Do Shareholders Listen? M&A Advisor Opinions and Shareholder Voting

David Becher Department of Finance, Drexel University; Wharton Financial Institutions Center, University of Pennsylvania; Philadelphia, PA 19014, USA; tel: (215) 895-2274; e-mail: becher@drexel.edu

Jay Cai Department of Finance, Drexel University, Philadelphia, PA 19014, USA; tel: (215) 895-1755;

e-mail: jaycai@drexel.edu

Wenjing Ouyang

Department of Finance, Drexel University, Philadelphia, PA 19014, USA; tel: (215) 895-1741; e-mail: wenjing.ouyang@drexel.edu

July 2010

Abstract

Recent studies find that merger advisors, in particular acquirer advisors, often face conflicts of interest and present biased opinions about an underlying deal. It is not clear, however, how shareholders react to these opinions. Using a sample of mergers announced from 2000 to 2006, we examine whether target and acquirer advisors’ opinions (valuation of target equity, long-term earnings forecast, and affiliated analyst recommendations) impact acquirer shareholders votes on mergers. Our results show that target advisor opinions, but not those of acquirer advisors, significantly impact shareholder voting. Further, if a deal receives higher shareholder support, the merger advisor is more likely to be retained in future deals. We conclude that shareholders are able to discern the potential bias in the opinions of merger advisors and follow the advice of the less-biased target advisors. Our study provides important evidence for the ongoing debate about regulatory reform governing investment banking transactions.

Keywords: Merger and Acquisition, Shareholder Voting, Investment Banking, Financial Advisor JEL Classification: G24, G30, G34

We thank Anup Agrawal, John, Coates, Fabrizio Ferri, Dalida Kadryzhanova, Kate Litvak, Adam Pritchard, Roberta Romano, Katsiaryna Salavei, and seminar participants at the 2010 American Law and Economics Association meetings, the 2010 Eastern Financial Association meetings, and Drexel University.

1 

 

1. Introduction

In merger transactions, acquirer and target firms frequently seek fairness opinions from financial

advisors. Kisgen, Qian, and Song (2009) document that 45% of acquirers and 82% of targets obtain

financial advisors’ opinions during 1,175 mergers from 1994-2003. In their letters to the boards and

shareholders of the merging companies, financial advisors systematically come to the conclusion that

the merger under consideration is fair to the target or acquirer firms. Prior studies, however, document

that financial advisor fees are likely contingent upon deal completion (Kisgen et al. 2009; Makhija

and Narayanan, 2007; Becher and Jeurgens, 2010). As a result, financial advisors often provide

biased opinions to increase the likelihood of deal completion, rather than to maximize acquirer

shareholder wealth (McLaughlin, 1990; Rau, 2000; and Kolasinski and Kothari, 2008). This potential

conflict of interest has attracted attention from regulators. In 2007, the Securities and Exchange

Commission (SEC) approved Rule 2290 that requires Financial Industry Regulatory Authority

(FINRA) member firms to fully disclose any potential conflict of interest in the process of reaching

their fairness opinions presented to shareholders of public firms involved in acquisitions.

Nevertheless, mergers, on average, do not create wealth for acquirer shareholders (Travlos, 1987;

Amihud, Lev, and Travlos, 1990; Moeller, Schinglemann, and Stulz, 2005). For example, Moeller et

al. (2005) detail that acquirer shareholders lose nearly $216 billion in the 1990s. Kisgen et al. (2009)

document that merger announcement returns are 2.3% lower for acquirers with fairness opinion.

Many mergers, however, require shareholder approval and shareholders can block a deal if the

majority of them believe the merger is not in their best interest. Thus, it is a puzzle why these value-

destroying mergers receive shareholder support and are completed.

2

 

In this study we examine whether advisor opinions influence shareholder voting on mergers. The

answer to this question is important. If shareholders have full knowledge about a merger’s value and

vote on its merit, financial advisors’ biased opinion will have no impact on how shareholders vote. In

this case, financial advisors’ potential conflict of interest has little consequence on shareholder wealth.

On the contrary, if merger advisors’ opinion can substantially affect how shareholders reach their

voting decision, these opinions can significantly affect shareholder wealth. In this case, additional

regulations and/or disclosure may be warranted to help reducing the conflict of interests.

Based on the existing literature, we propose three hypotheses to examine the relation between

financial advisors’ opinions and shareholder voting on mergers: the passive listener, uninterested

listener, and active listener hypotheses. The Passive Listener hypothesis postulates that acquirer

shareholders listen to financial advisors’ opinions regardless of potential conflicts of interest. This

hypothesis predicts higher shareholder support for a deal when the financial advisor’s opinion is more

favorable. The Uninterested Listener hypothesis, however, suggests that acquirer shareholders do not

listen to their financial advisors’ opinions and vote without considering these opinions. This

hypothesis predicts no significant relation between shareholder voting and financial advisor’s opinion.

Finally, the Active Listener hypothesis proposes that acquirer shareholders are able to recognize the

potential conflict of interest and bias in financial advisors’ opinions and are more likely to listen to

less biased opinions. Previous studies, as well as our own empirical evidence, suggest that target firm

advisors are more conservative and less biased.1 This hypothesis, therefore, predicts that target advisor

opinions are more likely to impact acquirer shareholders votes.

                                                               

1 See, for example, Kesner, Shapiro, and Sharma (1994), Allen, Jagtiani, Peristiani, and Saunders (2004), and Cain and Denis (2008).

3

 

The focus of our analyses is on the acquirer shareholders’ approval rate, which is measured as the

percent of shares held by outside shareholders that support a deal. We exclude managerial and board

ownership and shares held by financial advisors as we expect that these parties always vote for a deal,

given our sample of friendly negotiated mergers. Our acquirer shareholders’ approval rate, therefore,

reflects the outside shareholders’ assessment of the deal. Furthermore, Burch et al. (2004) find some

shareholders may not cast a vote when they are not satisfied with the deal. They conclude that the

approval rate based on outstanding voting rights rather than votes cast better represents shareholders’

attitude to a deal. As a result, we use outstanding votes held by outside shareholders as the base.

To measure financial advisors’ opinions, we collect three sets of information. First, we collect

financial advisors’ valuation on target firm equity. Since financial advisors often provide a valuation

range for target equity, we use the scaled difference between the offer price and the midpoint of target

equity valuation as a measure of whether an acquirer over- or under-pays. Second, we measure

financial advisors’ opinions with their long-term EPS forecasts for the combined firm. Blockholders

and institutional investors likely focus on long-term profits, rather than the short term gains or losses

(Chen, Harford, and Li, 2007; Holmstrom and Tirole, 1993). As a result, acquirer shareholders may

more likely vote for a deal if they perceive an increase in long-term EPS. Third, we examine a merger

advisor’s affiliated analysts’ recommendations after the deal announcement. Affiliated analysts work

for the same investment bank as the merger advisor. Their recommendations, therefore, may be

motivated by deal completion (Kolasinski and Kothari, 2008; Becher and Juergens, 2010).

Using hand-collected data of financial advisors’ opinions on 148 mergers announced from 2000

to 2006, we show that acquirer shareholders listen to target advisors but tend to value acquirer

advisors’ opinions less. Based on multivariate regressions, we document acquirer shareholders’

4

 

approval rate increases 3.2% when target advisors’ equity valuation increases one standard deviation

and 6.6% when they provide EPS forecasts. This rate increases by 11.3% when target-affiliated

analysts update recommendations after the deal is announced. We, however, find no evidence that

acquirer shareholder voting is related to their own advisor’s opinion. These results support the active

listener hypothesis in that acquirer shareholders appear more likely to respond to less biased opinions.

In addition, we provide evidence on whether merger advisors’ opinions are biased. We find that

the acquirer advisors are more optimistic in their valuation of target firm equity. Their affiliated

analysts also provide optimistically biased recommendations when the market reacts negatively to a

deal announcement. Target advisors and their affiliated analysts, on the other hand, do not appear to

be optimistically biased. Finally, we examine whether shareholder support for a merger impacts the

likelihood that an advisor is retained in future deals. We find that a one standard deviation increase in

the shareholder voting ratio increases the probability of advisor retention for future deals by 13%.

This suggests a merger advisor’s ability to garner shareholder support is highly valued.

Overall, our results imply that acquirer shareholders are able to discern that acquirer advisors’

opinions are biased and listen only to target advisors’ opinions. Our paper contributes to the

discussion about financial advisors’ conflict of interests on shareholder wealth as well as provides

evidence on the efficacy of shareholder voting.2 The remainder of the article is organized as follows.

Section 2 discusses related literatures and develops three hypotheses. Section 3 describes the data and

defines the main variables. Section 4 presents empirical results while Section 5 concludes.

                                                               

2 Studies on acquirer shareholder voting find that shareholder voting is significantly related with deal characteristics, indicating that acquirer shareholder voting has a monitoring effect (Hamermesh, 2003 and Hsieh and Wang, 2008). Our results provide further evidence that the monitoring effect from acquirer shareholder voting is robust to the presence of financial advisors’ biased opinions.

5

 

2. Literature review and hypotheses development

This paper is related to two streams of literature: financial advisors’ role in mergers and

acquisitions and shareholder voting on corporate decisions. The discussion about financial advisors’

role focuses on the benefits and potential conflict of interests of their advisory services. The debate on

shareholder voting centers on how shareholders vote and the efficacy of their votes. In this section, we

review these studies and propose three hypotheses.

2.1 Financial advisor opinions

Financial advisors possess certain expertise that may reduce information asymmetry in financial

transactions (Bowers and Miller, 1990). Servaes and Zenner (1996) note that acquirers are more likely

to hire financial advisors when deals are more complex and information on targets is less available.

Kisgen et al. (2009) find the presence of a fairness opinion is positively related to deal complexity as

well as board monitoring, which suggests that fairness opinions provide valuable information to

management and shareholders as well as legal protection to managers and boards of directors.

The reputation of the financial advisors appears to benefit the shareholders. Bowers and Miller

(1990) detail that total combined returns are higher when either party (target or bidder) hires a top-tier

investment bank.3 Kale, Kini, and Ryan (2003) document that shareholders gain a higher portion of

merger synergies when their financial advisors have higher reputation relative to the counter party.4

Bao and Edmans (2009) find that stock returns from an investment bank’s prior deal predict returns on

future deals advised by the same investment bank.

                                                               

3Bowers and Miller (1990) define first-tier bankers as those listed directly below the manager and co-manager in the tombstones placed in the financial section of newspapers. 4The authors define reputation as financial advisors’ investment banking market shares in the year of the merger.

