vary depending on the strategies that business units pursue. Further investigation will be required to determine whether different management structures for related and unrelated hospital firm business units contributes to financial performance.


We would like to thank Jeffrey Alexander, Ph.D., John R.C. Wheeler, Ph.D., and the anonymous reviewers for helpful comments on earlier drafts. Any errors or omissions remain the responsibility of the authors.


1. A strength of the proposed model is its parsimony. On the other hand, the model does not focus on some variables that could affect a business unit's financial performance. One such factor is management ability. This and other factors should be considered in future research.

2. Multihospital systems are counted as one firm.

3. Although other states, such as California, do require hospitals to report financial data, the data are limited to the hospital unit of the firm. Data from separately incorporated subsidiaries are not reported. The state of Florida has recently collected data on joint ventures. Our data base is broader, including wholly owned subsidiaries as well as joint ventures.

4. Extreme MARGIN and ROA outliers are deleted.

5. Including outpatient care in our definition of diversification does not influence our empirical results.

6. Since diversification is a relatively new area of study in health services research, we also evaluated the effect of defining "related diversification" as that related to health care delivery or related services and "unrelated diversification" as that not related to health care delivery. Although the two different classifications overlapped in many places, there were also several differences. For example, data processing is related diversification in the first typology. Since it is not health care delivery, however, it is unrelated diversification in the alternative typology. In contrast, most real estate services are unrelated in both. The results of the analyses with the alternative typology are not reported here because the variables are not statistically significant in any of the multivariate analyses.

7. The amount of resources devoted to the related and unrelated activities cannot be identified from the data.

8. An outlier business unit, created in 1915, is omitted from the analyses.

9. Recall, however, that the majority of the subsidiaries pursue activities that are health care related.

10. We also evaluated two other common measures of demand and competition, per capita income and physicians per 1,000 population. The results of the regression analyses do not differ when these measures were substituted for URATE and HERFBEDS.

11. Some observations had to be dropped from the analyses due to missing data. These observations, for the most part, do not differ from the observations included in the analyses. The only major difference is that the percentage of for-profit subsidiaries of nonprofit firms is somewhat higher in the missing group.

12. The difference in asset size for diversified or nondiversified subsidiaries compared to related or unrelated diversification is not statistically significant.

13. The difference in asset size is statistically significant (p |is less than~ .05).


Abell, D. F. Defining the Business: The Starting Point of Strategic Planning. Englewood Cliffs, NJ: Prentice Hall, 1980.

Alexander, J. A. "Diversification Behavior of Multihospital Systems: Patterns of Change." Hospital & Health Services Administration 35, no. 1 (1990): 83-102.

Alexander, J. A., L. L. Morlock, and B. D. Gifford. "The Effects of Corporate Restructuring on Hospital Policymaking." Health Services Research 23, no. 2 (1988): 311-37.

Ansoff, H. I. Corporate Strategy: An Analytic Approach to Business Policy for Growth and Expansion. New York: McGraw-Hill, 1965.

Barney, J. B. "Strategic Factor Markets: Expectations, Luck, and Business Strategy." Management Science 32, no. 10 (1986): 1231-41.

Bass, F. M., P. J. Cattin, and D. R. Wittink. "Market Structure and Industry Influence on Profitability." In Strategy + Structure = Performance. Edited by H. B. Thorelli. Bloomington: Indiana University Press, 1978.

Bentson, G. J. "The Validity of Profits-Structure Studies with Particular Reference to the FTC's Line of Business Data." American Economic Review 75, no. 1 (1985): 37-67.

Biggadike, E. R. Corporate Diversification: Entry, Strategy, and Performance. Cambridge, MA: Harvard University Press, 1979.

Bower, J. L. "Business Policy in the 1980s." Academy of Management Review 7 (1981): 630-38.

Capon, N., J. M. Hulbert, J. U. Farley, and L. E. Martin. "Corporate Diversity and Economic Performance: The Impact of Market Specialization." Strategic Management Journal 9, no. 10 (1988): 61-74.

