Repeated LBOs: the case of multiple LBO transactions.

Introduction

A sequence of actions is required to classify a firm as undergoing a repeated leverage buyout (LBO). The firm must be traded publicly, then must go private via an LBO, subsequently must obtain public status again through a new initial public offering, and finally must decide

to go private again. This is a rare occurrence. Studies of LBOs aggregate all LBO transactions and do not distinguish between multiple transactions by the same company. These multiple LBO firms may exhibit different characteristics than first-time LBO firms because they have had experience in both the private and public markets. There are significant differences for firms that are undergoing repeated LBOs because their recent completion of the public to private cycle lends credibility to their action. It may be that there is less uncertainty for these finns because they are returning to the private sector, ostensibly to capitalize on some feature of being private that is not available in the public market. This reduction in uncertainty may be perceived by firm management because they have already experienced an LBO transaction.

LBO transactions typically have involved an owner/founder seeking to cash out his or her investment and transfer the firm to managers or younger family members, who would put up a small amount of capital and borrow the rest. Today, LBOs differ from their earlier incarnation because they involve companies that are large and publicly traded with many potential motivations. No single motivation for LBOs has been identified, but potential hypotheses include the alignment of managerial interests with stakeholder objectives, monitoring by sponsors/buyout specialists, dissemination of free cash flow, tax savings, and the elimination of public reporting, exchange registration, and listing expenses. In addition, the original motivation may not be realized because many LBOs subsequently return to the equity capital markets as a new initial public offering.

The appearance of repeated LBOs as a special category of LBO is important because the motivation for these LBO transactions may be clearer than for first-time LBOs. A repeated LBO provides more than one opportunity to observe the motivation for an LBO. The individuals involved may be the same in both cases, thereby providing a pattern of evidence regarding the motivation for the LBO. Halpern, Kieschnick, and Rotenberg (1999) suggest that not all finns do an LBO for the same reason. This group of firms may be a specific subset of companies, but, unfortunately, there is little research that has examined this area. Kosedag and Lane (2002) examine the applicability of the tax savings and free cash flow hypotheses of going private transactions to the repeated LBO decision. They replicate the Lehn and Poulsen (1989) study of going-private transactions and find that firms that repeat LBOs are less likely to have undistributed cash flow as a motivator to go private. This contrasts the Lehn and Poulsen finding that the mitigation of agency problems associated with Jensen's (1986, 1988) free cash flow hypothesis is a main motivation in LBO transactions. Repeated LBOs are not the same.

The market reaction to the announcement of a firm going private in a repeated LBO transaction is important for a number of reasons. First, as a subset of all LBO transactions, the market reaction can indicate if repeated LBOs engender a different reaction than first-time LBOs. We hypothesize that repeat LBOs will he viewed more positively because these are firms with expertise in a usually unrepeated transaction. Second, market reaction can be linked to publicly available accounting data hypothesized to explain LBO transactions. The correlation between the magnitude of the reaction and the possible explanatory variables may lend insight into the perceived benefits of the repeated LBO. Third, differentiating between first-time and repeated LBOs is important from an economic perspective because perceptions of repeated exploitation may trigger public outcry and possible regulation.

Leveraged Buyouts

Although we focus on repeated LBO transactions, understanding LBO transactions is important because some of the same motivations may be present in both cases. A common thread of many earlier LBO studies has been to report the market reaction to LBO announcements and explain the premiums or abnormal returns paid or accrued to target shareholders. While the related studies present differing hypotheses to assign an economic rationale for LBOs, they all document (wherever applicable) Significant positive excess returns associated with these transactions. Table 1 reports the main results from selected LBO papers.

Depending on the time period being studied, the average excess returns for the event window [-1, 0] ranges from 11.95 percent (Bae and Simet, 1998) to 16.20 percent (Travlos and Millon, 1987). In general, results are conclusive in that announcements of going-private transactions are associated with significant positive excess returns. The studies diverge in the hypotheses that they offer to justify the LBOs. There are six possible explanations identified in the literature.

