Operating Performance Changes Associated with Corporate Mergers and the Role of Corporate Governance

Abstract We present new evidence on the association between operating performance changes associated with corporate mergers, and the governance characteristics of bidders. We find that board ownership is positively associated with performance improvements but the overall influence of board ownership is attenuated by the dispersion of ownership across board members. We also find that the presence of large outside blockholdings is positively associated with performance improvements but that larger boards are associated with smaller improvements. When there is a mismatch in the qualities of the management team, improvements suffer. Likewise when the bidder and target are of similar size improvements are smaller, potentially due to what Williamson (1985) has termed “diminishing returns to management.” Interestingly, while governance and managerial incentives’ related factors significantly impact on ex-post industry adjusted cash-flow returns, the market, in its ex-ante revaluation of the firms involved, appears to systematically underestimate the impact of these factors, whether positive or negative. As a result, merger-related market revaluations significantly predict future industry adjusted cash-flow returns only if we control for these governance and managerial incentives’ related factors. JEL classification: G34; L25 Keywords: Mergers; Corporate Governance; Operating Performance

1

Operating Performance Changes Associated with Corporate Mergers and the Role of Corporate Governance

1. Introduction

An important and open question is how the incentives of managers and the governance profiles of

companies influence the operating outcomes of a merger. This question is important because many

hypotheses about the motivation for a takeover, and the conclusions drawn from the capital market

revaluations of the firms involved, devolve from arguments about fundamental value and operating

effects that accompany the combination. In this paper we directly assess how the incentives of acquiring

firm managers and the governance profiles of acquiring firms influence the operating performance of

mergers. Our primary focus is on the relations between the governance and managerial incentives’

related factors of bidders in mergers and the merger-related operating performance changes associated

with those combinations. In conjunction, we also investigate the relations between these operating

performance changes and the capital market revaluations of the partners, and how those revaluations are

influenced by the same governance and incentive factors.

The relationships between company governance characteristics, including managerial incentives,

and the capital market’s revaluations of bidders and targets have been explored extensively. However, as

emphasized by Healy, Palepu and Ruback (1992), analyzing financial market revaluations associated with

takeover announcements permits inferences to be drawn not about changes in post-merger operating

performance but only about changes in the beliefs of investors about this operating performance.1 An

example will help illustrate this point. Safeway PLC, a leading UK firm, received a purchase offer from

Morrison Supermarkets PLC in 2003. Morrison, a firm tightly controlled by Sir Ken Morrison, projected

about £250 million annual merger-related savings and synergies. On paper, Morrison also appeared a

“better” company with a Q-ratio of 2.12 against Safeway’s 1.20. The capital market agreed with

Morrison’s pronouncements, with a value-weighted average increase of 7.8% in the combined value of

the two firms (net of changes in the relevant market index) from the initial announcement through

approval of the deal in 2004. However, the combined company soon announced exceptional merger-

1 Bhagat, Shleifer and Vishny (1990) in their study of 62 hostile takeovers from the 1984-86 period go further to suggest “…the value gains may simply reflect the market’s overestimation of the value of strategic combinations, just as the market overestimated the gains to conglomerate mergers in the 1960’s” (p. 57). Excellent surveys of the empirical evidence include Jensen and Ruback (1983) and Jarrell, Brickley and Netter (1988), and Andrade, Mitchell and Stafford (2001). See also, Weston, Chung and Hoag (1990), Weston, Chung and Siu (1998) and Weston, Mitchell and Mulherin (2004). Recent empirical studies have begun to raise serious doubts about the value-maximizing consequences of corporate takeovers (see Moeller, Schlingemann and Stulz, 2005).

2

related costs of £140 million, and the weighted-average industry-adjusted annualized operating cash-flow

returns for the two firms fell from 1.7% pre-merger to -0.5% post-merger, accompanied by a post-merger

market value loss through September 2005 of £2470 million, even though the market index had

appreciated 14% over the same interval. Despite the market initially revaluing Morrison and Safeway

upward, the operating performance of the merger worsened. We demonstrate later using a simple linear

model explaining merger operating performance changes based upon a sample of UK mergers that

predates and excludes the Morrison/Safeway merger, a decrease in operating performance would have

been (correctly) predicted ex-ante for the Morrison-Safeway merger at the time of the merger’s

announcement. Instead, judging from the market revaluations handed out, the market apparently expected

an increase. We show that the difference is driven largely by the market not adequately estimating the

effect of governance-related factors on expected operating performance.

Several features of our study are new to the literature. First, we present an investigation of the

empirical associations between managerial incentives and governance characteristics of bidders, and, the

actual operating performance changes accompanying mergers. Our study also makes a new and more

general contribution to the measurement of directors’ influence on company choices and the outcomes of

those choices by presenting evidence on the influence of the dispersion of ownership across corporate

board members on operating performance. Second, as many hypotheses about merger outcomes are

unrelated to governance issues, we also present a comprehensive investigation and new evidence on the

associations between other amalgamating firm characteristics, the nature of the underlying deals, and the

changes in operating performance realized by those deals. Our study differs from the work of Healy,

Palepu and Ruback (1992), and subsequent studies of the operating performance consequences of

corporate takeovers (for example, Andrade, Mitchell and Stafford, 2001; Ghosh, 2001; Linn and Switzer,

2001; Heron and Lie, 2002; Megginson, Morgan and Nail, 2004) by both integrating prior results in the

literature as well as extending the analysis to specifically include factors associated with managerial

incentives and governance in explaining changes in operating performance following a merger.2 A third

2 The determinants of the actual cash-flow effects of takeovers have been directly examined in relatively few studies. Healy, Palepu and Ruback (1992) find evidence of a general improvement in the operating performance of firms involved in 50 large corporate takeovers, Subsequent studies utilizing similar methods report average improvements in operating performance following corporate takeovers for which the method of payment is cash (Linn and Switzer, 2001) and for which the degree of industrial relatedness between the amalgamating pairs of firms is high (Heron and Lie, 2002; Megginson, Morgan and Nail, 2004) and the quality of acquiror management is high (Heron and Lie, 2002). Ghosh (2001) and Powell and Stark (2005) are mainly concerned with the specific approach used by Healy, Palepu and Ruback (1992) to measure changes in operating performance. In contrast to Linn and Switzer (2001), Heron and Lie (2002) find no evidence that the method-of-payment is associated with operating performance changes. Andrade, Mitchell and Stafford (2001) report positive changes associated with the mergers in their sample.

3

new contribution of our study is an analysis of the correspondence between the factors associated with

operating performance changes and those associated with the equity market revaluations of merger

partners; in the context of factors that specifically include managerial incentives and governance

characteristics.3

We find evidence of an economically and statistically significant industry-adjusted change in

operating cash flow return following the corporate takeovers in our sample. The financial market also

seems to expect such gains, as is evidenced by the positive combined stock market revaluation for the

amalgamating firms at takeover announcement. Our results also indicate that the corporate governance

characteristics of acquiring firms significantly impact the magnitude of the change in operating

performance for the amalgamating pairs of firms. We find that operating performance changes are

positively related to board ownership but that the positive effect of ownership is attenuated at high levels

of ownership when ownership is concentrated in the hands of a few officers and directors .4 We also find

that large outside blockholdings have a positive association with operating performance changes.

However, board size is significantly and negatively related to operating performance changes. Our results

on the positive association between large outside blockholdings and changes in operating performance

suggest strategic activism on the part of blockholders not generally identified in studies of blockholder

influence (Holderness and Sheehan, 1988; Parrino, Sias and Starks, 2003) but which is broadly consistent

with the beneficial effects of blockholdings suggested by Hotchkiss and Strickland (2003). The relation

between board size and operating performance is consistent with general findings on the influence of

board size on company value (Yermack, 1996). The associations between these governance

characteristics and changes in operating performance persist after controlling for other firm and deal

characteristics and after addressing the issue of board ownership endogeneity.

In addition to operating performance changes being influenced by managerial incentives and

governance, we find that a number of other firm-specific characteristics are also associated with

3 Long-term stock performance studies show that the financial market needs, on average, to significantly revise its expectations as to the gains from corporate takeovers, and that these instances are most prevalent for takeovers with specific types of amalgamating firm or deal characteristics (see, for example, Loughran and Vijh, 1997 and Rau and Vermaelen, 1998). 4 Stulz (1988) suggests that firm value is non-linerarly related to ownership. Several authors have explored the relation empirically (amongst others, Morck, Shleifer and Vishny, 1988; McConnell and Servaes, 1990). Several papers have suggested the relation between insider ownership and either firm value or the stock market’s reaction to some event (notably a merger announcement) is spurious and once endogeneity of the ownership choice is accounted for the relation vanishes (for instance, Loderer, and Martin, 1997; Cho, 1998; Palia, 2001). Later we test for the endogeneity of board ownership in our sample and conclude that we can reliably treat board ownership as an exogenous variable in our sample.

4

performance changes in mergers. Operating performance changes are adversely effected by a mismatch

of managerial quality in which the quality of bidding firm managers is greater than the quality of target

managers. The result suggests bad managers may be difficult to assimilate and/or that acquiring

managers may be potentially inflicted with hubris associated with the belief that they can turn around

poorly performing firms when the chance of this occurring is small. We also find that when bidders have

relatively more excess cash as compared with the target, operating performance falls, consistent with

bidders misusing resources. Finally we find that a mismatch of leverage between the amalgamating firms

has a beneficial impact on the operating performance effects of corporate takeovers possibly suggesting

that the potential for capital structure rearrangements may be value enhancing. We also find that the

relative sizes of amalgamating pairs of firms, and bids for which the method of payment is common stock,

have pronounced negative effects on operating performance changes. The latter result is consistent with

the results reported in Linn and Switzer (2001) and with the theory proposed by Fishman (1989).

Interestingly, with the exception of board ownership, the governance characteristics of the bidder

are not significantly related to the combined revaluation of the amalgamating firms. It seems that while

governance relations impact operating choices ex post, the stock market does not factor these governance

characteristics into prices in an unbiased manner. It is possible that the economic impact of these factors

is regarded by the market as being too uncertain prior to the merger. Specifically, for our data, the market

appears to have systematically underestimated the impact, whether positive or negative, of governance

related variables like board ownership, the interaction of board ownership with ownership dispersion,

board size, and block votes. As a result, merger-related abnormal asset revaluations did not significantly

predict actual industry-adjusted cash-flow returns, but the predictive ability became significant after

controlling for the operating performance changes attributable to governance-related and other factors.

Section 2 describes our sample of corporate takeovers and defines the measures we use for

calculating operating performance changes and combined revaluations associated with the sample firms.

Section 3 defines the governance and managerial incentive variables, and our other hypothesized

determinants for operating performance changes and asset revaluations, and also provides our motivations

for their choice. Section 4 and section 5 report empirical results on the economic factors influencing

merger-related operating performance changes and abnormal asset revaluations respectively. Section 6

relates expected and actual operating performance changes. Section 7 presents our conclusions.

5

2. The Sample and Measures of Performance

2.1 The sample

Our sample is drawn from domestic mergers in the UK during the period 1985-94, and thus

covers a period comparable to several studies based on US data. Corporate takeovers are identified from

the Securities Data Company (SDC) mergers and acquisitions data base.5 A completed corporate

takeover is defined as one where the offer is declared wholly unconditional and where the previously

independent firm being acquired subsequently has its stock exchange listing cancelled and is fully

amalgamated with the acquiring firm. We verify the announcement date of the offer, the unconditional

date of the corporate takeover, and the effective date of the amalgamation using various sources including

reports in the Times of London. Amalgamating firms must be listed on the London Stock Exchange for at

least five years before the corporate takeover, and the amalgamated firm must continue to be listed for a

minimum of five years after the amalgamation. The listing condition is necessary because the measure of

operating performance used in our analysis is scaled by company market value.

UK corporate combinations provide an excellent experimental setting for the examination of how

ownership and board structure of the bidder influence operating performance outcomes. UK companies

are required by law to obtain supermajority (75%) approval of shareholders for the adoption of poison

pills and other anti-takeover measures. This makes the recommendation of such provisions and their

adoption very rare, and makes the UK a setting in which the confounding actions by managers aimed at

thwarting takeovers using anti-takeover measures are minimized. The restrictions on the adoption of

antitakeover provisions by UK companies potentially sets the stage for the economic incentives of bidder

firm managers being as unmitigated as possible, in contrast to takeover events in the U.S. (Masulis, Wang

and Xie, 2006), and allows such incentives to more cleanly motivate the choice of target firm, and more

importantly, to influence the management of the combined firm and the performance outcomes of the

merger.6

Datastream is our source for accounting data. Amalgamating firms must have end-of-year

financial accounting data available for at least five years before that of the corporate takeover, and the

5 Under UK stock exchange rules, close to half of the corporate takeovers in our sample are classified as significant amalgamations requiring shareholder approval. 6 Cremers and Nair (2005) and Core, Guay and Rusticus (2006) have shown that, in general, firms with stronger shareholder protection (for instance, firms facing fewer restrictive anti-takeover provisions) are associated with better long-run stock performance; and firms with more restrictions (less shareholder protection), as measured by the Gompers, Ishii and Metrick (2003) index of shareholder rights, are associated with operating performance that is worse than firms with greater shareholder rights. See also the recent work by Bebchuk, Cohen, and Ferrell (2004) and Bebchuk and Cohen (2005).

6

amalgamated firm must continue to have end-of-year data available for a minimum of five years after that

of the corporate takeover. A representative pre-merger benchmark to use as a baseline for measuring

operating performance change is an important ingredient in the analysis of such changes. An additional

important measurement issue involves the uncertainty about the actual time between amalgamation and

when the operating performance effects of the combination emerge. Thus the choice of the length of

before-and-after time periods for evaluation has important implications for the interpretation of calculated

changes in operating performance accompanying a corporate combination. Clearly, using shorter before-

and-after periods will guarantee a sample-size larger than that based upon longer pre-merger and post-

merger data requirements. However, shorter pre-merger periods may not provide a benchmark for pre-

merger performance that is sufficiently representative, and shorter post-merger periods may not capture

the full consequences of the corporate combination. We are cognizant of the tradeoff and have opted for

the clearly more conservative choice of longer rather than shorter measurement period.7 Yet, in spite of

this conservative choice, our primary results also tend to meet the relatively more demanding standard of

being statistically significant at p-values of 0.01 or lower.

We exclude amalgamations involving firms operating in highly regulated industries, such as the

financial (including real estate) and utility sectors. Datastream is used to establish the primary industry in

which the amalgamating firms operated.

We also exclude cases in which the bidder was engaged in a completed public takeover (using the

same definition as above) during the 5 years before or after the sample amalgamation case. Finally, as

one of our principal interests is how the incentives faced by bidding firm managers influence the

operating performance consequences of corporate takeover decisions, mergers of equals resulting in the

creation of newly listed firms are excluded from the sample because the acquiring firms are not always

clearly defined in such situations. We use company public announcements to identify all confounding

corporate takeover events and mergers of equals.