6

 

While financial advisors may create value, conflicts of interests may help to explain why

numerous studies document non-positive acquirer shareholder returns around mergers. McLaughlin

(1990, 1992) argues that contingent fee payment contracts may lead to financial advisors’ conflicts of

interest. Rau (2000) finds that a higher proportion of top tier investment banks’ advisory fees are

contingent and that acquirers advised by these advisors are more likely to complete a deal, but pay

higher premiums. Further, financial advisors’ conflict of interest may affect their affiliated analysts’

recommendations. Kolasinski and Kothari (2008) detail that analysts affiliated with advisors provide

biased recommendations to maximize financial advisors’ expected fee revenue. Becher and Juergens

(2010) provide evidence advisors’ affiliated analysts issue biased recommendations to ensure deal

completion. Collectively, this evidence suggests that financial advisors may face conflicts in mergers

and provide biased opinions to maximize their own benefit at a cost to shareholders.

Acquirer and target firms as well as their financial advisors may all have different interests in

merger transactions. Specifically, acquirer firms want to pay a lower acquisition price, while target

firms want the opposite. Kesner, Shapiro, and Sharma (1994) document that the positive relation

between financial advisors’ compensation and merger premium aligns interests between shareholders

and target advisors, but creates conflict of interest for acquirer advisors. Allen, Jagtiani, Peristiani, and

Saunders (2004) find that target advisors with a prior lending relation to the target serve a certification

effect while acquirer advisors do not have such effect. This is because acquirer advisors tend to

provide biased opinions in order to garner future loan commitments from their clients while the target

advisor’s client ceases to exist after the merger is completed. Cain and Denis (2008) examine financial

advisors’ valuation of the target firm and find that only acquirer advisors issue optimistically biased

opinions. Using the methodology of Cain and Denis (2008), we find similar results in our sample

7

 

(untabulated). Further, we examine the choice of financial advisor using a sub-sample of combined

firms that engage in another merger within three years of the sample deal. We find that in the future

deals these firms are nearly four times more likely to hire the current acquirer advisors than the

current target advisor. Taken together, these studies suggest that the target advisors may provide less

biased opinions than the acquirer advisors.

2.2 Shareholder voting in corporate decisions

Mergers often require shareholder approval. Easterbrook and Fischel (1983) and Harris and

Raviv (1988) argue that shareholder voting have limited efficacy due to free-rider and agency

problems. Matvos and Ostrovsky (2008) find mutual funds that own both target and acquirer stocks

are more likely to vote for a merger at an acquirer shareholder meeting, even though the deal reduces

acquirer shareholder value. Bethel and Gillan (2002) find managers tend to classify proposals as

routine, rather than non-routine, to validate brokers’ uninstructed votes for a better outcome. Finally,

the evidence that shareholder votes have little impact on future performance suggests shareholder

voting may have limited efficacy (Gillan and Starks, 2007; Karpoff, Malatesta, and Walkling, 1996).

Other studies, however, detail that some shareholders cast votes based on firm performance,

providing a potential threat to management. Cai, Garner, and Walkling (2009) and Gordon and Pound

(1993) document a significant relation between shareholder voting and prior firm performance. Martin

and Thomas (2000) examine voting on stock option plans and show that shareholders are more likely

to support such plans at poorly performing firms. They conclude shareholders’ voting may help to

align managers’ interests. Martin and Thomas (2005) note changes in compensation are significantly

lower if shareholders strongly oppose management-sponsored stock option plans, suggesting boards

react to shareholders voting outcomes. Moreover, significant shareholder opposition in voting

8

 

outcomes often leads to governance and management reforms (DeAngelo and DeAngelo (1989),

Dodd and Warner (1983), Mulherin and Poulsen (1998), and Cai, et al. (2009)).

Overall, the evidence from these studies suggests that shareholder voting can be an effective

mechanism to protect shareholders’ interests in major corporate events. In this study, therefore, we

examine whether financial advisors can affect shareholder voting decisions in mergers.

2.3. Hypotheses

We develop three hypotheses to examine whether acquirer shareholder voting behavior is

impacted by target or acquirer financial advisors’ opinions. First, the Passive Listener hypothesis

states that acquirer shareholders rely on financial advisors’ opinions in their voting decisions. This

hypothesis postulates that acquirer shareholders listen to their financial advisors’ opinions regardless

of potential conflicts of interest. Financial advisors may have expertise in identifying merger partners

with higher potential synergy as well as valuing such merger gains. Bowers and Miller (1990) find the

choice of financial advisors affects combined target and acquirer returns in a merger. Bao and Edmans

(2009) show persistent acquirer firm announcement returns among different deals advised by the same

bank and conclude that investment bank skills affect their client’s shareholder wealth. Kisgen, Qian,

and Song (2009) and Servaes and Zenner (1996) document that the probability of hiring a financial

advisor increases with deal complexity and when the potential legal risk of conflict of interest is

higher. As a result, shareholders may be unable to recognize the bias in advisor opinions and rely on

advisors’ opinions for their voting decisions. This hypothesis predicts a positive relation between the

financial advisor’s opinion on a merger and shareholder voting support for this merger.

Next, the Uninterested Listener hypothesis centers on the notion that acquirer shareholders

recognize their financial advisors face potential conflicts of interest. In particular, shareholders are

9

 

aware that financial advisors provide opinions intended for deal completion rather than shareholder

wealth maximization. McLaughlin (1990, 1992) argues a contingent advisory fee payment structure

may trigger financial advisors’ conflicts of interest for deal completion regardless of quality. In

addition, financial advisors may take advantage of their position in a deal to the detriment of current

shareholders (Bodnaruk, Massa, and Simonov, 2009). Financial advisors’ affiliated analysts may also

issue biased recommendations to ensure maximum fee revenue (Becher and Juergens, 2010;

Kolasinski and Kothari, 2008). Due to these potential conflicts of interest, acquirer shareholders may

deem all financial advisors’ opinions biased and not take any advisor’s opinion into consideration

when evaluating mergers.

Finally, the Active Listener hypothesis proposes that acquirer shareholders may be able to

recognize when financial advisors issue biased opinions and only react to unbiased opinions. Several

studies document that acquirer financial advisors’ contingent fee payments and future business

opportunities lead to conflicts of interest and, consequently, biased opinions. Target advisors, however,

face less severe conflicts of interest and their opinions may be less biased. The target ceases to exist

after a merger, and as a result, the target advisor is less concerned about future business opportunities

(Davidoff, 2006; Cain and Denis, 2008).

In contrast, acquirer advisors have an incentive to form good relationship with acquirer managers

for future business opportunities. To this end, these acquirer advisors likely present over-optimistic

opinions to help management close a deal.5 In addition, to convince target management and board to

accept an offer, target advisors have an incentive to provide a lower valuation of the target and more

conservative forecasts. Although this downward bias may hurt target shareholders with reduced                                                                

5Acquirer managers often gain from completing deals despite shareholder losses. Grinstein and Hribar (2004) find acquirer managers receive a substantial merger bonus even if shareholders lose value in a deal.

10

 

premium, the acquirer shareholders may find such conservative opinions valuable since these opinions

may represent a lower bound of the target value and merger outcome. This hypothesis predicts,

therefore, acquirer shareholders are more likely to listen to the target advisors’ opinions.

3. Data and sample selection

3.1 Sample selection

We start with mergers announced between 2000 and 2006 on the Securities Data Corporation

(SDC) Mergers & Acquisition database. Since we focus on acquirer shareholder voting, the acquirer

must be a U.S. public firm, while the target may be public, private, or a foreign. Next, we obtain

shareholder voting data from Institutional Shareholder Services (ISS), SEC 8-k filings, or Factiva

News search. These data requirements yield a sample of 153 mergers announced between 2000 and

2006. The major U.S. exchanges all require shareholder approval if a firm issues over 20% of

outstanding shares in a merger.6 This requirement results in a sample of relatively large target firms.

Table 1 details that the average transaction value-to-acquirer size ratio is 0.80.7

Next, we collect financial advisors’ opinions from the proxy statement regarding the merger in

SEC filings (form S-4 and various proxy filings). Appendix A illustrates two examples of financial

advisor opinions provided in the proxy statements. We are able to obtain either target or acquire

advisor opinions in 148 of the 153 deals. These 148 deals announced between 2000 and 2006

represent our final sample. Stock price data are obtained from CRSP and augmented with data from

Yahoo Finance (for targets not on CRSP). All accounting data are collected from COMPUSTAT.8

                                                               

6NYSE Company Manual section 312.03, AMEX Company Guide section 712, and NASDAQ Marketplace Rules section 4350. 7Burch et al. (2004) find average target-to-acquirer size of 0.55 for a 1990-2000 sample while Hsieh and Wang (2008) report average relative size of 0.76 for all stock deals and 1.19 for mixed-payment deals (1990 to 2005). 8Shareholder voting may be endogenous. If an acquirer manager is not confident of shareholder support, she may pay with cash to avoid a vote. Our focus, however, is how shareholders respond to advisor opinions rather

11

 

3.2 Voting Measures

Our main dependent variable is acquirer shareholder approval rate for merger proposals, which

measures outside shareholders’ support. Since all deals in our sample are friendly, we assume acquirer

management and institutions affiliated with the financial advisors always vote for a deal and exclude

their shares when calculating acquirer shareholder approval rates.9

Burch et al. (2004) indicate that shareholders not in favor of a deal often vote “abstain” or do not

cast a vote. If the beneficiary owner does not vote, her shares are recorded as broker non-votes if these

shares held under street name. We, therefore, use the total number of voting shares held by outside

investors, rather than votes cast, as the base for the acquirer shareholder approval rate. This measure

focuses on outside shareholder support for a deal rather than voting outcome.

acquirer shareholder approval rate

= shares voted for - shares held by management and financial advisors

(1) shares outstanding - shares held by management and financial advisors

Table 1 details that, on average, 96.35% of shares cast vote in favor of a merger. However, only

an average of 68.39% of the outstanding shares vote for the mergers in our sample.10 After excluding

the shares owned by management and institutions affiliated with financial advisors, on average only

61.54% shares held by outside shareholders support these deals.

3.3 Financial Advisors’ Opinions

                                                                                                                                                                                                  

than how shareholder voting impacts deal completion. See Hsieh and Wang (2008) for an analysis of shareholder voting on deal completion. 9 The incentive of a target financial advisor may be different in hostile deals if her fee is contingent on the failure of the deal. Since our sample includes no hostile deals, the opinion of financial advisors in hostile deals is beyond the scope of this study. 10Bethel et al. (2009) find that from 1999 to 2005, the corresponding percentage is 70%.

