Long Term Performance of Divesting Firms and the Effect of Managerial Ownership. Robert C. Hanson

Long Term Performance of Divesting Firms and the Effect of Managerial Ownership Robert C. Hanson Department of Finance and CIS College of Business Eastern Michigan University Ypsilanti, MI 48197 Moon H. Song Department of Finance College of Business Administration San Diego State University San Diego, CA 92182-8236 First Draft: January 8, 1996 Current Draft: March 12, 1997

Long Term Performance of Divesting Firms Abstract Shareholders benefit when firms divest assets, but where do the gains come from? We find that the long term performance of the parent firm improves when assets are sold to another firm, but not when assets are sold to unit management. Our evidence suggests that gains from interfirm divestitures arise from within the parent firm as negative synergies are eliminated. Divestitures to unit managers are different. Gains from these divestitures do not arise from removing negative synergies, but likely arise from more efficient contracting. Our evidence shows, however, that divestitures to unit management generate normal returns, suggesting that the parent firm fails to capture a significant part of these gains.

Long Term Performance of Divesting Firms and the Effect of Managerial Ownership 1. Introduction Corporations that divested assets during the 1980s produced significant gains for their shareholders. The source of these gains has drawn the attention of researchers in recent years but remains an open empirical issue. Early studies mainly analyzed the wealth effects of divestitures and suggested, but did not specifically analyze motives for asset sales. 1 Recent cross-sectional studies of divestitures have analyzed specific sources of gains such as improving corporate focus, undoing failed acquisitions, or reducing the costs of asymmetric information. In this paper we pursue this issue by investigating whether shareholder gains are created within the divesting firm or are bargained away from the buying firm. Accumulated research on asset sales shows that divesting assets benefits the sellers' shareholders whether the buyer is another corporation (Rosenfeld 1984; Jain 1985; Hite, Owers and Rogers 1987) or division management (Hite and Vetsuypens 1989). In addition, research shows that returns to corporate buyers are generally positive (Jain 1985; Hite, Owers, and Rogers 1987; Rosenfeld 1984; Sicherman and Pettway 1987), suggesting that both buyer and seller share in the gains from asset sales. More recent studies have focused on cross-sectional variations in returns. Kaplan and Weisbach (1992) study the success of acquisitions and show that nearly half of their sample of acquisitions were later divested, and about a third these are classified as unsuccessful. They find positive returns at the announcement of the selloff, but returns are significantly higher for divestitures they classify as unsuccessful, suggesting that the divested assets impeded the firms other operations. Lang, Poulsen, and Stulz (1995) argue that divestitures are used to raise capital and that gains arise by reducing information asymmetries and the costs of managerial discretion. They show that divestiture announcement returns depend on the use of the proceeds, and that positive returns occur when proceeds are paid 1 See, for example, Alexander, Benson, and Kampmeyer (1984) or Hite, Owers and Rodgers (1987).

2 out to creditors or shareholders. When proceeds are retained, announcement returns are insignificantly negative. John and Ofek (1995) look to increasing corporate focus as the source of gain. They show that returns are higher, and the firm experiences increased operating performance, when divestitures lead to a more focused firm. This research extends the cross-sectional analysis of divestitures by examining whether asset sales precipitated improved long term performance. We suggest that gains from divestitures originate in the selling firm and arise from removing operations that cause negative synergies. An alternative, but not mutually exclusive, hypothesis suggests that gains are transferred from the buyer. Under this view, gains to divesting firms merely reflect a one-time bargained for gain when assets are transferred to a buyer with complementary resources or a more efficient contracting scheme. Examining long term performance allows us to distinguish between these alternatives because only by removing negative synergies should we expect improved long term performance. We find that on average long term performance as measured by industry adjusted Tobin's q increases in each of the three years following the divestiture. Moreover, significant increases in long term performance occurred only when assets were divested to another firm. Evidence suggests that interfirm divestitures increases the parent firm's Tobin's q from below to above industry average by the second year following the divestiture. When assets were sold to unit management, however, the parent firm showed no increase in industry adjusted Tobin's q. Our evidence suggests that removing negative synergies motivates interfirm divestitures, and that gains arise from the improved performance that follows the divestiture. Divestitures to unit managers are different. Our evidence supports the notion that removing negative synergies does not motivate divestitures to unit managers. These divestitures seem motivated by the prospect of capturing gains associated with more efficient contracting. Announcement period abnormal returns, however, are indistinguishable from zero, suggesting that the parent firms fails to capture a significant portion of these gains.

