EXECUTIVE SUMMARY

CORPORATE

ENTREPRENEURSHIP AND

FINANCIAL PERFORMANCE:

THE CASE OF MANAGEMENT

LEVERAGED BUYOUTS

SHAKER A. ZAHRA Georgia State University

Leveraged buyouts (LBOs) have created much controversy in the literature, centering on their potential effect on a company's ability to innovate, engage in new ventures, and support entrepreneurial projects. Some believe that post-LBO debt reduces the financial resources available for entrepreneurial activities. Conversely, others argue that, despite the burden of debt, some LBOs provide executives with an opportunity to innovate and take risks

(Malone 1989). However, past studies have focused primarily on changes in R&D spending and ignored other corporate entrepreneurship (CE) activities afirm might pursue. Additionally, these studies have not documented the changes in CE after a management-led LBO. Thus, past studies offer only a snapshot of the effect of LBOs on CEo Finally, earlier studies did not directly examine the association between changes in entrepreneurial activities after the LBO and changes in performance.

To address these shortcomings, this study adds to the literature in three ways. First, it compares a company's pre- and post-LBO commitment to corporate entrepreneurship (CE), measured along two dimensions: innovation and venturing. Second, the study compares pre- and post-company performance levels. Third, the study shows how changes in a company's entrepreneurial activities impact performance. Therefore, the following hypotheses were tested:

HI: A company's commitment to CE activities increases after an LBO.

H2a: Company peiformance will be higher after an LBO than before the transaction.

H2b: Post-LBO company peiformance will surpass that ofthejinn's industry average.

Address correspondence to Shaker A. Zahra, Department of Management, College of Business Administration, Georgia State University, Atlanta, GA 30303.

The author acknowledges with appreciation the helpful comments of Donald Neubaum, Dean Schroeder, and two anonymous reviewers. M. Fescina, J. Barham, and J. Cavallaro conducted interviews with executives. The assistance of 1. Puri and R. Jayrarnan in collecting secondary data is also gratefully acknowledged.

Journal of Business Venturing 10, 225-247 © 1995 Elsevier Science Inc., 655 Avenue of the Americas, New York, NY 10010

0883-9026/95/$9.50 ssm 0883-9026 (94) 00024-0

226 S.A. ZAHRA

H3: Changes in post-LBO CE activities (venturing and innovations) are positively associated with company financial performance.

Data were collected from 47 LBO companies. Data were gatheredfrom secondary sources to augment and validate interview data. Data collected from two executives per firm were averaged and the mean was then used in all analyses. Data covered the three years immediately before and three years immedi­ately after the LBO. Companies in the sample were selected based on three criteria. First, at least 25% of the value of the LBO should have been paid by the firm's senior executives (vice presidents and higher). Second, the firm must have gone through a single LBO and must have remained private through the period of the study that eliminated reverse LBOsfrom consideration. Finally, the firm must have undergone its LBO at least three years before data collection. This provided an opportunity to observe changes in a company's CE and performance.

Multivariate analysis of covariance (MANCOVA), analysis of covariance (ANCOVA), and multiple regression were used to test the study's hypotheses. The results supported H1, showing that LBO companies increased their commitment to developing new products; placed greater emphasis on commer­cializing their technology; enhanced the quality and size of their R&D function; and intensified their new venture efforts. However, R&D spending did not change significantly after the LBO. The results also showed that post-LBO company performance was significantly higher than pre-LBO levels and the industry's average, thus supporting H2a and H2b. Finally, as stated in H3, changes in a company's CE activities were significantly associated with changes in post-LBO company performance.

Leveraged buyouts (LBOs) have created considerable controversy among scholars, executives and public policy-makers. Concern centers on the potential impact of an LBO on a company's willingness to innovate and engage in entrepreneurial ventures. To date, little empirical research has been devoted to documenting the changes that occur in a company's entrepreneurial behavior after an LBO and to linking these changes to company performance. Using data from 47 LBOs, this study's results lead to three conclusions. First, companies reported increases in their product development, technology-related alliances, R&D staff size and capabilities, and new business creation activities. Moreover, these firms did not change their R&D spending or the focus of R&D projects. Second, post-LBO company perfor­mance was higher than pre-LBO levels. Third, changes in corporate entrepreneurship activities after LBOs were significantly and positively associated with changes in company performance.

INTRODUCTION Today, U. S. corporations are undergoing fundamental change to enhance their ability to compete globally by increasing their entrepreneurial capacity and improving their productiv­ity. This restructuring requires changing the firm's product portfolio through innovation and venturing, promoting strategic renewal, and encouraging senior executives to take risks by increasing their ownership in their companies.

Of the various approaches to corporate restructuring and renewal, leveraged buyouts (LBOs) have been the most controversial (Zahra and Fescina 1991). In an LBO "the company's equity is bought up and removed from publicly traded security markets" (Fox and Marcus 1992, p. 62). LBOs represent about one-third of all restructuring transactions in the U.S. (Hall 1989). Some executives have financed their firms' LBOs, believing that they will be freer to initiate changes to improve their company's performance by investing in long-term projects such as R&D and new ventures.

The effect of an LBO on managerial commitment to innovation and venturing has been the subject of much debate. Because LBOs are heavily financed by debt, they may pressure executives to focus on short-term, tactical changes that generate cash to pay the principle and interest on the debt. However, Jensen (1986) suggests that increased debt also strengthens the power of LBO finance specialists in monitoring executives' actions. This vigilance en-

CORPORATE ENTREPRENEURSHIP AND FINANCIAL PERFORMANCE 227

hances managers' discipline by increasing their attention to long-term value creating activities. Heavy debt also forces executives to rid companies of weak or unrelated businesses, thus eliminating a major source of inefficiency and waste. Frequently, managers' ownership in their companies increases after the LBO, tying executives' wealth to the company's long-term performance. This increased ownership empowers and motivates executives to invest in long-term activities.

Fox and Marcus (1992) observe that the performance of LBOs has not been thoroughly examined, and the sources of value in LBOs are not well understood. To date, research has documented the effect of LBO announcements on shareholders' wealth. This research has not identified sources of real value after the LBO, especially increases in entrepreneurial activities. This study focuses on these changes, because LBOs may allow managers to take risks (Bowman and Singh 1993; Zahra and Fescina 1991).

Past research has examined the effect ofLBOs on R&D spending, ignoring other indica­tors of a company's entrepreneurship. Thus, there is a need for an examination of post-LBO entrepreneurial changes and their impact on company performance. These changes, which go beyond R&D spending, include product development, commercialization of new technolo­gies, and business creation. Therefore, this study responds to Low and MacMillan's (1988) call for research on sources of value in new organizational forms such as LBOs.

The study examines three questions: (1) how does a company's entrepreneurial activities change after an LBO; (2) does the company's performance change after an LBO; (3) are changes in a company's entrepreneurial activities associated with changes in its performance. In answering these questions the study focuses on corporate entrepreneurship (CE)-the sum of a company's innovation, renewal, and venturing efforts. Innovation involves creating and commercializing products and technologies, providing financial and human resources for innovative projects, and maintaining an appropriate infrastructure for innovation. Renewal means revitalizing a company's business through innovation and changing its competitive profile. Venturing requires creating and nurturing new business in current and new industries. By focusing on CE, the study documents LBOs' commitment to R&D and other entrepreneurial activities. Though past research has focused on R&D spending, other CE activities can profoundly affect a company's performance (Zahra 1993a, 1993b; Zahra and Covin 1993, 1994, 1995).

The remainder of the article consists of five sections. The first reviews the literature on the theoretical links among LBO, CE, and company performance. This review leads to three hypotheses. The second section introduces a study that tests these hypotheses. The t.ltird presents the study's analytical techniques. The fourth summarizes the study's results. The final section discusses the implications of the study's findings for theory and practice.

THEORY AND HYPOTHESES

LBO and CE Commitment

Theoretical explanations of LBOs' sources of value fall into three perspectives: tax savings, managerial opportunism, and value creation. Lowenstein (1985) suggests that post-LBO tax savings are the major source of value in these transactions. Because of the heavy debt involved and the high interest payment required, LBOs create tax shields that can be a major source of cash flows. However, tax savings cannot fully explain the value created by LBOs. Because the LBO's transaction cost may exceed the direct tax savings from going private, managers

228 S.A. ZAHRA

may lack the motivation to sponsor LBOs, especially when the tax savings are available to other third parties.