12

 

Financial advisors’ opinions are detailed in the “Opinion of financial advisors” section in proxy

statement of a merger. We collect target firm equity valuation from discounted cash flow analysis and

combined firm EPS forecasts from pro-forma earnings analysis.11 We obtain financial advisors’

affiliated analyst recommendations from both I/B/E/S and First Call.

3.3.1 Target firm equity valuation

In the proxy statement, financial advisors often provide an estimation of target firms’ equity

value, usually in a valuation range. The offer price relative to the valuation range may indicate

whether the acquirer is overpaying. Following Cain and Denis (2008), we define an equity valuation

(EV) ratio based on the target’s relative selling price:

EV ratio = average target equity valuation - offer price

(2) offer price

A positive EV ratio indicates that the estimated target firm value is higher than the offer price, or

that the acquirer underpays the target, while a negative EV ratio indicates overpayment.

Table 1 details that, on average, acquirer advisors value the target equity 3% above the offer price,

while the target advisors value the target equity 7% below offer price. The mean and median

differences between acquirer and target advisors’ EV ratios are 9% and 10%, respectively, and both of

these differences have a p-value of 0.01. This evidence shows that acquirer advisors are relatively

more optimistic in target firm valuation.

3.3.2 Advisors’ EPS forecasts

                                                               

11 In our sample, a total of 95 investment banks serve as financial advisors, among which 18 banks advise more than five deals. Moreover, none of these18 banks either always provides specific opinions on the merger or never provides opinion. Therefore, the banks do not appear to have rules on always or never provide specific opinions. Thus, we do not include advisor fixed effect in the regression analysis. 

13

 

Merger advisors’ earnings accretion/dilution forecasts usually cover from deal completion to

several years after the merger. In the first couple of years after a merger, the earnings of the combined

firm often include one-time merger charges and integration cost. The value of such earnings forecast

in long-term valuation is questionable. Management and their financial advisors, however, often

suggest that the earnings forecast in the last year of the forecast period can be extrapolated into later

years. Thus, in our analysis, we focus on the financial advisors’ earnings forecast for the last year of

the forecast period.

EPS forecasts can be numerical or descriptive. In our sample, numerical EPS projections are

available from the acquirer advisors in 42 cases, 39 cases from the target advisors, and 19 cases from

both advisors. Descriptive EPS projections, such as accretive, neutral, or dilutive, however, are

available from acquirer advisors in 94 cases, 66 cases from the target advisors, and 48cases from both

advisors. Due to limited sample size, we measure EPS forecasts with three categories: accretion,

neutral, and dilution. Table 1 reflects that in the majority of cases, advisors predict accretive EPS for

the acquirer after the merger. Financial advisors often provide a range of EPS forecasts under different

scenarios of merger synergies. Table 1 details that acquirer (target) advisors provide non-dilutive EPS

forecasts in 78 (50) deals, provide at least one dilutive EPS forecasts in 14 (10) deals, and do not

provide EPS forecasts in 44 (78) deals. Compared with acquirer advisors, target advisors are less

likely to provide EPS forecasts.

3.3.3 Affiliated analyst recommendations

We define an analyst as affiliated if she works for the same investment bank as the financial

advisors. From I/B/E/S and First Call, we obtain analyst recommendations for each acquirer firm from

three years before the merger announcement to shareholder voting date. Recommendations have five

14

 

levels, ranging from “1” for strong buy to “5” for sell. Following Kolasinski and Kothari (2008) and

Becher and Juergens (2010), we classify analysts’ affiliations by manually matching their names with

financial advisors’ names. We classify analysts into three affiliation groups: acquirer-affiliated, target-

affiliated, and unaffiliated.

We define recommendation updates as the difference between the analyst recommendation after

the merger announcement but before the shareholder voting and the last recommendation before the

merger announcement. The variable “drop recommendation” equals one if at least one analyst makes

a recommendation during the three years prior to the merger announcement but no recommendation

afterwards. Irvine (2003) finds that the market responds more positively to analysts’ initiations than to

other recommendations. The variable “initiate recommendation” equals one if an acquirer firm is

covered by analyst recommendations after deal announcement but not before.

Table 1 indicates after merger announcements acquirer affiliated analysts provide significantly

more favorable recommendation than the unaffiliated analysts, but the target affiliated analysts do not.

Among the 86 acquirer advisors’ affiliated analysts that cover the acquirer pre-merger announcement,

44 drop coverage after the announcement. Similarly, among the 55 target advisors’ affiliated analysts

who cover the acquirer pre-merger, 32 stop coverage post-announcement. Since their affiliated

investment banks are serving as financial advisors, these analysts may choose to keep quiet rather

than give pessimistic recommendations. Thus, their silence may signal poor prospects for the merger.

4. Empirical Results

In this section, we examine the relation between acquirer shareholder voting and financial

advisors’ opinions. As noted, financial advisors and their affiliated analysts tend to provide optimistic

opinions on mergers they advise. When advisors choose not to express their opinions, the absence of

15

 

a positive opinion may send a negative signal about the merger. We, therefore, examine both the

presence of a financial advisor opinion about a merger as well as the specific context of such opinion.

4.1 Acquirer shareholder voting and financial advisors’ valuation of target firm equity

We first examine whether the financial advisors’ valuation of the target firm equity impacts how

outside acquirer shareholders vote. Such valuation is routinely presented in the merger proxy

statement that shareholder vote on and signals whether an advisor believes the acquirer over- or

under-pays for the target. Panel A in Table 2 shows that when the target firm advisor provides an

estimation of the target equity value, the average and median increase in acquirer shareholder support

for the merger equals 8.4% and 7.3%, respectively, and both figures have a p-value of 0.01. When the

acquirer firm advisor provides such estimation, the corresponding figures equal 6.3% and 1.6% and p-

values are 0.03 and 0.08, respectively. The presence of a target advisors’ opinion appears to have a

stronger effect on shareholder holding than that of an acquirer advisor.

We next examine whether advisors’ specific opinions affect shareholder voting. Panel B of Table

2 shows that when the target advisor provides a positive EV ratio, i.e. the target is undervalued, the

mean and median acquirer shareholder support increases by 6.9% and 6.8%, respectively (mean

p-value of 0.11 and median 0.08). In contrast, when the acquirer advisor presents an opinion that the

target is under-paid, mean and median increases in shareholder support are 2.6% and 3.9%,

respectively, and both figures are insignificant. This evidence suggests that the acquirer shareholders

regard the target advisor’s valuation more highly than that of their own advisors. In addition, we find

target advisors are more conservative in their valuation. In the 67 mergers where both advisors

provide valuation figures, target advisors opine that the offer undervalues the target in 12 cases (18%)

while the acquirer advisor do so in 29 cases (43%).

16

 

We next present multivariate results in Table 3, where the dependent variable is the acquirer

shareholder approval rate (since this variable is bounded between zero and one, we estimate all

models with a tobit specification). We include several deal, firm, and advisor characteristics as control

variables. In Model (1) the main independent variable is whether an advisor opines about the equity

value of the target. When the target advisor provides such opinions, acquirer shareholder support for a

deal increases by 6.2% (p-value 0.03). The presence of an acquirer advisor opinion on valuation,

however, has no significant impact on shareholder voting.

In Model (2) of Table 3, we include only those cases where both advisors (acquirer and target)

provide a specific valuation of the target firm. The main independent variable in this model is the EV

ratio. The results demonstrate that the acquirer shareholders’ approval rate increases 3.17% when the

target advisor’s valuation increases by one standard deviation. The acquirer advisor’s specific opinion

on valuation, however, again has no effect on shareholder voting.

Finally, in Model (3) of Table 3, we classify advisor opinions into three categories: no opinion on

valuation, negative EV ratio, and positive EV ratio, assigning each a value of -1, 0, and 1, respectively.

We then use this categorical variable as the main independent variable. We again find that the target

advisor opinion has a significant effect on how acquirer shareholders vote (p-value of 0.01), while the

acquirer advisor opinion has a much smaller effect. These results provide consistent evidence that

acquirer shareholders listen to target advisors’ opinions rather than those of acquirer advisors.

4.2 Acquirer shareholder voting and advisors’ EPS forecasts

Financial advisors often forecast the accretion or dilution effect of a merger on acquirer firm

earnings in the proxy statement of a merger. These earnings growth forecast, in particular the last

figure in the forecast window which is often projected in to the infinite future, may signal a merger’s

17

 

long-term effect on firm performance. In this section, we examine whether these earnings forecasts

by financial advisors affect acquirer shareholder voting on the merger.

Panel A in Table 4 details that the acquirer shareholders’ approval rate increases by an average of

6.5% when the target advisor provide earnings forecasts, which has a p-value of 0.01. In contrast, the

average approval rate increases by only 3.3% when the acquirer advisor provides earnings forecasts,

and this figure is not significant.

We next divide the sample by whether the advisors present non-dilutive earnings forecasts. Panel

B of Table 4 details that the average shareholder approval rate increases by 5.1% if target advisors

predict that the merger deal will be non-dilutive to acquirer firm earnings (p-value of 0.04). The

average approval rate, however, increases only 3.1% if acquirer advisors predict an accretive EPS, and

this figure is again statistically insignificant.

These earnings forecasts by financial advisors, however, may be correlated with the quality of a

merger, which may independently affect shareholders’ voting decision. To control for this possible

confounding factor, we estimate multivariate regressions of shareholder voting in Table 5. The

dependent variable is the acquirer shareholder approval rate and the main independent variables are

whether financial advisors provide EPS forecast as well as whether they forecast earnings dilution.

We include several deal, firm, and advisor characteristics as controls.

Results from Model (1) of Table 5 indicate that the acquirer shareholder approval rate increases

by 6.6% when target advisors provide earnings forecast (p-value of 0.01). Model (2) demonstrates that

acquirer shareholder support for a merger increases by 4.5% when acquirer advisors forecast earnings

(p-value of 0.10). In Model (3), we include both target and acquirer financial advisors’ opinions on

EPS. The acquirer shareholders’ approval rate increases by 5.1% when target advisors provide EPS

18

 

forecasts (p-value of 0.04). In contrast, the coefficient of acquirer financial advisor providing EPS

forecast loses statistical significance.