3 Additional results show a strong direct relation between managerial ownership and changes in long term performance associated with interfirm divestitures. Our results agree with the notion that higher levels of ownership gives managers incentive to sell assets that create negative synergies, and that ownership provides managers negotiating power for the hard bargaining that is necessary to transfer gains that arise from the buyer's complementary resources. Managerial ownership, however, plays no significant role in explaining changes in long term performance for interfirm divestitures. Again, our evidence suggests that divestitures to unit managers differ from interfirm divestitures. In section 2 we review hypothesized sources of gains to divestitures and testable hypotheses. Section 3 describes the sample and empirical methods. We present results in section 4, and offer a summary and conclusion in section 5. 2. Source of Gains and Hypotheses Theory suggests that asset sales create value for essentially three reasons. First, asset sales improve the operational efficiency of the parent firm by eliminating negative synergies. Second, asset sales are an efficient means of raising capital for firms suffering from severe costs associated with asymmetric information or managerial discretion. Third, assets are sold to another firm that is willing to pay more for the assets than their value to the parent firm. In the first two instances, higher values originate within the selling firm itself, while in the third case value originates in the buying firm and is transferred to the seller. In order to determine the source of shareholder wealth gains when firms sell assets we examine the post divestiture performance of these firms. Our main hypothesis is that gains from divesting assets arise primarily from within the selling firm because the asset sale removes negative synergies. These gains reflect anticipated improvements in long term performance. We further suggest that selling assets merely to capture a relatively high

4 price that another firm or management team is willing to pay will not in itself improve the selling firm's long term performance. In the paragraphs that follow, we tie the theorized source of gains to testable hypotheses about long term performance.

5 2.1 Sources of Gains An important source of gain in divestitures is the improvement in long term performance that ensues when negative synergies are eliminated. 2 Negative synergies arise, for example, when assets unrelated to core operations prevent the firm from focusing on core competencies. Other examples include assets from a failed acquisition (Kaplan and Weisbach 1992) or operations that chronically lose money and that increasingly draw valuable resources from other parts of the firm. Here, as asset sales remove the negative synergies, gains arise from improvements in other parts of the firm. 3 These synergy gains are not transferred from the buyer through the bargaining process. This leads to our main hypothesis that long term performance will improve after synergy motivated divestitures. 2 See Hite, Owers, and Rogers (1987) and more recently John and Ofek (1995). 3 We are using a vary narrow definition of synergy gain. In this view, the total gain from selling the asset consists of the improvement in the selling firm's other operations, which we call the synergy gain, plus the increment in price over the value of the assets to the selling firm that the buyer is willing to pay. For example, say that the assets are worth $100 to the selling firm, but also reduce the value of other operations by $20. A buyer is willing to pay $115 for the assets. Then the total gain would be $35 = $20 (synergy gain) + $15 (price premium).

6 A second endogenous source of gain is when the divestiture reduces costs related to asymmetric information and managerial discretion. Lang, Poulsen, Stulz (1995) suggest that managers value firm size and control, and sell assets to raise capital when these costs make an equity offering too expensive. Shareholders benefit because the announcement reduces information asymmetries and resolves uncertainty about the true value of the assets, they receive additional gains from reduced agency costs when proceeds are used to pay down debt or are distributed to shareholders. Reducing costs of asymmetric information, unlike removing negative synergies, produces a one-time gain and no necessary improvement in operating performance. In the extreme case where managers retain funds from asset sales and use them to pursue their own objectives, long term performance might actually suffer. Thus, long term performance will not improve, or may even deteriorate, when divestitures are used to raise capital. Gains could also originate from the buyer. Hite, Owers, and Rogers (1987), among others, develop the efficient deployment argument which suggests that divestitures move assets to higher valued uses and shareholders capture some of the gains through effective bargaining. The buyer creates gains from combining the divested assets with complementary resources, or by managing the assets with a more efficient organizational structure or a more efficient contracting scheme. 4 The efficient deployment argument suggests that the selling firm merely extracts part of the gains generated by the buyer, and the argument offers no suggestions about improvements in the selling firm's long term performance. Thus, our second hypothesis suggests that selling assets to the highest bidder will not necessarily lead to improvements in long term performance. 4 We discuss asset sales to unit managers in more detail below.