Our final sample contains 81 corporate takeovers which meet all of the necessary conditions

described above and which have data available for constructing the firm-level governance, and other,

explanatory variables used in the analysis (as discussed in section 3).

2.2 Measuring operating performance effects of corporate takeovers

A measure of merger performance must be capable of revealing any operating efficiency changes,

synergy effects, as well as detrimental operating effects that are a consequence of the combination.

7 For further discussion of the methodology for the evaluation of operating performance changes see Barber and Lyon (1996).

7

Following Healy, Palepu and Ruback (1992), we measure operating performance in any financial year as

operating income before depreciation and non-recurring items. By construction, this measure of

operating cash flow is neither affected by the method of accounting for the amalgamation nor is it

influenced by the choice of financing for the merger. We deflate operating cash flows by an estimate of

the market value of assets defined as stock price times the number of shares outstanding, plus the book

value of preferred stock and long- and short-term debt, all measured at the end of the relevant financial

year. This scaling produces a cash-flow operating return that is comparable across mergers.

We define performance change as the change in the industry-adjusted operating performance

return where the post-merger performance measure is based upon the amalgamated firm and the pre-

merger measure is a value-weighted average of the industry-adjusted performance measures of the bidder

and target companies. We construct industry portfolios for each bidder and each target for each of the 5

financial years before and after that of the merger.8 The operating cash flow returns of the amalgamating

firms for each of the 10 years are industry-adjusted by subtracting the median operating cash flow return

for the comparable industry portfolio (excluding the sample firm) from the sample firm’s operating cash

flow return. Median values are used instead of averages because the number of comparable firms for the

industries varies across time and the sample, and because we also want to avoid the influence of outliers.

In each year of the pre-amalgamation period, we compute the performance measure for the

pseudo-combined firms as a value-weighted average of the industry-adjusted operating cash flow returns

for the firm being acquired and the acquiring firm. The weight for the acquiror is its market value as

defined above, divided by the sum of the market values of the acquiror and the acquired firm. The weight

for the acquired firm is one minus the acquiror’s weight. In each year of the post-amalgamation period,

the comparison industry performance measure for the combined firms is computed as a value-weighted

average of the industry median operating cash flow returns for the industry portfolio measures of the

amalgamated firms, where the weights are again computed as just described. The post-merger industry-

adjusted operating cash flow return is computed by subtracting the weighted average industry benchmark

return from the operating cash flow return of the actual combined firm.

Our calculations give us a 10-year history of industry-adjusted operating cash flow returns for

each merger, 5 years before and 5 years after the combination. We then determine the median industry-

adjusted operating return for the per-merger 5 year period and the median for the post-merger 5 year

period for each merger. In all of the ensuing analyses, abnormal operating performance changes for each

8

merger are measured as the difference between the post- and pre-amalgamation 5-year median industry-

adjusted operating cash flow returns. When measuring the change in performance we exclude

performance during the year of the combination, which we consider to be a transition period and not

representative of the post merger performance of the combined firms. The results discussed below reveal

that during the pre-merger period the amalgamating firms are not superior performers relative to their

industry counterparts.

Panel A of Table 1 presents descriptive statistics for the industry-adjusted operating cash flow

returns around the corporate takeovers. The pseudo-combined amalgamating firms do not perform

significantly better or worse than the industry counterparts over the 5 financial years before the merger.

The mean and median industry-adjusted annual operating cash flow returns for the pre-merger period are

both roughly -0.9%, neither of which is significantly different from zero at conventional levels of

statistical significance.9 In the post-amalgamation period, however, the amalgamated firms significantly

outperform their industry counterparts by, on average, 9.7%. The fact that the median for the post-merger

period is smaller than the mean suggests the industry-adjusted operating cash flow returns are skewed.

Notwithstanding this, the median abnormal operating performance return during the post-merger period

equals 4.4%. This result is significantly different from zero at the 1% level using a variety of tests.10

Seventy-seven percent (77%) of the corporate takeovers in the sample have positive industry-adjusted

operating cash flow returns in the 5 post-amalgamation years, a result that is significantly different from

the proportion expected in the absence of corporate takeover at the 1% level.

Given the results for the separate periods surrounding the mergers, it is not surprising that the

results for the change in industry-adjusted operating cash flow returns from the pre- to the post-

amalgamation period provide strong support for the conclusion that the mergers in our sample are

associated with positive operating performance changes. We compute the changes in performance for

each merger as described earlier. The median of the changes in industry-adjusted operating cash flow

returns across the sample mergers is a positive 6.4% and the percentage of cases registering positive

changes in performance is 81%. Both of these results are significant at the 1% level. The mean of the

changes is equal to 10.7%. The operating performance effects for our sample of corporate takeovers

8 Our approach is similar to the approach followed by Healy, Palepu and Ruback (1992) with the added quality that the pre-merger performance of our sample firms is not significantly different from our benchmark comparison firms (see Barber and Lyon, 1996 and Ghosh, 2001). 9 In results not reported, we also find that amalgamating firms are indistinguishable from industry counterparts based on operating performance for each of the pre-amalgamation years. We test the null hypothesis that the median equals zero using the Wilcoxan signed rank test and the van der Waerden test (see Conover, 1980). 10 See footnote 13.

9

corroborate, in particular, those of Healy, Palepu and Ruback (1992).

In Panel B of Table 1, we present cross-sectional results from an examination of the association

between the median of the pre-merger industry-adjusted operating cash flow returns and the median of the

post-merger industry-adjusted cash flow returns for the sample cases. We transform the median operating

cash flow returns to continuously compounded returns by taking the natural log of 1 plus the return. The

transformation reduces skewness but does not materially affect the results. We use the notation

i,ePrOPCFRET to define the pre-merger value for merger i and i,PostOPCFRET for the post-merger value

and for convenience refer to these as the industry-adjusted operating cash flow returns. The first

regression reveals an almost perfect positive correlation between post- and pre-amalgamation industry-

adjusted operating performance. Recall we define the change in performance as the difference between

the 5-year post-merger median for merger i and the associated 5-year pre-merger median. We use the

notation iOPCFRET∆ to represent the natural log of 1 plus the change for merger i. In the second

regression, iOPCFRET∆ is regressed on the pre-merger value i,ePrOPCFRET revealing no association.

These results are consistent with those reported by Healy, Palepu and Ruback (1992) and Ghosh (2001).

Based upon these results, the change in industry-adjusted operating cash flow returns is used in all

subsequent analyses as our measure of performance change.11

2.3 Measuring market revaluations

We measure the market’s revaluations of the merger partners by computing the abnormal asset

returns of the merged firms. All price and market value data are obtained from Datastream. We first

compute abnormal stock returns as continuously compounded market model adjusted total returns from 5

days, for acquiring firms, and 30 days, for the firms being acquired, preceding the announcement dates of

the offers to 5 days following the corporate takeovers being declared wholly unconditional. Market

model parameters are estimated using stock return data from 500 to 101 days before the relevant

announcement dates and using the FTSE All Share index total returns as the proxy for the market return.12

Abnormal stock returns for the target company are measured relative to the earliest bid announcement

date which, if the acquiring firm faced competition from a rival bidder, could have been the date of the

rival’s first bid if it preceded the acquiring company’s first bid. A longer pre-announcement run up for

the firms being acquired is used to capture the well-documented leading revaluation effects associated

11 Specifically, for each merger we compute the difference between the pre and post-merger annual medians. We then transform the change by taking the natural log of 1 plus the change. 12 We also computed abnormal stock returns for the acquiring firms and for the firms being acquired using a simple market-adjusted model (ignoring beta). The results using this approach are almost identical to those found using market model residuals.

10

with target firms. Abnormal stock returns are measured to beyond the unconditional dates because we

want the revaluations to reflect the market’s overall beliefs about the completed deals.13

Because the operating performance measure is an ‘unlevered’ cash flow, we convert the abnormal

stock returns into unlevered returns. We make the assumption that bondholder wealth changes are on

average zero when computing the unlevered returns for the bidders and the targets.14 The assumption

implies the unlevered (asset) revaluations of the individual amalgamating firms equal the abnormal stock

returns for each firm weighted by the respective ratio of the market value of equity for the firm in

question to its total market value. Specifically, if in general the total abnormal asset value change is given

by DD

VD

SS

VS =

VV ∆

+∆∆ , where S equals the equity market value, D equals the market value of debt

and, V represents total firm value and ∆ is used to denote the abnormal change in a variable, and if

0D =∆ then SS

VS =

VV ∆∆ . We measure the equity to value ratio at the financial year end before

amalgamation. Asset revaluations for the pseudo-combined firms are then defined as the weighted-

average of the abnormal asset returns for the acquiring firms and for the firms being acquired, where,

consistent with the operating cash flow returns, the weights are based on the market values of the

amalgamating firms. This is similar to the approach used by Bradley, Desai and Kim (1988) and others to

compute pseudo-combined stock returns.15

Table 2 presents asset revaluations for the pseudo-combined firms comprising the sample of

corporate takeovers, as well as the component asset returns for the acquiring firms and for the firms being

acquired. Consistent with the findings of numerous other studies acquired firms experience a positive and

significant asset revaluation. The positive and significant asset revaluations for the pseudo-combined

firms imply that overall the market expected value increasing changes to be generated by the corporate

takeovers in our sample. The mean and median market revaluations equal 3.3% and 2.3%, respectively.

Both the mean and the median asset revaluations are significantly different from zero at the 1% level. A

significantly greater number of pseudo-combined firms than would be expected by chance have positive

13 Loughran and Vijh (1997) conclude that the market revaluation effects of corporate takeovers should be measured over a time frame extending into the post-amalgamation period to incorporate what are found to be significant revisions in synergy expectations. We agree. Our focus however is on the market’s beliefs as reflected in the revaluations up to the point just prior to the merger actually commencing operation, thus reflecting beliefs about anticipated performance only. 14 Healey, Palepu and Ruback (1992) employ this approximation as well. The assumption is consistent with several empirical results in the literature. See for instance Dennis and McConnell (1986). 15 Bradley, Desai and Kim (1988) for instance weight the bidder and target abnormal returns using equity value weights. See also Berkovitch and Narayanan (1990) and Andrade, Mitchell and Stafford (2001).

11

revaluations. The magnitudes of the revaluations for our sample of corporate takeovers are comparable to

the revaluations reported by Moeller, Schlingemann and Stulz (2005) for the U.S. market over the time

period represented in our study.

The remainder of the paper focuses on an examination of the determinants of the operating cash

flow changes for our sample of corporate takeovers, the association between these determinants and the

market revaluations for these firms, and finally, the relationship between operating cash flow changes and

market revaluations in the context of these determinants.

3. Explanatory Variables

3.1 Motives for takeovers

Broadly speaking, there are three primary motives for mergers. The first arises from expected

synergies or efficiency enhancing reasons with the principal motive being to create value.16 This should

result in merger-related operating performance gains. The incentives which drive such behavior are not

uniquely specified under such a hypothesis, but one argument is that these incentives allow managers to

share in the value created. The second motive arises from acquiring firm managers pursuing their own

personal interests, which can manifest themselves in several forms - for example, empire building, asset

(size) growth or real asset diversification - and these choices do not necessarily create value, and may

even destroy value.17 In this scenario, managers’ incentives to maximize firm value do not dominate their

utility from maximizing other personal interests unrelated to firm value. Mergers motivated by such

personal interests should lead to degradation in operating performance. The third motive arises from

acquiring firm managers exhibiting hubris, and pursuing mergers because of mistaken, inflated estimates

of the operating performance benefits of the combinations.18 Hubris could ex-ante be exasperated by

16 See Weston, Chung and Hoag (1990, Chapter 8), Weston, Chung and Siu (1998, Chapter 5), Weston, Mitchell and Mulhern (2004). 17 See (Jensen and Meckling, 1976; Williamson, 1985; Shleifer and Vishny, 1988, 1989; Morck, Shleifer and Vishny, 1990). 18 See (Roll, 1986).Value creation for bidder shareholders may also stem from cases in which the target is undervalued due to capital market inefficiency, so that purchase is possible at a price less than true value. While redistribution would occur from target shareholders to bidder shareholders in such case, if the merger causes the market to revise the value of target upward, the combined value of the bidder and target will appreciate and overall value will be created. We lump actions motivated by these circumstances with actions motivated generally by the desire to create value. An alternative motive is the redistribution of wealth from bidder firm bondholders to bidder firm shareholders as a result for instance of the purchase increasing the overall riskiness of the firm. If managers’ interests are not aligned with shareholder interests, but are aligned with bondholder interests, then this motive may be active. Managers whose compensation (wealth) is not aligned with share prices, but lets say involves only a fixed payment salary (no bonus of any kind), will in principle act like bondholders and select actions that guarantee their salaries. This of course is an agency problem and we would argue will manifest itself in the merger choices

12

contractual arrangements in which managers believe the direct or indirect incentives they face will lead to

personal wealth gains as a result of the merger proposal they are evaluating, despite the fact that they

mistakenly have inflated expectations about the merger gains. The hubris hypothesis should predict that,

at best, the ex-post operating performance of the combined firm will be unchanged relative to how the

partners would have otherwise been predicted to perform. However, if scarce resources are used up in

accomplishing the merger without any consequent return for their use, then operating performance of the

combined firm would decline.

We accordingly propose a set of explanatory variables intended to capture the potential incentive

forces influencing managerial choice in mergers as well as the characteristics of firms or deals likely

suggested in the literature as correlates with each of the three motives described above. Many of these

characteristics have been found to explain the cross-sectional variation in the abnormal stock returns of

acquiring firms. Some firm and deal characteristics are also related to the post-takeover, long-term stock

returns of acquiring firms.

3.2 Factors influencing operating performance and asset revaluations

We classify potential determinants of merger performance into four categories: (1) factors

associated with the corporate governance arrangements and managerial incentives of the acquiring firms;

(2) factors associated with the acquiring firm; (3) factors associated with the acquiring firm relative to the

firm being acquired; and, (4) factors associated with the nature of the underlying transaction.

Table 3 presents the names and definitions of the explanatory variables employed in our analysis

along with descriptive statistics for each variable. Appendix A presents a table with the corresponding

motivations for each of the explanatory variables along with a brief review of the extant literature

associated with each variable, as is relevant to mergers and takeovers. Even though these motivations are

absolutely crucial for an adequate appreciation of nuances of the selected variables, but, for the sake of

compactness and clarity, we have elected not to repeat the motivations and expected effects of these

variables in the main body of the paper. The reader needs to refer to the corresponding discussion in

Appendix A while interpreting any discussion of the relevant variables later in the paper. Two specific

comments are however necessary.

Our study makes a new and more general contribution to the measurement of directors’ influence

on company choices and the outcomes of those choices than previously seen in the literature. The

made in much the same way as the general manager-stockholder agency problem when it comes to selection of merger candidates. The qualifier is that such actions will not necessarily destroy value, but value destruction is a possibility, all that is required is that the positive change in bondholder wealth be at least equal to the absolute value of the loss in shareholder wealth.