Next, we examine whether advisor opinions on EPS dilution affect shareholder voting. Model (4)

of Table 5 details that when target advisors provide non-dilutive EPS forecasts shareholder support for

the merger is 5% higher (p-value of 0.05). Similarly, model (5) shows that shareholder support is also

5% higher (p-value of 0.06) when the acquirer advisor provides non-dilutive EPS forecast. In model

(6), results are qualitatively similar when we include both advisors’ opinions on EPS dilution. In

addition, we note that the target advisor’s reputation has a significantly positive effect on shareholder

voting in all specifications.12 In contrast, the coefficient of acquirer advisor reputation decreases by

more than half and loses statistical significance when included together with target advisor variables

in models (3) and (6). The results of Table 5 suggest that shareholder voting is more related to the

target advisor’s opinions than those of the acquirer advisors.

4.3 Acquirer shareholder voting and affiliated analysts’ recommendations

A number of studies propose that changes in analyst recommendations provide more information

to the market than recommendation levels. For example, Jegadeesh, Kim, and Lee (2004) find that

changes in recommendations, rather than levels, have predictive power for returns while Kolasinski

and Kothari (2008) suggest analyst recommendation changes convey more information than levels.

Further, Becher and Juergens (2010) note that the 2002 Global Research Analyst Settlement led to

rescaling of recommendation levels and focus on changes in recommendation to avoid a bias in results.

                                                               

12 The coefficients are negative because we measure advisor reputation with their market share ranking where one is the highest rank and 26 is the lowest rank.

19

 

Thus, we examine how acquirer shareholder voting is related to affiliated analysts’ recommendation

issuance, changes, and terminations.

Panel A in Table 6 examines the certification effect of affiliated analysts’ recommendations. The

data shows that in more than half of the mergers the target-affiliated analysts stop providing acquirer

recommendations after the deal is announced. The median (mean) acquirer shareholders’ approval rate

of these deals are 5.65% (5.99%) lower than those deals where the target affiliated analysts continue

to provide stock recommendation. The difference in the median value is statistically significant at the

10% level. 13 As documented, acquirer-affiliated analysts tend to be more optimistic. Acquirer

shareholder votes, however, are (insignificantly) lower when acquirer affiliated analysts

maintain/initiate recommendations.

In Panel B of Table 6, we categorize recommendation changes into two groups: one group

includes initiated, upgraded and same recommendations, while the second group includes

recommendation downgrades as well as instances when analysts stop providing recommendations

after the deal is announced. The results are similar to those from Panel A. When the groups are formed

by recommendations of target affiliated analysts, the median acquirer shareholders’ approval rate is

5.36% lower in the second group compared to the first (p-value of 0.10). The average shareholder

approval rate is similar in magnitude although statistically insignificant. When groups are formed by

recommendations of acquirer affiliated analysts, both the average and median shareholder approval

rates are not different between the two groups, both statistically and economically. Collectively, these

results suggest acquirer shareholders are less likely to listen to their own analysts, but appear to alter

their voting behavior based on the recommendations of the target affiliated advisors.                                                                

13 Zhu and Zhao (2008) find that 57% of the analysts covering the acquirer stop providing recommendations after the deal announcement.

20

 

We next examine the relation between shareholder voting and affiliated analysts’

recommendations in a multivariate setting. Table 7 details three models: post-announcement

recommendation changes, ranked recommendations, and ranked recommendations allowing for

multiple changes in levels. Model (1) highlights acquirer shareholders’ approval rate increases over

11% when target-affiliated analysts provide post-announcement recommendations (p-value of 0.04).

In contrast, the presence of an acquirer-affiliated analyst recommendation appears to have no effect on

how shareholders vote (p-value 0.42). Moreover, repeat business with a firm may cause an advisor to

have a greater conflict of interest (more bias in their recommendations) to garner future business.

Consistent with this, acquirer shareholder approval rates decrease 14% (p-value of 0.01) if the

acquirer financial advisors has prior business relations with the acquirer firm.

In Model (2), we segment affiliated analysts’ post-announcement recommendations into five

different categories. The category equals one for new recommendations, two for upgraded

recommendations, three for the same recommendations, four for downgraded recommendations, and

five if analysts stop providing recommendations post-announcement. Acquirer shareholders’ approval

rate increases 4.3% when the rank of target-affiliated analysts’ recommendations increases one level

after the merger announcement (p-value of 0.07). This suggests that shareholders are more likely to

vote for a deal when target-affiliated analysts view the deal favorably. In contrast, the

recommendations of the acquirer affiliated analysts have no significant effect on shareholder voting.

Finally, in Model (3) we segment further to distinguish whether the analysts increases (decreases)

recommendations by one or more than one level. Even allowing for these multiple level changes, we

find that the acquirer shareholders’ approval rate increases 2.7% when the target-affiliated analysts’

category (multiple) increases one level. Again, we find no significant relation between the

21

 

recommendations of the acquirer affiliated analysts and the acquirer shareholder support for a deal.

These results are consistent with those findings from univariate tests suggesting that only target-

affiliated analysts’ recommendations certify deal quality.

4.4 Additional tests

4.4.1 Acquirer announcement returns and advisor opinions

Prior studies as well as our results suggest that acquirer advisors’ opinions are biased to

promote deal completion. To further investigate this result, we examine acquirer affiliated advisor

recommendations conditional on the initial market reaction at the deal announcement. We expect

these affiliated analysts to behave differently depending on what they see in the market reaction. If

the initial market sentiment is negative, affiliated acquirer advisors should provide more positive

recommendations after a merger is announced in order to ensure deal completion.

In untabulated tests, we find that in deals where the acquirer announcement return is negative,

acquirer-affiliated analysts’ recommendations post-announcement are significantly more optimistic

than those from unaffiliated analysts (p-value 0.01). We do not, however, find such evidence of target-

affiliated analysts’ post-announcement recommendations. These results indicate that acquirer advisors

are more likely to provide over-optimistic opinions, particularly when market sentiment is negative on

a deal and provide further evidence of the bias in their recommendations.

4.4.2 Shareholder vote and advisor retention

Our evidence suggests that acquirer shareholders do not listen to their own financial advisors’

opinions. In fact, as noted in Table 7 above, shareholder approval rates are nearly 14% lower when

acquirer affiliated advisors have a prior business relation with the acquirer firm. It is not clear,

however, how or if this voting behavior impacts future advisor business. As a result, we examine

22

 

whether the percentage of shareholder support for a merger impacts the likelihood that an advisor is

retained in future takeover deals.

In Table 8, we examine 158 future mergers by 58 acquirers from our acquirer shareholder

voting sample that made at least one acquisition within three years after a deal’s effective date. We

measure acquirer advisor retention based on whether the acquirer uses the same advisor in the future

deal as in the previous deal. Surprisingly, acquirer shareholder approval rate is significantly related

with advisor retention. A one standard deviation increase in the shareholder voting ratio increases the

probability of advisor retention for future deals, on average, by 13% (p-value of 0.02).

Results from our earlier tables suggest that shareholders can see through affiliated advisors’

conflicts of interest. Managers, however, appear to believe that the choice of the advisor positively

impacts shareholder votes. In particular, managers are more likely to retain an advisor if the vote in a

previous merger advised by the same advisor is higher.

4.4.3. Financial advisor opinion and expected probability of merger completion

As previously detailed, financial advisors typically receive the bulk of their fees conditional

upon merger completion. These advisors, therefore, may have stronger incentives to promote merger

completion if they believe that the deal may not be completed. Similarly, if ex ante shareholder

support for a merger is high, e.g. when management stock ownership is high, the financial advisors

may not need to provide over-optimistic opinions in an effort to persuade shareholders. In this

situation, the management also may not have the need to hire the financial advisor with the most over-

optimistic opinions. As a result, the opinions of financial advisors may be related to the ex ante

expected probability of merger completion.

23

 

To address this potential endogenous issue, using a sample of 564 deals from 2000 to 2006,

we estimate a model of the probability of merger completion similar to that of Bates, Becher, and

Lemmon (2008). 14 The dependent variable of the regression is a binary variable for merger

completion. The independent variables include toehold, target termination fee, cash payment, relative

size, hostility, horizontal merger and tender offer dummies, and acquirer managerial and institutional

ownership. We next estimate the correlation between the expected probability of merger completion

and the advisor opinion variables. If lower ex ante expected probability of merger completion leads to

more optimistic financial advisor opinions, we expect a negative correlation between the expected

probability and the opinion variables. We find that 12 out of 14 correlation estimates are statistically

insignificant, and one is positive with a p-value of 0.09. The expected probability of completion is

negatively correlated with only one of the 14 variables (acquirer advisor’s analyst recommendation)

with a p-value of 0.08.15 This evidence suggests that the financial advisor opinion is not endogenously

determined by the expected probability of merger completion.

5. Conclusion

Voting on merger and acquisitions are arguably one of the most important corporate decisions

shareholders have to make. Whether this shareholder decision is influenced by the opinions of the

financial advisors has important consequence for shareholder wealth and the effectiveness of

regulatory oversight. Further, previous studies show that shareholders on average experience

significant losses when a firm undertakes an acquisition while the merger advisors stand to gain

                                                               

14 The number of observation in this analysis is higher because we do not require voting and financial advisor data. From the financial advisor’s point of view, they are also likely to consider all available mergers in estimating the probability of completion of the deal they advise. 15 This negative coefficient is consistent with the results in Section 4.4.1 that acquirer advisor’s affiliated analysts are more likely to issue optimistic recommendation when the merger announcement return is negative for the acquirer firm.  

24

 

substantial fees when a deal is closed. The apparent conflict of interest between the merger advisors

and the shareholders prompt us to examine the relation between acquirer shareholder voting and

financial advisor opinions.

It is possible that shareholders listen to financial advisors’ opinions regardless of the potential

bias and are more likely to support a deal when the merger opinion is more favorable. Alternatively,

shareholders may not value their financial advisors’ opinions and no relation will exist between

shareholder voting and financial advisor opinions. Finally, it is plausible that acquirer shareholders are

able to recognize the more severe conflict of interest of their own advisors but are more willing to

follow the less biased opinions of target advisors.

Our principal result is that acquirer shareholders apparently are able to see through their own

financial advisors’ conflicts of interest and do not alter their voting decision based on their opinions.

These results hold whether the opinions are in the form of valuation of target equity, long-term

earnings forecast, or affiliated analyst recommendations. Acquirer shareholders, however, do appear

to listen to the opinions of target financial advisors. These results support the active listener

hypothesis in that acquirer shareholders appear more likely to consider less biased opinions when

making their voting decisions. Our evidence has important implications for shareholder wealth and

potential regulatory reforms.

25 

 

References

Allen, L., Jagtiani, J., Peristiani, S., Saunders, A., 2004. The role of bank advisors in Mergers and

Acquisitions. Journal of Money, Credit & Banking 36, 197-224.