7 2.2 Sales to Unit Management The nature of the gains to the selling firm will depend on whether the buyer is another firm or unit management. Interfirm asset sales create value by combining sold-off assets with complementary resources, while asset sales to unit managers create value through more efficient contracting. We suggest that assets that will most benefit from complementary resources are likely the same assets that create negative synergies in the divesting firm. On the other hand, assets or operations that would benefit from a more efficient contracting scheme are unlikely the same assets that produce negative synergies. Research on leverage buyouts (LBO) and management buyouts (MBO) shows that firms with strong cash flow and few growth opportunities make the best buyout candidates (Jensen 1989, Palepu 1990), and that gains come from more efficient contracting and reduced agency costs (Kaplan 1989). Assets divested to managers will likely have traits similar to LBO/MBO firms. Assets that produce strong cash flow but suffer from agency problems typically do not produce negative synergies, but instead may merely suffer from parent firm neglect or an overly bureaucratic organizational structure (Baker and Wruck, 1989). In other words, unit managers will buy the assets they oversee when these assets will benefit from a more efficient organizational form, but unit managers are unlikely to buy assets that produce negative synergies, such as chronic losing operations with poor prospects 5. Therefore, we conjecture that increases in long term performance are an unlikely source of gain when assets are divested to unit management. the lemons. 5 In the context of Akerlof's (1970) lemons story, better informed unit managers would not bid for

8 2.3 Ownership Structure and Gains Stock ownership provides managers an incentive to sell assets that cause negative synergies, and the incentive to improve operations once these assets are gone. 6 Stock ownership will, in addition, give managers the power and the incentive to bargain away gains created by the buyer. This notion parallels the arguments of Stulz (1988) and Stulz, Walkling, and Song (1990) who show that target firm managerial stock ownership influences the bid premium and division of gains between buyer and seller in takeovers. Thus, stock ownership by parent firm managers will have a positive influence on the division of gains created by the buyer under the efficient deployment and efficient contracting arguments. 6 In a related issue, Tehranian, Travlos, and Waegelein (1987) show that sell-off returns are significantly positive when long-term compensation plans are present, while returns are insignificantly negative without long-term plans. Long-term compensation plans provide incentives similar to stock ownership.

9 3. Data and Methods 3.1 Sample We collect our sample from Mergers and Acquisitions and select transactions classified as divestitures, have a value listed, and were also announced in the Wall Street Journal. We identify from Mergers and Acquisitions whether the assets were sold to a group that included division or parent firm management or sold to another firm. The initial sample consists of 151 divestitures to management and 634 divestitures to interfirm buyers announced from July 1980 to June 1991. From Q-File7 proxy statements for the year of the divestiture we collect information about stock ownership by the Chief Executive Officer (CEO), by officers and directors as a group, and outside blockholders. We found proxy data for 114 divestitures to management and 401 divestitures to interfirm buyers. 3.2 Long Term Performance We use Tobin's q, the firm's market value divided by its replacement cost, as our measure of performance. 7 For our purpose of measuring the actual change in long term performance Tobin's q has some advantages over using excess stock returns or accounting statement-based measures of profitability as measures of performance. Tobin's q captures more than just the unexpected gains to shareholders. It captures the performance of the whole firm, an important consideration since asset sales will affect bondholders and preferred stockholders along with shareholders. Standard event study excess returns will not provide accurate measures of performance when there are wealth transfers between security holders and risk shifts. For example, a wealth transfer from bondholders to stockholders will show up as positive excess returns even though no value was created. Even if excess returns accurately measure gains, they do not identify the source of the gains, whether from market inefficiencies or improved operations. Accounting statement-based measures of performance suffer 7 Other studies using Tobin's q to measure performance include Lindenberg and Ross (1981), Morck, Shliefer, and Vishny (1988), Lang, Stulz, and Walkling (1989), Servaes (1990).