In addition to an immediate savings of public ownership expenses (DeAngelo, DeAngelo, and Rice, 1984), LBOs have often been perceived to take advantage of the tax-shield potential. The tax advantage of LBOs is supported by Lehn and Poulsen (1988); Marais, Schipper, and Smith (1989); Kaplan (1989a); and Muscarella and Vetsuypens (1990). For example, Kaplan (1989a) estimates a median value of tax-driven benefits (both from interest and depreciation deductions) that is between 21 percent and 142.6 percent of the premium paid to shareholders. LBOs are also perceived as a mechanism of wealth expropriation from bondholders in an option-pricing framework. Lehn and Poulsen (1988), Mamis, Schipper, and Smith (1989), Asquith and Wizman (1990), and Cook, Easterwood, and Martin (1992) consider the wealth transfers from the bondholders of the target firm to stockholders as a possible portion of premiums paid to shareholders but the evidence provided by these studies is inconclusive, and the magnitude of the bondholders' losses is insufficient to explain shareholders' gains.

There is an apparent conflict of interest if management is involved in the LBO, because the purchasing managers have an information advantage over both current stockholders and other potential purchasers, it is plausible, therefore+ to posit information asymmetry as an explanation for going-private transactions. Both Smith (1990) and Ofek (1994) cast doubt on the information asymmetry hypothesis as their findings attribute performance improvements to completed buyout proposals only. For example, Ofek (1994) reports that, at cancellation announcements of MBO proposals, returns drop to 2 percent (from 27 percent measured over one month before the buyout announcement to the day after the MBO offer announcement), which is insignificantly different from zero, and persist there for the following two years.

It is also often indicated in the literature that managers may even distort the operating data through manipulations of accounting information to reduce the acquisition price. (1) While limited, the evidence does not support the management manipulation hypothesis (Kaplan, 1989b; and DeAngelo, 1986).

According to Jensen (1986, 1988), many of the benefits in going-private transactions stem from the mitigation of agency problems associated with free cash flow. Free cash flow is defined as cash flow in excess of that required to fund all positive net present value (NPV) projects of a firm. An implication of the free cash flow argument for going-private transactions is that in order to be an LBO candidate, a firm must have substantial free cash flow at the discretion of management. Accordingly, firms or divisions of large firms that have stable business histories and low growth prospects are more likely to be subject to LBOs. Empirical observations are consistent with these predictions. Most LBOs take place in mature industries (Lehn and Poulsen, 1988; Kieschnick, 1989), and the growth rates and capital expenditures of LBO firms are lower than those of comparable firms in the same industry (Kaplan, 1989a; Lehn and Poulsen, 1988, 1989). Moreover, Lehn and Poulsen (1989) show that, consistent with Jensen's theory, their proxy for free cash flow and the premium paid in LBOs are positively related. Similarly, Opler and Titman (1993) conclude that the main characteristics of firms initiating LBOs are unfavorable investment opportunities (low Tobin's q) and high cash flows.

In a more recent study, Bae and Simet (1998) use a sample of LBO and leverage recapitalization (LR) firms in comparing two important aspects of these two highly levered transactions: (1) the returns accruing to stockholders and (2) the attributes of firms associated with these gains. The announcements of both transactions result in significant positive average excess returns, with the LBO group experiencing substantially larger returns than the LR group. Additional analyses show that the level of a firm's free cash flow is responsible for this difference. That is, while free cash flow and stockholders' return on both transactions are positively correlated, the association is more pronounced for LBO group. These findings suggest that firms with free cash flow benefit more via an LBO rather than a LR on the reduction of associated agency costs.