Amihud, Y., Lev, B., Travlos, N., 1990. Corporate control and the choice of investment financing: The

case of corporate acquisitions. Journal of Finance 45, 603-616.

Bao, J., Edmans, A., 2009. Do investment banks have skill? Performance persistence of M&A

advisors. Ohio State University and Wharton Financial Institutions Center working paper.

Bates, T. W., Becher, D. A., Lemmon, M. L., 2008. Board classification and managerial entrenchment:

Evidence from the market for corporate control. Journal of Financial Economics 87, 656-677.

Becher, D. A., Juergens, J. L., 2010. M&A Advisory Fees and Analyst Conflicts of Interest. University

of Texas-Austin working paper.

Bethel, J. E., Gillan, S., 2002. The impact of the institutional and regulatory environment on

shareholder voting, Financial Management 31, 29-54.

Bodnaruk, A., Massa, M., Simonov, A., 2009. Investment banks as insiders and the market for

corporate control. Review of Financial Studies 22, 4989-5026.

Bowers, H., Miller, R., 1990. Choice of investment banker and shareholders wealth of firms involved

in acquisitions. Financial Management 19, 34-44.

Burch, T. R., Morgan, A. G., Wolf, J. G., 2004. Is acquiring-firm shareholder approval in stock-for-

stock mergers perfunctory? Financial Management 33, 45–69.

Cai, J., Garner, J. L., Walkling, R. A., 2009. Electing directors. Journal of Finance 64, 2389-2421.

Cain, M., Denis, D., 2008. Do Fairness opinion valuation analyses contain useful information? Purdue

University working paper.

Chen, X., Harford, J., Li, K., 2007. Monitoring: Which institutions matter? Journal of Financial

Economics 86, 279-305.

Davidoff, S., 2006. Fairness opinions. American University Law Review 55, 1557–1625.

DeAngelo, H., DeAngelo, L., 1989. Proxy contests and the governance of publicly held corporations.

Journal of Financial Economics 23, 29-59.

Dodd, P., Warner, J. B., 1983. On corporate governance: A study of proxy contests. Journal of

Financial Economics 11, 401-438.

Easterbrook, F., Fischel, D., 1983. Voting in corporate law. Journal of Law and Economics 26, 375-

390.

Gillan, S. L., Starks, L. T., 2007. The evolution of shareholder activism in the United States. Journal

of Applied Corporate Finance 19, 55-73.

Gordon, L., Pound, J., 1993. Information, ownership structure, and shareholder voting: Evidence from

shareholder-sponsored corporate governance proposals. Journal of Finance 48, 697-718.

26

 

Grinstein, Y., Hribar, P., 2004. CEO compensation and incentives: evidence from M&A bonuses.

Journal of Financial Economics 73, 119-143.

Hamermesh, L. A., 2003. Premiums in stock-for-stock mergers and some consequences in the law of

director fiduciary duties. University of Pennsylvania Law Review 152, 881-915.

Harris, M., Raviv, A., 1988. Corporate governance: Voting rights and majority rules. Journal of

Financial Economics 20, 203-235.

Holmstrom, B., Tirole, J., 1993. Market Liquidity and Performance Monitoring. Journal of Political

Economy 101(4), 678–709.

Hsieh, J., Wang, Q., 2008. Shareholder voting rights in mergers and acquisitions. Georgia Institute of

Technology working paper.

Hunter, W. C., Jagtiani, J., 2003. An analysis of advisor choice, fees, and effort in mergers and

acquisitions. Review of Financial Economics 12, 65 – 81.

Irvine, P., 2003. The incremental impact of analyst initiation of coverage, Journal of Corporate

Finance 9, 431-451.

Jegadeesh, N., Kim, J., Krische, S., Lee, C. M. C., 2004. Analyzing the analysts: When do

recommendations add value? The Journal of Finance 59, 1083 – 1124.

Kale, J. R., Kini, O., Ryan, H. E., 2003. Financial advisors and shareholder wealth gains in corporate

takeovers. Journal of Financial and Quantitative Analysis 38, 475 – 501.

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

initiatives: Empirical evidence. Journal of Financial Economics 42, 365-395.

Kesner, I. F., Shapiro, D. L., Sharma, A., 1994. Brokering mergers: An agency theory perspective on

the role of representatives. Academy of Management Journal 37, 703-721.

Kisgen, D. J., Qian, Q. J., Song, W., 2009. Are fairness opinions fair? The case of mergers and

acquisitions. Forthcoming Journal of Financial Economics.

Kolasinski, A., Kothali, S. P., 2008. Investment banking and analyst objectivity: Evidence from

analysts affiliated with M&A advisors. Journal of Financial and Quantitative Analysis 43,

817-842.

Louis, H., 2005. Acquirers’ abnormal returns and the non-big 4 auditor clientele effect. Journal of

Accounting and Economics 40, 75−99.

Makhija, A. Narayanan, R., 2007. Fairness opinions in mergers and acquisitions. Ohio State

University working paper.

Matvos, G., Ostrovsky, M., 2008. Cross-ownership, returns, and voting in mergers. Journal of

Financial Economics 89, 391-403.

McLaughlin, R., 1990. Investment-banking contract in tender offers: an empirical analysis. Journal of

Financial Economics 28, 209-232.

McLaughlin, R., 1992. Does the form of compensation matter? Investment banker fee contracts in

tender offers. Journal of Financial Economics 32, 223-260.

27

 

Moeller, S., Schlingemann, F., Stulz, R., 2005. Wealth destruction on a massive scale? A study of

acquiring-firm returns in the recent merger wave. Journal of Finance 60, 757–782.

Mulherin, J. H., Poulsen, A. B., 1998. Proxy contests and corporate change: Implications for

shareholder wealth. Journal of Financial Economics 47, 279-313.

Rau, P. R., 2000. Investment bank market share, contingent fee payments, and the performance of

Acquiring Firms. Journal of Financial Economics 56, 293-324.

Servaes, H., Zenner, M., 1996. The role of investment banks in acquisition. Review of Financial

Studies 9, 787-815.

Stigler, G., 1964. A theory of oligopoly. Journal of Political Economy 72, 44-61.

Martin, K. J., Thomas, R. S., 2000. The determinants of shareholder voting on stock option plans.

Wake Forest Law Review 35, 31-82.

Martin, K. J., Thomas, R. S., 2005. When is enough, enough? Market reaction to highly dilutive stock

option plans and the subsequent impact on CEO compensation. Journal of Corporate Finance 11,

61-83.

Travlos, N., 1987. Corporate takeover bids, method of payment, and bidding firms’ stock returns.

Journal of Finance 42, 943-963.

Zhao, M., Zhu, J., 2008. Analyst coverage around mergers & acquisitions. 2007. Annual Meeting of

the American Accounting Association Paper.

28 

 

Appendix A.1

The following paragraphs come from the joint proxy statement issued by Boise Cascade Corporation (acquirer) and OfficeMax, Inc. (target). This proxy statement was filed with SEC on Nov 4, 2003 and then mailed to shareholders on Nov 7, 2003. The fairness opinions are presented to both firms' boards of directors and officially dated on Jul 13, 2003. Goldman Sachs is the financial advisor for Boise (acquirer).

Opinion of Goldman Sachs

In connection with Boise's consideration of the proposed acquisition of OfficeMax, Boise received financial advice from Goldman Sachs. Goldman Sachs rendered its opinion to Boise's board of directors that, as of July 13, 2003, and based upon and subject to the factors and assumptions set forth therein, the stock and cash consideration to be paid by Boise in the merger, taken in the aggregate, is fair to Boise from a financial point of view. For purposes of Goldman Sachs' opinion, the "stock consideration" refers to the number of shares of Boise common stock equal to the exchange ratio, as described in "The Merger — General," and the "cash consideration" refers to $9.00 in cash per share, subject to the adjustments described in "The Merger Agreement — Consideration to be Received in the Merger."

The full text of the written opinion of Goldman Sachs, dated July 13, 2003 is attached as Annex B to this joint proxy statement/prospectus. Goldman Sachs provided its opinion for the information and assistance of Boise's board of directors in connection with its consideration of the merger. The Goldman Sachs opinion is not a recommendation as to how any holder of Boise's common stock should vote with respect to the merger. Goldman Sachs has not been asked to render an updated fairness opinion.

……

Discounted cash flow analysis. Goldman Sachs performed a discounted cash flow analysis to determine an illustrative range of implied present values per OfficeMax common share. All cash flows were discounted back to October 31, 2003. In performing this analysis, Goldman Sachs used projections supplied by the management of Boise that did not include any of the synergies or integration costs expected to result from the transaction. Using discount rates ranging from 8% to 12%, and terminal 2007 EBITDA multiples ranging from 4.0x to 6.0x, this analysis resulted in an illustrative range of implied present values, excluding synergies, of $5.76 to $9.27 per OfficeMax common share.

Using the same set of projections, Goldman Sachs also performed a sensitivity analysis to illustrate the effect of different assumptions for same store sales growth and increases or decreases in EBIT margin from Boise management projections. The analysis was based on a 10% discount rate and a terminal EBITDA multiple of 5.0x, and used a range of same store sales growth rates of 1.0% to 5.0% and a range of changes in EBIT margin from (2.0)% to 2.0%. This analysis resulted in a range of implied present values, excluding synergies, of $(3.50) to $14.44 per OfficeMax common share.

……..

Pro forma merger analysis. Goldman Sachs prepared pro forma analyses of the financial impact of the merger to Boise (a) using earnings estimates for Boise and OfficeMax for 2004 and 2005 prepared by Boise's management; (b) based on each of a 70%/30% stock/cash consideration mix and a 55%/45% stock/cash consideration mix; and (c) including 100% of the synergy and integration cost estimates for

29 

 

2004 and 2005 prepared by Boise's management. In addition, Goldman Sachs analyzed the results based upon (i) the implied exchange ratio of 0.4268, which is the exchange ratio at the low end of the proposed collar; (ii) the base price exchange ratio of 0.3841 (assuming that the Boise common stock price is $23.43); and (iii) the implied exchange ratio of 0.3492, which is the exchange ratio at the high end of the collar. For each of the years 2004 and 2005, Goldman Sachs compared the projected earnings per share of Boise common stock on a standalone basis to the earnings per share of Boise common stock after giving effect to the proposed transaction for each consideration mix and each exchange ratio. The following table presents the results of these analyses:

Additionally, Goldman Sachs performed the merger analysis described above based upon each of the following scenarios: (a) assuming no impact of the estimated integration costs or synergies as projected by the Boise management and (b) assuming no impact of the estimated integration costs and full impact of the synergies as projected by the Boise management. The table below displays the results of these analyses:

The preparation of a fairness opinion is a complex process and is not necessarily susceptible to partial analysis or summary description. Selecting portions of the analyses or of the summary set forth above, without considering the analyses as a whole, could create an incomplete view of the processes underlying Goldman Sachs' opinion. In arriving at its fairness determination, Goldman Sachs considered the results of all of its analyses and did not attribute any particular weight to any factor or analysis considered by it. Rather, Goldman Sachs made its determination as to fairness on the basis of its experience and professional judgment after considering the results of all of its analyses. No company or transaction used in the above analyses as a comparison is directly comparable to Boise or OfficeMax or the contemplated transaction.