10 from their arbitrary nature and the discretion that managers have in determining these values. In addition, Tobin's q will capture changes in performance that result from both immediate improvements in operations and cash flow, and long-term improvements such as an expanding investment opportunity set and better growth opportunities that are unlikely to generate immediate increases in cash flow. We calculate an approximate Tobin's q as ( S + P + D - NWC ) q = TA where S equals the market value of equity, P equals the liquidating value of preferred stock, D equals the book value of long term debt, NWC represents net working capital, and TA equals the book value of total assets. 8 We calculate an industry q-ratio as the average q-ratio of all companies within a four-digit SIC code. If there are less than three companies in the four-digit SIC, we use a three-digit SIC code to define the industry. Missing financial data when calculating q using Compustat7 files reduces the sample size to 225 divestitures of which 177 were interfirm divestitures and 48 were divestitures to unit managers. 8 Chung and Pruitt (1994) report that this proxy for Tobin's q explains at least 96.6% of the variability in q values calculated using the more elaborate method of Lindenberg and Ross (1981).

11 4. Empirical Results 4.1 Sample Description Table 1 presents descriptive statistics for our sample of divestitures. Divested assets were valued at $216.0 million on average, at $213.7 million when the buyer was another corporation, and at $224.7 million when unit management participated in the purchase. The difference between the two groups is not significant. The mean values of the transactions in our sample are nearly four times larger than that reported by Hite and Vetsuypens (1989) and about twice the value reported by Lang, Poulsen, and Stulz (1995). Different time period and different sample selection procedures account for the size differences. Although the value of the transaction is about the same between buyers, the relative size differs. Divestitures to unit managers are twice as large in relative terms since parent equity is about half that of firms selling assets to another company. Table 1 also shows that neither Tobin's q, nor leverage (the ratio of long term debt to market value of equity plus preferred stock), nor managerial stock ownership differs between buyers. Officers and directors as a group, for example, own 8.58% of the firms stock when assets are divested to another firm, and they own an insignificantly lower 7.86% when assets go to unit management. These levels of ownership are lower than that reported by Lang, Poulsen, and Stulz (1995), but not unexpectedly so given the larger firms in our sample. We find, consistent with previous studies, that divestiture announcements produce significantly positive excess returns on average (0.674%, t=2.25), and when assets are sold to other firms (0.836%, t=2.37). We find, in contrast to some earlier studies, that divestitures to unit management generate normal returns (0.075%; t=0.14). 9 9 Hite and Vetsuypens (1989), for example, report significant positive returns surrounding

12 4.2 Long Term Performance We use Tobin's q as a measure of performance and examine changes in q for the three years following the divestiture. A divestiture will not necessarily increase performance, since the sell off could simply replaces physical assets with financial assets. But performance could improve, as described by the synergy argument, if divesting assets also exorcises negative synergies. Table 2 panel A reports Tobin's q and changes in q for the full sample of 225 divestitures. In each of the three years following the divestiture a Wilcoxon signed ranks test shows that median q is significantly higher than it was the year before the divestiture. Over the four year period starting the year before the divestiture, the increase in Tobin=s q suggests that the divestiture significantly enhanced the market=s perception of the selling firm=s value, from selling at a discount to replacement value to selling at a premium. Our results agree with John and Ofek (1995) who found that profitability increased after divestitures, although they found an increase only in firms for which the divestiture increased focus. Our results show that performance clearly increased after the divestiture, inconsistent with the notions that gains are transferred from the buyer or arise from reducing the costs of managerial discretion or asymmetric information, but consistent with our main hypothesis that gains originate within the firm when negative synergies are eliminated. Descriptive statistics show significant positive excess returns when the buyer was another firm, but only announcements of management buyouts of divisions. Our results could differ because of our later sample period (through 1990 versus 1985) and smaller relative size (9.1% versus 16.6%).