13

primary focus in the literature has been on aggregate officer and director share ownership and its

association with corporate decision-making.19 Recent evidence suggests that the positive effects of

increased ownership by officers and directors may be attenuated at high levels of ownership (see Demsetz

and Villalonga, 2001, for a review), that is, a nonlinear relation may exist between firm value and

ownership. We suggest that an overlooked dimension of board ownership, the dispersion of ownership

stakes across board members, may potentially influence the incentives of the board as a whole and

contribute to the attenuation of the positive benefits of ownership. Suppose the total number of shares

held by a group of officers and directors is fixed but the holdings of individual directors are unequal. The

greater the inequality of board ownership the more the shares are concentrated in the hands of a relatively

small number of officers and directors. Two polar extremes are the cases in which, first, one board

member owns all the shares held by the board as a whole and, second, all of the board members hold an

equal number of shares. Clearly, these cases represent significantly different structures and could, ceteris

paribus, lead to significantly different decision-making histories. In the context of mergers, several

outcomes are possible as a result of the inequality of ownership across board members. First, greater

inequality implies that a relatively larger number of board members have low levels of ownership and

thus arguably have lesser incentive to expend effort on value-maximizing corporate decisions. This may

lead to greater abuse of power and the wasting of corporate assets, especially if those who hold this power

gain utility from overseeing larger firms assembled through takeovers. Hence, for a given level of total

holdings by the board, those firms with greater inequality of shareholdings within the group of officers

and directors may be associated with poorer decisions and thus poorer outcomes. On the other hand, cases

in which officer and board ownership is more equally distributed give no single individual undue power

and hence the negative consequences of large aggregate ownership are likely to be mitigated.20 We

explore the relation between officer and director ownership within the bidder’s board and the operating

19 Jensen and Meckling (1976) is generally regarded as the progenitor of this line of thought. Much of the empirical work in this area is reviewed in Shleifer and Vishny (1997) and Weston, Chung and Hoag (1990), Weston, Chung and Sin (1998) and Weston, Mitchell and Mulherin (2004). Llewellen, Loderer and Rosenfeld (1985), Hubbard and Palia (1995) and Datta, Iskandar-Datta and Raman (2001) investigate revaluation and ownership structure; Byrd and Hickman (1992) investigate the role of board structure. A recent paper by Masulis, Wang and Xie (2006) is the first to analyze ownership and board structure characteristics jointly in the context of expected gains for acquiring firms. Loderer and Martin (1997) present evidence suggesting that no connection exists between managerial share ownership and share price revaluations in takeovers. 20 It is also conceivable that officers and directors do truly pursue a goal of value maximization. In that scenario, greater inequality might imply that decision-making authority is concentrated in the hands of those who have a high incentive to allocate resources wisely. Hence, for a given level of total holdings by the group, those firms with greater inequality of shareholdings within the group of officers and directors may be associated with better takeover outcomes.

14

performance changes observed for the combination, as well as the association with the market’s

revaluation of the target and bidder, controlling for the distribution of share holdings across officers and

directors.

Our measure of managerial incentives is based exclusively on direct stock ownership because

consistent corporate disclosure of managerial compensation was not available for UK public companies

throughout the time period of our study. We examine the impact of the aggregate level of board

ownership (as captured by the variable BOARDOWN). However, in a departure from previous studies,

the relationship between board ownership and performance is not modeled as a one-dimensional factor,

but additionally also through the distribution of board ownership across the individual board members (as

captured by the variable DISPBOARDOWN). We hypothesize that a greater degree of concentration of

aggregate ownership in the hands of one or a relatively small fraction of the board will enable those board

members to effectively exert more control over decisions. This may have a detrimental effect on

performance when aggregate board ownership is high because of the potential for some members to

extract private benefits of control. High aggregate board ownership, however, does not necessarily lead

to value-reducing decisions when the ownership is relatively uniformly distributed across board members.

Our specification of the board ownership/ performance relationship provides an alternate and different

interpretation of the curvilinear association suggested by Morck, Shleifer and Vishny (1988) and others.

The dispersion of board ownership across the individual board members is measured using the Thiel

index of dispersion (Thiel, 1967), which is the preferred measure when dealing with variables having

decreasing marginal effects, such as BOARDOWN. The Thiel index value (T) for the dispersion of

BOARDOWN across the acquiror board is defined in the notes to table 3. A T-value of 0 indicates

BOARDOWN is uniformly dispersed across the acquiror board. In finite samples, T has an upper bound of

ln(n). DISPBOARDOWN is interacted with BOARDOWN2 in the models we estimate to capture the

hypothesized multi-dimensional curvilinear effect of BOARDOWN. The variable DISPBOARDOWN

approaches zero as ownership becomes equally distributed across board members, and conversely

approaches one as ownership becomes concentrated in the hands of a few. Therefore when ownership is

uniform the nonlinear effect of ownership is attenuated while it is maximized when ownership is

concentrated.

Our second specific comment relates to director independence and governance regulation. The

potential importance of board structure in monitoring managerial decisions is emphasized by Fama and

Jensen (1983) and Jensen (1993). However, the empirical evidence does not suggest overwhelming

support for the hypothesis that firms with more independent boards perform better (John and Senbet,

1998; Bhagat and Black, 1999; Hermalin and Weisbach, 2003). Byrd and Hickman (1992) do find that

15

acquiror stock returns are positively influenced by outsider-dominated boards but Masulis, Wang and Xie

(2006) find no statistically significant association between such board structure and acquiror returns in

mergers. For much of the time period of our study, UK company boards were dominated by insiders and

there was not an active market for outside directors. Non-executive (yet alone independent outside)

director status was not consistently disclosed in UK company annual reports in the earlier half of our

sample period. We therefore decided against including a measure of the fraction of the board represented

by outside directors as we felt the data were unreliable in this regard.

That said, approximately half of the corporate takeovers in our sample occurred at a time when

there was increasing activism in the UK for improved corporate governance regulation. This activism

was concentrated at the end of the 1980s and throughout the first half of the 1990s. In 1992 the Cadbury

Report “Code of Best Practice” in corporate governance was published in the UK. The report was

anticipated for several years prior to its publication as public debate over these issues evolved. The

proposal, which was at the time nonbinding on UK public companies, emphasized the importance of

independent, non-executive, directors on boards. The London Stock Exchange made the

recommendations binding in 1993. Most public companies quickly endorsed the proposal. We account

for this general trend in governance change and the increased emphasis on board independence by

including a control variable defined as YEAR, which equals the year of the merger announcement. One

prediction is that increases in governance intensity which arose chronologically in time during our sample

period, will be associated with better performing mergers because enhanced governance results in better

decisions. On the other hand, if forced changes in governance practices, a ‘one size fits all’ regulatory

environment, is not optimal, then the influence of forcing a board structure on a company when that

structure is suboptimal, may in turn lead to suboptimal decisions by that board. We capture the changing

effects of increased governance intensity with a time variable YEAR which is the year of the merger.

3.3 Correlations between the explanatory variables

The correlations between the explanatory variables for our sample cases reveal no surprises

(Appendix B). Acquiring firms with lower levels of industry-adjusted debt capital are associated with

greater industry-adjusted growth prospects (QRATIO’s) broadly consistent with evidence in the literature

(e.g. Smith and Watts, 1992). Acquiring firms that are more levered than their industry counterparts are,

in turn, correlated with corporate takeovers of a lesser relative size. Aggregate board ownership of

acquiring firms is positively correlated with industry-adjusted growth opportunities and negatively

correlated with firm size. Board size is positively related to the size of the firm. Hostile takeovers are

associated with larger acquiring firms. Finally, industry-adjusted growth prospects of the acquiror

relative to the firm being acquired is negatively related to the difference in cash liquidity between these

16

firms. Although it is beyond the scope of this paper to draw conclusions as to the direction of causation,

all of the above more extreme correlations are consistent with elements of existing corporate finance

theory.

4. Factors Influencing Operating Performance in Corporate Takeovers

4.1 Results

Column 1 of Table 4 presents the results of a cross-sectional linear regression of the change in

industry-adjusted operating cash flow returns on all of the hypothesized determinants of the change in

operating performance. We use the natural log of 1 plus the change in industry-adjusted operating cash

flow returns as the dependent variable for all of the results reported in Table 4. The transformation

reduces skewness but results using the untransformed data (not reported) are qualitatively similar.

We begin with a discussion of the role that corporate governance plays in how operating

performance changes for the mergers in our sample. We examine the relation between the operating

performance effects of corporate takeovers and the share ownership of the acquiring firm’s board.

Aggregate board ownership (BOARDOWN) is included in its own right. We also account for any

nonlinearities as well as the influence of the dispersion of ownership across board members by

incorporating the square of BOARDOWN interacted with the measure of ownership dispersion

DISPBOARDOWN. The hypothesis that board owners make decisions which result in improvements and

are consistent with shareholder interests predicts a positive sign on the estimated coefficient of

BOARDOWN. However, this effect may be attenuated (Morck, Shleifer and Vishny, 1988; Stulz, 1988)

if at higher levels of ownership there is a propensity for own interest (private benefits of control) to

dominate. Our arguments are that this latter effect is itself attenuated when board ownership is equally

dispersed across board members, or conversely, when board ownership is concentrated in the hands of a

few members, when the self-interest problem is most acute.

The change in industry-adjusted operating cash flow returns is significantly related to the level of

board ownership. The coefficient on aggregate board ownership is positive and statistically different

from zero at the 1% level. This result is consistent with the hypothesis proposed by Jensen and Meckling

(1976) and explored by many others that board members with positive incentives to create value will

make corporate takeover decisions that result in performance benefits (Llewellen, Loderer and Rosenfeld,

1985). However, we also find that the overall impact of aggregate board ownership is conditional on how

such ownership is dispersed across board members. The negative and statistically significant (at the 1%

level) coefficient for the interaction between aggregate board ownership (squared) and the level of

dispersion implies that for the interests of the board to be aligned with those of shareholders, board

17

ownership needs to be relatively equally dispersed across board members. At the one extreme, if board

ownership is high and uniformly dispersed across the board members then the beneficial effects of

aggregate board ownership manifest themselves fully in operating performance improvements. At the

other extreme, if board ownership is high but concentrated in the hands of one board member then the

negative effects of high board ownership will overpower the positive effects, resulting in corporate

takeover decisions that do not necessarily further the best interests of stockholders. These results suggest

that the negative effects of high board ownership (entrenchment and private benefits of control) are

conditional on the distribution of ownership across officers and board members. Our results could

potentially explain the vast array of curvilinear effects that have been observed in studies relating

managerial ownership to either firm value or the valuation effects of a corporate decision, as is

graphically illustrated in Demsetz and Villalonga (2001).

Board size, has a negative and significant association with the operating performance effects of

corporate takeovers, consistent with the predictions and general findings of Yermack (1996).21 The sign

of the estimated coefficient on the CEOCHAIR dummy variable is negative but the estimate is not

statistically significantly different from zero. We infer that, on average, top managers who also chair

their boards neither enhance nor impair operating performance changes following the mergers in our

sample. The results indicate the presence of outside voting blocks is associated with operating

performance improvements. The estimated coefficient on the variable BLOCKVOTES is positive and

statistically significant at the 1% level. The result is consistent with the hypothesis that blockholders

bring important influence to bear on operating decisions (Shleifer and Vishny ,1986). Finally, attenuating

these influences we find that mergers occurring later in the sample are associated with lower operating

performance changes. The latter result suggests that the effect of increased intensity of the governance

regulatory environment was negative, consistent with greater regulation or pseudo-regulation imposing

costs on performance.

On the other hand, more (product market) regulated acquiring firms are associated with

significant improvements in operating performance compared to lesser regulated firms. This result may

be due to the franchise granted by regulation to some firms being protected by the regulatory rules.

Managerial quality as proxied by the variable QRATIO, cash liquidity, industry-adjusted leverage, and

the size of the acquiring firm have no statistically significant association with the change in industry-

21 The natural logarithm of BOARDSIZE and SIZE is used to capture a decreasing marginal effect for both of these acquiror explanatory variables and to give SIZE a more symmetrical distribution.

18

adjusted operating cash flow returns.22

The coefficient on the variable RELLEV (the absolute difference in relative debt-use between the

acquiror and acquired firms) has a positive and statistically significant (at the 1% level) relation with the

operating performance effects of corporate takeovers. This result is consistent with the conjecture that a

mismatch in relative debt allows the possibility of new debt and consequent operating improvements from

a reduction in the agency cost of free cash flow, and/or the ability to exploit growth opportunities that had

been restricted due to a lack of funding. Arguably, this effect and our interpretation is also related to the

corporate governance of the merged firm assuming that debt is intentionally used to reduce the agency

costs of free cash flow. This interpretation would be broadly consistent with the findings of Ghosh and

Jain (2000) and Heron and Lie (2002) that acquiring firms tend to take on more debt following corporate

takeovers.

The coefficient on the variable RELQRATIO is negative and statistically different from zero at

the 1% level. The variable RELQRATIO is the ratio of the acquiror’s QRATIO to that of the acquired

and is intended to reflect the relative differences in managerial quality between the two firms. If

operating improvements come primarily from better management teams taking control of firms that have

been managed poorly, then we would expect the coefficient on RELQRATIO to be positive. The

negative sign on the coefficient suggests that this disciplinary motive is not supported for our sample of

corporate takeovers. Two potential explanations for the negative coefficient present themselves. The

negative and significant coefficient on RELQRATO could be consistent with the hypothesis that

operating performance changes are adversely effected by a mismatch of managerial quality because

poorly managed firms may be difficult to assimilate into well managed firms. Alternatively, the result

could be consistent with the view that acquiring managers are inflicted with hubris (Roll, 1986) associated

with the belief that they can turn around poorly performing firms when the chance of this occurring is

small. The coefficient on the variable RELCASHLIQ is also significantly and negatively related to the

operating performance effects of corporate takeovers. Taken together these results suggest the possibility

that, as the mismatch in managerial quality (which, in turn, is highly correlated with the mismatch in cash

liquidity) becomes more pronounced, the managers of acquiring firms become overly optimistic about

their abilities to turn around poorly performing firms.23

The relation between the relative size of the mergers in the sample (RELSIZE) and their operating

22 QRATIO and industry-adjusted leverage, LEV are negatively correlated, ρ = -.48. However, removing either variable from the multivariate analysis does not change our conclusion about the influence of the other variable in the regression.

19

performance changes is negative and statistically significant in the regression results. This result is

consistent with post-merger integration difficulties being related to the size of the firms involved. The

variable FOCUS is not significantly related to the change in industry-adjusted operating cash flow

returns.