30 

 

Appendix A.2.

The following paragraphs come from the joint proxy statement issued by New York Community (acquirer) and Roslyn's (target). This proxy statement was officially files with SEC on Sep 23, 2003, and then mailed or otherwise delivered to both firms' shareholders on Sep 25, 2003. Citigroup Global Markets Inc. is the financial advisor of New York Community (acquirer), and Goldman Sachs and Sandler O’Neill are Roslyn's (target) financial advisors. The financial opinions are presented to both firms' boards of directors on Jul 26, 2003 and officially dated on Jul 27, 2003.

Opinion of Citigroup Global Markets Inc. to New York Community

Citigroup Global Markets Inc. was retained to act as financial advisor to New York Community in connection with a potential combination transaction with Roslyn. Pursuant to Citigroup Global Markets’ letter agreement with New York Community, dated May 30, 2003, Citigroup Global Markets rendered an opinion to the New York Community board of directors on June 26, 2003, to the effect that, based upon and subject to the considerations and limitations set forth in the opinion, Citigroup Global Markets’ work described below and other factors it deemed relevant, as of that date, the exchange ratio in the proposed merger was fair, from a financial point of view, to New York Community.

…..

Discounted Cash Flow Analysis. Citigroup Global Markets performed a discounted cash flow analysis to estimate a range for the implied equity value per share of Roslyn common stock as of June 25, 2003, including certain potential expenses and cost savings forecasted by management to result from the merger. Citigroup Global Markets performed this analysis both including and excluding the potential impact on Roslyn’s forecasted earnings of the $3.5 billion downsizing of Roslyn’s securities portfolio expected by management to be effected following the merger. In this analysis, Citigroup Global Markets assumed a weighted average cost of capital of 10.6% and used a range of 8.5% to 12.5% to derive the present values of (1) Roslyn’s estimated free cash flows available to stockholders from 2004 to 2008, plus (2) Roslyn’s terminal value at the end of 2008. Terminal values for Roslyn were calculated based on a range of 9.0x to 12.0x estimated 2009 EPS. In performing this analysis, Citigroup Global Markets used IBES estimates of EPS as of June 25, 2003 for Roslyn and an estimated long-term annual growth rate for Roslyn’s EPS (also obtained from IBES) of 9.0%. EPS data were adjusted to account for certain restructuring charges anticipated by management to result from the merger and management’s assumptions of cost savings resulting from the merger of 35% of Roslyn’s pre-tax controllable overhead expense, with an annual growth rate of such cost savings of 3% per year after 2004. In determining cash flows available to stockholders, Citigroup Global Markets used forecasted dividend payout ratios (in other words, percentages of adjusted EPS payable to stockholders), which assume the maintenance of a constant ratio of tangible common equity to tangible assets of 5.0% and an asset growth rate of 10% per annum. The results of these analyses are set forth below:

31 

 

Based on these results, Citigroup Global Markets derived a reference range for the implied equity value per share of Roslyn common stock without taking into account the potential impact of the expected downsizing of Roslyn’s securities portfolio and also derived a reference range for the implied equity value per share of Roslyn common stock taking the potential impact of the expected downsizing into account. Citigroup Global Markets noted that the implied value per share of Roslyn common stock of $20.86 based on the exchange ratio in the merger and the closing price of New York Community common stock on June 25, 2003 and the implied value per share of Roslyn common stock of $20.33 based on the exchange ratio in the merger and the closing price of New York Community common stock on June 26, 2003 were below both reference ranges for the implied equity value per share of Roslyn common stock derived by Citigroup Global Markets in its discounted cash flow analysis. The following table summarizes the results of these analyses:

……

Forecasted Pro Forma Financial Analysis. Citigroup Global Markets analyzed the estimated financial impact of the merger on New York Community’s 2004 and 2005 estimated EPS and 2004 and 2005 estimated cash EPS (“CEPS”). CEPS is determined by adding per share amortization of acquisition-related intangible assets to EPS. In the course of this analysis, Citigroup Global Markets used IBES estimates of EPS for 2004 and 2005, utilizing the IBES forecasted long-term EPS growth rate of 13.5%, and assumed, based on management forecasts, that the merger will result in cost savings equal to 35% of Roslyn’s pre-tax controllable overhead expense. Based on its analysis, Citigroup Global Markets determined that the merger would be accretive to New York Community’s estimated EPS and estimated CEPS for 2004 and 2005 and noted that the tangible common ratio would improve to surpass pre-transaction levels during that period.

……

32

Table 1 Summary Statistics

This table includes 148 mergers announced from 2000 to 2006 that require acquirer shareholder approval. Acquirer (Target) firm size equals the market value of equity on 21 days before the announcement (AD – 21). Relative size equals the transaction value divided by acquirer firm market value of equity. Acquirer announcement return is the cumulated abnormal return [AD-20, AD+1], with CRSP value–weighted index as the benchmark. Financial advisor reputation is measured with Thomson Financial rankings (1-25) of completed merger transaction value in the year prior to the announcement date. If an advisor is not included in the Thomson ranking, we assign a rank of 26. Acquirer shareholder approval rate equals the ratio of outside shareholders voting for a merger divided the total number of shares held by outside shareholders. We exclude shares held by executives, board of directors, and institutions affiliated with financial advisor from the calculation of outside shareholder holdings. Financial advisors’ equity valuation (EV) ratio equals the difference between the midpoint of valuation range and the offer price divided by the offer price. Financial advisors’ earnings forecasts refer to their prediction of whether the merger is accretive or dilutive to the acquirer firms’ stand-alone earnings. Financial advisors often provide earnings forecast for multiple years after merger completion. Since one-time merger changes can often affect the earnings in the first year after a deal, we use last year in the forecast period. We use the average of GAAP and Cash earnings forecast if the acquirer is a bank, and only GAAP earnings otherwise. In Panel C, the subgroup of “No dilutive EPS forecasts” includes deals where neither EPS forecasts with synergies or without synergies is dilutive. The subgroup of “At least one dilutive EPS forecasts” includes deals where EPS forecast with synergies and/or without synergies is dilutive. Affiliated analysts work for the same investment bank as the financial advisors. We obtain analyst recommendations from I/B/E/S and First Call: 1 for “strong buy,” 2 for “buy,” 3 for “hold,” 4 for “underperform,” and 5 for “sell.” We classify an analyst recommendation as “initiate recommendation” if the analyst does not provide recommendation before a merger announcement but does so after the announcement and as “drop recommendation” she provides recommendation before but not after the announcement.

Variable N Mean Median Min Max Std.

Panel A: Firm and deal characteristics

Acquirer market value of equity ($ millions) 148 4,482 355 12 140,12

6 17,29

9Target market value of equity ($ millions) 110 2,305 223 7 49,434 7,415Deal transaction value ($ millions) 148 2,326 180 1 58,761 8,111Transaction value/acquirer mkt val of equity 148 0.80 0.61 0.00 10.30 0.93Acquirer announcement return (%) 148 -2.02 -1.80 -36.62 48.53 12.60Acquirer advisor reputation 143 14.16 11.00 1.00 26.00 9.75Target advisor reputation 141 14.98 15.00 1.00 26.00 10.34Number of acquirer advisors 143 1.18 1.00 1.00 2.00 0.39Number of target advisors 141 1.17 1.00 1.00 3.00 0.43

Panel B: Acquirer shareholder voting

Shares vote-for / votes cast (%) 148 96.35 98.49 50.95 100.00 6.27Shares vote-for / shares outstanding (%) 148 68.39 70.01 12.13 93.02 11.09Management ownership (%) 148 14.11 10.05 0.02 84.30 14.34Acquirer advisor ownership (%) 148 0.56 0.00 0.00 23.74 2.11Target advisor ownership (%) 148 0.59 0.00 0.00 14.12 1.90

Acquirer shareholder approval rate (%) 148 61.54 64.88 0.00 92.57 15.04

33

Table 1 (continued) Summary Statistics

Panel C: Financial advisors’ opinions

Advisor equity valuation ratio (%) N Mean Median Min Max Std. (1) Acquirer advisors 87 3.27 0.18 -67.61 115.57 25.59 (2) Target advisors 98 -7.34*** -9.21*** -59.21 58.36 20.82 Difference (1) – (2) 67 8.83*** 10.27*** -72.84 83.37 24.61

Advisor EPS Accretion/ Dilution forecasts

No dilutive

EPS forecasts

1+ dilutive EPS

forecasts EPS forecasts not reported

EPS forecasts by acquirer advisors 78 14 44 EPS forecasts by target advisors 50 10 78

Analyst recommendations after deal announcement Initiate Update Drop Acquirer advisor-affiliated analyst 2 42 44 Target advisor-affiliated analyst 2 23 32 Unaffiliated analyst 2 120 15

Average analyst recommendation after deal announcement N Mean Median Min Max Std. (1) Acquirer advisor

affiliated analyst 44 2.07 2.00 1.00 3.00 0.74 (2) Target advisor

affiliated analyst 25 2.26 2.00 1.00 4.00 0.83 (3) Unaffiliated analyst 122 2.47 2.50 1.00 5.60 0.77 Difference (1) – (3) 41 -0.25* -0.20

Difference (2) – (3) 25 -0.21 -0.02

34

Table 2 Acquirer Shareholders Voting and Advisor Target Equity Value Estimations

This table details summary statistics on acquirer shareholder voting and advisor target equity value estimations. Acquirer shareholder approval rate equals the ratio of outside shareholders voting for a merger divided the total number of shares held by outside shareholders. We exclude shares held by executives, board of directors, and institutions affiliated with financial advisor from the calculation of outside shareholder holdings. Panel A includes 136 deals where both the acquirer and target firms have financial advisors. Panel B focuses on 67 deals where both the acquirer and target advisors provide valuation estimations for target firm equity. Target firm equity valuation (EV) ratio equals the difference between the midpoint of an advisor’s valuation range and the offer price divided by the offer price. P-values are reported in parentheses.