13 normal returns when the buyer was unit management. We investigate changes in Tobin's q to see if q changes in parallel fashion for each category of buyer. Table 2 panel B shows that for 177 interfirm buyers, the divesting firm's median q is significantly higher than year t-1 in all three years following the divestiture (t+1 to t+3), similar to results shown in panel A. From year t-1 to t+3, for example, the mean q-ratio increases significantly (t=4.78) from 0.865 (median 0.731) to 1.080 (median 0.815). With interfirm buyers, results suggest that expectations of improved operations as suggested by increases in Tobin=s q explain the positive announcement period excess returns shown in Table 1, although some gains could still be transferred from the buyer as assets are moved to higher value uses. Table 2, panel C presents results for 48 divestitures to unit management. Here mean Tobin's q increases from 0.773 (median 0.589) at year t-1 to 0.884 (median 0.693) at year t+3 following the divestiture, but does not become significantly higher until the second year (t+2), and remains below 1.0 at year t+3. These point estimates suggest that although operations improved as suggested by higher q-ratios, the increases were not strong enough to produce significant positive announcement period gains. Normal returns shown in table 1 suggest that any gains arising from reduced costs of managerial discretion or asymmetric information about divested assets were not significant, and any gains from more efficient contracting were not transferred back to the parent firm.. These results suggest that divestitures to unit management do not seem motivated by a desire to eliminate negative synergies, and therefore fundamentally differ from divestitures to another firm. 4.3 Industry Adjusted Long Term Performance In table 3 we show industry adjusted changes in Tobin=s q to consider the possibility that increases in Tobin=s q shown in table 2 merely reflect industry wide changes rather than firm specific changes. We found industry adjusted changes in Tobin=s q by subtracting from each firm=s q-ratio the average q-ratio of all firms with the same four-digit SIC code (three-digit SIC if there were fewer than three firms). Industry averages consist of

14 4 to 142 firms with a median of 10 firms. Panel A shows that for all 225 divestitures, parent firms start with below industry mean and median values of Tobin=s q, but q increases each year such that the mean is above the industry average two years after the divestiture. Median q is still below the industry median at year t+3, but is significantly higher than year t-1 (z=1.75). Panel B reports industry adjusted changes for 177 interfirm divestitures, and shows that the year before the divestiture parent firms have below mean and median q-ratios. Following the divestiture, the xq-ratio shows that parent firms' mean q-ratio increases to match the industry average by year t+1, and continues to increase through year t+3, significantly higher (t=2.88) than at year t-1. These increases suggest that operating performance improved after the divestiture, even after adjusting for industry changes, and are consistent with the notion that the divestiture removed negative synergies. In contrast, panel C shows no similar increases in industry adjusted Tobin=s q for divestitures to management. Parent firm q-ratios are below industry mean and median the year before the divestiture, and remain below industry values even three years after the divestiture. These results, unlike divestitures to other firms, suggest that divestitures to management are not designed to remove negative synergies. If gains do occur, they must arise from more efficient contracting or from reducing the costs of managerial discretion. But excess returns shown in table 1 suggest that these gains are not significant or are not transferred to the parent firm.

15 4.4 Managerial Ownership's Affect on Performance Higher levels of stock ownership provide managers increasingly more incentive to sell assets that produce negative synergies, and to bargain hard to capture gains that originate with the buyer. We investigate these issues by first examining the effect that managerial ownership has on long term performance. We regress the industry adjusted change in Tobin's q from years t-1 to t+3 relative to the selloff announcement on the fraction of shares owned by management. We control for size by including the logarithm of the market value of equity and we control for the effect of capital structure on long term performance by including a change in leverage control variable that equals the change from year t-1 to t+3 in the ratio of long term debt to the sum of preferred stock and market value of common stock. Regressions in Table 4 show a strong positive relation between the industry adjusted change in Tobin's q and managerial ownership, before (regression 1) and after (regression 2) controlling for firm size. This relation supports the notion that ownership encourages managers to sell assets that have negative synergies. The significantly positive coefficient for the change in leverage control variable reflects the positive role that debt plays in controlling agency costs (Jensen, 1986) and monitoring managers (Maloney, McCormick, and Mitchell, 1993). In regression 3 we include only interfirm divestitures and show that there remains a strong positive relation between ownership and the change in industry adjusted Tobin's q. In regression 4 we consider only divestitures to management and in contrast find no similar significant relation between ownership and industry adjusted changes in Tobin's q. This result is consistent with our argument that divestitures to unit managers are not motivated by negative synergies, but by a desire to implement a more efficient contracting scheme. Regression estimates in table 4 support the notion that for interfirm asset sales higher levels of

16 managerial ownership provide more incentive to remove negative synergies and thereby improve long term performance. When assets are sold to unit managers, however, managerial ownership plays no role because these divestitures are not motivated by removing negative synergies. Here, unit managers are buying essentially good divisions that unfortunately suffer from information asymmetries, parent firm neglect, or inefficient contracting. Managerial ownership will influence gains received by the seller when these gains are bargained away from buyers that create value through complementary resources or a more efficient contracting scheme. Managers will more likely drive a hard bargain, and will less likely fritter away potential gains during the bargaining process if they share in the gains. For sell offs to unit management, we suggest that gains must be transferred from the buyer back to the parent firm. The success of the negotiation process will depend in part on parent firm management's stake in the outcome, which will help ensure an arms-length agreement and also increases the likelihood of capturing some of the gains.

17 5. Summary and Conclusions We investigate the long term performance of firms that divested assets to another firm or to unit management. Our results show that long term performance, measured by Tobin's q, improves when assets are sold to another firm, that the change in performance is firm specific, and that the change is strongly related to managerial ownership. These results suggest that removing negative synergies generates the gains from interfirm divestitures. Our results do not support the notion that gains are transferred from a buyer that more highly values the assets, although we cannot reject the notion that some gains may be transferred from the buyer. In contrast, when assets are divested to unit management, long term performance of the selling firm does not improve once we control for industry-wide changes. Our results suggest that gains from divestitures primarily arise from within the firm as long term performance improves following the elimination of assets that cause negative synergies. Although we cannot rule out that some gains are transferred from a buyer with complementary resources, or that some gains may arise from reducing the costs of managerial discretion or asymmetric information, our evidence suggests that these sources of gains cannot be the main source. Our evidence also shows that asset sales to unit managers are different. These divestitures are not motivated by removing negative synergies, but likely by an effort to capture gains that arise from more efficient contracting. But evidence shows that for our sample of divestitures to unit management, the parent firm failed to capture a significant share of these gains.

References Akerlof, G., 1970, The market for 'lemons': Quality uncertainty and the market mechanism, Quarterly Journal of Economics 84, 488-500. Baker, G., K. Wruck, 1989, Organizational changes and value creation in leveraged buyouts: The case of The O.M. Scott & Sons Company, Journal of Financial Economics 25, 163-190. Boot, A., 1992, Why hang on to losers? Divestitures and takeovers, Journal of Finance 47, 1401-1424. Chung, K. and S. Pruitt, 1994, A simple approximation of Tobin's q, Financial Management 23, 70-74. Hite, G. L., Owers, J. and R. Rogers, 1987, The market for interfirm asset sales, partial sell-offs and total liquidations, Journal of Financial Economics 18, 229-252. Hite, G. L. and M. Vetsuypens, 1989, Management buyouts of divisions and shareholder wealth, Journal of Finance 44, 953-970. Jain, P., 1985, The effect of voluntary sell-off announcements on shareholder wealth, Journal of Finance 40, 209-224. Jensen, M., 1986, The agency costs of free cash flow: Corporate finance and takeovers, American Economic Review 76, 323-329. Jensen, M., 1989, Eclipse of the public corporation, Harvard Business Review, September-October, 61-74. Jensen, M. and W. Meckling, 1976, Theory of the firm: Managerial behavior, agency costs and ownership structure, Journal of Financial Economics 3, 303-360. John, K. and E. Ofek, 1995, Asset sales and increase in focus, Journal of Financial Economics 37, 105-126. Kaplan, S., 1989, The effects of management buyouts on operating performance and value, Journal of Financial Economics 24, 217-254. Kaplan, S. and M. Weisbach, 1992, The success of acquisitions, Evidence from divestitures, Journal of Finance 47, 107-139. Lang, L., A. Poulsen, and R. Stulz, 1995, Asset sales, firm performance, and the agency costs of managerial discretion, Journal of Financial Economics 37, 3-37. Lang, L., R. Stulz, and R. Walkling, 1989, Tobin's q and the gains from successful tender offers, Journal of Financial Economics 24, 137-154. Lindenberg, E. and S. Ross, 1981, Tobin's q ratio and industrial organization, Journal of Business 54, 1-32. Maloney, M., R. McCormick, and M. Mitchell, 1993, Managerial decision making and capital structure, Journal of Business 66, 189-218.

Morck, R., A. Shleifer, and R. Vishny, 1988, Management ownership and market valuation: An empirical analysis, Journal of Financial Economics 20, 293-315. Palepu, K., 1990, Consequences of leveraged buyouts, Journal of Financial Economics 27, 247-262. Park, S. and M. Song, 1995, Employee stock ownership plans, firm performance, and monitoring by outside blockholders, Financial Management 24, 52-65. Rosenfeld, J., 1984, Additional evidence on the relation between divestiture announcements and shareholder wealth, Journal of Finance 39, 1437-1448. Servaes, H., 1990, Tobin's Q and the gains from takeovers, Journal of Finance 46, 409-419. Shleifer, A. and R. Vishny, 1986, Large shareholders and corporate control, Journal of Political Economy 95, 461-488. Sicherman, N. and R. Pettway, 1987, Acquisition of divested assets and shareholder wealth, Journal of Finance 42, 1261-1273. Song, M. and R. Walkling, 1993, The impact of managerial ownership on acquisition attempts and target shareholder wealth, Journal of Financial and Quantitative Analysis 28, 439-457. Stulz, R., 1988, Managerial control of voting rights: Financing policies and the market for corporate control, Journal of Financial Economics 20, 25-54. Stulz, R., R. Walkling and M. Song, 1990, The distribution of target ownership and the division of gains in successful takeovers, Journal of Finance 45, 817-833. Tehranian, H., N. Travlos, and J. Waegelein, 1987, The effects of long-term performance plans on corporate selloff-induced abnormal returns, Journal of Finance 42, 933-942. 19

Table 1 Descriptive statistics for a sample of 177 interfirm divestitures and 48 divestitures to unit management from 1980 to 1991. Value is the reported price paid for the divested assets. Parent equity is market value for the fiscal-year end prior to the announcement year. Tobin's q and leverage are for the year prior to the divestiture. Leverage is ratio of long term debt to the sum of preferred stock and market value of common stock. Managerial ownership is by Officers and Directors. Buyer All Cases Interfirm Management Value of transaction ($ millions) 216.0 213.7 224.7 Parent equity value ($ millions) 4205.5 4678.5 2461.1 ** Tobin's q.845.865.773 Leverage.746.739.768 CEO ownership (%) 3.45% 3.55% 3.07% Managerial ownership (%) 8.38% 8.53% 7.86% Number of 5% blockholders.40.36.54 ** Excess returns (days -1,+1) 0.674% (t=2.25) 0.836% (t=2.37) 0.075% (t=0.14) Number of cases 225 177 48 Asterisks indicate that Management value differs from Interfirm value at the five (**) or ten (*) percent level of significance.

Table 2 Tobin's q for Divesting Firms Change in long term performance as measured by changes in Tobin's q from the year before to three years following the divestiture announcement. The sample consists of 177 interfirm divestitures and 48 divestitures to unit management from 1980 to 1991. Tobin's q and changes in q ratio for divesting firms Year relative to divestiture t-1 t t+1 t+2 t+3 Panel A: All divestitures (N= 225) q-ratio mean median 0.845 0.731 0.856 0.731 0.936 0.776 1.001 0.794 1.038 0.800 Increase from year t-1 mean (t-statistic) (Wilcoxon z-statistic) 0.011 (0.69) (0.59) 0.090 (3.91) (3.60) 0.155 (5.17) (4.66) 0.193 (5.22) (5.03) Panel B: Interfirm divestitures (N=177) q-ratio mean median 0.865 0.743 0.872 0.746 0.973 0.830 1.036 0.817 1.080 0.815 Increase from year t-1 mean (t-statistic) (Wilcoxon z-statistic) 0.007 (0.39) (0.75) 0.108 (3.86) (3.65) 0.171 (4.62) (4.07) 0.215 (4.78) (4.68) Panel C: Management divestitures (N=48) q-ratio mean median 0.773 0.589 0.798 0.657 0.797 0.707 0.869 0.716 0.884 0.693 Increase from year t-1 mean (t-statistic) (Wilcoxon z-statistic) 0.025 (0.76) (0.22) 0.024 (0.77) (0.66) 0.096 (2.59) (2.39) 0.111 (2.09) (1.86)

Table 3 Industry Adjusted Tobin's q for Divesting Firms Change in long term performance as measured by industry adjusted changes in Tobin's q (xq-ratio) from the year before to three years following the divestiture announcement. Industry adjusted Tobin's q are firm Tobin q less the mean Tobin's q for the firm's four or three digit SIC code industry. The sample consists of 177 interfirm divestitures and 48 divestitures to unit management from 1980 to 1991. Industry adjusted Tobin's q and changes in q ratio for divesting firms Year relative to divestiture t-1 t t+1 t+2 t+3 Panel A: All divestitures (N= 225) xq-ratio mean median -0.102-0.086-0.072-0.052-0.023-0.049 0.005-0.029 0.018-0.044 Increase from year t-1 mean (t-statistic) (Wilcoxon z-statistic) 0.030 (1.58) (0.90) 0.078 (2.89) (3.00) 0.107 (3.06) (2.94) 0.120 (2.86) (1.75) Panel B: Interfirm divestitures (N=177) xq-ratio mean median -0.096-0.044-0.056-0.047 0.000-0.045 0.027 0.006 0.052-0.018 Increase from year t-1 mean (t-statistic) (Wilcoxon z-statistic) 0.040 (1.90) (1.30) 0.096 (3.00) (2.84) 0.123 (2.93) (2.70) 0.147 (2.88) (1.70) Panel C: Management divestitures (N=48) xq-ratio mean median -0.124-0.168-0.133-0.125-0.111-0.112-0.075-0.078-0.104-0.054 Increase from year t-1 mean (t-statistic) (Wilcoxon z-statistic) -0.009 (-0.23) (0.46) 0.014 (0.35) (0.96) 0.049 (0.92) (1.26) 0.021 (0.31) (0.54)

Table 4 Regression analysis of industry adjusted changes in Tobin's q from year t-1 to t+3 on measures of managerial ownership, market value of equity, change in leverage. Regressions are shown for the full sample and samples of interfirm divestitures and divestitures to management. Ownership is by officers and directors as a group. Change in leverage is the change from year t-1 to t+3 in the ratio of long term debt to the sum of preferred stock and market value of common stock. Dependent Variable: Industry Adjusted Change in Tobin=s q Divestitures All (1) All (2) Interfirm (3) Mgmt (4) Constant.0679 (1.42) -.1978 (-1.01) -.1307 (-0.57) -.4021 (-1.12) Ownership (Officers and Directors).0064 (2.17)**.0084 (2.57)**.0092 (2.40)**.0023 (0.41) Change in leverage from year t-1 to t+3.0411 (3.17)***.0415 (3.20)***.0390 (2.78)**.0827 (1.52) Log Equity Value.0352 (1.40).0280 (0.95).0618 (1.52) adjusted R 2.05.05.05.06 F-test p-value.00.00.01.12 Sample size 225 225 177 48 t-statistics in parentheses. Asterisks indicate significance at the 10% (*), 5%(**), or 1%(***) level.

Table 6 (Not revised from old version) Regression analysis of three day announcement period excess returns on managerial ownership and market value of equity. Ownership is by officers and directors as a group in regressions 1, 2, and 4, and ownership by the CEO in regressions 3 and 5. Regression All sell offs Interfirm sell offs Management buyouts 1 2 3 4 5 Constant.4638 (1.15).5583 (1.33).6494 (1.69)*.2819 (0.24).6081 (0.56) Ownership (Officers and Directors).0450 (2.07)**.0275 (1.19).0975 (1.73)* Ownership (CEO).0430 (1.22).1752 (2.01)** Equity Value (10-5 ) -1.600 (-0.40) -1.314 (-0.35) -1.456 (-0.39) -11.57 (-0.51) -11.95 (-0.54) adjusted R 2.01.00.00.03.05 F-test p-value.09.42.40.17.10 Number of cases 275 217 217 58 58 t-statistics in parentheses. Asterisks indicate significance at the 10% (*), 5%(**), or 1%(***) level.