Corporate takeovers for which the method of payment is a pure exchange of stock are negatively

and statistically associated with the change in industry-adjusted operating cash flow returns. The result is

consistent with the theory presented by Fishman (1989). Hostile takeovers do not lead to significant

reductions in performance. Finally, competition for the firm being acquired has no effect on subsequent

operating performance

The adjusted R-squared for the model shown in column 1 of Table 4 is about 30%. Moreover, the

null hypothesis that the coefficients across the model are jointly equal to zero is rejected at the 1% level.

However, because many of the variables have little explanatory power, their inclusion in the model

depresses the adjusted R-squared of the regression and the magnitude and hence significance of the F-

statistic. In column 3 of Table 4 we present the results from a regression of the change in industry-

adjusted operating cash flow returns on only those variables that add explanatory power to the model.

Notwithstanding that the adjusted R-squared and F-statistic increase in magnitude, the coefficients and

standard errors for the explanatory variables included in the reduced model are qualitatively the same as

those shown in column 1 of the table.

Diagnostic tests (not reported) on the reduced model reveal that the results are not unduly

influenced by outliers, that variance inflation factors pertinent for assessing multicollinearity are within

acceptable limits, and that the model residuals are linearly related to the change in industry-adjusted

operating cash flow returns. However, the coefficients for the corporate governance variables could be

biased if unobserved factors related to these variables are also influential in determining the operating

performance effects of corporate takeovers, that is, if ownership and performance are jointly determined

endogenous variables. We use Durbin-Wu-Hausman tests to ascertain whether the null hypothesis that

the estimated coefficients on the corporate governance variables are unbiased can be rejected. We follow

the approach proposed by Davidson and MacKinnon (1993) when formulating the test (see also Greene,

2003, p. 80-83). Following the results in Demsetz and Lehn (1985) and Yermack (1996) we use firm size

and the volatility of stock returns to model aggregate board ownership, blockholder control, and board

size in independent regressions. The dispersion of board ownership is not modeled, and is therefore

23 The variables RELQRATIO and RELCASHLIQ are correlated. The results presented in Table 5 remain qualitatively the same when we exclude either from the regression and retain the other.

20

excluded from the tests, because it is uncorrelated with aggregate board ownership and with the

instruments. Based upon these tests (not reported) we do not reject the null hypotheses that the

coefficients on board ownership, blockholder control and board size are unbiased in the results shown in

column 3 of Table 4.

4.2 Economic significance

We end this section by first presenting results on the economic significance of changes in the

statistically significant factors shown in Table 4 (column 3). We assess the change induced for each

applicable explanatory variable by computing the predicted operating performance change using the

values of the explanatory variable at the 10th and 90th percentiles of the sample empirical distribution for

the variable, holding the other variables fixed at their respective mean levels. The results of these

calculations are presented in Table 5. The results presented in Table 5 reveal that a change in each of the

variables from their level at the 10th percentile to the 90th percentile results in an average absolute change

in the operating cash flow return change of roughly 787 basis points. The effect of the change in board

ownership is to induce a 743 basis point change, while a change for board size induces a -1059 basis point

change. The largest influence arises from a change in the relative size of the target to the bidder, inducing

a -1251 basis point change.

We have intentionally excluded an assessment of the dispersion of board ownership from the

results in Table 5. The influence of ownership dispersion across board members depends upon the level

of total board ownership and is better illustrated graphically over a range of values for board ownership.

Total board ownership for the sample ranges from close to zero to roughly 50%. In Figure 1, we illustrate

the curvilinear relationship between the change in industry-adjusted operating cash flow returns and board

ownership, as predicted by the reduced model in Table 5. We plot three curves in Figure 1. One of the

curves holds the level of ownership dispersion (DISPBOARDOWN) at its sample mean level, while the

other two curves hold the variable at its 10th percentile and 90th percentile levels based upon the variable’s

empirical distribution. The value DISPBOARDOWN at its 90th percentile represents the case in which

dispersion is at its lowest amongst the three cases (ownership is concentrated in the hands of a few).

Conversely, the 10th percentile case represents the most dispersed board ownership case. The figure

clearly shows the negative consequences of higher levels of ownership are reached more quickly when

ownership across the board is highly concentrated in the hands of one or a relatively small number of

directors. Accounting for this effect therefore may be a partial explanation for the array of relationships

between ownership and firm value or performance that have been documented in the literature (see

Demsetz and Villalonga , 2001).

21

5. Factors Influencing Market Revaluations

In this section we investigate the association between the hypothesized determinants of the

operating performance consequences of corporate takeovers and the market revaluations of the companies

involved. We then seek to establish whether the component of market revaluation not explained by these

hypothesized determinants has any association with operating performance changes. This allows us to test

whether unobserved or other variables accounted for in the market’s revaluations, contribute to explaining

changes in operating performance.

5.1 Results

We estimate the expected effects of corporate takeovers using the pseudo-combined asset

revaluations of the amalgamating pairs of firms.

Column 1 of Table 6 reports coefficient estimates for a model in which total abnormal asset

revaluation is regressed on all of our explanatory variables. Only two of the hypothesized determinants of

the operating performance consequences of corporate takeovers are related to the expected performance

effects in the same way. First, board ownership is positively and significantly related to revaluation. The

coefficient on the interaction variable that accounts for potential private benefits of control at higher

levels of ownership as well as the influence of the dispersion of board ownership is negative and

significant. This reinforces the hypothesis that it is not simply the positive alignment effect alone (as

suggested by Lewellen, Loderer and Rosenfeld, 1985) of board ownership that is expected to significantly

influence performance effects, but the overall effect of board ownership depends crucially on its level and

on how it is dispersed across board members. Specifically, the adverse effects postulated to influence

managerial decision making at higher ownership levels, (Morck, Shleifer and Vishny, 1988; Hubbard and

Palia, 1995) are attenuated by how equal the ownership stakes are of the board members. Second, the

negative and significant coefficient on RELQRATO is consistent with the hypothesis that operating

performance changes are expected to be adversely effected by a mismatch of managerial quality because

bad managers may be difficult to assimilate. The result is also consistent with the hypothesis that that the

market expects acquiring managers may be potentially inflicted with hubris associated with the belief that

they can turn around poorly performing firms when the chance of this occurring is small. For this

hypothesis to be a credible explanation it would require that poorly selected mergers ultimately consume

scarce resources, and that the market understands this will happen.

Revaluation is positively related to the size of the acquiring firm and to the relative size of the

target to the acquiror, both generally consistent with the market holding the belief that larger firms have

some advantage. On the other hand revaluation is negatively related to the extent excess free cash flow is

present for acquiring firms with growth opportunities below those of firms in the same industry (a result

22

that is consistent with the theories of Jensen, 1986).

Neither board size nor the concentration of blockholder control, nor any of the other influential

determinants of the operating performance consequences of corporate takeovers are significantly related

to the revaluations at the 5% level. A quarter of the cross-sectional variation in the pseudo-combined

asset revaluations is explained by the model, and the null hypothesis that the coefficients are jointly equal

to zero is rejected at the 1% level. In column 3 of Table 6, we exclude those regressors that have

coefficients insignificantly different from zero and/or that have little or no explanatory power in the full

model. This has the effect of increasing the overall explanatory power of the model by approximately

10% and producing an F-statistic that is also significantly larger than for the full model. The conclusions

for the reduced model remain qualitatively the same as those for the full model.

In assessing the value consequences of corporate takeovers, the market does seem to incorporate

the possibility of private benefits of control at high levels of board ownership, as is evidenced by the

results for the variable measuring board ownership. However, the market significantly overestimates the

impact of both board size and relative firm size on the value effects of corporate takeovers, with larger

boards and larger targets in actual fact being associated with lesser improvements in operating

performance. When revaluing the amalgamating firms in our sample the market does not seem to take

into account that the operating performance benefits are less pronounced when the form of payment is

stock only. However, the market underestimates the impact of other firm and transaction specific

variables in gauging its expectations of the operating performance consequences of corporate takeovers.

The asset revaluations do not impound the beneficial effect of concentrated blockholder control on the

operating performance effects, nor do they reflect that acquiring firms that primarily operate in industries

attracting more regulation generate more pronounced operating performance improvements. Finally, the

operating performance benefits of amalgamating firms with a more extreme mismatch of leverage are also

not recognized by the market.

5.2 Impact of unobserved factors on the operating performance effects of corporate takeovers

The under- and over-estimation by the market of the effects of publicly observable acquiring firm

characteristics on the operating performance effects of corporate takeovers would seem to imply that there

is little connection between the actual and the expected effects. However, our analysis has so far ignored

the possible existence of unobserved factors associated with the operating performance changes, which

are however reflected in the market revaluations. We address this issue by using the residuals from the

asset revaluation regressions as estimates of the aggregate effect of any unobserved factors. If there are

unobserved factors that explain the cross-sectional variation in operating performance changes but which

are implicitly incorporated in the revaluations, then we would expect to observe a statistically significant

23

relation between the residuals from the asset revaluation regression and the changes in operating

performance returns.

In Table 7, we present coefficient estimates for models in which we regress the change in

industry-adjusted operating cash flow returns on the residual asset revaluation obtained from the models

presented in Table 6. Column 1 presents results from estimating a model for the change in industry-

adjusted operating cash flow returns which includes the residual asset revaluation only. Column 3 of

Table 8 presents results for the complete operating performance model including the residuals from the

asset revaluation regression presented in Table 6, Column 3. The coefficient for the asset revaluation is

not significantly different from zero at conventional levels in both models. All of the observed

hypothesized determinants of the operating performance effects of corporate takeovers remain almost

identical to the results presented in column 1 of Table 4.

The asset revaluation residual from the reduced model presented in Table 6 column 3 also has an

insignificant coefficient in the reduced operating performance model results presented in Column 7 of

Table 7. Once again, results for the other variables are of comparable magnitudes and significance levels

as for the model presented in column 3 of Table 4. We therefore conclude that, whatever the other factors

driving the market revaluations, they are not associated with changes in operating performance.

6. Actual versus Expected Performance Effects of Corporate Takeovers

Merger related abnormal asset revaluations should arguably reflect the market’s consensus

expectations (at the time of the merger) of the future operating cash-flow benefits or costs associated with

the amalgamation. In this context, Healy, Palepu and Ruback (1992) find that the operating performance

effects for their sample of corporate takeovers are systematically related to asset revaluations, but their

result differs from the findings reported by Ghosh (2001) who finds no relation after an equivalent

analysis. Given that the issue clearly needs further re-examination, we investigate whether the market

expectation reflected in abnormal asset revaluations is a reasonable and unbiased estimate of future

operating cash-flow performance-related effects. In particular, we examine if the market adequately

factors in the effect of governance and managerial incentive variables.

We first report, in Panel A of Table 8, the results of a cross-sectional regression of the change in

industry-adjusted operating cash flow returns on the pseudo-combined firm asset revaluations. Although

the coefficient for the asset revaluation explanatory variable is positive it is not statistically different from

zero at conventional significance levels. For robustness, we also report in Table 8 Panel B, the results of

regressing the change in industry-adjusted operating cash flow returns on the pseudo-combined equity

asset revaluations (that accounts for leverage), and also report in Table 8 Panel C, an alternative

24

specification in which we use standardized values (i.e. values obtained after subtracting the mean and

dividing by the standard deviation) of the operating performance and asset revaluation variables.

Irrespective of what transformations we use, we find that the realized abnormal cash flow performance is

not reflected, whether in an unbiased sense or otherwise, in merger related abnormal asset revaluations.

We further examine the issue by ranking the announcement period asset revaluations cross-

sectionally and dividing them into four equal categories: strongly positive, mildly positive, strongly

negative and mildly negative. The proportion of positive changes in industry-adjusted operating cash flow

returns is then calculated for each of these sub-samples and the results are presented in Table 8 Panel D. It

is clear that the proportion of positive cash flow return realisations are higher in the strongly negative

abnormal asset revaluations group relative to the strongly positive abnormal asset revaluations group!

Similarly, we rank the changes in industry-adjusted operating cash flow returns cross-sectionally, and

divide them into four equal categories: strongly positive, mildly positive, strongly negative and mildly

negative. The proportion of positive changes in announcement period asset revaluations is then calculated

for each of these sub-samples and the results are presented in Table 8 Panel E. Once again, except for the

quartile of strongly positive industry-adjusted operating cash flow returns, there is little correspondence

between industry-adjusted operating cash flow returns and announcement period asset revaluations.

The above results are consistent with the facts of the Morrison-Safeway merger reported in our

introduction section where the abnormal market value revaluation was large and positive, but the realized

cash-flow return large and negative. The results appear to suggest that the operating performance effects

for our sample of corporate takeovers are not systematically related to the asset revaluations for the

mergers in our sample.

However, we note that the results of Tables 4 and 7 indicate that many of our hypothesized

explanatory variables, and in particular, our governance and managerial incentive variables, impact

industry-adjusted operating cash flow returns and announcement period asset revaluations very

differently, suggesting the possibility that those variables are not unbiasedly factored in by the market in

forming its expectations of future cash flow performance. To test this directly, Table 9 reports the results

of estimating a regression of the change in industry-adjusted operating cash flow returns on not only the

pseudo-combined firm asset revaluations but also the hypothesized determinants we have earlier used in

Table 4 and Table 6. Table 9 reports all the variables in the first two columns, and then estimates a

reduced model featuring only the variables that have a significant relationship with the dependent

variable. Very interestingly, after accounting for our hypothesized determinants, the coefficient of

abnormal asset revaluations is now statistically significant, albeit at the 10% level, a result that is found to

be robust to different transformations or subsets of the variables involved. Our results show that the

25

market does not appear to have adequately accounted for board ownership, the interaction of board

ownership with ownership dispersion, board size, block votes, regulated industrial sectors, relative q-

ratio, relative cash liquidity, relative leverage, relative size and payments as stock. And since the signs

with which each of these variables appears in Table 9 are the same as the signs with which they appear in

Table 5, the market appears to have underestimated the impact of each of the above variables, whether

positive or negative. Hence, the market under-estimated the positive impact on cash flow performance of

board ownership, block votes, regulated industrial sectors, and relative leverage; and has also under-

estimated the negative effects of the interaction of board ownership with ownership dispersion, board

size, relative q-ratio, relative cash liquidity, relative size and payments as stock.

Table 6 shows that the market does recognize governance and other deal specific factors in

merger revaluations, but does not fully recognize all the factors influencing changes in operating

performance documented in Tables 4 and 9. Also interesting to note is that the significance of the positive

effect of board ownership (as captured by BOARDOWN) is weakened by the inclusion of abnormal

returns, but not so much it's negative effect (as captured by BOARDOWN2 * DISPBOARDOWN). The

market does indeed recognize both effects, but more so the standard alignment effect than the refined (by

dispersion) private benefits of control/entrenchment effect.