Acquirer Shareholder Approval Rate

Panel A: Whether advisors give equity valuation estimations

All cases

Financial advisor does not provide EV

estimation

Financial advisor provides EV estimation

Difference (p-value)

(1) (2) (2) - (1)

Target advisor

N 136 39 97

Mean 61.51% 55.20% 63.58% 8.38% (0.01)Median 64.81% 58.74% 66.00% 7.26% (0.01)

Acquirer advisor

N 136 53 83 Mean 61.18% 57.36% 63.61% 6.25% (0.03)Median 63.92% 63.38% 64.95% 1.57% (0.08)

Panel B: Whether advisors give positive or negative equity valuation ratio (EV ratio)

All cases

Financial advisor provides negative

EV ratio

Financial advisor provides positive

EV ratio

Difference (p-value)

(1) (2) (2) - (1)

Target advisor

N 67 55 12 Mean 65.14% 63.90% 70.82% 6.92% (0.11)Median 66.92% 64.99% 71.81% 6.82% (0.08)

Acquirer advisor

N 67 38 29 Mean 65.14% 64.01% 66.61% 2.60% (0.46)Median 66.92% 65.63% 69.53% 3.90% (0.17)

35

Table 3 Acquirer Shareholder Voting and Advisor Target Equity Value Estimations

This table details a tobit regression of acquirer shareholder voting on advisor target equity value estimations. The dependent variable is acquirer shareholder approval rate (%) which equals the ratio of outside shareholders voting for a merger divided the total number of shares held by outside shareholders. We exclude shares held by executives, board of directors, and institutions affiliated with financial advisor from the calculation of outside shareholder holdings. Target firm equity valuation (EV) ratio equals the difference between the midpoint of an advisor’s valuation range and the offer price divided by the offer price. Category of equity valuation ratio equals -1 if an advisor does not provide valuation, 0 if EV ratio is negative, and 1 if EV ratio is positive. Financial advisor prior business relation with the acquirer firm equals one if this financial advisor served as an advisor in a prior merger within the past three years, and zero otherwise. Acquirer announcement return is the cumulated abnormal return [AD-20, AD+1], with CRSP value–weighted index as the benchmark. Financial advisor reputation is measured with Thomson Financial rankings (1-25) of completed merger transaction value in the year prior to the announcement date. If an advisor is not included in the Thomson ranking, we assign a rank of 26. Acquirer industry adjusted ROA equals acquirer ROA (DATA13 / DATA6 from Compustat) minus industry median ROA in the last fiscal year before deal announcement. Relative size equals deal the transaction value from SDC divided by acquirer firm market value of equity. Stock payment dummy equals one if the deal is 100% equity financed, and zero otherwise. For each institution that owns both the acquirer and target stock, we calculate the institution’s combined announcement return from cross ownership [AD-20, AD+1] and multiply combined returns with the institution’s percent ownership in the acquirer firm and sum this product for all institutions with such cross ownership. This variable is included in all regressions to control for these institutions’ tendency to vote for the mergers. Same industry dummy equals one if the acquirer and the target are in the same Fama and French 48 industry classification, and zero otherwise. P-values are in parentheses.

36

Table 3 (continued) Acquirer Shareholder Voting and Advisor Target Equity Value Estimations

Dependent Variable: Acquirer Shareholder Approval Rate (1) (2) (3) Intercept 65.09 58.97 70.94

(0.00) (0.00) (0.00) Target advisor provides equity valuation 6.20

(0.03) Target advisor equity valuation ratio 15.87

(0.07) Category of target advisor equity value ratio 4.89

(0.01) Acquirer advisor provides equity valuation 3.53

(0.18) Acquirer advisor equity valuation ratio 9.70

(0.14) Ranking of acquirer advisor equity value ratio 2.68

(0.08) Target advisor prior business relation with the acquirer

-10.48 -15.85 -10.19 (0.20) (0.02) (0.21)

Target advisor reputation -0.16 -0.37 -0.13 (0.29) (0.07) (0.40)

Number of target advisors -0.01 8.74 0.47 (0.99) (0.02) (0.88)

Acquirer advisor prior business relation -3.76 -5.46 -4.01 with the acquirer (0.30) (0.17) (0.27) Acquirer advisor reputation -0.10 0.20 -0.05

(0.58) (0.33) (0.78) Number of acquirer advisors -0.51 6.64 0.84

(0.88) (0.11) (0.81) Acquirer firm announcement return 4.12 22.15 3.35

(0.69) (0.15) (0.75) Acquirer firm industry adjusted ROA 15.83 -12.65 18.18

(0.06) (0.40) (0.03) Relative size -1.38 -2.92 -1.83

(0.34) (0.38) (0.20) Stock payment dummy -2.23 -0.42 -2.30

(0.39) (0.90) (0.38) Institutional ownership* Combined return 20.67 -53.30 18.68 from cross-ownership (0.42) (0.09) (0.46) Toehold -0.14 -34.81 0.33

(0.99) (0.06) (0.98) Same industry dummy 1.67 2.28 2.22

(0.57) (0.55) (0.46) Year fixed effect Yes Yes Yes N 136 67 136 p value of chi square test 0.0077 0.1157 0.0042

37

Table 4 Acquirer Shareholder Voting and Advisor EPS Forecast

This table includes 136 deals where both the acquirer and target firms hire financial advisors. Acquirer shareholder approval rate equals the ratio of outside shareholders voting for a merger divided the total number of shares held by outside shareholders. We exclude shares held by executives, board of directors, and institutions affiliated with financial advisor from the calculation of outside shareholder holdings. Financial advisors’ earnings forecasts refer to their prediction of whether the merger is accretive or dilutive to the acquirer firm earnings. Financial advisors often provide earnings forecast for multiple years after merger completion. Since one-time merger changes can often affect the earnings in the first year after a deal, we use last year in the forecast period. We use the average of GAAP and Cash earnings forecast if the acquirer is a bank, and only GAAP earnings otherwise. Panel B further segments into subgroups according to whether EPS forecasts are dilutive. The subgroup of “No dilutive EPS forecasts” includes deals where neither EPS forecasts with synergies or without synergies is dilutive. The subgroup of “Dilutive EPS forecasts or no EPS forecast” includes deals where EPS forecast with synergies and/or without synergies is dilutive or financial advisors do not provide EPS forecasts. P-values are reported in parentheses.

Acquirer Shareholder Approval Rate

Panel A: Whether advisors give EPS forecasts

    All

No EPS forecast

EPS forecast

Difference   (p-value)

  (1) (2) (2) - (1) Target N 136 78 58

advisor Mean 61.18% 58.40% 64.92% 6.52% (0.01)

Median 63.92% 61.39% 66.13% 4.74% (0.04) Acquirer N 136 44 92 advisor Mean 61.18% 58.97% 62.23% 3.26% (0.24)

Median 63.92% 56.51% 65.74% 9.23% (0.13)

Panel B: Whether advisors give dilutive EPS forecasts

    All

Dilutive or no EPS forecasts

No dilutive EPS forecasts

Difference (p-value)

  (1) (2) (2) - (1) Target N 136 86 50

advisor Mean 61.18% 59.32 64.37% 5.05% (0.04)

Median 63.92% 62.15% 66.00% 3.85% (0.11) Acquirer N 136 58 78 advisor Mean 61.18% 59.41% 62.49% 3.08% (0.23)

Median 63.92% 59.22% 65.87% 6.65% (0.08)

38

Table 5 Acquirer Shareholders Voting and Advisor EPS Forecasts

This table details a tobit regression of acquirer shareholder voting on advisor earnings per share (ESS) forecasts. The dependent variable is acquirer shareholder approval rate (%) which equals the ratio of outside shareholders voting for a merger divided the total number of shares held by outside shareholders. We exclude shares held by executives, board of directors, and institutions affiliated with financial advisor from the calculation of outside shareholder holdings. Financial advisors’ earnings forecasts refer to their prediction of whether the merger is accretive or dilutive to the acquirer firms’ stand-alone earnings. Financial advisors often provide earnings forecast for multiple years after merger completion. Since one-time merger changes can often affect the earnings in the first year after a deal, we use last year in the forecast period. We use the average of GAAP and Cash earnings forecast if the acquirer is a bank, and only GAAP earnings otherwise.. Advisors provide EPS forecasts with synergies and/or without synergies. “Financial advisor provide at least one dilutive EPS forecast” means that the EPS forecast with synergies and/or EPS forecast without synergies is dilutive. Financial advisor prior business relation with the acquirer firm equals one if this financial advisor has served as an advisor in previous mergers during the last three years, and zero otherwise. Financial advisor reputation is measured with Thomson Financial rankings (1-25) of completed merger transaction value in the year prior to the announcement date. If an advisor is not included in the Thomson ranking, we assign a rank of 26. Acquirer announcement return is the cumulated abnormal return [AD-20, AD+1], with CRSP value–weighted index as the benchmark. Acquirer industry adjusted ROA equals acquirer ROA (DATA13 / DATA6 from Compustat) minus industry median ROA in the last fiscal year before deal announcement. Relative size equals deal the transaction value from SDC divided by acquirer firm market value of equity. Stock payment dummy equals one if the deal is paid entirely with stocks, and zero otherwise. For each institution that owns both the acquirer and target stock, we calculate the institution’s combined announcement return from cross ownership [AD-20, AD+1] and multiply combined returns with the institution’s percent ownership in the acquirer firm and sum this product for all institutions with such cross ownership. This variable is included in all regressions to control for these institutions’ tendency to vote for the mergers. Same industry dummy equals one if the acquirer and the target are in the same Fama and French 48 industry classification, and zero otherwise. P-values are in parentheses.