Still, managers may gain from making the firm private ifthey know that it is undervalued by the market. Accordingly, the premium paid to shareholders in an LBO will be less than the real value of the firm. However, there is no credible evidence that corporate assets are deliberately undervalued by management before the LBO (Palepu 1990). Further, the market for corporate control would reduce the likelihood of significant gains by executives after the LBO. Competition from other bidders can dilute most of these gains (Singh 1990). Neither the tax savings nor managerial opportunism can explain the major operational and strategic changes that follow an LBO.

The value creation perspective offers a compelling explanation of post-LBO changes. DeAngelo et al. (1984) suggest that reduction of costs associated with being a public corpora­tion (e.g., costs of regulatory compliance) can create value after the LBO. Unencumbered by bureaucracy and the demands of public disclosure, the LBO firm can devote its resources to creating value by supporting CEo

Post-LBO changes in agency and governance relations are another important source of value. By becoming owners, managers' and shareholders' interests are more aligned than found when the firm was publicly held. Because managers' wealth increases through the maximization of the firm's long-term value, they have an incentive to initiate strategic changes that eliminate waste and create value. After an LBO, subsidization of weak units is usually discontinued and excess cash can be used to create new business (Jensen 1989a), thereby promoting CEo

The heavy debt associated with the LBO also forces executives to pursue focused strategic choices. Many LBO companies divest unprofitable units that do not match the firm's strategic thrust (Jensen 1989b), which refines the firm's business definition and generates funds to retire the debt. This is consistent with Jensen's (1986) proposition that high LBO debt increases managerial discipline and erodes their power over the use of post-LBO cash flows. New projects must compete for external funding, further reducing managers' abuses of cash flows.

Agency theory suggests that managerial opportunism exists because of the diffused ownership of public corporations, leading to the free-rider phenomenon (Kosnik 1990). Because of their small percentage of shares of company stock, individual owners may not monitor managerial actions-giving executives an opportunity to control companies. Con­versely, in an LBO "managers are responsible to a small but powerful group of shareholders" (Gupta and Rosenthal 1991, p. 70). Because of their large investments and their small num­bers, LBO owners closely monitor executives-encouraging them to focus on value-creating activities while reducing managerial opportunism. An LBO encourages executives to max­imize firm cash flow to repay debt and interest; innovation and paying debt can thus go hand in hand.

LBOs can combine increased managerial monitoring with revamped executive compen­sation systems. After an LBO, executives can develop compensation contracts that are directly tied to company performance (DeAngelo et al. 1984). These contracts align the interests of shareholders and managers, thus reducing agency problems by curtailing managers' abuses of cash flows. Because executive compensation is tied to growth in company size, managers of public corporations may use free cash (cash in excess of the net present value of the firm's investments) to support marginal projects, those with zero or even negative net present values (Jensen 1986). Under an LBO, this practice vanishes because the executives' and shareholders' interests are aligned by adopting long-term, performance-based executive compensation.

CORPORATE ENTREPRENEURSHIP AND FINANCIAL PERFORMANCE 229

Post-LBO changes in corporate governance align the interests of owners and executives. Singh (1990) reports that after an LBO, there is a significant increase in the percentage of institutional owners and stockholding managers on the board, accompanied by a significant reduction in the percentage of specialists (e.g., lawyers) and prominent citizens among direc­tors. Thus, equity holders are better represented on the company's post-LBO board, giving them greater discretion in making decisions.

In summary, according to the value creation perspective, several changes follow an LBO. The first is the increased ownership by managers in their firms enhancing their power to undertake strategic changes. Coupled with changes in governance structures, this power also allows managers to rewrite employment contracts that tie compensation to the firm's long-term value, which motivates executives to focus on value-adding activities. With so much at stake, executives rid the firm of unprofitable businesses and change the company's competitive strategy and organizational structure. After an LBO, managers are forced to change their focus-from increasing earnings to maximizing cash flows (Easterwood et al. 1989) because of the high levels of debt the LBO firm must service. This is achieved by generating revenue from existing businesses or venturing into new fields. Innovation improves performance in existing businesses by creating (or modifying) products, processes, and orga­nizational systems. Venturing into new businesses can enhance a company's profit base. Collectively, these changes spur post-LBOs' commitment to CE (Malone 1989; Wright et al. 1992).

Consistent with the value creation perspective, Bull (1989b, p. 69) notes that a major source of "improvement in real performance is the introduction of the entrepreneurial dimen­sion." Bull concluded that" ... the evidence is convincing that management does change significantly after an LBO. The management change is interpreted as one of becoming entre­preneurial, subject to the severe constraint that a high level of debt must be serviced" (p. 89). Kaplan (1989) agrees, adding that debt servicing forces managers to become innovative to generate cash. If they succeed, managers become rich (p. 95). Grundfest (1989, p. 243) also states that after an LBO, "Managements are. . . motivated to adopt the kinds of efficiency enhancing measures that are rarely if ever implemented in publicly traded firms with diffuse ownership."

Three factors explain managers' increased commitment to CE after an LBO. First, after a management-financed LBO, escalation of commitment intensifies among managers. Incumbent managers (owners) will retain their commitment to ongoing CE activities because they possess private information about these projects or because they are professionally associated with ventures.

Second, agency theory proposes that post-LBO changes in ownership can reduce mis­alignments of interests between shareholders and executives. Ownership offers an incentive for retaining R&D and innovation projects that enhance the firm's value. When managers have a vested interest in the fate of their firms (through ownership), they tend to invest more heavily in innovative ventures (Hitt, Hoskisson, and Ireland 1990). The firm's value usually rises as a result of the announcement of new R&D projects or the introduction of new products (Zahra 1993c). Because their wealth is tied to their company's long-term value, LBO executives will emphasize CEo Profits from new products can also help in retiring debt (Jensen 1989a, 1989b).

Third, corporate control theory suggests that in large, public firms where the multidivi­sional structures prevail, companies emphasize financial reporting in measuring performance (Baysinger and Hoskisson 1990). This is achieved by using formal budgets and information systems. Specific, quantitative performance targets are established. To meet these short-term

230 S.A. ZAURA

targets, managers may sacrifice investments in projects that require long payback periods such as CE activities. After an LBO, strategic controls generally replace financial controls because owners will be intimately involved in managing the company and making key deci­sions. These controls encourage long-term investments in projects that influence the firm's value. Strategic controls encourage spending on innovation and CE activities (Hitt et al. 1990).

Consistent with the value creation perspective, research shows an increase in some post-LBOs' CE activities. In a study of 56 LBOs, Malone (1989) reported that 50% of the companies increased their new product introductions, 50% increased their customer base, and 41 % pursued new markets. Another study of 182 European firms (Wright et al. 1992), found that over 62 % of these firms have introduced new products that, they believed, they would not have introduced except for the LBO. Further, most companies modified their existing products, whereas others ceased to make some products. Most LBO companies revised their product portfolios without additional R&D investments. Finally, Zahra and Fescina (1991) reported that some LBO firms emphasized applied R&D projects to enhance new product introduction and commercialization. Other companies upgraded their R&D infrastructure and staff. All of these changes reflect managers' interests in creating cash flow to repay the debt by promoting CE and risk-taking.

The previous discussion suggests that after management-led LBOs, some firms increase CE because of changes in governance, ownership, and managerial power and incentives. Increased CE may also create a need to redeploy organizational resources by changing the product portfolio to service debt (Jensen 1989a). Heavy debt can increase a firm's appreciation of its external environment and markets, infusing a renewed sense of entrepreneurship in its operations by altering R&D priorities and emphasizing product development. Therefore:

H1: A company's commitment to CE activities increases after an LBO.

Post-LBO Company Performance

The value creation perspective suggests that company performance will improve after an LBO, primarily because of improved managerial incentives (Baker and Wruck 1989; Kaplan 1989; Smith 1990). Indeed, "research suggests that the primary source of ... [post-LBO] gains is new value created through significant operating performance improvements" (Palepu 1990, p. 249). These improvements result from strong managerial incentives after an LBO (Lichtenberg and Siegel 1990). LBOs increase managers' stake in the firm and its value, and also empower managers to initiate changes that improve performance. These managers invest in creating new business (Seth and Easterwood 1993), divesting unprofitable units (Bowman and Singh 1993), adopting new competitive strategies (Baker and Wruck 1989), and increasing their company's overall CEo In addition, the alignment of the interests of managers with the shareholders will reduce waste and enhance performance.

Managerial shirking also declines after an LBO. In its place, the executive owners have an incentive to build and develop the organization, initiate strategic changes, and reposition their company. They also have the incentive to take these risks because their own compensation (and wealth) are tied directly to company performance. Changes in governance structures (e.g., board composition) also enable bondholders to monitor managerial actions, thus reduc­ing agency costs (Jensen 1989a).

Further, post-LBO strategic changes can enhance a company's performance (Easterwood et al. 1989). These changes can alter a firm's business definition and competitive approach.

CORPORATE ENTREPRENEURSHIP AND FINANCIAL PERFORMANCE 231

Baker and Wruck (1989) suggest that changes in a company's definition of products and markets and corresponding competitive strategies frequently improve its post-LBO perfor­mance. Smith (1990) also reports that changes in the company's inventory, cash management policies, and plant reorganization are additional sources of post-LBO value.

Revamping the post-LBO company structure also improves performance. It reduces administrative overhead by removing layers of hierarchy, decreasing communication bottle­necks, speeding up the flow of information in the company, and quickening the pace of decision-making (Baker and Wruck 1989; Palepu 1990). Simplifying the firm's organizational structure instills a renewed sense of commitment to CE activities, another source of value creation.

Evidence of post-LBO company performance is mixed. A study of 83 reverse LBOs concluded that post-LBO sales and operating profits were not significantly higher than the industry average (Mohan 1990). However, Kaplan (1989) found that the average return to the post-buyout investors was about 40% higher than market return. Singh (1990) also documented significant increases in sales and operating income relative to their industries' averages. Smith (1990) observed increases in the companies' operating cash flow to operating assets; the sales-to-working capital; and the operating cash flow-to-employee. Bull (1989b) noted that post-LBO increases in the return on equity and the sales-to-asset ratio. Muscarella and Vetsuypens (1990) found that sales increased after an LBO. Opler (1992) reported in­creases in cash flow and operating profits after the LBO. Overall, theory and empirical research suggest that post-LBO improvements in managerial incentives can enhance post-LBO company performance, surpassing their pre-LBO levels and their respective industry aver­ages. Therefore:

H2a: Company performance will be higher after an LBO than before the transaction.

H2b: Post-LBO company performance will surpass that of the firm's industry average.

Three factors suggest a need to examine H2a and H2b. First, the studies cited earlier have reported contradictory conclusions on a company's post-LBO performance, highlighting a need for additional studies to reconcile past findings. Indeed, the bulk of the evidence on LBOs' performance is based on data from reverse LBOs, which are companies that underwent an initial public offering. These studies may suffer from an upward bias in terms of post-LBO performance (Seth and Easterwood 1993). Second, many of the studies have relied on the capital asset pricing model in estimating returns to investors from LBOs. This model primarily focuses on short time frames, and long-term returns are overlooked in these analyses. Addi­tionally, event studies do not fully capture post-LBO changes that create value, especially CEo Third, past studies have not explored the specific association between a company's CE efforts and its performance.

CE and Post-LBO Performance

A shortcoming of past LBO studies is lack of attention to the association between CE and performance. Instead, researchers examined operational changes as the source of post-LBO financial improvements (Bull 1989a; DeAngelo et al. 1984; Lichtenberg and Siegel 1990; Wright et al. 1992), often without directly documenting post-LBO CE changes and corre-

232 S.A. ZAHRA

sponding changes in performance. Only Long and Ravenscraft (1993) have examined the association between post-LBO changes in R&D and concurrent changes in company perfor- . mance. They have found LBOs that spent more on R&D had higher performance levels than those that did not spend as heavily on R&D.

Still, it is unclear if other post-LBO changes in CE activities are directly associated with company performance. This is a limitation of the literature because different CE activities are believed to improve company performance (Zahra 1991, 1993a, 1993b; Zahra and Covin 1994b). For instance, R&D spending enhances the company's ability to capture market share (Franko 1989). It also improves the company's productivity by increasing its process and product innovations, offsetting reductions in other resources by opening up markets for existing or new products. CE activities also increase the firm's market responsiveness by commercializing innovative products or by establishing strategic alliances with other firms. These alliances overcome shortages in slack resources and internal capabilities. Changes in CE are, therefore, expected to be positively associated with concurrent changes in the firm's post-LBO performance. This proposition does not imply causality between post-LBO CE and performance. Indeed, the reverse relationship is also plausible because successful LBOs may generate funds required for CE. Rather, this study focuses on the contemporaneous association between changes in post-LBO CE and company performance. Therefore:

H3: Changes in post-LBO CE activities (venturing and innovations) are positively associated with company financial performance.

In summary, the previous discussion shows that empirical evidence on the implications of LBOs for a company's CE and performance is needed. The study's hypotheses were tested, as described later.

METHOD

Sample and Data The lack of detailed secondary data on a company's post-LBO activities, especially those associated with CE, has been acknowledged in the literature (Fox and Marcus 1992). Conse­quently, data were collected through interviews with two senior executives in 47 LBO firms. Interview data were then validated with information from secondary sources. After the inter­views, contacts were made with some respondents to clarify and verify responses.

Initially, 125 manufacturing LBO firms, located in the southern and southeastern regions of the U.S., were contacted by phone and mail requesting their participation in the study. To qualify, a firm had to meet three criteria. (1) To ensure that managers are assuming risk, at least 25 % of the value of the LBO should have been paid by the firm's senior executives (defined as vice presidents and higher). When managers own 20% or more of the company's stock, they control its strategic agenda and determine its strategic choices (Amit et al. 1989).1 (2) The firm must have gone through a single LBO and must have remained private throughout the time of the study. This condition eliminated reverse LBOs. (3) The firm must have

'The 25% cutoff point was also based on past LBO research. Jensen (1989b) found that managers owned about 30% of their companies after the LBO. Bull (1989b) reported that management ownership after the LBO averaged 27% of a company's stock. Gupta and Rosenthal (1991) reported that ownership by managers averaged 16.9%. Kaplan (1989) reported that ownership by the management team had a median of 22.63 %. Thus, although a formal cutoff point does not exist, the 25 % ownership stake shows managers' serious commitment to the LBO firm.

CORPORATE ENTREPRENEURSHIP AND FINANCIAL PERFORMANCE 233

TABLE 1 An Overview of the Study's Measures, Sources, and Evidence of Validity

Measure Type" Data Sourceb Post-LBO Validation

Variables Pre Post Pre Post Sources r n

R&D spending 1 A&C I & A A, B, D, T, R & W 0.91 10 R&D focus 2 2 Prod development A, I, S, T & W A, B, S, T& W 0.86 9 Modified products I 1 A, I, S & W I&A B, S, T& W 0.85 10 Commercialization 2 2 I External R&D 2 2 B, S, R& W 0.78 11 R&D staff quality 2 2 R&D staff size A& I A &1 B,.R&W 0.89 16 New business reve- A, C & I D & I D, R, T&W 0.86 14

nue New bus. entered A, C & I D & I A,T&W 0.84 13 New segments A, C & I D&I A,T&W 0.81 12 Productivity A, C&D A, D&I B, D, T& W 0.86 17 Sales-to-assets A, C&D A&I A, D, T&W 0.86 12 ROI 1 1 A, C&D A &1 D,T&W 0.88 10 Tech opportunities 2 2 I I M&C 0.91 45 Size A&C D&I D 0.75 11 Age A, C&D A, D & I D Debt A&C A& I D 0.84 13

• Type: 1 = a three-year arithmetic average and 2 = overall index covering a three-year period. b A = Annual Reports; B = Business Week; C = COMPUSTAT; D = Dun's Million Dollar; I = Interviews; M = Commerce

Department Publications; R = Forbes; S = Funk & Scott; T = Fortune; & W = Wall Street Journal.

undergone its LBO at least three years before data collection, allowing observation of changes in CE activities and performance without introducing noise into the data because of significant changes in the external environment (Bull 1989a). Executives were encouraged to check their responses against corporate documents. Also, responses of the two executives from each company were averaged and then used in the analyses.

The average company in the sample had $789.1 million (SD = $567.4) in assets in 1992, was 46.7 years in age (SD = 14.3), and employed 6734 people (SD = 1457). T-tests compared the sample to nonparticipating companies; the two groups did not significantly differ in age, assets, or employees.

Measures

Data were collected for CE, company performance, and control variables. As shown in Table 1, data for some measures were collected from secondary sources; others were provided by executives. Data were collected for the three years before and the three years after the LBO. Because many activities attenuate performance during the year of the LBO (Bull 1989a), this year was ignored.

The Appendix presents the study's measures and their Cronbach's coefficient n. It also shows sources of data and evidence of the measures' validity. The validity was established by correlating interview and secondary data for the same measure. Because the Appendix presents the study's measures, this section only explains the rationale for each. Table 1 lists the sources of the measures, and the correlation between secondary and interview data-an indication of validity.

234 S.A. ZAHRA

As shown in Table 1, two types of measures were used. The first mossier used the three-year arithmetic average to gauge R&D spending, major product development, modifi­cations in existing products, R&D staff size, new business revenue, new business entered, new segments entered, productivity, sales-to-assets, ROA, size, age, and debt. The second used the two scale-based ratings provided by the executives: one for the three years pre-LBO and the other for the three years post-LBO. This approach covered R&D focus, external R&D, commercialization, R&D staff quality, and technological opportunities. These compro­mises were necessary because companies were reluctant to provide annual figures for each variable.

Corporate Entrepreneurship

As the Appendix shows, CE was defined as having two dimensions: innovation and venturing (Guth and Ginsberg 1990; Zahra 1993b), using the following multiple indicators.

Innovation: Data were collected for: (1) R&D spending, (2) R&D focus, (3) radical product innovation, (4) product modification, (5) commercialization, (6) using external R&D sources, (7) improving R&D staff quality, and (8) increasing R&D staff size.

The previous eight measures were included in the study because of their relevance to CEo (1) R&D spending is a major determinant of product and process innovations (Franko 1989)-a major indicator of CE (Zahra 1993a, 1993b). (2) R&D focus (type), including basic, applied, and developmental R&D, also determines a company's innovation and CE (Zahra and Das 1993). Basic R&D covers projects that push the frontiers of science and create new knowledge. Applied projects focus on transforming scientific concepts into products. Developmental projects involve the creation of prototypes and models. (3) Product develop­ment is necessary for venturing (Guth and Ginsberg 1990). A new product is defined as one that was never offered before by the responding company (Zahra 1993c). (4) Modifying existing products helps to meet changing customer needs and increases the value derived from past R&D investments (Porter 1985). (5) Commercialization reflects a company's proac­tiveness, a major CE indicator (Miller 1983). It shows a company's emphasis on the speedy introduction of products to the market (Zahra, Nash, and Bickford 1995). (6) External R&D­related activities indicate a company's focus on external sources of innovation to replace or complement its internal skills. This measure also shows a willingness to take risks. (7) R&D staff quality influences a company's ability to create and commercialize new products. (8) Staff size influences the effectiveness of a company's innovation.

Venturing: This CE component was measured along three dimensions, all measured by the three-year arithmetic average. The first was the percent of revenue from new business or industries which showed a company's ability to expand operations to achieve profitability (Block and MacMillan 1993). The second was the number of new businesses the company has entered that showed an increase in the emphasis on redefining the company's business concept. The third was the number of new market segments served by the company that gauged the increase in the scope of operations (Porter 1985).

Company Performance: Annual data were collected on four indicators, each measured as the three-year arithmetic average. The first was employee productivity, which is a key performance measure (Lichtenberg and Siegel 1990). The second was sales-to-beginning assets, which showed the company's ability to use its assets effectively (Bull 1989b). The

CORPORATE ENTREPRENEURSHIP AND FINANCIAL PERFORMANCE 235

third was return on investment (ROI) , which is a key measure of performance (Kaplan 1983), showing how well the company was being managed. 2 The fourth indicator was the earnings before interest and tax (EBIT)-to-assets ratio. It showed a company's ability to generate a profit.

Unadjusted and adjusted performance data were used in the analyses. Data were adjusted by dividing a firm's data by its industry's average, weighted by its share in different industries. Next, factor analysis was performed on the performance measures, using the average of the three-years' pre-and post-LBO, separately. Each analysis uncovered two factors. The first, named profitability, consisted of sales-to-assets, earnings after interest and tax, and ROI (eigenvalues = 1.45 and 1.27, for pre- and post-LBO). The second had the single item "productivity" (eigenvalues = 1.03 and 1.07, for pre- and post-LBO).

Control Variables: Four variables were included as controls-technological opportuni­ties, size, age, and level of debt-because they were expected to influence the association between CE and performance.

Technological Opportunities: This variable refers to the extent an industry offers oppor­tunities for innovation. Mature and declining industries provide few opportunities for R&D, whereas growing industries typically offer many such opportunities. There is a strong, positive association between an industry's technological opportunities and firm-level R&D investment (Baysinger and Hoskisson 1990). High technological opportunities are expected to promote CEo Still, although some consider technological opportunities important for CE (Fox and Marcus 1992), others disagree (Hall 1989) . These conflicting perspectives suggest that techno­logical opportunities should be considered when examining CEo

Company Size: Research has examined the association between a company's size and CE (Kamien and Schwartz 1982). Though the effect of company size remains a controversial issue, smaller companies are believed to be more innovative than their larger counterparts (Scherer 1980). Smaller companies remain close to their markets and become aware of opportunities quickly. Also, these companies' simple structures allow them to respond quickly to changing markets. Smaller companies also need innovative products and technologies to survive, which forces them to support CEo However, some small companies may lack financial resources for CE activities.

Several countervailing forces may influence large companies' post-LBO commitment to CEo Although larger firms' bureaucratic and highly formalized organizational structures inhibit innovation, these firms typically have slack resources for CEo Research shows a positive, significant association between company size and R&D spending (Hoskisson and Johnson 1992). There are also some innovations that only the largest companies can support. Further, as mentioned, because large companies emphasize financial controls, R&D and CE spending is already rationed and only the most essential projects are supported. Consistent with

2 As mentioned in the Appendix, the data for ROI were adjusted by withholding interest payments. Similarly, data for the EBIT-to-assets ratio were adjusted, by withholding both interest and tax payments. Two separate runs of the analyses reported in the article were performed. The first included productivity, sales-to-assets, ROI (unadjusted for interest), and EBIT-to-assets (unadjusted for interest and taxes). The second run included the four variables, after making the relevant adjustments. The analyses yielded similar results. Therefore, the results using adjusted data were reported because they indicated that post-LBO performance reflected real improvements, and were not the result of accounting changes. Finally, additional analyses were conducted using the cash flow-to-sales and cash flow-to-assets measures for a subset of 19 companies that provided the data. The results were also consistent with those reported later in the article.

236 SA ZAHRA

this point, Long and Ravenscraft (1993) found that post-LBO larger companies experienced smaller reductions in R&D compared to smaller firms. Overall, a negative association is expected between company size and post-LBO CE activities.

Company Age: Older companies are believed to be less innovative than their younger rivals (Acs and Audretsch 1988). Older companies have highly formalized structures that discourage innovation. They also tend to focus on modifying their existing products and technologies, using marketing as a means of establishing their position. Moreover, as compa­nies become mature, they venture into new industries by acquiring other companies. As diversification escalates, resources for internal innovation decline (Hitt et al. 1990). Thus, a negative association is expected between company age and CE.

Level of Debt (Leverage): There are two views on the effect of debt on CEo The first suggests that debt weakens the organization, forcing it to emphasize short-term investments and reduce CEo Long and Ravenscraft (1993) highlight four reasons for this prediction. The first is the loss of tax credit associated with CE activities after the LBO. The second is managers' preoccupation with operational and financial changes after the LBO, leaving little time for CEo The third is because outside finance sources typically do not have detailed information about the ongoing CE projects, they are unwilling to provide funds for such purposes. Finally, some CE activities may require specialized assets that are typically firm­specific. Because debtors cannot dispose of these assets quickly, they may not support CEo Accordingly, high debt ratio is expected to be negatively associated with post-LBO CEo

The second view holds that debt disciplines the company and its managers, causing them to carefully weigh the benefits and risks of CE. Therefore, only CE projects with a positive net present value will receive support (Jensen 1986, 1989a). CE investments are thus brought into closer alignment with the norms of the external capital market. In an LBO, debtors are also owners who closely oversee managers' decisions and have access to detailed information on CE projects, which allows them to exercise strategic controls. Therefore, CE does not automatically decline because of debt. Rather, as mentioned, because managers are pressured to create long-term value, post-LBO CE activities may increase.

Clearly, the exact effect of debt on CE is an empirical issue. Debt can destroy or create value, depending on its amount and terms. When debt far exceeds the industry's norms or the company's ability to repay it, it can stifle CEo Thus, significant increases in post-LBO debt compared to the industry norm are expected to be negatively associated with post-LBO changes in CE.

ANALYSIS Before performing statistical tests, all measures were standardized (mean = 0, SD = 1). Factor analysis was used to identify five CE dimensions whose reliability was established using Cronbach n. HI was then tested using multivariate analysis of covariance (MAN­COY A). After the significant MANCOV A results, univariate analyses of covariance (ANCO­VAs) were also conducted to identify significant changes in CE dimensions from pre- to post-LBO levels. MANCOVA, with repeated measures, and ANCOVAs determined ifpost­LBO financial performance was significantly higher than before the LBO (H2a and H2b). Finally, multiple regression analysis tested H3 to determine if changes in post-LBO CE activities were positively associated with concurrent changes in post-LBO performance.

CORPORATE ENTREPRENEURSHIP AND FINANCIAL PERFORMANCE 237

TABLE 2 Results of Factor Analysis of Corporate Innovation Measures

Factors Pre-LBO' Post-LBO'

2 3 4 2 3 4

R&D spending 0.89 0.23 0.28 0.17 0.88 0.21 0.25 0.11 Basic R&D 0.71 0.29 -0.34 0.27 0.76 0.18 -0.30 0.19 Applied R&D -0.63 0.21 -0.26 0.25 -0.67 0.17 0.29 0.28 # of new products 0.25 0.88 0.11 0.27 0.21 0.77 0.28 0.23 # of modified products 0.27 0.75 0.28 0.26 0.29 0.83 0.22 0.20 Commercialization 0.19 0.20 0.79 -0.18 0.23 0.26 0.74 -0.25 External R&D -0.32 0.16 0.58 0.23 0.27 0.23 0.53 0.21 R&D staff size 0.28 0.12 -0.33 0.54 0.33 0.22 -0.21 0.51 Quality R&D staff 0.19 0.14 0.11 0.66 0.25 0.29 0.09 0.73

Eigenvalue 2.08 1.60 1.48 1.11 2.21 1.63 1.24 1.11 % of variance explained 29.13 21.15 13.45 11.26 30.25 19.95 14.15 10.97 a 0.77 0.65 0.72 0.73 0.74 0.65 0.71 0.77

• Factor labels: I = R&D spending; 2 = product development; 3 = commercialization; 4 = R&D quality and capability.

RESULTS

Factor Analysis of CE Initially, the measures of innovation and the renewal component of CE were factor-analyzed. Two analyses were conducted: one for innovation and the second for venturing.

Innovation

To reduce the overlap among the study's measures of innovation, factor analysis with an orthogonal rotation was performed twice: once for the pre-LBO data, and the other for the post-LBO period. Each analysis yielded four factors (eigenvalues> 1.0), as shown in Table 2. For each period, indices were created by multiplying loadings by their respective raw scores. Items with absolute loadings of 0.50 or higher were used in developing these indices.

Factor i: intensity of basic R&D spending. This factor reflected a company's investment in R&D activities, especially basic research.

Factor 2: Product development. This factor showed a company's willingness to modify its existing products or create new products that meet the changing needs of its markets.

Factor 3: Commercialization and external linkages. This factor reflected a com­pany's commitment to creating conditions conducive to the timely commercializa­tion of its products and technology. It also showed a company's willingness to engage in strategic alliances and licensing agreements to enhance its innovation efforts and technological base.

Factor 4: R&D quality and capability. This factor showed a company's commit­ment to retaining a large and productive R&D unit.

Venturing

A second factor analysis helped to determine the distinctiveness of three measures of ventur­ing. Two runs were performed, resulting in one significant factor (pre-LBO: eigenvalue =

238 S.A. ZAHRA

TABLE 3 Intercorrelations Among the Study's Variables: Pre-LBO.

Variables

Variables 2 3 456 7 8 9 10 11 12 13

Pre-LBO: 1 R&D spending 2 Commercialization .31' -3 Product development .28" .23" -4 R&D staff quality .29" .27" .30" -5 New business development .18 -.10 -.14 -.23" -6 Profitability: unadjusted .14 .16 .18 .11 .16 -7 Productivity: unadjusted .16 .14 .13 .19 .09 .33" -8 Profitability: adjusted .11 .12 .17 .14 .19 .59b .3D" 9 Productivity: adjusted .13.09 .15 .17 .13 .22" .26" .21 -

10 Size (log) .15 - .26" .23" .25" .08 .28" .19 .11 .18 11 Age -.20 -.17 -.19 .25" .09 .13.13 .09 .12 .26" 12 Technological opportunities .33" .21 .27" .31" .11 .19 .08 .16.11 .08 -.19 -13 Debt-to-assets -.36" .26" .19 .18 -.16-.31" .17 -.26" .21 .30" .23"-.16-

Post-LBO: 1 R&D spending 2 Commercialization .35" -3 Product development .29" .29" -4 R&D staff quality .34" .26" .32" -5 New business development .24" -.11 .12 - .16 -6 Profitability: unadjusted .18 .16 .26" .15 .22 -7 Productivity: unadjusted .19 .14 .20 8 Profitability: adjusted .17 .15 .25" 9 Productivity: adjusted .17 .15 .20

10 Size (log) .12 - .24" .22 11 Age -.18 -.19 -.18 12 Technological opportunities .32" .22 .33" 13 Debt-to-assets - .38" .27" - .23"

.16 .03 .30" -

.13 .21 .62b .28"

.18 -.04 .27" .25"

.23"-.09 .21 .16

.19 .04 .17 .09

.38b .19 .18 .09

.26" - .22 - .23" .21

Mean (standardized): pre-LBO .73 -.17 .45 - .36 .12 .31 .21 SD: pre-LBO .12 .28 .15 .14 .22 .26 .17

Mean (standardized): post-LBO .89 1.34 1.67 2.56 1.37 .74 .58 SD: post-LBO .29 .47 .49 .67 .38 .41 .30

'p< .05. bp < .01.

.21 -

.16 .14

.11 .09 .26"

.18 .16 .12 -.29" --.19 .18 .35b .30"-.19-

.66 .68 .43 .17 .05 .12

.54 .21 .26 .21 .09 .15

1.59 1.17 .45 .29 .13 .19 .41 .39 .26 .16 .20 .18

1.61, a = 0.72 and post-LBO: eigenvalue = 1.76; a = 0.76). Factor scores were used in subsequent analyses.

Intercorrelations among the variables were examined. Table 3 presents the means, stan­dard deviations, and intercorrelations for the pre- and post-LBO periods, respectively.

Comparisons of Pre- and Post-LBOs CE (HI)

MANCOVA, with repeated measures, was used to determine ifCE activities changed signifi­cantly after the LBO. MANCOV A was significant (Wilks' A = 0.45; p < .(01). Next, ANCOV A helped to identify the variables that contributed to the MANCOV A differences, as reported in Table 4. Four CE variables increased significantly after the LBO: commercial­ization, product development, R&D quality and capability, and business venturing. However, consistent with some past research (Hall 1989; Lichtenberg 1989), the R&D spending factor

CORPORATE ENTREPRENEURSHIP AND FINANCIAL PERFORMANCE 239

TABLE 4 A Comparison of Pre- and Post-LBOs' CE Performance

Pre-LBO Post-LBO

Mean SD Mean SD

CE Activities· R&D spending 0.73 0.12 0.89 0.29 Commercialization -0.17 0.28 1.34 0.47 Product development 0.45 0.15 1.67 0.49 R&D quality and capability -0.36 0.14 2.56 0.67 New business 0.12 0.22 1.37 0.38

Unadjustedb

Profitability 0.29 0.26 0.74 0.41 Productivity 0.21 0.17 0.58 0.30

Adjustedc

Profitability 0.66 0.54 1.59 0.42 Productivity 0.69 0.21 1.11 0.35

'Wilks' y = 0.47···. 'Wilks' y = 0.31···. 'Wilks' Y = 0.26··· . • p < .05 . •• p < .01. *.'p < .001.

Univariate ANOV As

0.86 6.78*** 2.81 * 4.06** 4.01 **

3.73** 3.91 ***

3.79*** 1.98*

did not change significantly after the LBO. Consistent with HI, CE increased significantly after LBOs.

Pre- Versus Post-LBO Performance (H2a and H2b)

H2a and H2b posited that companies' post-LBO performance will be significantly higher than their pre-LBO levels and higher than the industry's average. MANCOV A and ANCOV As tested this hypothesis, using unadjusted data (H2a) and adjusted data (H2b). The results are presented in Table 4.

MANCOV A showed that performance was significantly higher after the LBO (Wilks' A = 0.31;p< .001). ANCOVA showed that the post-LBO company profitability and produc­tivity were significantly higher than pre-LBO levels, supporting H2a. Next, after adjusting the performance measures for the pre- and post-LBO periods, MANCOV A and ANCOV As were used. MANCOVA results were significant (Wilks' A = 0.26; p < .001). Compared to their industries' average, LBO companies' performance on profitability and productivity improved after the LBO, supporting H2b.

CE and Company Performance (H3)

According to H3, changes in CE will be positively associated with concurrent changes in post-LBO performance. Therefore, change scores in CE and performance were calculated as follows:

Unadjusted change in a variable = [(the variable's post-LBO value - the variable's pre-LBO value)/the variable's pre-LBO value] x 100

Adjusted change in a variable = ([(the variable's post-LBO value - the variable's

pre-LBO value)/the variable's pre-LBO value]/[(Industry average of the variable' post-LBO value - Industry average for the variable's pre-LBO]] x 100

240 S.A. ZAHRA

TABLE 5 Changes in CE and Performance Improvements: Regression Results

Dependent Variables Unadjusted Data Industry-Adjusted Data

Profitability Productivity Profitability Productivity

Intercept 2.23b .91" 1.69- 1.01'

Organizational size -.12 .15 -.17 .13 Organizational age 0.08 0.14 0.08 0.12 Technological opportunity 0.25- 0.11 0.33b 0.16-Debt-to-assets ratio 0.36b 0.32b -0.10 -0.07

R&D spending 0.12 0.15 0.26- 0.23-Commercialization 0.49b 0.58< 0.61< O.4Ob

Product development 0.52b 0.46b 0.55< 0.14 R&D quality and capabilities 0.33- 0.24- O.4Ob 0.33-Net business 0.66< 0.11 0.78< 0.15

R2 0.38 0.20 0.39 0.24 F 7.88< 3.93< 9.02< 2.04-

'p < .05. bp < .01. 'p < .001.

Because multiple regression was used to test H3, the intercorrelations among the study's variables were first examined. The tests suggested by Neter, Wasserman, and Kutner (1989) suggested the absence of serious multicolinearity. The four control variables were entered first, followed by CE, thereby gauging the effect of CE after controlling for other factors. Table 5 presents the results.

All regression equations in Table 5 are significant, suggesting that post-LBO changes in CE activities are significantly associated with concurrent changes in company performance. Because of the short time frame involved, the results in Table 5 are best interpreted as implying a significant correlation between contemporaneous changes in post-LBO CE and company performance.

The results also showed that increases in commercialization and R&D quality and capa­bilities are positively associated with changes in company performance in the four equations. With the exception of adjusted profitability, product development is also significant in three of the four regressions. Business creation is also associated with changes in profitability, but not in productivity. R&D spending is significantly associated with industry-adjusted changes in profitability and productivity. Of the 20 I3s relating to CE, 15 are positive and significant, thereby supporting H3. Changes in post-LBO CE activities are associated with, or accompanied by, concurrent changes in company performance. 3

3 Additional analyses were conducted to understand the associations between CE and performance. First, pre-LBO CE was used to "predict" post-LBO performance, and vice versa. The results suggested that CE was a stronger "predictor" of performance, rather than the other way around. Second, changes in CE and performance were considered. Changes in CE were used to "predict" changes in performance, and vice versa. This analysis, too, suggested that CE changes were a stronger "predictor" of performance changes, rather than the other way around. Third, lagged regressions of CE and performance changes were conducted. One-, two-, and three-year lags were attempted, indicating that CE was a stronger "predictor" of performance, rather than the other way around. The results did not change when debt was used as a control variable. Still, it is unwise to suggest causation between CE and performance because of the study'S short-term time frame, and managers' motives for supporting CE are unclear. Similarly, post-LBO changes in companies' strategies, governance, and structures are not incorporated directly in the analyses.

CORPORATE ENTREPRENEURSHIP AND FINANCIAL PERFORMANCE 241

Focusing on control variables, changes in company age are not significant in the four regressions, perhaps because of the short time frame of the study. Changes in company size are also insignificant. Because the average company size declined after the LBO, the results indicate that performance improvements are not at the expense of short-term asset reductions (Kaplan 1989; Smith 1990). Further, as predicted, technological opportunities are significant and positive in three of the four regression equations; the exception is unadjusted productivity. Finally, the debt -to-asset ratio is positively and significantly associated with the unadjusted profitability and productivity. However, though the debt ratio has negative ~s, it lacks signifi­cance with changes in adjusted measures of profitability and productivity. These results signal the increased (but insignificant) rise in the post-LBO debt ratio. It is also possible that companies are diverting some of their profits to service the debt, hence the negative coefficient.

DISCUSSION LBOs are a controversial means of corporate restructuring and renewal. Despite wide interest in the topic, the sources of value in LBOs are not well documented. Low and MacMillan (1988) call for research on the sources of value in new organizational forms. This study has responded to this call, focusing on post-LBO changes in a company's CE activities and performance.

Hypothesis 1

The results (Table 4) support HI, which suggests that a company's commitment to CE increases after an LBO. A comparison of pre- and post-LBO CE levels suggests that companies have increased their emphasis on commercialization and R&D-related external alliances, enhanced the quality and capabilities of the R&D function, and intensified their venturing activities. These results, which are consistent with past research (Malone 1989; Wright et al. 1992), may reflect reduced agency and transaction costs (DeAngelo et al. 1984), improved managerial incentives because of ownership, and increased managerial discretion in initiating strategic changes (Easterwood et al. 1989). Changes in CE may also reflect the pressures of LBO debt, which force managers to explore innovative ways to quickly generate cash to repay debt and interest (Baker and Wruck 1989). It should be noted that sample companies were slightly below their industries in the debt-to-asset ratio before the LBO. In the year immediately after the LBO, these companies had a significantly higher debt ratio than their industries (t = 2.03; p < .05). By year 3, their debt ratio was in line with the industry average (t = 1.09; p = .23). LBOs might have forced companies to make effective use of debt without overburdening them.

Long (1989) suggests that, after the LBO, companies reduce their R&D staff and capabili­ties. The results do not support this proposition and, instead, show that companies increase their R&D units' size and capabilities. Table 4 also shows a decline (but statistically insignifi­cant) in the R&D spending factor-a composite of R&D spending as a percent of sales, commitment to basic R&D, and emphasis on applied R&D. The results support some past research (Hall 1989; Lichtenberg 1989; Lichtenberg and Siegel 1990) , but appear to contradict the National Science Foundation's (1987) finding that R&D spending declines after LBOs. It appears that, over the span of this study, companies in the sample did not significantly reduce their R&D spending. Still, because past research has examined R&D spending as a percent of company sales, a t-test was run on this variable alone. No significant differences were found between pre- and post-LBO periods, supporting the results of the overall R&D

242 S.A. ZAHRA

factor. This suggests that post-LBO CE improvements have occurred without increasing R&D spending.

Hypothesis 2 H2a posited that post-LBO company performance will exceed their pre-LBO performance levels. The data in Table 4 support this hypothesis, indicating a significant improvement in both the profitability and productivity measures. This finding supports Smith's (1990) conclu­sion that LBO firms have higher performance levels than before the transaction. The results are also consistent with those of Lichtenberg and Siegel (1990) who reported significant improvements in post-LBO plant productivity.

Adjusting a company's performance relative to its industry, LBO companies had higher gains in their profitability and productivity than the average firm in their industries. This finding supports H2b and is consistent with past research (Singh 1990). Caution is necessary in interpreting the current results because performance is a complex construct and several of its dimensions are not examined here. Similarly, the long-term (five years or longer) effects of LBOs on performance are not documented.

Hypothesis 3

MANCOVA results show that post-LBO changes in CE are associated with, or accompanied by, concurrent changes in company performance. As mentioned, these results do not imply that CE changes cause performance changes. The results indicate that post-LBO changes in the two sets of variables are positively associated after the LBO. Past research has failed to link these variables directly. Thus, although researchers speak in general terms of increased post-LBO R&D and risk-taking, this study shows a positive correlation between a company's post-LBO's CE activities and concurrent changes in performance. Finally, the current results add to the accumulating evidence on the positive association between CE and short-term company performance (Zahra 1991, 1993b; Zahra and Covin 1995).

This study's results should be interpreted with caution. Even though the results suggest that short-term improvements occur in CE and company performance after the LBO, longer time frames are necessary to establish any long-term changes. There is the possibility that senior executives might have overestimated post-LBO changes for social desirability reasons. However, the data in Table 1 suggest that absence of bias. This is consistent with Palepu (1990). Caution is also necessary because observed changes in both CE and performance may be the result of actions taken by the firm before the LBO.

The causes of post-LBO changes in CE found in this study are also unclear. Were they made to retire the debt quickly? Were they a part of a major strategic reorientation? Answers to these questions must await future research in order to clarify the sources of post-LBO changes and also establish cause-effect relationships. Additionally, this study's relatively small sample size-which is typical of LBO studies (Fox and Marcus 1992)-suggests that generalizations to all LBOs are inappropriate. Finally, executives owned a large portion of the LBOs examined in the study, possibly influencing the current results. It is unwise to generalize the results to LBOs that did not benefit from increased managerial ownership and corresponding changes in incentives and governance structures.

There are many future research opportunities on the link between CE and company performance and the nature of these relationships among LBOs. For example, the current results should be validated using data from other LBO companies, with different levels of

CORPORATE ENTREPRENEURSHIP AND FINANCIAL PERFORMANCE 243

managerial ownership, to ensure the generalizability of the findings. Replications may also use other indicators of CE and performance such as cash flow. Researchers need also to clarify managers' motivations in revamping CE activities. How do these changes relate to a company's strategy? Does the focus of these changes vary over time-away from paying debt and into reconfiguring the firm's business concept? Answers to these and similar questions can help in validating the propositions of the value creation perspective on LBOs.

Clearly, concern over LBO financing has overshadowed scholarly research on the impli­cations of these transactions for corporate entrepreneurship. The time has come to assess the impact of LBOs on a company's ability to innovate, take risks, and initiate significant venturing activities. It is hoped that the current results will encourage future research on the consequences of LBOs for corporate entrepreneurship and financial performance.

REFERENCES

Acs, Z., and Audretsch, D.B. 1988. Innovation in large and small firms: An empirical analysis. American Economic Review 78:678-690.

Amit, R., Livant, J., and Zarowin, P. 1989. The mode of corporate diversification: internal ventures versus acquisitions. Managerial & Decision Economics 10(2):89-100.

Baker, G.P., and Wruck, K.H. 1989. Organizational changes and value creation in leveraged buyouts: the case of the O.M. Scott & Sons Company. Journal of Financial Economics 25(2):163-190.

Baysinger, B., and Hoskisson, R.E. 1990. The composition of boards of directors and strategic control: effects on corporate strategy. Academy of Management Review 15(1):72-87.

Block, Z., and MacMillan, I. 1993. Corporate Venturing: Creating New Business Within the Firm. Cambridge, MA: Harvard Business School Press.

Bowman, E.H., and Singh, H. 1993. Corporate restructuring: reconfiguring the firm. Strategic Manage­ment Journal 14(special issue):5-14.

Bull, I. 1989a. Financial performance ofleveraged buyouts: an empirical analysis. Journal of Business Venturing 4(4):263-279.

Bull, I. 1989b. Management performance in leveraged buyouts: an empirical analysis. In Y. Amihud, ed., Leveraged Management Buyouts: Causes and Consequences. Homewood, IL: Dow Jones­Irwin, pp. 69-94.

DeAngelo, H., DeAngelo, L., and Rice, E. 1984. Going private: the effect of a change in corporate ownership structure. Midland Corporate Finance Journal 2:35-84.

Easterwood, J.C., Seth, A., and Singer, R.F. 1989. The impact of LBOs on strategic directions. California Management Review 23(1):30-43.

Fox, I., and Marcus, A. 1992. The causes and consequences of leveraged management buyouts. Academy of Management Review 17:62-85.

Franko, L. 1989. Global corporate competition: who's winning, who's losing, and the R&D factor as one reason why. Strategic Management Journal 10:449-474.

Grundfest, J .A. 1989. Management buyouts and leveraged buyouts: are the critics right? In Y. Amihud, ed., Leveraged Management Buyouts: Causes and Consequences. Homewood, IL: Dow Jones­Irwin, pp. 241-261.

Gupta, A., and Rosenthal, L. 1991. Ownership structure, leverage, and firm value: the case ofleveraged recapitalizations. Financial Management 20(3):69-83.

Guth, W., and Ginsberg, A. 1990. Guest editor's introduction: corporate entrepreneurship. Strategic Management Journal l1(special issue):5-16.

Hall, B. 1989. The Impact of Corporate Restructuring on Industrial Research and Development. U niver­sity of California, Berkeley: The National Bureau of Economic Research.

Hitt, M.A., Hoskisson, R.E., and Ireland, R.D. 1990. Mergers and acquisitions and managerial commit­ment to innovation in M-form firms. Strategic Management Journal 11(s):29-47.

244 S.A. ZAHRA

Hoskisson, R.E., and Johnson, R.A. 1992. Corporate restructuring and strategic change: the effect on diversification strategy and R&D intensity. Strategic Management Journal 13:625-634.

Jensen, M. 1986. The takeover controversy: analysis and evidence. Midland Corporate Finance Journal 4(2):6-32.

Jensen, M.C. 1989a. Eclipse of the public corporation. Harvard Business Review 67(5):61-74. Jensen, M. C. 1989b. LBOs and the reemergence of institutional monitoring of managers. In Y. Amihud,

ed., Leveraged Management Buyouts: Causes and Consequences. Homewood, IL: Dow Jones­Irwin, pp. 263-268.

Kamien, M., and Schwartz, N. 1982. Market Structure and Innovation. Cambridge, U.K.: Cambridge University Press.

Kaplan, R.S. 1983. Measuring manufacturing performance: a challenge for managerial accounting research. The Accounting Review 58(4): 686-705.

Kaplan, S.N. 1989. Sources of value in management buyouts. In Y. Amihud, ed., Leveraged Manage­ment Buyouts: Causes and Consequences. Homewood, IL: Dow Jones-Irwin, pp. 95-101.

Kosnik, R.D. 1990. Effects of board demography and directors' incentives on corporate greenmail decisions. Academy of Management Journal 33: 129-150.

Lichtenberg, F. 1989. Statement before the House Committee on Science, Space, and Technology, Subcommittee on Science, Research and Technology, Washington, DC.

Lichtenberg, F.R., and Siegel, D. 1990. The effects of leveraged buyouts on productivity and related aspects of firm behavior. Journal of Financial Economics 27: 165-194.

Long, W.F. 1989. Statement Before the House Committee on Science, Space, and Technology, Wash­ington, DC.

Long, W.F., and Ravenscraft, D.J. 1993. LBOs, debt and R&D intensity. Strategic Management Journal 14:119-135.

Low, M.B., and MacMillan, I. C. 1988. Entrepreneurship: past research and future challenges . Journal of Management 14:139-161.

Lowenstein, L. 1985. Management buyouts. Columbia Law Review 85:730-784. Malone, S. 1989. Characteristics of smaller company leveraged buyouts. Journal of Business Venturing

4:349-359.

Miller, D. 1983. Entrepreneurship correlates in three types of firms. Management Science 29:770-791. Mohan, N. 1990. Do LBOs sustain efficiency gains? Akron Business & Economic Review 21(3):91-99. Muscarella, C.J., and Vetsuypens, M.R. 1990. Efficiency and organizational structure: a study of

reverse LBOs. Journal of Finance 45:1389-1413. National Science Foundation. 1987. Instructions for Survey of Industrial Research and Development.

Washington, DC: US Department of Commerce. Neter, J., Wasserman, W., and Kutner, M. 1989. Applied Linear Regression Models, 2nd ed. Home­

wood, IL: Irwin. Opler, T. 1992. Operating performance in leveraged buyouts: evidence from 1985-1989. Financial

Management 21(1):27-34. Palepu, K. 1990. Consequences of leveraged buyouts. Journal of Financial Economics 27:247-262. Porter, M.E. 1985. Competitive Advantage. New York, NY: Free Press.

Scherer, F.M. 1980. Industrial Market Structure and Economic Performance, 2nd ed. Chicago, IL: Rand McNally College Publishing Co.

Seth, A., and Easterwood, J. 1993. Strategic redirection in large management buyouts: the evidence from post-buyout restructuring activity. Strategic Management Journal 14(4):251-273.

Singh, H. 1990. Management buyouts: distinguishing characteristics and operating changes prior to public offering. Strategic Management Journal11(s):I11-131.

Smith, A. 1990. The effects of leveraged buyouts. Business Economics 25: 19-25. Wright, M., Thompson, S., and Robbie, K. 1992. Venture capital and management-led, leveraged

buy-outs: a European perspective. Journal of Business Venturing 7(1):47-71. Zahra, S. 1991. Predictors and financial outcomes of corporate entrepreneurship: an exploratory study.

Journal of Business Venturing 6(4):259-285.

CORPORATE ENTREPRENEURSHIP AND FINANCIAL PERFORMANCE 245

Zahra, S. 1993a. A conceptual model of entrepreneurship as firm behavior: a critique and extension. Entrepreneurship: Theory and Practice 14(4):5-2l.

Zahra, S. 1993b. Environment, corporate entrepreneurship and financial performance: a taxonomic approach. Journal of Business Venturing 8:319-340.

Zahra, S. 1993c. New product innovation in established companies: associations with industry and strategy variables. Entrepreneurship: Theory and Practice 18(2):47-69.

Zahra, S., and Covin, J. 1993. Business strategy, technology policy and firm performance. Strategic Management Journal 14:451-478.

Zahra, S., and Covin, J. 1994. The financial implications of fit between innovation types and sources and competitive strategy. Journal of High Technology Management Research 5(2): 183-212.

Zahra, S., and Covin, J. 1995. Contextual influences on the corporate entrepreneurship-company performance relationship in established firms: a longitudinal analysis. Journal of Business Ventur­ing 10(1): 43-58.

Zahra, S. , and Das, S. 1993. Innovation strategy and financial performance in manufacturing companies: an empirical analysis. Production and Operations Management 2(1):15-37.

Zahra, S., and Fescina, M. 1991. Will leveraged buyouts kill U.S. corporate research and development. Academy of Management Executive 5(4):7-2l.

Zahra, S., Nash, S., and Bickford, D. 1995. Creating a competitive advantage from technological pioneering. Academy of Management Executive (in press).

APPENDIX This Appendix presents the study's measures and sources of data. Table 1 also summarizes the source of measures, type of data collected, interrater reliability, and the sources of validation data.

CORPORATE ENTREPRENEURSIDP Both innovation and venturing were measured in this study.

Innovation

Data were collected for eight measures of company innovation, as follows:

1. R&D Spending was measured by spending in company-sponsored R&D, as a percentage of company sales. Pre-and post LBO means were: 1.89 and 1.73%, respectively.

2. R&D Focus (Type). Both pre- and post-LBO data were collected from interviews. Execu­tives distributed 100 points among three R&D types to show their company's emphasis on each. A high score showed high emphasis. Executives provided two overall scores: one for the 3-year period before the LBO and the other covered the 3-year period after the LBO.

Focus (Type)

Basic R&D refers to those projects that aim to push the frontiers of science forward by creating new knowledge.

Applied projects focus on transforming scientific concepts into products and goods.

Developmental projects focus on the creations of prototypes and models.

Total

Points

100

246 S.A. ZAHRA

Because none of the firms allotted any points to the developmental category, analyses empha­sized basic and applied R&D. Pre-and-post LBO means for basic R&D were 37 % and 17 %, and 63% and 83% for applied R&D.

3. Product Development (Major). Executives provided the number of new products intro­duced to the market. Pre-and- post LBO means were 4.67 and 7.03, respectively.

4. Modifying Existing Products. Executives provided data for the pre- and post-LBO; means were 7.21 and 10.02, respectively.

5. Commercialization. Pre- and post-LBO commercialization were each measured by a three­item index. Executives rated their overall company's situation twice: once for pre-LBO and the other for the post-LBO. Each item followed a five-point scale (l = very low to 5 = very high). Items were: "emphasis on commercializing new technologies," "emphasis on being among the first companies to introduce new products to the market," and "empha­sis on introducing products to the market faster than the competition." Pre-and-post LBO means were 2.2 and 3.4, respectively.

6. External R&D-Related Activities. Using a four-item index, with a five-point scale (l = very low to 5 = very high), executives rated the items twice-for the pre-and-post LBO. Items were: "participation in R&D-related joint ventures," "emphasis on collaborating with universities/research centers in R&D," "number of R&D-related joint ventures with your suppliers," and "emphasis on acquiring innovations for other companies." Pre-and-post LBO means were 2.1 and 3.2, respectively.

7. R&D Staff Quality. Executives rated three items twice, using a five-point scale (1 = very low to 5 = very high). Items were: "productivity of your R&D group," "quality of your R&D staff," "spending on R&D staff training and development," and "quality of your R&D staff compared to that of your key rivals." Pre-and-post LBO means were 2.4 and 2.9.

8. R&D Staff Size. The average number of scientists and engineers who work on R&D related functions was used as the measure. Pre-and-post LBO means were .13 and .21, respectively.

Venturing Executives provided annual data for the three years pre-and-post LBO, as follows:

1. Percent of revenue generated from industries that the company entered this year. The three-year average was 7.2% and 11.3% for pre-and-post LBO, respectively.

2. Number of new businesses the company has entered this year. The three-year average was 1.3 and 2.3 for pre-and-post LBO, respectively.

3. New market segments served. The three-year average pre-and-post LBO figure for this variable was 1 and 2.04, respectively.

COMWANYPERFORMANCE Four performance measures were used:

1. Employee Productivity was calculated annually as the firm's total sales divided by the number of full-time employees. Pre-and-post LBO means were $67,112 and $83,007, respectively.

CORPORATE ENTREPRENEURSHIP AND FINANCIAL PERFORMANCE 247

2. Sales-to-Beginning Assets Ratio was defined as the company's sales divided by beginning assets (Bull 1989b), showing the company's ability to create value from using its assets. It averaged 1.02 and 2.47, for the pre-and-post LBO periods.

3. Return on Investment (ROI) was defined as the ratio of earnings divided by total assets. Pre-and-post LBO means were 3.47% and 7.06%. To ensure valid comparisons, interest was withheld from the numerator of ROI, for both the pre-and-post LBO periods. Means were 2.91 % and 6.11 %, respectively.

4. Earnings before interest and tax (EBIT)-to-assets ratio was also used. It showed the company's ability generate a profit. Pre-and-post LBO means were 1.04 and 2.81, respec­tively. However, to ensure that performance improvements reflected real gains in profit­ability, rather than being the result of accounting changes, this ratio was recalculated­by withholding interest and taxes from earnings. The ratios for pre-and-post LBO were 93 and 1. 97, respectively.

CONTROL VARIABLES Data were collected on the following four control variables:

1. Technological Opportunities. Executives rated their industries twice, using a three-item index (1 = very low to 5 = very high). Items were: "rate of innovation in your industry," "opportunities for product innovation in your industry," and "opportunities for technologi­cal innovation in your industry." Companies averaged 2.81 and 2.85 for the pre-and-post LBO periods, respectively.

2. Company Size. For each year, the log of company assets measured company size; it averaged 2.13 and 1.65, for pre-and-post LBO, respectively.

3. Company age was measured by the number of years the firm has been in existence. 4. Debt was defined as the ratio of total debt to total assets (Baysinger & Hoskisson 1990).

Pre- and post-LBO three-year averages were 43.8% and 68.1 %.