Reverse Leveraged Buyouts

Reverse LBOs have received some measure of attention in the recent literature. For example, Muscarella and Vetsuypens (1989) use a sample of 74 firms re-entering the public market to examine the underpricing phenomenon of initial public offerings (IPOs). Their study supports the information asymmetry explanation of underpricing in IPOs because average initial returns for firms re-entering the public markets (which presumably are known more by public) are lower than those of IPOs. (2) In a follow-up study on the same sample, Muscarella and Vetsuypens (1990) examine the performance of these companies during the period the firm was privately held. For the full sample, they report a 268.4 percent median annualized rate of return on equity, based on comparisons of the going-private price and going-public price. Muscarella and Vetsuypen's (1990) also detect a decline (1) in leverage (from 93.4 percent prevalent at the buyout to 78.6 percent prior to the IPO) under private ownership (3) and (2) in insider ownership-ownership of executive officers and directors (from 63.4 percent prior to the IPO to 44.5 percent after the IPO) following going public again. (4)

DeGeorge and Zeckhauser (1993) find that reverse LBOs display superior performance in the pre-offering year (compared with other firms and continuing LBOs) and a disappointing performance in the year following the IPO (compared with the previous year and with the performance of control firms over the same period). This result is consistent with the authors' information asymmetry hypothesis. (5)

Contrary to DeGeorge and Zeckhauser (1993), Mian and Rosenfeld (1993) report 31 percent cumulative abnormal returns in 25 months after LBO firms went public. Further examination of the sample, however, reveals that the long-term positive abnormal performance following the reverse LBOs is mainly due to takeover premiums. They conclude that going public provides initial liquidity and a showcase (to buyout specialists) that facilitates a subsequent sale to an outside party.

Finally, Holthausen and Larcker (1996) also find no evidence of abnormal common stock performance after the reverse LBO while they document significantly better accounting performance for reverse LBOs (compared to their industries at the time of the IPO and for the following four years).

Data and Method of Analysis

We identified 21 repeated LBOs that occurred between 1980 and 1995 using a variety of sources. We first compiled an initial sample of 450 reverse LBO firms. (6) Next, we searched The Wall Street Journal Index for each individual reverse LBO firm to determine if the firm experienced a second leveraged buyout following the reverse LBO.

Table 2 reports descriptive statistics, grouped by year, for the 21 repeated LBO firms. Although the study covers a wide period, 1980 to 1995, repeated LBO firms are clustered in the years 1986 through 1989, reaching a peak of nine transactions in 1988. The average length of time that the firms spend as a free-standing public firms before their second going-private transaction is 30.74 months (median 22.27 months). Total equity value is computed as the number of common shares outstanding times the closing price of common stock at the end of the year immediately preceding the repeated LBO transaction. Total equity value for the full sample is $2,199,498,000, with a corresponding median equity value of $82,356,000. An interesting statistic (not reported) is the degree of leverage present in these firms prior to their second buyout. Leverage is measured as the ratio of total debt to total assets. The average leverage for the full sample is 0.6212 (median 0.5972). This observation is important because most explanations of the value created in LBO transactions are based on bringing the unused capacity of debt financing to an unusually high level that already seems to be present in repeated LBO firms. This suggests that usual sources that enhance the firm value through leveraging in a typical LBO may not be available the second time around.

Empirical Findings

We examine the market reaction to the repeated LBO announcement adjusting for risk by calculating prediction errors (or abnormal returns) as follows:

(1) P[E.sub.i,t]= [R.sub.i,t] - ([a.sub.i] + [b.sub.i][R.sub.m,t])

where:

[R.sub.i,t] and [R.sub.m,t] = Daily returns for the stock of firm i and of the market portfolio at time t, respectively.

The source for the return data is the Center for Research in Security Prices (CRSP), and we use the CRSP equally-weighted index as the market proxy. For each firm, parameters [a.sub.i] and [b.sub.i] are estimated with an OLS regression over a 150 day interval from day -170 to day -21 relative to the announcement date. The relevant null hypothesis is that the average prediction error [APE] across all firms in the sample is zero for the period of interest. We use three different statistics to test whether estimated APEs significantly differ from zero: standardized residual z-test of Patell (1976), the traditional t-test proposed by Brown and Warner (1980), and the standardized cross-sectional t-test suggested by Boehmer, Musumeci, and Poulsen (1991). (7)