39

Table 5 (continued) Acquirer Shareholders Voting and Advisor EPS Forecasts

Dependent Variable: Acquirer Shareholder Approval Rate

(1) (2) (3) (4) (5) (6)

Intercept 79.28 73.41 74.57 76.86 77.40 77.26 (0.00) (0.00) (0.00) (0.00) (0.00) (0.00)

Target advisor provide EPS forecast

6.61 5.13 (0.01) (0.04)

Target advisor provide non-dilutive EPS forecast

4.95 4.27 (0.05) (0.09)

Acquirer advisor provide EPS forecast

4.55 3.70 (0.09) (0.19)

Acquirer advisor provide non-dilutive EPS forecast

5.02 4.63 (0.06) (0.08)

Target advisor reputation -0.29 -0.28 -0.34 -0.30 (0.01) (0.07) (0.01) (0.05)

Number of target advisors 0.10 -0.46 -0.12 0.12 (0.97) (0.89) (0.97) (0.97)

Target advisor prior relationship with acquirer

-6.15 -7.45 -6.28 -8.46 (0.46) (0.37) (0.45) (0.31)

Acquirer advisor reputation -0.26 -0.11 -0.29 -0.11 (0.06) (0.52) (0.03) (0.54)

Number of acquirer advisors 1.10 0.31 0.93 0.62 (0.76) (0.93) (0.79) (0.86)

Acquirer advisor prior relationship with acquirer

-3.01 -2.91 -3.16 -3.51 (0.41) (0.42) (0.40) (0.33)

Acquirer announcement return 4.17 5.39 4.36 5.21 6.25 4.54 (0.68) (0.61) (0.68) (0.62) (0.56) (0.67)

Relative size -1.50 -1.55 -1.35 -1.35 -1.14 -1.29 (0.26) (0.22) (0.32) (0.32) (0.38) (0.34)

Stock payment dummy -3.63 -3.94 -3.59 -3.57 -4.22 -4.09 (0.15) (0.12) (0.17) (0.17) (0.11) (0.12)

Institution combined return from cross-ownership

26.37 26.83 25.88 27.03 23.01 23.35 (0.30) (0.29) (0.31) (0.29) (0.37) (0.36)

Toehold 0.11 -2.46 2.69 -0.62 -2.67 1.88 (0.99) (0.84) (0.82) (0.95) (0.82) (0.87)

Same industry dummy -0.65 0.60 0.86 -0.31 0.04 -0.18

(0.81) (0.84) (0.77) (0.91) (0.98) (0.95)

Year fixed effect Yes Yes Yes Yes Yes Yes N 141 143 136 141 143 136 p value for chi square test 0.015 0.077 0.036 0.032 0.091 0.041

40

Table 6 Acquirer Shareholders Voting and Affiliated Analyst Recommendations

Acquirer shareholder approval rate equals the ratio of outside shareholders voting for a merger divided the total number of shares held by outside shareholders. We exclude shares held by executives, board of directors, and institutions affiliated with financial advisor from the calculation of outside shareholder holdings. Affiliated analysts work for the same investment bank as the financial advisors. We obtain analyst recommendations from I/B/E/S and First Call: 1 for “strong buy,” 2 for “buy,” 3 for “hold,” 4 for “underperform,” and 5 for “sell.” We classify an analyst recommendation as “initiate recommendation” if the analyst does not provide a recommendation pre-merger announcement but does after the announcement and “drop recommendation” if she provides a recommendation before but not after the announcement. This table includes 45 deals where both the target and acquirer financial advisors’ affiliated analysts make at least one recommendation during a period from three years before a merger announcement to the shareholder voting date. P-values are reported in parentheses.

Acquirer Shareholders Approval Rate

Panel A: Whether affiliated analysts provide recommendation

All cases

Affiliated analysts maintain or initiate recommendation

Affiliated analysts drop

recommendation

Difference (p-value)

(1) (2) (2) - (1)

Target-affiliated analyst

N 45 20 25  

Mean 67.45% 70.78% 64.79% -5.99% (0.21)Median 70.44% 73.46% 67.81% -5.65% (0.10)  

Acquirer-affiliated analyst

N 45 26 19  

Mean 67.45% 65.23% 70.50% 5.27% (0.22)Median 70.44% 68.85% 71.52% 2.67% (0.51)

Panel B: Whether affiliated analysts improve recommendation

   

All cases

Affiliated analysts initiate, upgrade, or do

not change in recommendation

Affiliated analysts downgrade or drop recommendation

Difference (p-value)

(1) (2) (2) - (1)

Target-affiliated analyst

N 45 19 26 Mean 67.45% 70.75% 65.04% -5.71% (0.23)Median 70.44% 73.75% 68.39% -5.36% (0.10)

Acquirer-affiliated analyst

N 45 21 24 Mean 67.45% 69.51% 65.65% -3.86% (0.40)Median 70.44% 70.44% 70.50% 0.06% (0.62)

41

Table 7 Acquirer Shareholder Voting and Affiliated Analysts’ Recommendations

This table details a tobit regression of acquirer shareholder voting on affiliation analyst recommendations in 45 deals where both the target and acquirer financial advisors’ affiliated analysts make at least one recommendation during a period from three years before a merger announcement to the shareholder voting date. Acquirer shareholder approval rate (%) equals the number of outside shareholder shares that vote for a merger divided the total number of shares held by outside shareholders. We exclude shares held by executives, board of directors, and institutions affiliated with financial advisor from the calculation of outside shareholder holdings. Affiliated analysts work for the same investment bank as the financial advisors. Analyst recommendations are from I/B/E/S and First Call: 1 for “strong buy,” 2 for “buy,” 3 for “hold,” 4 for “underperform,” and 5 for “sell.” We classify an analyst recommendation as “initiate recommendation” if the analyst does not provide recommendation before a merger announcement but does so after the announcement and as “drop recommendation” she provides recommendation before but not after the announcement. In model (2), the category of analyst recommendations equals one if the analysts initiate recommendations after the deal announcement, two if they upgrade recommendations, three if they maintain the same recommendations, four if they downgrade recommendations, and five if they drop recommendations. In model (3), the category of analyst recommendations equals one if the analysts initiate recommendation after the deal announcement, two if their average recommendation upgrade is more than one level, three if their average upgrade is one level or less, four if they maintain the same recommendations, five if their average recommendation downgrade is one level or less, six if their average downgrade is more than one level, and seven if they drop recommendations. Financial advisor prior business relation with the acquirer firm equals one if this financial advisor has served as an advisor in previous mergers during the last three years, and zero otherwise. Financial advisor reputation is measured with Thomson Financial rankings (1-25) of completed merger transaction value in the year prior to the announcement date. If an advisor is not included in the Thomson ranking, we assign a rank of 26. Acquirer announcement return is the cumulated abnormal return [AD-20, AD+1], with CRSP value–weighted index as the benchmark. Acquirer industry adjusted ROA equals acquirer ROA (DATA13 / DATA6 from Compustat) minus industry median ROA in the last fiscal year before deal announcement. Relative size equals deal the transaction value from SDC divided by acquirer firm market value of equity. Stock payment dummy equals one if the deal is financed entirely with equity, and zero otherwise. For each institution that owns both the acquirer and target stock, we calculate the institution’s combined announcement return from cross ownership [AD-20, AD+1] and multiply combined returns with the institution’s percent ownership in the acquirer firm and sum this product for all institutions with such cross ownership. This variable is included in all regressions to control for these institutions’ tendency to vote for the mergers. Same industry dummy equals one if the acquirer and the target are in the same Fama and French 48 industry classification, and zero otherwise. P-values are in parentheses.

42

Table 7 (continued) Acquirer Shareholder Voting and Affiliated Analysts’ Recommendations

Dependent Variable: Acquirer Shareholder Approval Rate

(1) (2) (3) Intercept 59.07 92.94 88.26

(0.00) (0.00) (0.00) Dummy for target-affiliated analyst issue recommendations after deal announcement

11.29 (0.04)

Category of target-affiliated analyst recommendations after deal announcement

-4.32 (0.07)

Category (multiple) of target-affiliated analyst recommendations after deal announcement

-2.72 (0.08)

Dummy for acquirer-affiliated analyst issue recommendations after deal announcement

4.47 (0.42)

Category of acquirer-affiliated analyst recommendations after deal announcement

-2.18 (0.28)

Category (multiple) of acquirer-affiliated analyst recommendations after deal announcement

-1.16 (0.40)

Target advisor reputation 0.04 0.01 0.01 (0.92) (0.97) (0.98)

Acquirer advisor reputation 0.73 0.75 0.71 (0.16) (0.13) (0.15)

Acquirer firm announcement return 25.46 28.55 26.96 (0.32) (0.29) (0.31)

Acquirer industry adjusted ROA 28.13 26.12 26.63 (0.20) (0.24) (0.24)

Relative size -2.37 -2.40 -2.46 (0.14) (0.13) (0.12)

Stock payment dummy -1.73 -1.87 -1.28 (0.80) (0.79) (0.86)

Institutional ownership* Combined returns 55.89 43.23 44.34 from cross-ownership (0.14) (0.28) (0.27) Toehold -11.59 -10.54 -11.02

(0.49) (0.52) (0.50) Acquirer advisor prior business relation with the acquirer

-13.96 -13.51 -13.50 (0.01) (0.01) (0.01)

Target advisor prior business relation with the acquirer

-3.01 -4.36 -3.35 (0.79) (0.70) (0.77)

Same industry dummy 0.74 1.59 1.59 (0.88) (0.76) (0.76)

Year fixed effect Yes Yes Yes N 45 45 45 p value for chi square test 0.2054 0.2343 0.2452

43

Table 8

Acquirer Shareholder Voting and Future Advisor Retention This table details 158 future merger transactions by 58 acquirers that make at least one acquisition within three years after the deal effective date from our acquirer shareholder voting sample. Acquirer advisor retention dummy equals one if the acquirer firm uses the same advisor in the future deal as in the previous voting deal and equals zero if future transactions do not include financial advisors or financial advisors are not the same as before. If the future transaction has more than one acquirer advisors, we set the advisor retaining dummy equals to one if any of these advisors are from the previous transaction in our voting sample. Financial advisor reputation is measured with Thomson Financial rankings (1-25) of completed merger transaction value in the year prior to the announcement date. If an advisor is not included in the Thomson ranking, we assign a rank of 26. CEO turnover equals one if the CEO changed after the previous deal; otherwise, it equals zero. Relative size equals deal value divided by acquirer market capitalization 21 days before deal announcement. P-values are in parentheses while marginal effects are in brackets. Marginal effects are the change in the retention of an acquirer advisor for a one standard deviation change in a continuous variable or a shift from zero to one for an indicator variable.

Dependent Variable: Acquirer Advisor Retention

(1) (2)

Intercept -2.37 -2.65 (0.00) (0.00)

Acquirer shareholder approval rate 2.22 3.24 (0.02) [0.53]

(0.02) [0.85]

Previous return in voting sample 0.97 (0.47) [0.25]

Previous advisor reputation one year before deal announcement

-0.01 (0.36)

[-0.00]

CEO turnover 0.08 (0.86) [0.02]

Relative size -0.09 (0.80)

[-0.02]

N 158 104

p value of chi square test 0.08 0.00