Carter, J. R. "In Search of Synergy: A Structure-Performance Test." Review of Economics and Statistics 59 (May 1977): 279-89.

Christensen, H. K., and C. A. Montgomery. "Corporate Economic Performance: Diversification Strategy versus Market Structure." Strategic Management Journal 2, no. 4 (1981): 327-43.

Clement, J. P. "Does Hospital Diversification Improve Financial Outcomes?" Medical Care 25 (October 1987): 988-1001.

-----. "Vertical Integration and Diversification of Acute Care Hospitals: Conceptual Definitions." Hospital & Health Services Administration 33, no. 1 (1988): 99-110.

Day,D. L., I. C. MacMillan, and D. C. Hambrick. "An Empirical Test of the Inherent Differences between Consumer and Industrial Product Businesses." Working paper, Columbia University Graduate School of Business, 1983.

Dubofsky, P., and P. Varadarajan. "Diversification and Measures of Performance: Additional Empirical Evidence." Academy of Management Journal 30, no. 3 (1987): 597-608.

Ernst and Whinney. Corporate Reorganization: A Survey of Hospitals That Have Reorganized. Cleveland, OH: Ernst and Whinney, 1982.

Fisher, F. M., and J. J. McGowan. "On the Misuse of Accounting Rates of Return to Infer Monopoly Profits." American Economic Review 73, no. 1 (1983): 82-97.

Friedman, B., and S. Shortell. "The Financial Performance of Selected Investor-owned and Not-for-Profit System Hospitals before and after Medicare Prospective Payment." Health Services Research 23, no. 2 (1988): 237-67.

Gerber, L. Hospital Restructuring: Why, When, and How. Chicago: Pluribus Press, 1983.

Gort, M. Diversification and Integration in American Industry. Princeton, NJ: Princeton University Press, 1962.

Govindarajan, V., and J. Fisher. "Strategy, Control Systems, and Resource Sharing: Effects on Business-Unit Performance." Academy of Management Journal 33, no. 2 (1990): 259-85.

Grant, R. M. "On 'Dominant Logic', Relatedness, and the Link between Diversity and Performance." Strategic Management Journal 9 (1988): 639-42.

Grant, R. M., A. P. Jammine, and H. Thomas. "Diversity, Diversification, and Profitability among British Manufacturing Companies, 1972-1984." Academy of Management Journal 31, no. 4 (1988): 771-801.

Hambrick, D. C., and D. Lei. "Toward an Empirical Prioritization of Contingency Variables for Business Strategy." Academy of Management Journal 23, no. 4 (1985): 763-68.

Holzmann, O. J., R. M. Copeland, and J. Hayya. "Income Measures of Conglomerate Performance." Quarterly Review of Economics and Business 15, no. 3 (1975): 67-78.

Keats, B. W., and M. A. Hitt. "A Causal Model of Linkages among Environmental Dimensions, Macro Organizational Characteristics, and Performance." Academy of Management Journal 31, no. 3 (1988): 570-98.

Lubatkin, M., and R. C. Rogers. "Diversification, Systematic Risk, and Shareholder Return: A Capital Market Extension of Rumelt's 1974 Study." Academy of Management Journal 32, no. 2 (1989): 454-65.

Melicher, R. W., and D. F. Rush. "The Performance of Conglomerate Firms: Recent Risk and Return Experience." Journal of Finance 28 (1973): 381-88.

Michel, A., and I. Shaked. "Does Business Diversification Affect Performance?" Financial Management 13, no. 4 (1984): 18-24.

Mick, S. S., and D. A. Conrad. "The Decision to Integrate Vertically in Health Care Organizations." Hospital & Health Services Administration 33, no. 3 (1988): 345-60.

Montgomery, C. A. "Product-Market Diversification and Market Power." Academy of Management Journal 28, no. 4 (1985): 789-98.

Nathanson, D. A., and J. S. Cassano. "Organization, Diversity, and Performance." The Wharton Magazine 6, no. 4 (1982): 19-26.