The APEs around the announcement of a repeated LBO are reported in Table 3. Panel A displays APEs, together with relevant statistics, for each single day over the period [-5 to +5]. Panel B has cumulative APEs and associated test statistics for various windows. On the announcement day (day 0) 71.43 percent of the events generated positive prediction errors averaging 16.07 percent. This is economically meaningful and significant using all three statistical tests. The market perceives an important gain accruing to target shareholders, not unlike earlier LBO studies. The magnitude, however, is much larger than in any previous LBO study, irrespective of the event window compared, presented in Table 1. We compare our results to the Bae and Simet (1998) study because they use a similar time period and sample size. Bae and Simet report a cumulative abnormal return of 11.95 percent compared to our 17 percent for the two-day [-1,0] event window. For the [-1, +1] event window the difference is even more pronounced with a 14.77 percent reaction in the Bae and Simet study compared to the 25.57 percent reaction for repeated LBOs.

The larger response to the repeated LBO announcement may reflect the perceived likelihood of success of the transaction because the company has experience with this transaction. We also can link the market reaction to a number of potential explanatory variables that may help to explain the reaction to the repeated LBO. We estimate a regression model using four measures to explain the large market reaction. We include a tax-savings measure (TXTA), a free cash flow measure (FCFTA), a Tobin's Q free cash flow measure (TQ), and an information asymmetry measure (SIZE).

Proponents of tax-based explanations argue that the only major source of value enhancement in LBO transactions is the savings in taxes due to the tax-shield generated by leveraging. We measure the tax-savings potential of the firm following Lehn and Poulsen (1989) and normalize it by the firm's total assets. If, as in typical LBOs, the tax-savings potential is a source of gain in repeated LBOs, TXTA will be related positively to the abnormal returns.

The rationale for including FCFTA as an explanatory variable is Jensen's (1986, 1988) argument offered as an explanation for LBO transactions m general. Firms with little or no growth opportunities are more likely to accumulate and carry free cash which, in turn, makes them a viable LBO candidate. These are the firms who benefit more from the dissemination of cash to shareholders through the LBO mechanism, suggesting a direct relationship between the abnormal returns and the free cash flow measure. We compute the free cash flow potential of the firm following Lehn and Poulsen (1989) and normalize it by the firm's total assets.

Tobin's Q is used to further test the free cash flow hypothesis. A low q (measured as the sum of the market value of equity and the book value of debt divided by the book value of total assets) may suggest lower growth opportunities and hence a higher cash flow potential. A negative relationship between the abnormal returns and TQ should exist to support the free cash flow argument in repeated LBOs.

Finally, we include the log of the firm's total assets, SIZE, among the independent variables. Small firms are more likely to be associated with a greater degree of informational asymmetry and the potential to be undervalued. DeAngelo et al. (1984) suggest that, by going private, small firms benefit more than large firms from the cost savings on shareholder-relations. Both of these hypotheses predict a negative relationship between abnormal returns and SIZE.

The announcement day (day 0) prediction errors (PE) are explained in a regression model of the following form:

(2) PE = [b.sub.0] + [b.sub.1] TXTA + [b.sub.2]FCFTA +[b.sub.3]TQ + [b.sub.4]SIZE.

The regression results reported in Table 4 reveal that that tax-savings fail to explain the variation in cross-sectional abnormal returns of repeated LBO announcements. This finding is in contrast to the findings by Kosedag and Lane (2002). The estimated coefficient for TXTA carries a counter-intuitive minus sign and is not statistically significant at conventional levels. Our results may be due to the small sample size.

Variable FCFTA, pertaining to Jensen's free cash flow hypothesis, is also not significantly different from zero. The findings on TQ are also discouraging for the free cash flow hypothesis. While TQ emerges as the sole significant variable of the model, its sign is inconsistent with the hypothesized prediction. As mentioned earlier when reviewing summary statistics, repeated LBO firms are already highly levered from their first LBO. They have little or nothing to gain from re-leveraging. It is not surprising that the two leverage based explanations of typical LBO transactions tested here, free cash flow and tax-savings, fail to justify the large magnitude of abnormal returns.

In the absence of sources discussed above, one expects to find support for the savings on shareholder relations and/or undervaluation as possible motives and sources of gain in repeated LBOs. Variable SIZE generates a statistically insignificant coefficient, however, suggesting that the former of the two explanations is not related to the observed abnormal returns.

The estimated coefficient for the TQ variable is statistically significant at 1 percent level (i.e., p-value = 0.013). Its positive coefficient indicates that firms with higher (lower) TQ ratio experience larger (smaller) abnormal returns. A high TQ ratio is consistent with favorable growth prospects for the firm that raises the firm's market value relative to its book value. Therefore, the abnormal returns or a value gain observed at the announcement of the repeated LBO transactions reflect the favorable future growth prospects of repeated LBOs. These firms are experiencing their second buyouts after spending a relatively short period in the public arena fop lowing the first buyout and organizational changes experienced in the first round might have enlarged their investment opportunity set, suggesting favorable future prospects and future growth potential for the firms. This may also have enhanced the ability of management to assess the value of the firm. It is thus not surprising that the majority of the repeated LBOs occur following the stock market crash of 1987. Management in that period may perceive that the low market value represents an opportunity to create value, and, based on prior experience, they can repeat the going private transaction to capitalize on the value enhancement.

Conclusion

The stock market reaction to the announcement of a repeated LBO is larger in magnitude than first-time LBOs. We find a 16.07 percent abnormal return, compared to LBO abnormal returns of approximately 12-16 percent. We find that two widely cited sources of value gain in typical LBO transactions, tax savings and free cash flow, do not hold for the repeated LBO sample and that the large reaction is associated with the future growth potential of the finn. This is confirmed by noting that the majority of repeated LBO transactions occur following the 1987 stock market crash, suggesting that expertise from previously being privately run influenced the decision to go private. Simply put, the low publicly-traded price undervalued the firm and past experience with the going private transaction offered a method of realizing the higher value.

Table 1--Summary of LBO Abnormal Returns in Prior Studies

                                   Number of
                       Period of   Buyouts in     Event      CAPE (b)
Study                    Study     the Sample   Window (a)      (%)

                                                  [-1]          2.13

Bae and Simet (1998)                               [0]          9.82

                       1985-1990        21       [-1,+1]       14.77

Amihud (1999)          1983-1986        15       [-20,0]       19.60
                                                 [-1,+1]       16.30

Lehn and               1980-1987       244      [-10,+10]      19.90
  Poulsen (1989)
                                                [-20,+20]      20.50
                                                 [-1,0]        13.93

Lehn and               1980-1984        93      [-10,+10]      20.76
  Poulsen (1988)

Marais, Schipper,      1974-1985        79       [-1,0]        13.00
  and Smith (198)
                                                 [-1,0]        16.20

Travlos and            1975-1983        56      [-10,+10]      19.24
  Millon (1987)

Torabzadeh and         1982-1985        48      [Month 0]      18.64
  Bertin (1987)
                                                 [-1,0]        14.04

Grammatikos and        1975-1984       131       [-10,0]       19.52
  Swary (1986)

(a) event window is reported in days around the announcement

(b) CAPE refers to the cumulative abnormal prediction error with
significant positive excess returns

Table 2--Repeated LBO Firms' Summary Statistics

                           Average Public      Average Equity
           Number of           Life (a)           Value (b)
Year      Repeated LBOs       (months)            (millions)

1986           2                17.66               50.122
1987           3                33.38              162.814
1988           9                23.90               49.753
1989           4                41.71               99.260
1990           1                41.16               99.416
1991           1                71.26               31.341
1994           1                15.57              635.000
Full           21               30.74              104.738
Sample                         (22.27) (c)         (82.356) (c)

(a) Average time that a reverse LBO firm spends in the public markets
before being subject to a second LBO transaction

(b) Number of common shares outstanding multiplied by the closing
price of a share at the end of fiscal year preceding the calendar year
of second going private transaction

(c) Corresponding median values for the full sample.

Table 3--Average Prediction Errors (APEs) for Various Windows Around
the Repeated LBO Announcements

Panel A: Daily APEs from 5 Days Before to Five Days After the
Announcement

                             Percentage    Traditional
Day             APE           Positive      t-statistic

-5             0.0307         0.6667        3.7812 **
-4             0.0211         0.5238        2.6005 *
-3             0.0056         0.2857        0.6912
-2             0.0118         0.4762        1.4548
-1             0.0107         0.5238        1.3136
0              0.1607         0.7143       19.7697 **
+1             0.0844         0.6190       10.3830 **
+2             0.0006         0.6190        0.0687
+3             0.0076         0.4762        0.9323
+4             0.0059         0.5238        0.7332
+5             0.0016         0.5238        0.2018
Panel B: Cumulative APEs for Various Windows
[-10, +1]      0.2910         0.9048       10.3379 **
[-1, 0]        0.1713         0.7619       14.9081 **
[-1, +1]       0.2557         0.8571       18.1670 **
[-10, +10]     0.3355         0.9048        9.0083 **

Day            Patell-z        BMP-t

-5             3.3443 **      1.7138
-4             2.4692 *       1.8200
-3             1.8243         0.7628
-2             1.4693         1.1042
-1             0.8689         0.6437
0             26.8191 **      3.4044 **
+1            11.9411 **      1.6170
+2            -0.2971        -0.4013
+3             0.8035         0.9457
+4             0.5069         0.9475
+5             0.3860         0.5565
Panel B: Cumulative APEs for Various Windows
[-10, +1]     13.8059 **      4.7345 **
[-1, 0]       19.5784 **      3.5122 **
[-1, +1]      22.8799 **      4.1282 **
[-10, +10]    11.2464 **      5.3946 **

** and * mark 1 percent and 5 percent level of statistical
significance, respectively, for two tailed tests

Notes: Average prediction errors for 21 repeated LBO announcements
are reported using three different statistics to determine if
estimated APEs differ significantly from zero: the traditional t-test
of Brown and Warner (1980), the z-test of Patell (1976), and a
standardized cross-sectional t-test of Boehmer, Musumeci, and Poulsen
(1991). Panel A reports average prediction errors calculated on a day
by day basis over the two weeks surrounding the event date, while Panel
B reports cumulative prediction errors for different windows

Table 4--Cross-Sectional Regression Results

Variable        Coefficient Estimate      p-value

INTERCEPT         -0.3892                 0.2575
TXTA              -1.7603                 0.5659
FCFTA             -2.8031                 0.2515
TQ                 0.4988                 0.0130
SIZE               0.0345                 0.4986
F (a)              2.5420                 0.0803
R-square (b)       0.3886

F-statistic for the overall significance of the model. Corresponding
adjusted R-square figure is 0.2357

Notes: Cross-sectional regressions are reported for the sample of 21
repeated LBOs using announcement day (day 0) abnormal returns as the
dependent variable. The explanatory variables are defined as follows:
TXTA is the firm's tax obligations divided by total assets within the
year preceding the second LBO announcement. FCFTA is the firm's free
cash flow (measured as income before depreciation expenses minus taxes,
interest expenses, and any cash dividends paid) divided by total
assets in the year preceding the second LBO announcement. TQ is a
variant of Tobin's q measure and is obtained by dividing the sum of
market value of equity and the book value of total debt with the book
value of total assets for the year preceding the LBO announcement. SIZE
is the firm size as measured by the log of firm's total assets in the
year preceding the second LBO announcement

(1) The CEO (of Regina), who held about 50 percent of the stock during the buyout, sold one-tenth of his stake for $2.1 million. Regina exhibited very strong stock price performance in the first two years following its IPO. In 1988, the CEO abruptly resigned and confessed to "having manipulated the firm's reported results" (DeGeorge and Zeckhauser, 1993).

(2) In evaluating a previously bought-out company returning to public, investors, in addition to the information provided by the prospectus, have access to stock-price history and prior financial statements. In another reverse-LBO study, Kosedag and Michayluk (2000) find that dividend initiations in reverse-LBO firms do not engender the usual positive market reaction also, possibly due to a lower information asymmetry.

(3) Coupled with the fact that only a few companies in the sample intend to use the IPO proceeds for capital expenditures, their finding suggests that going public lowers a company's leverage rather than expands its asset base. Indeed, Muscarella and Vetsuypens (1990) report a median post-IPO leverage of 55.5 percent.

(4) Similarly, DeGeorge and Zeckhauser (1993) find that, on average, insiders sell 10 percent of their holdings in the IPO associated with the reverse LBO.

(5) Information asymmetry, in their setting, translates into management's manipulation of performance, their extraordinary effort before the IPO, or performance borrowing from the future (i.e., discounting prices to boost the sales or deferring R&D expenses).

(6) Going Public-The IPO Reporter (a publication of Investment Dealers Digest); Mergers & Acquisitions; W.T. Grimm's Mergerstat Review; and The Yearbook on Corporate Mergers, Joint Ventures and Corporate Policy. This initial list was augmented by a comprehensive list of reverse LBOs (1980 to 1996) prepared by Investment Dealers Digest ODD) Information Services. Finally, Professor Chris J. Muscarella supplied a list of reverse LBOs generated by Kidder, Peabody & Co. in 1988.

(7) For detailed coverage of these test statistics, see Boehmer et al. (1991).

References

[1.] Amihud, Y., "Leveraged Management Buyouts and Shareholders' Wealth," in Y. Amihud (d.), Leveraged Management Buyouts: Causes and Consequences (Homewood, IL: Dow Jones, Irwin, 1989).

[2.] Asquith, P., and T. Wizman, "Event Risk, Covenants, and Bondholder Returns in Leveraged Buyouts," Journal of Financial Economics, 27, no. 1 (September1990), pp. 195-213.

[3.] Bae, C.S., and D. Simet, "A Comparative Analysis of Leveraged Recapitalization Versus Leveraged Buyout as a Takeover Defense," Review of Financial Economics. 7, no. 2 (Winter 1988), pp. 157-172.

[4.] Boehmer, E., J. Musumeci, and A. B. Poulsen, "Event-Study Methodology Under Condition of Event-Induced Variance," Journal of Financial Economics, 30, no. 2 (December 1991), pp. 253-272.

[5.] Brown, S., and J. Warner, "Measuring Security Price Performance," Journal of Financial Economics. 8, no. 3 (September 1980), pp. 205-258.

[6.] Cook, D.. J, Easterwood, and J. Martin, "Bondholder Wealth Effects of Management Buyouts." Financial Management. 21, no. 1 (Spring 1992), pp. 102-113.

[7.] DeAngelo, L., "Accounting Numbers as Market Valuation Substitutes: A Study of Management Buyouts of Public Shareholders," The Accounting Review, 61, no. 3 (July 1986), pp. 400-420.

[8.] DeAngelo, H.. L, DeAngelo, and E. Rice, "Going Private: Minority Freeze-Outs and Stockholder Wealth," Journal of Law and Economics, 27, no. 2 (October 1984), pp. 367-402.

[9.] DeGeorge, F., and R. Zeckhauser, "The Reverse LBO Decision and Firm Performance: Theory and Evidence," Journal of Finance. 45, no. 4 (September 1993), pp. 1323-1348.

[10.] Grammatikos, T., and I. Swary, "Incentives for Public Firms to Go Private: Superior Information or Organizational Efficiency," (1986) unpublished working paper.

[11.] Halpern, P., R. Kieschnick, and W. Rotenberg, "On the Heterogeneity of Levered Going Private Transactions," Review of Financial Studies, 12, no. 2 (Summer 1999), pp. 281-309.

[12.] Holthausen, R., and D. Larcker, "The Financial Performance of Reverse Leveraged Buyouts," Journal of Financial Economics, 42, no. 3 (November 1996), pp. 293-332.

[13.] Jensen, M., "Agency Costs of Free Cash Flow, Corporate Finance and Takeovers," American Economic Review, 76, no. 2 (May 1986), pp. 323-339.

[14.] Jensen, M., "Takeovers: Their Causes and Consequences," Journal of Economic Perspectives, 2, no. 1 (Winter 1988), pp. 21-48.

[15.] Kaplan, S., "Management Buyouts: Evidence on Taxes as a Source of Value," Journal of Finance, 44, no. 3 (July 1989a), pp. 611-632.

[16.] Kaplan, S., "The Effects of Management Buyouts on Operating Performance and Value," Journal of Financial Economics. 24, no. 2 (October 1989b), pp. 217-254.

[17.] Kieschnick, R., "Management Buy-Outs of Public Corporations: An Analysis of Prior Characteristics," in Y. Amihud (ed.), Leveraged Management Buyouts (Homewood, IL: Dow Jones-Irwin, 1989).

[18.] Kosedag, A., and W. Lane, "Is It Free Cash Flow, Tax Savings, or Neither? An Empirical Confirmation of Two Leading Going-Private Explanations: The Case of ReLBOs," Journal of Business Finance & Accounting. 29, no. 1/2 (January 2002), pp. 257-271.

[19.] Kosedag, A., and D. Michayluk, "Dividend Initiations in Reverse-LBO Firms," Review of Financial Economics, 9, no. 1 (Spring 2000), pp. 55-63.

[20.] Lehn, K., and A. Poulsen, "Leveraged Buyouts: Wealth Created or Wealth Redistributed," in Murray Weidenbaum and Kenneth Chilton, (eds.) Public Policy Towards Corporate Takeovers (1988) (New Brunswick, NJ: Transaction Publishers, need year).

[21.] Lehn, K., and A. Poulsen, "Free Cash Flow and Stockholder Gains in Going Private Transactions", Journal of Finance 44, no. 3 (July1989), pp. 771-787.

[22.] Marais L., K. Schipper, and A. Smith, "Wealth Effects of Going Private for Seniors Securities," Journal of Financial Economics, 23, no. 1 (Jane 1989), pp. 155-191.

[23.] Mian, S., and J. Rosenfeld, "Takeover Activity and the Long-Run Performance of Reverse Leveraged Buyouts," Financial Management, 22, no. 4 (Winter 1993), pp. 46-57.

[24.] Muscarella, C., and M. Vetsuypens, "The Underpricing of 'Second' Initial Public Offerings," Journal of Financial Research, 12, no. 3 (Fall 1989), pp. 183-192.

[25.] Muscarella, C., and M. Vetsuypens, "Efficiency and Organizational Structure: A Study of Reverse LBOs," Journal of Finance. 45, no. 5 ( December 1990), pp. 1389-1413.

[26.] Ofek, E., "Efficiency Gains in Unsuccessful Management Buyouts," Journal of Finance, 49. no. 2 (June 1994), pp. 637-654.

[27.] Opler, T., and S. Titman, "The Determinants of Leveraged Buyout Activity: Free Cash Flows vs. Financial Distress Costs," Journal of Finance, 48, no. 5 (December 1993), pp. 1985-1999.

[28.] Patell, J., "Corporate Forecasts of Earnings Per Share and Stock Price Behavior: Empirical Tests," Journal of Accounting Research. 14, no. 2 (Autumn 1976), pp. 246-276.

[29.] Smith, A., "Corporate Ownership Structure and Performance: The Case of Management Buyouts." Journal of Financial Economies. 27, no. 1 (September 1990), pp. 143-164.

[30.] Torabzadeh, K., and W. Bertin, "Leveraged Buyouts and Shareholder Returns," Journal of Financial Research, 10, no. 4 (Winter 1987), pp. 313-319.

[31.] Travlos, N. G., and M. H. Millon, "Going Private: Buy-Outs and Determinants of Stockholders Returns" (1987), unpublished working paper.

Arman Kosedag

Sabanci University

David Michayluk

University of Rhode Island