Our result that the market did not adequately account for the positive or negative effects of

governance related and other characteristics explains the difference between the results obtained for

different samples, for example the samples of Healy, Palepu and Ruback (1992) and Ghosh (2001). We

further investigate the above result by considering the Morrison-Safeway merger that we described in the

introduction. We use the model estimated ex-ante in Table 4 for UK mergers over the 1985-94 period to

predict the industry adjusted cash flow return that we would expect for the Morrison-Safeway merger.

Table 10 provides the value of the relevant variables for the Morrison-Safeway merger, and finds that the

predicted industry adjusted cash flow return is a negative 2.1%, reasonably consistent with what actually

happened, and a far cry from the associated 7.8% abnormal market-value return. And the reduction in

performance is driven mainly by two factors: first, board dispersion is much lower than its median value

of about 1; and second, the relative Q-ratio (hubris) contributes negatively. This again underscores, in the

context of this specific example, the importance of governance and managerial incentives in determining

actual merger-related operating performance.

7. Conclusions

Compared to the analysis of market revaluations of bidders and targets, relatively little is known

about the determinants of the operating performance changes associated with corporate takeovers.

26

Several studies beginning with Healy, Palepu and Ruback (1992) generally report statistically significant

improvements in the abnormal operating performance of firms involved in completed corporate takeovers.

However, the attempt to explain the economic determinants of the operating performance consequences

of corporate takeovers or the full extent of the association between the actual and expected performance

effects has received little attention. This study has sought to address these issues. Our contribution to the

literature comes in three parts. First, we explore in detail the cross-sectional determinants of operating

performance changes associated with mergers. We examine the association between characteristics of the

partners, characteristics of the merger deals and changes in operating performance of the deals, for a

sample of completed mergers. In particular, we investigate the relations between the corporate

governance characteristics of the bidders and operating performance changes. Our second contribution is

more general and explores an added dimension of ownership structure and its impact on the relation

between board ownership and valuation effects. Specifically, we propose that the dispersion of

ownership across board members acts to attenuate (equal ownership) or accentuate (ownership

concentrated in the hands of a few) the propensity to waste corporate resources in the pursuit of private

benefits from control when ownership is high. Our results show that both operating performance changes

as well as market revaluations of the bidders and targets are positively related to aggregate board

ownership but are negatively related to a factor that conditions the nonlinear effects of high board

ownership (Stulz, 1988) on the dispersion of ownership within the board. Finally, we investigate whether

the factors associated with operating performance changes are also associated with the market

revaluations of the firms involved.

We find strong statistical and economic support for a curvilinear relationship between board

ownership and operating performance change (consistent with Morck, Shleifer and Vishny, 1988 for firm

value in general, and with Hubbard and Palia, 1995 for the expected takeover performance of acquiring

firms). We show that this relationship is multi-dimensional and that the level of board ownership at

which entrenchment/ private benefits of control takes effect depends on how dispersed ownership is

across board members. Our board ownership results imply that the opportunity for extracting private

benefits of control as board ownership increases are significantly impeded the more uniformly dispersed

is the ownership across the board. The corporate governance results hold after controlling for other

acquiror and amalgamation explanatory variables previously shown to be important in explaining the

expected performance effects for the individual firms involved in corporate takeover. We further find

that other governance factors are important in explaining changes in operating performance following

mergers. Consistent with Yermack (1996) for firm value in general, board size is inversely related to

performance change. This implies that board size can be an impediment to efficient operating strategy.

27

There is also evidence consistent with outside blockholders playing an important monitoring role

(Shleifer and Vishny, 1986), but the effect is less economically pronounced than that for board size or

board ownership.

We find that the operating performance changes following corporate takeovers are significantly

less when the form of payment is stock consistent with the theory presented by Fishman (1989) and is

smaller when the size of the firm being acquired is closer to that of the acquiring firm. The latter result is

consistent with Williamson (1985) who hypothesizes that inefficiencies may arise in larger firms due to

“diminishing returns to management.” The operating performance changes are also adversely effected by

a mismatch of managerial quality, suggesting bad managers may be difficult to assimilate and/or that

acquiring managers may be potentially inflicted with hubris associated with the belief that they can turn

around poorly performing firms when the chance of this occurring is small, and that this is taken into

account by the market when valuing corporate takeover decisions. We also find that a mismatch of free

cash flow between the amalgamating pairs of firms is negatively related to changes in operating

performance. Finally we find that a mismatch of leverage between the amalgamating firms has a

beneficial impact on the operating performance effects of corporate takeovers possibly suggesting that the

potential for capital structure rearrangements may be value enhancing.24 This result is also in line with

the finding by Ghosh and Jain (2000) that the leverage of acquiring firms increases significantly from

before to after a corporate takeover, irrespective of the form of payment. The degree of industry

regulation and the timing of corporate takeovers also impact operating performance effects.

While the board ownership and mismatch of managerial quality effects are also observed in our

analysis of market revaluations, the market does not appear to adequately account for many governance-

related and other factors associated with operating performance changes. Specifically, for our data, the

market appears to have systematically underestimated the impact, whether positive or negative, of

governance related variables like board ownership, the interaction of board ownership with ownership

dispersion, board size, and block votes. As a result, merger-related abnormal asset revaluations did not

significantly predict actual industry-adjusted cash-flow returns, but the predictive ability became

significant after controlling for the operating performance changes attributable to governance-related and

other factors.

24 This could be due to a disciplining role for debt as suggested by Jensen (1986) or a coinsurance-type effect (Kim and McConnell, 1977).

28

References

Andrade, G., M. Mitchell and E. Stafford, 2001. New evidence and perspectives on mergers. Journal of

Economic Perspectives 15, 103-120. Asquith, P., R.F. Bruner and D.W. Mullins, 1983. The gains to bidding firms from merger. Journal of

Financial Economics 11, 121-140. Barber, B.M. and J. D. Lyon, 1996. Detecting abnormal operating performance: The empirical power

and specification of test statistics. Journal of Financial Economics 41, 359-399. Bebchuk, L. A., and A. Cohen, 2005. The costs of entrenched boards. Journal of Financial Economics

78,409-433. Bebchuk, L. A., A. Cohen, and A. Ferrell, 2004. What matters in corporate governance? Harvard Law

School Olin Discussion Paper No. 491. Berger, P.G. and E. Ofek, 1995. Diversification's effect on firm value. Journal of Financial Economics 37,

39-66. Berkovitch, E. and M.P. Narayanan, 1990. Competition and the medium of exchange in takeovers.

Review of Financial Studies 3, 153-174. Bhagat, S., A. Shleifer and R.W. Vishny, 1990. Hostile takeovers in the 1980’s: The return to corporate

specialization. Brookings Papers on Economic Activity: Microeconomics 1990, 1-84. Bradley, M., A. Desai and E.H. Kim, 1988. Synergistic gains from corporate acquisitions and their

division between the stockholders of target and acquiring firms. Journal of Financial Economics 21, 3-40.

Brickley, J, J. Coles and G. Jarrell, 1997. Leadership structure: separating the CEO and chairman of the

board. Journal of Corporate Finance 3, 189-220. Byrd, J.W. and K.A. Hickman, 1992. Do outside directors monitor managers? Evidence from tender

offer bids. Journal of Financial Economics 32, 195-221. Cho, M., 1998. Ownership structure, investment, and the corporate value: An empirical analysis.

Journal of Financial Economics 47, 103-121. Comment, R. and G.A. Jarrell, 1995. Corporate focus and stock returns, Journal of FinancialEconomics

37, 67-88. Core, J.E., W.R. Guay and T.O. Rusticus, 2006. Does weak governance cause weak stock returns? An

examination of firm operating performance and investor expectations. Journal of Finance 61, 655-687.

Conover W.J., 1980. Practical nonparametric statistics. J. Wiley & Sons, New York, NY.

29

Cremers, K.J. and V.B. Nair, 2005. Governance mechanisms and equity prices. Journal of Finance 60, 2859-2894.

Datta, S., M. Iskandar-Datta, and K. Raman, 2001. Executive compensation and corporate acquisition

decisions. Journal of Finance 56, 2299-2336. Davidson, R. and J. MacKinnon, 1993. Estimation and inference in econometrics. Oxford University

Press, New York. Dennis, D. and J. J. McConnell, 1986. Corporate mergers and security returns. Journal of Financial

Economics 16, 143-187. Demsetz, H. and K. Lehn, 1985. The structure of corporate ownership. Journal of Political Economy 93,

1155-1177. Demsetz, H. and B. Villalonga, 2001. Ownership structure and corporate performance. Journal of

Corporate Finance 7, 209-233. Duggal, R. and J.A. Millar, 1999. Institutional ownership and firm performance: the case of bidder

returns. Journal of Corporate Finance 5, 103-117. Eckbo, E., R. Giammarino and R. Heinkel, 1990. Asymmetric information and the medium of exchange

in takeovers: Theory and tests. Review of Financial Studies 3, 651-676. Fama, E. and M. Jensen, 1983. Separation of ownership and control. Journal of Law and Economics 26,

301-325. Fazzzari, S.M., R.G. Hubbard and B.C. Petersen, 1988, Financing constraints and corporate investment.

Brookings Papers on Economic Activity 1, 141-195. Fishman, M.J., 1989. Pre-emptive bidding and the role of the medium of exchange in acquisitions.

Journal of Finance 44, 41-57. Franks, Mayer and Renneboog 2001. Who disciplines management in poorly performing companies?

Journal of Financial Intermediation 10, 209-248. Ghosh, A., 2001. Does operating performance really improve following corporate acquisitions? Journal

of Corporate Finance 7, 151-178. Ghosh, A. and P.J. Jain, 2000. Financial leverage changes associated with corporate mergers. Journal of

Corporate Finance 6, 377-402. Gillan, S. L., and L. T. Starks, 2000. Corporate governance proposals and shareholder activism: the role

of institutional investors. Journal of Financial Economics 57, 275-305. Gompers, P., J.Ishii and A. Metrick, 2003. Corporate governance and equity prices. The Quarterly Journal

of Economics 118, 107-155. Goyal, V.K. and C.W. Park, 2002. Board leadership structure and CEO turnover. Journal of Corporate

Finance, 8, 49-66.

30

Greene, W.H., 2003, Econometric analysis, Fifth Edition. Prentice Hall, New Jersey. Harford, J., 1999. Corporate cash reserves and acquisitions. Journal of Finance 54, 1969-1997. Healy, P.M., K.G. Palepu and R.S. Ruback, 1992. Do mergers improve corporate performance? Journal

of Financial Economics 31, 135-175. Hermalin, B.E. and M.S. Weisbach, 2003. Boards of directors as an endogenously determined institution:

a survey of the economic literature. Economic Policy Review 9, 7-26. Heron, R. and E. Lie, 2002. Operating performance and the method of payment in takeovers. Journal of

Financial and Quantitative Analysis 37, 137-155. Hirshleifer, D., 1995. Mergers and acquisitions: strategic and informational issues. In: Jarrow, R., et al.

(Eds.), Handbooks in Operations Research and Management Science, vol. 9, Finance. Elsevier Science B.V., Amsterdam.

Holderness, C. G. and D. P. Sheehan, 1988. The role of majority shareholders in publicly held

corporations: An exploratory analysis. Journal of Financial Economics 20, 317-346. Hotchkiss, E. S. and D. Strickland, 2003. Does shareholder composition matter? Evidence from the

market reaction to corporate earnings announcements. Journal of Finance 58, 1469-1498. Hubbard, G., 1998. Capital-market imperfections and investment. Journal of Economic Literature, 193-

225. Hubbard, R.G., A. K. Kashyap and T.M. Whited. 1995, Internal finance and firm investment, Journal of

Money, Credit, and Banking 27, 683-701. Hubbard, R.G. and D. Palia, 1995. Benefits of control, managerial ownership, and the stock returns of

acquiring firms. RAND Journal of Economics 26, 782-792. Jarrell, G., J.A. Brickley and J. Netter, 1988. The market for corporate control: the empirical evidence

since 1980. Journal of Financial Economics 2, 49-68. Jensen, M. C., 1986. Agency costs of free cash flow, corporate finance and takeovers. American

Economic Review 76, 323-329. Jensen, M.C., 1993. Presidential address: The modern industrial revolution, exit and the failure of

internal control systems. Journal of Finance, 48, 831-880. Jensen, M.C. and W.H. Meckling, 1976. Theory of the firm: managerial behavior, agency costs, and

ownership structure. Journal of Financial Economics 3, 305-360. Jensen, M. C. and R.S. Ruback, 1983. The market for corporate control: the scientific evidence. Journal

of Financial Economics 11, 5-50. John, K., and L.W. Senbet, 1998. Corporate governance and board effectiveness. Journal of Banking &

Finance 22, 371-403.

31

Kaplan, S. and L. Zingales, 1997, Do financing constraints explain why investment is correlated with cash flow, Quarterly Journal of Economics 112, 169-215.

Karpoff, J. M., P. H. Malatesta, and R. A. Walkling, 1996. Corporate governance and shareholder

initiative: empirical evidence. Journal of Financial Economics 42, 365-395. Kim, H.E. and J.J. McConnell, 1977. Corporate mergers and the coinsurance of corporate debt. Journal

of Finance 32, 349-365. Lajoux, A.R., 1998. The art of M&A integration : a guide to merging resources, processes, and

responsibilities. McGraw-Hill, New York, NY. Lang, L.H.P., R.M. Stulz and R.A. Walkling, 1989. Managerial performance, Tobin’s Q, and the gains

from successful tender offers. Journal of Financial Economics 24, 137-154. Lang, L.H.P., R.M. Stulz and R.A. Walkling, 1991. A test of the free cash flow hypothesis: the case of

bidder returns. Journal of Financial Economics 29, 315-336. Lewellen, W., C. Loderer and A. Rosenfeld, 1985. Merger decisions and executive stock ownership.

Journal of Accounting and Economics 7, 209-231. Linn, S.C. and J.A. Switzer, 2001. Are cash acquisitions associated with better postcombination

operating performance than stock acquisitions? Journal of Banking and Finance 25, 1113-1138. Lipton, M. and J. Lorsch, 1992. A modest proposal for improved corporate governance. Business Lawyer

48, 59. Loderer, C. and K. Martin, 1997. Executive stock ownership and performance: Tracking faint traces.

Journal of Financial Economics 45, 223-255. Loughran, T. and A.M. Vijh, 1997. Do long-term shareholders benefit from corporate acquisitions?

Journal of Finance 52, 1765-1790. Manne, H., 1965. Mergers and the market for corporate control. Journal of Political Economy 75, 110-

118. Masulis, R.W., C. Wang and F. Xie, 2006. Corporate Governance and acquiror returns. Working paper

(Vanderbilt University), forthcoming Journal of Finance. McConnell, J. J. and H. Servaes, 1990. Additional evidence on equity ownership and corporate value.

Journal of Financial Economics 27, 595-612. Megginson, W.L., A. Morgan and L. Nail, 2004. The determinants of positive long-term performance in

strategic mergers: corporate focus and cash. Journal of Banking and Finance 28, 523-552. Moeller, S.B., F.P. Schlingemann and R.M. Stulz, 2004. Firm size and the gains from acquisitions.

Journal of Financial Economics 73, 201-228. Moeller, S.B., F.P. Schlingemann and R.M. Stulz, 2005. Wealth destruction on a massive scale? A study

of acquiring-firm returns in the recent merger wave. Journal of Finance 60, 757-782.

32

Morck, R., A. Shleifer and R.W. Vishny, 1988. Managerial ownership and market valuation. Journal of Financial Economics 20, 293-315.

Morck, R., A. Shleifer and R.W. Vishny, 1990. Do managerial objectives drive bad acquisitions, Journal

of Finance 45, 31-48. Palia, D., 2001. The endogeneity of managerial compensation in firm valuation: A solution. Review of

Financial Studies 14, 735-764. Pi, L. and S.G. Timme, 1993. Corporate control and bank efficiency. Journal of Banking and Finance

17, 515-530. Powell, R.G. and A.W. Stark, 2005. Does operating performance increase post-takeover for UK

takeovers? A comparison of performance measures and benchmarks. Journal of Corporate Finance 11, 293-317.

Qiu, X., 2004. Which institutional investors monitor? Evidence from acquisition activity. Working

paper, Brown University. Rau, P. and T. Vermaelen, 1998. Glamour, value, and the post-acquisition performance of acquiring

firms. Journal of Financial Economics 49, 223-253. Roll, R., 1986. The hubris hypothesis of corporate takeovers. Journal of Business 59, 197-216. Servaes, Henri, 1991. Tobin’s q and the gains from takeover. Journal of Finance 46, 409–419. Shleifer, A. and R.W. Vishny, 1986. Large shareholders and corporate control. Journal of Political

Economy 94, 461-488. Shleifer, A. and R.W. Vishny, 1988. Value-maximization and the acquisition process. Journal of

Economic Perspectives 2, 7-20. Shleifer, A. and R.W. Vishny, 1989. Management entrenchment: The case of manager-specific

investments. Journal of Financial Economics 25, 123-139. Shleifer, A. and R.W. Vishny, 1997. A survey of corporate governance. Journal of Finance 52, 737–783. Smith, C. W. and R.L. Watts, 1992. The investment opportunity set and corporate financing, dividend

and financing policies. Journal of Financial Economics 32, 262-292. Smith, R.L. and J.-H. Kim, 1994. The combined effects of free cash flow and financial slack on bidder

and target stock returns. Journal of Business 67, 281-310. Spence, M. and R. Zeckhauser, 1971. Insurance, information and individual action. American Economic

Review 61, 380-387. Stulz, R., 1988. Managerial control of voting rights: Financing policies and the market for corporate

control. Journal of Financial Economics 20, 25-44. Theil, H., 1967. Economics and information theory. North Holland, Amsterdam.

33

Travlos, N.G., 1987. Corporate takeover bids, methods of payment, and bidding firms' stock returns. Journal of Finance 42, 943-963.

Wahal, S., 1996. Pension fund activism and firm performance. Journal of Financial and Quantitative

Analysis 31, 1-23. Weston, J.F., K.S. Chung and S.E. Hoag, 1990. Mergers, restructuring and corporate control. Prentice

Hall, Englewood Cliffs, NJ. Weston, J.F., K.S. Chung and J.A. Siu, 1998. Takeovers, restructuring, and corporate governance,

Second Edition. Prentice Hall, Englewood Cliffs, NJ. Weston, J.F., M.L. Mitchell, J.H. Mulherin, 2004. Takeovers, restructuring, and corporate governance,

Fourth Edition. Pearson/Prentice-Hall, Upper Saddle River, NJ. White, H., 1980. A heteroscedasticity-consistent covariance matrix estimator and a direct test for

heteroscedasticity. Econometrica 48, 7817-7838. Williamson, O.E., 1985. The economic institutions of capitalism: Firms, markets, relational contracting.

The Free Press, New York, NY. Yermack, D., 1996. Higher market valuation of companies with a small board of directors. Journal of

Financial Economics 40, 185-212.

34

Table 1 Operating performance of merger partners pre- and post-merger Panel A: Descriptive statistics for operating performance around merger

Industry-adjusted operating cash flow return Mean Median Proportion positive

Pre-merger period -0.009 -0.009 0.42 Post-merger period 0.097*** 0.043*** 0.77*** Merger-related change 0.106*** 0.063*** 0.81*** Panel B: Linear regressions for operating performance around merger The change in industry-adjusted operating cash flow return for merger i,

{ }( )i,ePri,Posti OPCFRETOPCFRET1lnOPCFRET −+=∆ . Numbers in parentheses are t-statistics computed using White (heteroskedasticity-adjusted) standard errors. Dependent variable ( )i,PostOPCFRETln iOPCFRET∆ Constant 0.092*** 0.093***

(5.72) (5.58) ( )i,ePrOPCFRETln 0.976*** 0.090

(2.93) (0.78)

F 13.49*** 0.11

Adj. R2 0.135 -0.011

Operating performance is computed as the value-weighted average industry-adjusted operating cash flow returns for the acquiror and acquired firm using 5 year medians pre and post the merger year. The merger-related operating performance change is the difference between the post- and pre-merger industry-adjusted operating cash flow returns. ***, **, * Values are significantly different from zero (for means, medians and regression coefficients), and significantly different from 0.5 for proportions, at the 1%, 5% and 10% level respectively.

Table 2 Asset revaluations of merger partners Descriptive statistics for merger-related abnormal asset revaluations

Abnormal asset revaluations Mean Median Proportion positive Acquirors -0.003 -0.008 0.41 Acquired firms 0.168*** 0.128*** 0.91*** Pseudo combined firms 0.032*** 0.022*** 0.63**

Asset revaluations are unlevered abnormal stock returns computed from 5 and 30 trading days for the acquiror and acquired firm respectively pre their initial bid announcements, through 5 trading days post-merger completion. Pseudo combined firm asset revaluations are the value-weighted average abnormal asset returns of the acquiror and acquired firm. ***, **, * Values are significantly different from zero (for means and medians), and significantly different from 0.5 for proportions, at the 1%, 5% and 10% level respectively.

35

Table 3 Hypothesized determinants of merger-related operating performance changes Explanatory variable Predicted sign of relationship with

operating performance changes Mean Median BOARDOWN: Proportion of acquiror outstanding common stock held by directors of acquiror board in year preceding merger. (LSE filings)

Positive 0.062 0.009

DISPBOARDOWN: Thiel index valuea for the dispersion of managerial ownership BOARDOWN across acquiror board. (LSE filings)

- 1.02 0.96

BOARDOWN2 ×DISPBOARDOWN: Measure of the potential for consumption of private benefits of control conditioned on dispersion of board ownership.

Negative 0.0149 0.0001

BOARDSIZEb: Number of directors comprising acquirer board in year preceding merger. (LSE filings) Negative 9 9

CEOCHAIR: Variable is one when the highest paid acquiror director has been chair of the acquiror board for at least two years before merger, and zero otherwise. (LSE filings) Positive or Negative 0.26 0.00

BLOCKVOTES: Herfindahl concentration indexc for the proportions of acquiror votes controlled by outside blockholders, i.e. stockholders with stake of at least 5% and not included in BOARDOWN, in year preceding merger. (LSE filings)

Positive 0.014 0.000

SIZEd: Market value of acquiror assets, i.e., market value of common stock combined with the book values of other stock and total debt, at the end of the year preceding amalgamation. SIZE is in units of £ million (Consumer Price Index adjusted to 1994 values). (Datastream)

Negative 1589 421

QRATIO: Valuation ratio for the acquiror (market value of assets as defined in SIZE over the book value of total assets) at the end of the year preceding amalgamation and after subtracting the median valuation ratio for other firms in the same primary industry as the acquiror. (Datastream)

Positive -0.05 -0.03

CASHLIQ: Cash liquidity for the acquiror (book value of cash and equivalents over the market value of assets as defined in SIZE) at the end of the year preceding amalgamation and conditional on QRATIO being negative (otherwise zero). (Datastream)

Negative 0.0467 0.0008

LEV: Leverage for the acquiror (book value of total debt over market value of assets as defined in SIZE) at the end of the year preceding amalgamation and after removing the median leverage for other firms in the same primary industry as the acquiror. (Datastream)

Positive -0.022 0.000

36

Explanatory variable Predicted sign of relationship with operating performance changes Mean Median

REGULATED: Value of one if acquiror is primarily in public transport, broadcasting, cable and satellite provision, newspaper publishing, or telecommunication services; zero otherwise. (Datastream)

- 0.12 0.00

FOCUS: Value of one if the amalgamating pair of firms are in the same primary industry, otherwise zero. (Datastream)

Positive 0.32 0.00

RELQRATIO: Valuation ratio for the acquiror over that for the acquired (valuation ratios for the amalgamating pair of firms are as defined in QRATIO) at the end of the year preceding corporate amalgamation.

Positive or Negative 1.21 1.15

RELCASHLIQ: The variable is a product of two terms. The first term is the difference between the cash liquidities of the bidder and target where cash liquidity is the variable CASHLIQ. The second term is the reverse difference in investment opportunities of the two firms as measured by the difference in their QRATIO’s.

Positive -0.0085 -0.0001

RELLEV: Absolute difference between the leverage for the amalgamating pair of firms (leverage for the acquiror and acquired is as defined in LEV) at the end of the year preceding amalgamation.

Positive 0.168 0.142

RELSIZE: Market value of assets of the acquired divided by the acquiror (market values of assets for the amalgamating pair of firms are as defined in SIZE) at the end of the year preceding amalgamation

Negative 0.297 0.221

STOCK: Value of one if common stock only is the form of payment for corporate amalgamation, otherwise zero. (Securities Data Corp.)

Negative 0.11 0.00

HOSTILE: Value of one if there is opposition from the acquired directors to amalgamation with the acquiror, otherwise zero. (Securities Data Corp.)

Positive or Negative 0.23 0.00

CONTEST: Value of one if there is another firm competing with the acquiror for amalgamation with the acquired, otherwise zero. (Securities Data Corp.) Positive or Negative 0.17 0.00

a The Thiel index value (T) for the dispersion of BOARDOWN across the acquiror board is computed as: ( ) ( )µµjxn

1jjx

n1 lnT ∑= = where n is the number of directors comprising the acquiror

board, xj is the number of acquiror outstanding common shares accounted for by the beneficial interest of the jth director on the acquiror board, µ is the mean interest for the n acquiror directors, and ln is the natural logarithm. A T value of 0 indicates BOARDOWN is uniformly dispersed across the acquiror board. In finite samples, T has an upper bound of ln(n) (Theil, 1967). DISPBOARDOWN is interacted with BOARDOWN2 in the models we estimate to capture the hypothesized multidimensional curvilinear effect of BOARDOWN. b The natural logarithm of BOARDSIZE is used in all subsequent analyses to account for a potentially decreasing marginal effect from adding more directors to the board. c The Herfindahl index value (H) for the concentration of the proportions of acquiror common votes controlled by outside blockholders is computed as follows: ∑= =

J1j

2jxH where xj is the

proportion of acquiror common votes controlled by the jth outside blockholder. The larger the value of H the higher is the concentration of the proportions of acquiror common votes controlled by outside blockholders. d The natural logarithm of SIZE is used throughout.

37

Table 4 Factors influencing merger-related operating performance changes: Linear regression coefficient estimates and model summary statistics Dependent variable is the change in industry-adjusted operating cash flow return for merger i, iOPCFRET∆ . Explanatory variable Full regression Reduced regression

(1) (2) (3) (4) Constant 33.766* (1.85) 34.041** (2.03) BOARDOWN 0.807*** (2.82) 0.698*** (2.83) BOARDOWN2 × DISPBOARDOWN -1.745*** (-3.62) -1.601*** (-4.18)

BOARDSIZE -0.152** (-2.54) -0.136*** (-2.93) CEOCHAIR -0.006 (-0.27) BLOCKVOTES 0.566*** (2.71) 0.580*** (2.83) YEAR -0.016* (-1.83) -0.016** (-2.01) SIZE 0.013 (1.08) QRATIO 0.003 (0.62) CASHLIQ -0.090 (-0.65) LEV -0.142 (-1.41) REGULATED 0.071* (1.69) 0.098** (2.19) FOCUS -0.010 (-0.45) RELQRATIO -0.085*** (-3.43) -0.061** (-2.65) RELCASHLIQ -0.316* (-1.84) -0.310* (-1.88) RELLEV 0.351*** (3.47) 0.332*** (3.47) RELSIZE -0.157*** (-2.67) -0.181*** (-3.83) STOCK -0.083*** (-3.04) -0.081*** (-3.27) HOSTILE -0.013 (-0.46) CONTEST 0.012 (0.39) F 2.83*** 4.69*** Adj. R2 0.303 0.336 The table reports OLS estimation statistics. Dependent variable: The change in industry-adjusted operating cash flow return for merger i, { }( )i,ePri,Posti OPCFRETOPCFRET1lnOPCFRET −+=∆ . All other variables are defined in Table 3, except YEARi, which is the announcement year for merger i. Estimated coefficient t-statistics are computed using White (heteroskedasticity-adjusted) standard errors and are reported in columns (2) and (4) in parentheses. ***, **, * Values are significantly different from zero at the 1%, 5% and 10% level respectively.

38

Table 5 Predicted values of the merger-related change in the industry-adjusted operating cash flow return Explanatory variable 10th percentile 90th percentile Difference BOARDOWN 0.074 0.148 0.074 BOARDSIZE 0.158 0.052 -0.105 BLOCKVOTES 0.102 0.122 0.019 YEAR 0.157 0.072 -0.084 REGULATED 0.098 0.197 0.098 RELQRATIO 0.143 0.076 -0.067 RELCASHLIQ 0.126 0.099 -0.026 RELLEV 0.063 0.167 0.103 RELSIZE 0.158 0.032 -0.125 STOCK 0.119 0.038 -0.081

Predictions are based upon evaluating the estimated model shown in column (3) of Table 4. Each row of the table shows the prediction computed using either the level of the respective variable at the 10th or 90th percentile of the variable’s empirical sample distribution while fixing all other variables in the model at their sample mean values.

39

Table 6 Factors influencing abnormal asset revaluations: Linear regression coefficient estimates and model summary statistics Dependent variable is the value-weighted average announcement period abnormal asset return of the acquiror and acquired firm for merger i, iASSETREV . Explanatory variable Full regression Reduced regression (1) (2) (3) (4) Constant 5.988 (1.06) -0.116*** (-3.35) BOARDOWN 0.616*** (3.49) 0.626*** (3.23) BOARDOWN2 × DISPBOARDOWN -0.934*** (-2.96) -0.923*** (-2.77)

BOARDSIZE 0.050 (1.56) CEOCHAIR 0.021 (1.32) BLOCKVOTES -0.185 (-0.95) YEAR -0.003 (-1.11) SIZE 0.014* (2.00) 0.021*** (3.84) QRATIO 0.001 (0.25) CASHLIQ -0.215** (-2.63) -0.226 ** (-2.56) LEV 0.025 (0.39) REGULATED -0.008 (-0.33) FOCUS 0.002 (0.12) RELQRATIO -0.029* (-1.77) -0.034** (-2.57) RELCASHLIQ 0.011 (0.12) RELLEV 0.017 (0.35) RELSIZE 0.170*** (2.67) 0.166*** (3.05) STOCK 0.040 (1.28) HOSTILE 0.003 (0.15) CONTEST 0.000 (0.00) F 2.28*** 6.06*** Adj. R2 0.233 0.275 The table reports OLS estimation statistics. Dependent variable: Combined announcement period abnormal asset revaluation for merger i. All other variables are defined in Table 3, except iYEAR , which is the announcement year for merger i. Estimated coefficient t-statistics are computed using White (heteroskedasticity-adjusted) standard errors and are reported in columns (2) and (4) in parentheses. ***, **, * Values are significantly different from zero at the 1%, 5% and 10% level respectively.

40

Table 7 Factors influencing merger-related operating performance changes: Regressionsafter controlling for residuals in abnormal asset revaluation regressions Dependent variable is the change in industry-adjusted operating cash flow return for merger i, iOPCFRET∆ . Explanatory variable Full regression Reduced regression (1) (2) (3) (4) (5) (6) (7) (8)

Constant 0.092*** (6.39) 33.766* (1.92) 0.092*** (6.25) 32.536* (1.80)

ASSETREVResidual 0.270 (1.20) 0.270 (1.62) 0.044 (0.20) 0.212 (1.26)

BOARDOWN 0.807*** (2.95) 0.681*** (2.54) BOARDOWN2 × DISPBOARDOWN -1.745*** (-3.84) -1.574*** (-3.85)

BOARDSIZE -0.152*** (-2.64) -0.145*** (-2.90)

CEOCHAIR -0.006 (-0.28)

BLOCKVOTES 0.566*** (2.85) 0.612*** (2.87)

YEAR -0.016* (-1.91) -0.016* (-1.78)

SIZE 0.013 (1.04)

QRATIO 0.003 (0.61)

CASHLIQ -0.090 (-0.66)

LEV -0.142 (-1.46)

REGULATED 0.071 (1.66) 0.100** (2.00)

FOCUS -0.010 (-0.46)

RELQRATIO -0.085*** (-3.46) -0.061** (-2.52)

RELCASHLIQ -0.316* (-1.85) -0.314* (-1.79)

RELLEV 0.351*** (3.55) 0.329*** (3.21)

RELSIZE -0.157** (-2.42) -0.183*** (-3.49)

STOCK -0.083*** (-2.84) -0.089*** (-3.22)

HOSTILE -0.013 (-0.47)

CONTEST 0.012 (0.41)

F 1.32 2.83*** 0.04 4.41***

Adj. R2 0.004 0.313 0.000 0.339

The table reports OLS estimation statistics. Dependent variable: The change in industry-adjusted operating cash flow return for merger i, { }( )i,ePri,Posti OPCFRETOPCFRET1lnOPCFRET −+=∆ . All other variables are defined in Table 3, except

i,sidualReASSETREV , which is the residual for merger i from either the regression for which results are presented in columns (1)-(2) or columns (3)-(4) of Table 7, and iYEAR , which is the announcement year for merger i. Estimated coefficient t-statistics are computed using White (heteroskedasticity-adjusted) standard errors and are reported in columns (2), (4), (6) and (8) in parentheses. ***, **, * Values are significantly different from zero at the 1%, 5% and 10% level respectively.

41

Table 8 Relation between the merger-related change in industry-adjusted operating cash flow return and abnormal revaluations The change in industry-adjusted operating cash flow return for merger i,

{ }( )i,ePri,Posti OPCFRETOPCFRET1lnOPCFRET −+=∆ . The abnormal revaluation for merger i, ASSETREVi (EQUITYREVi), is the value-weighted average announcement period abnormal asset (stock) return of the acquiror and acquired firm. Panel A: Linear regression of the change in industry-adjusted operating cash flow return on abnormal asset revaluations Numbers in parentheses are t-statistics computed using White (heteroskedasticity-adjusted) standard errors. ***, **, * Values significantly different from zero at the 1%, 5% and 10% level respectively.

∆OPCFRETi = 0.089*** + 0.070 ASSETREVi F: 0.15 Adj. R2: -0.010 (5.89) (0.50) Panel B: Linear regression of the change in industry-adjusted operating cash flow return on abnormal equity revaluations Numbers in parentheses are t-statistics computed using White (heteroskedasticity-adjusted) standard errors. ***, **, * Values significantly different from zero at the 1%, 5% and 10% level respectively.

∆OPCFRETi = 0.090*** + 0.024 EQUITYREVi F: 0.05 Adj. R2: -0.012 (6.01) (0.27) Panel C: Standardised regression of the change in industry-adjusted operating cash flow return on abnormal asset revaluations Numbers in parentheses are t-statistics computed using White (heteroskedasticity-adjusted) standard errors. ***, **, * Values significantly different from zero at the 1%, 5% and 10% level respectively.

∆OPCFRETi = -0.000 + 0.039 ASSETREVi F: 0.12 Adj. R2: -0.011 (0.00) (0.51) Panel D: Proportion of positive changes in industry-adjusted operating cash flow return for group levels of abnormal asset revaluations Numbers in parentheses are group level sizes. Abnormal asset revaluations

Strongly positive

Midly positive

Mildly Negative

Strongly negative

Proportion of positive changes in industry -adjusted operating cash flow return 0.81 0.88 0.67 0.87

(26) (25) (15) (15)

Panel E: Proportion of positive abnormal asset revaluations for group levels of changes in industry-adjusted operating cash flow return Numbers in parentheses are group level sizes.

Changes in industry-adjusted operating cash flow return

Strongly positive

Midly positive

Mildly Negative

Strongly negative

Proportion of positive abnormal asset revaluations 0.76 0.55 0.50 0.57

(33) (33) (8) (7)

42

Table 9 Regressions of merger-related operating performance changes on abnormal asset revaluations after including governance-related and other relevant factors Dependent variable is the change in industry-adjusted operating cash flow return for merger i, iOPCFRET∆ . Explanatory variable Full regression Reduced regression (1) (2) (3) (4) Constant 32.144* (1.82) 32.420* (1.99) ASSETREV 0.270 (1.62) 0.295* (1.86) BOARDOWN 0.640** (2.02) 0.565** (2.02) BOARDOWN2 × DISPBOARDOWN -1.492*** (-2.86) -1.391*** (-3.27)

BOARDSIZE -0.165*** (-2.77) -0.159*** (-3.31) CEOCHAIR -0.012 (-0.54) BLOCKVOTES 0.617*** (3.17) 0.657*** (3.41) YEAR -0.015* (-1.80) -0.016* (-1.97) SIZE 0.009 (0.74) QRATIO 0.002 (0.53) CASHLIQ -0.032 (-0.23) LEV -0.149 (-1.52) REGULATED 0.074* (1.71) 0.097** (2.17) FOCUS -0.011 (-0.48) RELQRATIO -0.077*** (-3.24) -0.056** (-2.57) RELCASHLIQ -0.319* (-1.87) -0.325* (-1.95) RELLEV 0.347*** (3.50) 0.332*** (3.52) RELSIZE -0.203*** (-3.23) -0.220*** (-4.90) STOCK -0.094*** (-3.21) -0.092*** (-3.54) HOSTILE -0.014 (-0.50) CONTEST 0.012 (0.41) F 2.83*** 4.63*** Adj. R2 0.313 0.352 The table reports OLS estimation statistics. Dependent variable: The change in industry-adjusted operating cash flow return for merger i, { }( )i,ePri,Posti OPCFRETOPCFRET1lnOPCFRET −+=∆ . All other variables are defined in Table 3, except iASSETREV , which is the combined announcement period abnormal asset revaluation for merger i, and iYEAR , which is the announcement year for merger i. Estimated coefficient t-statistics are computed using White (heteroskedasticity-adjusted) standard errors and are reported in columns (2) and (4) in parentheses. ***, **, * Values are significantly different from zero at the 1%, 5% and 10% level respectively.

43

Table 10 Predicted operating performance change for the Morrison Supermarkets PLC and Safeway PLC merger Fitted Table 4 reduced model predicting the change in industry-adjusted operating cash flow return for the Morrison-Safeway merger ∆OPCFRETi = 34.0413 + 0.6984 BOARDOWNi -1.6010 BOARDOWNi

2 × DISPBOARDOWNi -0.1369 BOARDSIZEi + 0.5808 BLOCKVOTESi -0.0169 YEARi + 0.0986 REGULATEDi -0.0614 RELQRATIOi -0.3107 RELCASHLIQi + 0.3325 RELLEVi -0.1811 RELSIZEi -0.0811 STOCKi = -0.0213

Explantory variable Morrison (acquiror) Safeway (acquired firm)

BOARDOWN 0.071 - DISPBOARDOWN 1.83 -

BOARDSIZE ln(7) - BLOCKVOTES 0.004 -

YEAR 1985 - REGULATED 0 -

QRATIO (not industry-adjusted) 2.12 1.20 CASHLIQ (not conditional on

QRATIO being negative) 0.0501 0.0367

LEV (not industry-adjusted) 0.176 0.476 SIZE 3772 5861

STOCK 0 -

44

-0.10

-0.05

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40

Predicted ∆OPCFRET

BOARDOWN

Fig. 1. Predicted merger-related change in industry-adjustedoperating cash flow return as a function of board ownership anddispersion of board ownership fitted from the Table 5 reduced model(all other variables held at their mean values).

DISPBOARDOWN held at its mean valueDISPBOARDOWN held at its 10th percentile valueDISPBOARDOWN held at its 90th percentile vlaue

i

Appendix A Hypothesized determinants of merger-related operating performance changes Panel A: Governance and managerial incentive variables Explanatory variable definitions (and data sources) Motivation and review of relevant literature

Managerial ownership, BOARDOWN Proportion of acquiror outstanding common stock held by directors of acquiror board in year preceding merger. (LSE filings)

Jensen and Meckling (1976) emphasize importance of aligning interests of managers and stockholders through greater ownership stake of managers. Studies documenting a positive relationship between managerial incentives and post-merger revaluation include Lewellen et al. (1985) on association with direct managerial ownership; and Datta et al. (2001), on association with equity-based compensation. Morck et al. (1988) propose and find that managerial ownership does not necessarily have a wholly positive linear relationship with firm value; and Stulz (1988) presents a theory of behavior supporting a non-linear relation between ownership and value. These authors argue that, with sufficiently high level of ownership, managers’ private benefits from control can exceed the costs they share as stockholders for their errant decision-making, with potentially negative consequences for firm value. Morck et.al. (1988) and Hubbard and Palia (1995) find empirical support for this non-linear relationship for corporate takeovers. However, Loderer and Martin (1997) have shown that an empirical relationship leading from ownership to firm value or changes in firm value from managerial actions can be spurious when the former is treated as an exogenous variable.

Dispersion of managerial ownership, DISPBOARDOWN Thiel index valuea for the dispersion of managerial ownership BOARDOWN across acquiror board. (LSE filings) BOARDOWN2 ×DISPBOARDOWN Measure of the potential for consumption of private benefits of control conditioned on dispersion of board ownership. BOARDSIZEb Number of directors comprising acquiror board in year preceding merger. (LSE filings)

Smaller boards may be arguably better at monitoring managers because it is easier and quicker to reach a consensus in smaller groups. Jensen (1993) and Lipton and Lorsch (1992) have argued that large boards are dysfunctional. Yermack (1996) documents a positive small board size effect on firm performance in general, but Masulis et al. (2006) find no relation between board size and acquiror stock returns in merger settings.

CEOCHAIR Variable is one when the highest paid acquiror director has been chair of the acquiror board for at least two years before merger, and zero otherwise. (LSE filings)

CEO’s who also chair the board have more power over board decisions. One hypothesis is that such an arrangement is good if it improves decision making by concentrating information and decision authority. However, such an arrangement can also be poor when too much power is concentrated in the hands of an individual set on maximizing her own self-interest. Jensen (1993, p. 866) for instance argues that “…for the board to be effective, it is important to separate the CEO and chairman positions.” Brickley et al. (1997) find that firms in which CEO is also Chair do not perform worse. Pi and Timme (1993), Baliga et al. (1996) and Rechner and Dalton (1991) present results suggesting that firms that combine the positions perform worse than firms that do not. Goyal and Park (2002) on the other hand find general support for the power-reducing notion of separating the positions of CEO and Chair. For takeovers, Masulis et al. (2006) do not observe that acquiror stock returns are influenced by how their board leadership arrangements are structured.

ii

Hypothesized determinants of merger-related operating performance changes (continued) Panel A: Governance and managerial incentive variables (continued) Explanatory variable definitions (and data sources) Motivation and review of relevant literature

BLOCKVOTES Herfindahl concentration indexc for the proportions of acquiror votes controlled by outside blockholders, i.e. stockholders with stake of at least 5% and not included in BOARDOWN, in year preceding merger. (LSE filings)

Large stockholders can either substitute or complement inside ownership in aligning the interests of managers and stockholders. If blockholders are more inclined than other stockholders to monitor boards because small stockholders find it is more cost effective to free-ride (Shleifer and Vishny, 1986), then, for fear of being disciplined, boards are more likely to make corporate takeover decisions that are in the interests of their stockholders when blockholders are present. The empirical evidence on the influence of large stockholders in the context of corporate takeover decisions is mixed. Duggal and Miller (1999) find no systematic relationship between institutional ownership and acquiror announcement period returns after accounting for the determinants of institutional ownership in a first stage regression. The passivity of institutional stockholders is also supported by the results of Masulis, Wang and Xie (2006), Cremers and Nair (2005), Gillan and Starks (2000), Karpoff, Malatesta and Walkling (1996), and Wahil (1996). However, Qiu (2004) finds that public pension funds, perceived to be more active than other types of institutional investors, are associated with corporate takeovers that generate improvements consistent with the conclusions drawn by Jensen (1993) on the importance of such investors. Qui reports however that there is no association between such holdings and the market revaluations of the bidder and target.

Panel B: Acquiror characteristics Explanatory variable definitions (and data sources) Motivation and review of relevant literature

SIZEd Market value of acquiror assets, i.e., market value of common stock combined with the book values of other stock and total debt, at the end of the year preceding amalgamation. SIZE is in units of £ million (Consumer Price Index adjusted to 1994 values). (Datastream)

Evidence in Moeller, Schlingemann and Stulz (2004) suggests that abnormal revaluations of large acquirers are more negative. There are several reasons why size may be an important determinant of a merger’s operating performance. First, size may proxy for the level of bureaucracy in the acquiror’s organization and hence be an indicator of how effectively any investment undertaken by the acquiror will be implemented (Williamson, 1985). Second, size may reflect the extent to which economies of scale or scope can be captured by the acquiror. Third, the results in Moeller et al. (2004) suggest that because larger acquirers pay higher premiums these firms may exhibit greater managerial hubris (Roll, 1986). As argued earlier, decisions based upon errors may consume scarce resources leading to declines in productivity. Fourth, large firms may be large because their equity is overvalued and may be more likely to pay with stock, eliciting a market response similar to that accompanying an equity issue (Myers and Majluf, 1984). Fifth, large firms may be associated with more surplus free cash flow, and hence, in those cases where managers are not fully disciplined, may be associated with investments that waste assets (Jensen, 1986). Finally, Masulis, Wang and Xie (2006) suggest that because larger firms are less likely to be takeover targets they are subject to less discipline from the market for corporate control and hence the managers of larger firms may be more inclined to make decisions based upon personal preferences rather than value creation.

iii

Hypothesized determinants of merger-related operating performance changes (continued) Panel B: Acquiror characteristics (continued)Explanatory variable definitions (and data sources) Motivation and review of relevant literature

QRATIO Valuation ratio for the acquiror (market value of assets as defined in SIZE over the book value of total assets) at the end of the year preceding amalgamation and after subtracting the median valuation ratio for other firms in the same primary industry as the acquiror. (Datastream)

Lang, Stulz and Walking (1989, 1991) and Servaes (1991) suggest that managerial quality may influence the outcomes of mergers and that a proxy for quality is Tobin’s q. We use a pseudo q ratio for an acquiror (ratio of pseudo market value to the total assets of the firm, where the market value of assets is computed as total assets minus the book value of equity plus the market value of equity). We account for quality controlling for industry effects. The results relating the pseudo q-ratio to acquiror returns is mixed. Lang, Stulz and Walkling (1989, 1991) and Servaes (1991) find a positive association between q and returns while Moeller, Schlingemann and Stulz (2004) and Masulis, Wang and Xie (2006) find a negative relation. Managers who have exhibited the ability to create more wealth than their industry counterparts are defined here as having superior quality and hence the hypothesis is that such managers will select to do mergers which exhibit the same potential and will be associated with improvements in performance.

CASHLIQ Cash liquidity for the acquiror (book value of cash and equivalents over the market value of assets as defined in SIZE) at the end of the year preceding amalgamation and conditional on QRATIO being negative (otherwise zero). (Datastream)

Jensen (1986) has argued that managers of firms with excess cash may engage in activities that waste assets such as expenditures or investments that bring them private benefits (utility) but which do not enhance shareholder wealth (see also Morck, Shleifer and Vishny, 1990). Jensen labels the value loss from such choices the agency cost of free cash flow. Lang, Stulz and Walkling (1991) have suggested that firms with poor growth opportunities but which are large cash generators are especially prone to such exploitation. Managers of acquiring firms with few investment opportunities but high cash flow may face incentives to invest in value destroying mergers if those managers receive private benefits from controlling larger firms. The CASHLIQ variable is a control for this agency problem. We define a dummy variable for managerial quality that takes the value 1 whenever the industry-adjusted QRATIO for the acquiror is less than the industry median (and zero otherwise). CASHLIQ is the product of managerial quality dummy and the acquiror’s cash liquidity ratio.

LEV Leverage for the acquiror (book value of total debt over market value of assets as defined in SIZE) at the end of the year preceding amalgamation and after removing the median leverage for other firms in the same primary industry as the acquiror. (Datastream)

The agency cost of free cash flow may be attenuated through the use of debt which forces the disgorging of cash, hence reducing the potential for wasteful expenditures (Jensen, 1986). We measure the extent to which debt is an enforcing mechanism in reducing the free cash flow problem in acquirers by the deviation of the acquiror’s relative debt financing choice from an estimate of a normal relative debt financing level. Attenuation of the free cash flow problem would suggest that acquisitions which waste assets are less likely, and hence that acquiror leverage, ceteris paribus, should be associated with improvements in operating performance.

iv

Hypothesized determinants of merger-related operating performance changes (continued) Panel B: Acquiror characteristics (continued) Explanatory variable definitions (and data sources) Motivation and review of relevant literature

REGULATED Value of one if acquiror is primarily a public transport firm or broadcasting contractor or cable and satellite provider or newspaper publisher or telecommunication services operator, zero otherwise. (Datastream)

Roughly 12% of the acquiring firms in our sample operated in the public transport, media, or telecommunication industries and are therefore deemed to be under greater regulatory scrutiny relative to other firms represented in the sample. We control for cases in which the acquiror operated at least partially in these regulated sectors with this variable.

Panel C: Relative attributes of acquiror and target Explanatory variable definitions (and data sources) Motivation and review of relevant literature

FOCUS Value of one if the amalgamating pair of firms are in the same primary industry, otherwise zero. (Datastream)

Amalgamating firms that operate primarily in closely related industries are in a better position to generate operational synergy through economies of scale and scope when combining their businesses (Weston, Chung and Hoag, 1990). Healy, Palepu and Ruback (1992) find weak evidence for such a relation, while Megginson, Lance and Nail (2004) and Heron and Lie (2002) find a significant positive relation between changes in operating performance and measures of whether the acquiror and the target operate in the same industry. Numerous authors present evidence suggesting that a lack of focus (eg. when a firm operates several divisions in unrelated lines of business) is associated with a value discount. See for instance Berger and Ofek (1995) and Comment and Jarrell (1995).

RELQRATIO Valuation ratio for the acquiror over that for the acquired (valuation ratios for the amalgamating pair of firms are as defined in QRATIO) at the end of the year preceding corporate amalgamation.

Operational improvements may arise from acquiring and turning around poorly performing firms as originally articulated by Manne (1965). This necessitates that the managerial quality of the acquiring firm be relatively better than that of the firm being acquired. Lang, Stulz and Walkling (1989, 1991) and Servaes (1991) use a measure of Tobin’s q to proxy for managerial quality and observe that acquiror stock returns around corporate takeover announcements are larger when acquiring firms have high market-to-book ratios and the firms being acquired have low market to book ratios. We suggest that if this hypothesis is true then when poorly performing (quality) firms are acquired by high quality firms operating performance should improve and value should be created. Support for the beneficial effects of a relative mismatch between the managerial performance of acquiring firms and firms being acquired is suggested by the results of Heron and Lie (2002). In contrast if poorly managed firms are difficult (costly) to assimilate, then the relation between RELQRATIO and the change in operating performance may be negative. Likewise if acquiring firm managers who have previously performed well are inflicted with hubris (Roll, 1986) associated with the belief that they can turn around poorly performing firms when the chance of this occurring is small, then the relation between RELQRATIO and the change in operating performance may be negative. That is, resources are wasted in an effort to turn around the acquired firm.

v

Hypothesized determinants of merger-related operating performance changes (continued) Panel C: Relative attributes of acquiror and target (continued)Explanatory variable definitions (and data sources) Motivation and review of relevant literature

RELCASHLIQ The variable is a product of two terms. The first term is the difference between the cash liquidities of the bidder and target where cash liquidity is the variable CASHLIQ. The second term is the reverse difference in investment opportunities of the two firms as measured by the difference in their QRATIO’s.

Opportunities for liquidity sharing may present themselves if some firms are cash constrained while others are not. While the evidence is not completely settled, there is empirical support for the hypothesis that some firms are cash constrained (Fazzzari, Hubbard and Petersen, 1988; Hubbard, Kashyap and Whited, 1995; Hubbard, 1998, but see Kaplan and Zingales (1997) for an alternative view). Situations may present themselves in which value-creating, unexploited investment opportunities may become feasible by bringing together a cash-rich firm with low growth opportunities and a cash-poor firm with high growth prospects, of which in this instance the acquiring firm and the firm being acquired could be either. Smith and Kim (1994) provide evidence to support this proposition in the context of the revaluation effects around announcements of corporate takeovers.

RELLEV Absolute difference between the leverage for the amalgamating pair of firms (leverage for the acquiror and acquired is as defined in LEV) at the end of the year preceding amalgamation.

Financial benefits may result from combining firms with a relative mismatch in their debt capacities if such a mismatch provides the opportunity for expanding debt use. Corporate takeovers have been shown to be debt-increasing events by Ghosh and Jain (2000) and Heron and Lie (2002). Debt increasing events may be associated with operating improvements for several reasons including a reduction in the agency cost of debt, and/or the ability to exploit otherwise liquidity restricted growth opportunities.

RELSIZE Market value of assets of the acquired divided by the acquiror (market values of assets for the amalgamating pair of firms are as defined in SIZE) at the end of the year preceding amalgamation.

The size of the target may matter in terms of the smooth integration of the two firms involved. The integration of a large firm into another large firm may be more difficult than the integration of a small firm into a large firm. Williamson (1985) argues that managerial incentives are impaired when market transactions are replaced by intra-firm transactions and that as a firm becomes larger it becomes more bureaucratic. He suggests there is a two-part cost of bureaucratization. First, well-intentioned managers regularly take on complex problems that are beyond their abilities to manage. Williamson suggests a manager’s cognitive abilities are not infinite and as the firm grows a manager’s finite abilities are stretched. Second, managers tend to pursue personal goals or a hidden agenda in larger more bureaucratic organizations (see also Shleifer and Vishny, 1988, 1989). It may also be the case that large amalgamations result in post-takeover integration problems with a concomitant adverse effect on combined performance. We control for this effect using the relative size of the target to the bidder.

vi

Hypothesized determinants of merger-related operating performance changes (continued) Panel D: Transaction-specific variables Explanatory variable definitions (and data sources) Motivation and review of relevant literature

STOCK Value of one if common stock only is the form of payment for corporate amalgamation, otherwise zero. (Securities Data Corp.)

Fishman (1989), Berkovitch and Narayanan (1990) and Eckbo, Giammarino and Heinkel (1990) have argued that cash financed amalgamations should be expected to generate larger performance improvements than corporate takeovers for which the means of payment is stock. One explanation is that because, with shareholders being more partial to cash offers and potential competing acquiring firms being less likely to enter the fray when cash is offered, cash offers are generally quicker to complete and the acquiring firm is therefore less likely to be forced to overpay for the expected benefits of the corporate takeover. Travlos (1987) finds that the method- of-payment is associated with takeover announcement period stock returns and Linn and Switzer (2001) find an association with merger operating performance changes. Heron and Lie (2002) however find no relation between the method-of-payment and changes in operating performance for a sample of U.S. takeovers.

HOSTILE Value of one if there is opposition from the acquired directors to amalgamation with the acquiror, otherwise zero. (Securities Data Corp.)

Hostile corporate takeovers are generally viewed as being a threat to inefficient target managers whereas friendly offers are often viewed as emerging when positive synergies are present and result in arrangements in which both the bidder and the target benefit (Morck, Shleifer and Vishny, 1988, 1989). Schwert (2000) argues that the choice to engage in a hostile offer is a strategic move and that economically there is little pre-bid difference between hostile and friendly offers. Corporate takeovers which are hostile may however result in post-amalgamation integration problems if trust between the people operating the organizations is diminished as a result of hostilities. This could dampen or eliminate any previously expected improvements in performance. Post merger human relations and integration problems are repeatedly cited by practitioners as the key reasons mergers fail (Lajoux, 1998). Alternatively, if hostile offers tend to be made by acquirers with better knowledge of the target and its industry as suggested by Bhagat, Shleifer and Vishny (1990), such offers could be associated with positive improvements in performance.

CONTEST Value of one if there is another firm competing with the acquiror for amalgamation with the acquired, otherwise zero. (Securities Data Corp.)

Bradley, Desai and Kim (1988) show that the division of takeover gains between the acquiror and the acquired firm is influenced by competition amongst bidders. Berkovitch and Narayanan (1990) find that total revaluations of targets and bidders taken together are greater when there is competition for the target. The relation between whether a contest for the target occurs and the change in operating performance may be either positive or negative. Competition may bring out bidders with more potential for creating synergies, but may also cause bidders to over pay (the winner’s curse or hubris). If over paying consumes valuable resources with no subsequent gain, then competition may reduce ex post operating performance.

vii

Appendix B Correlation matrix for hypothesized determinants of merger-related operating performance changes

[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18]

[2] 0.03 [3] -0.52* 0.16 [4] -0.15 0.04 0.19* [5] -0.01 -0.13 -0.17 0.01 [6] 0.06 -0.09 -0.05 0.11 0.11 [7] -0.66* 0.06 0.69* 0.23* -0.21* 0.10 [8] 0.29* 0.01 -0.23* -0.12 -0.08 -0.11 -0.13 [9] -0.09 -0.12 0.19* 0.10 0.02 0.23* 0.21* -0.23* [10] -0.27* -0.01 0.23* 0.12 -0.07 0.03 0.16 -0.48* 0.23* [11] -0.04 0.08 0.08 0.12 -0.12 -0.12 0.05 0.10 -0.11 -0.25* [12] 0.12 0.23* -0.03 -0.04 -0.11 -0.02 -0.16 0.15 -0.10 -0.22* 0.06 [13] 0.26* -0.15 -0.24* -0.04 -0.07 0.03 -0.05 0.45* -0.07 -0.26* -0.22* -0.10 [14] -0.07 0.09 0.01 0.06 0.07 -0.06 0.02 -0.12 0.06 0.13 -0.11 0.00 -0.31* [15] 0.07 -0.04 -0.02 -0.10 0.08 -0.03 -0.03 0.04 0.03 0.03 -0.13 0.02 0.13 -0.01 [16] 0.62* 0.03 -0.47* -0.14 0.18 -0.15 -0.63* 0.19* -0.02 -0.30* 0.05 0.06 0.21* -0.03 0.03 [17] 0.15 0.01 -0.07 -0.12 0.01 -0.15 -0.16 -0.01 -0.12 -0.07 -0.01 0.09 0.03 0.15 0.01 0.06 [18] -0.01 -0.10 0.26* 0.14 -0.11 -0.04 0.29* 0.01 0.19* -0.02 0.06 -0.07 0.05 0.08 -0.04 0.14 -0.01 [19] 0.00 0.24* 0.18 -0.12 -0.13 -0.20* 0.05 0.05 -0.14 -0.09 -0.07 0.04 0.09 -0.02 -0.20* 0.09 0.15 -0.02 Variables represented in the table: [1]BOARDOWN [2]DISPBOARDOWN [3]BOARDSIZE [4]CEOCHAIR [5]BLOCKVOTES [6]YEAR [7]SIZE [8]QRATIO [9]CASHLIQ [10]LEV [11]REGULATED [12]FOCUS [13]RELQRATIO [14]RELCASHLIQ [15]RELLEV [16]RELSIZE [17]STOCK [18]HOSTILE [19]CONTEST. The definitions (and data sources) for each variable are presented in Table 3, except [6]YEAR, which is the year of the merger announcement (from Securities Data Corp.). * Significantly different from zero at the 10% or less level.