Ouchi, W. G. "Markets, Bureaucracies, and Clans." Administrative Science Quarterly 25, no. 1 (1980): 129-41.

Palepu, K. "Diversification Strategy, Profit Performance and the Entropy Measure." Strategic Management Journal 6 (1985): 239-55.

Pitts, R. A., and H. D. Hopkins. "Firm Diversity: Conceptualization and Measurement." Academy of Management Review 7, no. 4 (1982): 620-29.

Prahalad, C. K., and R. A. Bettis. "The Dominant Logic: A New Linkage between Diversity and Performance." Strategic Management Journal 7 (1986): 485-501.

Ramanujam, V., and P. Varadarajan. "Research on Corporate Diversification: A Synthesis." Strategic Management Journal 10 (November-December 1989): 523-51.

Ravenscraft, D. J. "Structure-Profits Relationships at the Line of Business and Industry Levels." Review of Economics and Statistics 65, no. 1 (1983): 22-31.

Rhoades, S. A. "The Effect of Diversification on Industry Profit Performance in 241 Manufacturing Industries: 1963." Review of Economics and Statistics 55, no. 2 (1973): 146-55.

------. "A Further Evaluation of the Effect of Diversification on Industry Profit Performance." Review of Economics and Statistics 56, no. 4 (1974): 557-59.

Rumelt, R. P. "Diversification Strategy and Profitability." Strategic Management Journal 3, no. 4 (1982): 359-69.

------. Strategy, Structure and Economic Performance. Boston, MA: Harvard University Press, 1974.

Shortell, S. M., E. Morrison, and B. Friedman. Strategic Choices for America's Hospitals: Managing Change in Turbulent Times. San Francisco: Jossey-Bass, 1990.

Shortell, S. M., E. Morrison, and S. Hughes. "The Keys to Successful Diversification: Lessons from Leading Hospital Systems." Hospital & Health Services Administration 34, no. 4 (1989): 471-92.

Shortell, S. M., E. Morrison, S. Hughes, B. S. Friedman, and J. L. Vitek. "Diversification of Health Care Services: The Effects of Ownership, Environment, and Strategy." In Advances in Health Economics and Health Services Research. Vol. 7. Edited by R. M. Scheffler and L. F. Rossiter. Greenwich, CT: JAI Press, 1987.

Williamson, O. E. "The Economics of Organization: The Transaction Cost Approach." American Journal of Sociology 87 (1981): 548-77.

------. Markets and Hierarchies: Analysis and Antitrust Implications. New York: The Free Press, 1975.

Yavitz, B., and W. H. Newman. "What the Corporation Should Provide Its Business Units." Journal of Business Strategy 2 (1982): 14-19.

Zwanziger, J., and G. A. Melnick. "The Effects of Hospital Competition and the Medicare PPS Program on Hospital Cost Behavior in California." Journal of Health Economics 7 (1988): 301-20.

Address correspondence and requests for reprints to Jan P. Clement, Ph.D., Assistant Professor, Department of Health Administration, Medical College of Virginia Campus at Virginia Commonwealth University, Box 203, Richmond, VA 23298-0203. Thomas D'Aunno, Ph.D. is Associate Professor in the Department of Health Services Management and Policy, School of Public Health, University of Michigan, Ann Arbor; and Barbara Lou M. Poyzer, M.B.A. is a doctoral candidate in the Department of Health Administration, Medical College of Virginia Campus at Virginia Commonwealth University. This article, submitted to Health Services Research on January 14, 1991, was revised and accepted for publication on April 2, 1992.

@ @
COPYRIGHT 1993 Health Research and Educational Trust
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Clement, Jan P.; D'Aunno, Thomas; Poyzer, Barbara Lou M.
Publication:Health Services Research
Date:Feb 1, 1993
Previous Article:A national study of the efficiency of hospitals in urban markets.
Next Article:The relationship between adjusted hospital mortality and the results of peer review.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters