Mega vs. Boutique: Who is the Better Financial Advisor in Mergers and Acquisitions?

Transcription

1 Mega vs. Boutique: Who is the Better Financial Advisor in Mergers and Acquisitions? J. Diana. Wei * Weihong Song ** Abstract This paper examines the effect of using mega vs. boutique investment banks as financial advisors on wealth gains and deal outcomes in corporate takeovers. Boutique investment banks are defined as independent financial advisors whose focus is M&A advising. This is the first paper to examine the role of financial advisors from this perspective. We identify 202 boutique banks from 533 M&A advisory firms for a large sample of M & A deals from 1995 to 2006 and analyze how firms choose mega vs. boutique advisors. After controlling for the endogenous choice of financial advisors, we examine how investment banks expertise and independence affect deal premium, wealth gains, completion speed, and success rate. We find that deal size and deal attitude is an important factor that affects acquirer s choice of boutique versus mega advisors. We also find that on average, boutique advisors achieve a higher success rate while it takes them a longer time to complete deals. Boutique banks expertise in valuation is more appreciated than their independence by both the client and the market. They are better advisors in particular deals because of their expertise. However, due to the nature of the boutique, there is a limitation on what kind of deals that these niche players can exploit the most of their advantage. JEL Classification: G34, G39 Keywords: Financial advisor, boutique, mega, mergers and acquisitions * Ph.D. Candidate in Finance, College of Business, University of Cincinnati, OH Tel: weije@ .uc.edu ** Assistant Professor in Finance, College of Business, University of Cincinnati, OH Tel: weihong.song@uc.edu

2 Mega vs. Boutique: Who is the Better Financial Advisor in Mergers and Acquisitions? Abstract:This paper examines the effect of using mega vs. boutique investment banks as financial advisors on wealth gains and deal outcomes in corporate takeovers. Boutique investment banks are defined as independent financial advisors whose focus is M&A advising. This is the first paper to examine the role of financial advisors from this perspective. We identify 202 boutique banks from 533 M&A advisory firms for a large sample of M & A deals from 1995 to 2006 and analyze how firms choose mega vs. boutique advisors. After controlling for the endogenous choice of financial advisors, we examine how investment banks expertise and independence affect deal premium, wealth gains, completion speed, and success rate. We find that deal size and deal attitude is an important factor that affects acquirer s choice of boutique versus mega advisors. We also find that on average, boutique advisors achieve a higher success rate while it takes them a longer time to complete deals. Boutique banks expertise in valuation is more appreciated than their independence by both the client and the market. They are better advisors in particular deals because of their expertise. However, due to the nature of the boutique, there is a limitation on what kind of deals that these niche players can exploit the most of their advantage. 1. Introduction Financial advisors play a very important role in mergers and acquisitions. Their characteristics and roles influence the outcomes of the deals that these advisors are involved in. For example, previous studies (see Rau(2000), Hunter and Jagtiani (2003), Allen, Jagtiani, Peristiani, Saunders (2004)) show that the market share of financial advisors commercial banks and investment banks affects not only the deal premium, the speed of completion, the success rate, but also the total wealth gains created in these deals. However, to date there is no evidence on the impact that advisors expertise and independence have on the wealth gains and other deal outcomes. Financial advisors who are independent and focus solely on providing advisory services are considered 1

3 boutique investment banks by the industry. Their growing popularity is attested to by the hot IPO market for these boutique investment banks, especially in the last four years. Investor zest comes from boutique banks fast climbing market share. These small but specialized firms are stealing M&A business market share from their larger brethren at increasing rates. The combined market share of Goldman Sachs, Morgan Stanley and Merrill Lynch fell to 77% through August 2006 from 94% in Meanwhile mergers and acquisitions deals advised by Evercore, Lazard, and Greenhill are running at 28% through August 2006, up from 18% in While boutique investment banks have recently gained attention and popularity in the market, there is little (or no) evidence that merging firms benefit from deals advised by these specialized firms. On one hand, boutique banks are independent and free of conflict, which should have positive effect on wealth gains and deal outcomes. On the other hand, they are usually small, less diversified and not well-known and therefore do not have much reputation capital at stake, which in turn may be viewed as negative factors. This paper is the first to examine the effect of using specialized, independent advisors in M&A transactions. We attempt to address several unanswered questions in this paper. First, how much market share have boutique investment banks gained in the recent sizzling corporate takeover market? Second, is their growing popularity associated with superior performance at value creation? More generally, does the advisor s expertise and independence affect wealth gains and other deal outcomes? Lastly, we also investigate how initially firms choose either mega or boutique banks as their advisors. Using manually compiled data on a large sample of deals announced between 1995 and 1 See Wall Street Journal, Oct.2 nd,

4 2006, we examine whether the use of a boutique advisor, either by the target firm or by the acquiring firm, affects acquisition announcement returns, deal premiums, the probability of deal completion, and the speed of deal completion. As boutique investment bank Lazard states on its company website, one of the main competitive advantages of this kind of financial advisors is their independence. They are free of many of the conflicts that can arise at larger financial institutions as a result of their varied sales, trading, underwriting, research and lending activities. In recent years, the big, diversified firms have been hit hard by heavy investigations and allegations of conflicts of interest. Conflicts of interest involve stock research, mutualfund trading abuses, accounting fraud and proprietary trading, etc. The desire to avoid any potential future conflicts has increased the demand by management and boards of directors for trusted and unbiased advice. Moreover, while mega investment banks increasingly rely on their trading desks for profits and leave limited resources devoted to the traditional long-term investment-banking relationships, boutique banks thrive with their expertise on financial advice as these are their focus and main products. Although, mega banks still dominate the M&A advisory market, their market share has slipped in recent years. Diversification seems to be a double edged sword for the mega banks: on one hand, it creates potential conflicts of interest; on the other hand, prior relationships arising from other lines of business enhance the certification effect discussed in papers such as Allen et al (2004) 2. For example, if a mega bank has prior lending relationships with the acquirer or the target, they have access to private information about the firm s cash flows, financial resources and other crucial accounting 2 In the study of Allen et al. (2004), this certification effect takes the form of increased abnormal return to targets whenever their merger advisor is their own bank (with whom the target has had a prior lending relationship). 3

5 information which can assure a more accurate valuation. Although boutique banks as niche players have less conflicts of interest issues, they are by nature less diversified and surely don t benefit from the certification effect described above. Another advantage of mega investment banks is their reputation effect. When we try to examine how the choice of a boutique vs. a mega bank affects the wealth gain and other deal outcomes, we recognize that the choice can be endogenous. Although boutique banks have the advantage of being independent and focused, they tend to be smaller and less known. The reputation of mega firms may affect how acquirers pick their advisors. For example, if the acquirer s manager has a bad deal to push through, s/he might want to choose firms that are willing to bend the rules for her/his firm. Mega investment banks have their reputation at stake and are less likely to go along with management looking for private benefits through expansion. On the other hand, when the acquirer s management faces a good deal, they might want to choose a reputable advisor to signal their valuation. It is an empirical question whether boutique bank s expertise and independence create value and positively affect deal outcomes, given that firms may self-select certain financial advisors. This article empirically examines the effect of using mega vs. boutique investment banks as financial advisors on wealth gains and deal outcomes in corporate takeovers. Using the criterion of being independent and specialized in M&A advisory services, we manually classify financial advisors as boutique or mega investment banks 3. We further analyze how acquirers/targets choose mega vs. boutique advisors. After controlling for the endogenous choice of financial advisors, we focus on how investment banks expertise and independence affect takeover premium, wealth gains, deal 3 To our knowledge, this is the first paper to make this distinction between financial advisors. 4

6 completion speed, and completion rate. This paper contributes to the literature in several ways. First, we identify 202 boutique banks from 533 M&A advising firms; second, this is the first paper to analyze what drives the choice of boutique banks vs. mega banks. Deal size is one important factor that affects the choice, so is deal attitude. A boutique investment bank is more likely to be chosen as the advisor for hostile takeovers and when a deal is relatively small. Third, in terms of who is the better financial advisor, we find that on average, boutique advisors achieve higher success rates although it takes them longer time to complete deals. The boutique s expertise in valuation is more appreciated than their independence by both the client and the market. They are better advisors in particular deals because of their expertise. However, due to the nature of the boutique, there is a limitation on what kind of deals that these niche players can exploit the most of their advantage. 2. Literature Review Our paper is related to existing research on financial advisors in M&A. The previous literature has primarily focused on two aspects: the early literature analyzes the fee structure and contract terms of deals (McLaughlin (1990), Hunter and Walker (1990)); more recent studies investigate advisors reputation effects. Bowers and Miller (1990) and Michael et al. (1991) both find that high reputation investment bankers have the ability to detect better mergers, but do not provide any bargaining advantage. Servaes and Zenner (1996) study the acquiring firm s choice of using investment banks or not. They find that acquiring firms are more likely to use an investment bank when the acquisition is more complex and when they have less prior acquisition experience. However, returns are lower when investment banks are used. The distinction disappears 5

7 after controlling for the characteristics of the transaction. In other words, the returns earned by the acquirers do not depend on whether an investment bank is used, after controlling for the determinants of investment banking choice. This result shows that controlling for deal characteristics and the endogenous choice of financial advisors is very critical for studying financial advisor s effect on wealth gains. Rau (2000) investigates the determinants of the market share of the investment banks acting as advisors and how the investment banks market share affects their clients. He finds that acquirers advised by first-tier banks complete a significantly greater proportion of the tender offers, but similar proportions of mergers are completed across the different categories of investment banks. In mergers, clients of first-tier investment banks earn lower announcement-period excess returns. In tender offers, clients of firsttier investment banks earn higher announcement-period excess returns. Rau s (2000) study of the financial advisor s effect on deal outcomes and client s wealth gains characterizes advisors only from the perspective of market share. However market share, may be the result of a wide range of business characteristics, therefore we examine the question by focusing on the more fundamental characteristics that differentiate Mega vs. Boutique banks: financial advisors advisory focus and independence. Hunter and Jatianni (2003) also examine how market share affects deal outcomes. They find that top-tier advisors are more capable of completing deals. In terms of the speed of completing a deal, top tier advisors were found to be more efficient. But the announcement returns of acquiring firms hiring top tier investment banks are actually lower. In contrast, Kale, Kini and Ryan (2003) argue that one reason why previous studies do not find a significant role of advisor reputation may be that they failed to 6

8 control for the reputation of the opponent s advisor. After controlling for this and some other confounding factors, they document a significant advantage of employing prestigious financial advisors. However, it is not clear what the financial advisor s reputation means other than their market share in the context of takeovers. Bao and Edmans (2006) contend that past performance is a more appropriate measure of investment bank reputation instead of market share. They document significant longhorizon persistence in the average announcement returns to acquisitions advised by an investment bank. Speed and raw completion ratio are also persistent; moreover, banks that outperform with respect to these measures also exhibit superior shareholder value creation. In general, the previous literature focuses on studying the reputation of financial advisors defined as their market share. A notable exception is Allen et al (2004). They study the tradeoff between conflict of interests and bank certification effect of using commercial banks with prior lending relations as financial advisors in mergers and acquisitions. They find positive evidence of a net bank certification effect for target firms only. In contrast, acquirer s abnormal returns are either negative or insignificantly different from zero in all cases. Deviating from the perspective of reputation, this paper focuses on how financial advisors prior relation with clients affects wealth gains and deal outcomes. While the motivation of our paper is similar to theirs, our focus is on whether the expertise and independence of financial advisors affect wealth gains and deal outcomes. 3. Data and Methodology 7

9 The sample of M&A is collected from the Securities Data Corporation s (SDC) U.S. Mergers and Acquisitions Database. We select both completed and withdrawn domestic M&A with announcement dates between 1995 and 2006, and require both the acquirer and the target to be public firms. For earlier deals, information about their financial advisors is much scarcer and therefore difficult for us to define whether they were diversified financial advisors or boutique investment banks. We focus on deals in which the acquirer has at least one advisor. We include only acquisitions in which the acquiring firms control less than 50% of the shares of the target firms before acquisition announcements and that deal values are at least $10 million. These selection criteria yield 2,604 deals. We further exclude deals that have a premium of more than 200% or less than -50%. Our final sample consists of 2,351 acquisitions from 1995 to 2006, involving 533 separate investment banks in our sample. Among these investment banks, 202 (37.9%) advisors are classified as boutique investment banks 4. In terms of number of deals advised by boutique investment banks, 18.5% of the 2,351 deals have acquirers advised by boutique and 20.9% of them have targets advised by boutique. There are 348 tender offers and 2,003 mergers (See Table 1). Number of deals analyzed in various tests varies due to different sample selection criteria and data availability. Insert Table 1 Here 3.1. Univariate Analysis 4 For each investment bank, we find the introduction of the company on its website. If the company s website is not found, we search the internet for articles mentioning the company. Based on the introduction about the company and our definition on boutique banks, we classify investment banks into diversified mega banks or boutique banks. 8

10 Table 1 shows the distribution of the sample. The whole sample has 2,351 unique deals. There are 87 deals that only have acquirer advisor(s) and no target advisor, while the rest have both acquirer advisor(s) and target advisor(s). From the acquirer advisor s point, there are 1,712 (72.8%) deals advised by mega investment banks and 434 (18.5%) deals advised by boutique investment banks. From the target advisor s point, there are 1,458 (62%) deals advised by mega investment banks, and 491 (20.9%) deals advised by boutique investment banks. Additionally, we have 141 deals in which both the acquirer advisor and the target advisor are boutique investment banks. Insert Table 2 Here Table 2 provides a comparison of deal characteristics for different categories. Deals are divided into four categories according to the acquirer-target advisor combination: mega vs. mega, mega vs. boutique, boutique vs. mega, and boutique vs. boutique. From the simple comparison between these four groups, we can see that the average deal size, acquirer size, relative size, acquirer M/B ratio, target M/B ratio are all much bigger when both sides are advised by mega investment bank than when both sides are advised by boutique investment bank. For example, the average deal size for mega vs. mega is million dollars, while the average deal size for boutique vs. boutique is only million dollars. Interestingly, the percentage of hostile deals is much higher in the boutique vs. boutique category than in any other combinations. Table 2 suggests that the choice of boutique advisor is significantly related to the size of the deal and the size of the merging firms. Larger deals tend to be advised by mega investment banks at least on one side. Small deals are more likely to be advised by boutique banks. One possible reason for this difference is that boutique banks tend to be 9

11 small and less known, which might make them humble on the fees they charge. It is conceivable that smaller firms that cannot afford the high fees charged by mega investment banks might have to choose boutique banks. However, boutique advisors expertise seems to be more valuable in hostile takeovers. This can be seen in the higher percentage of hostile deals when both sides are advised by boutique investment banks. In other words, when deals are hostile, both sides are more likely to choose boutique investment banks as advisors. While the comparison on deal characteristics can shed some light on the reasons why firms may choose boutique vs. mega advisors, we are mainly interested in the effects of this choice on deal results. To see how boutique advisors differ from their larger brethren on deal outcomes, we compare the duration, the premium, announcement period returns, the combined wealth and the success rate of deals advised by mega or boutique banks. We make these comparisons for the whole sample and for mergers and tender offers separately, and the results are presented in Table 3. Insert Table 3 Here Deal duration refers to the number of days between deal announcement and the effective date (if the deal succeeds) or the withdrawn date (if the deal fails). In order to obtain as much and as accurate information as possible about the deal premium, we follow the method described in Officer (2000) to calculate the premium as follows. We first use the aggregate amount of each form of payment offered to target shareholders and compare it with the target s market value of equity 43 days prior to the bid announcement to compute the premium. For deals that have this number missing or premium calculated out of the range (-50% to 200%), we use the initial offer price per share of the target 10

12 stock and the percentage change of the final offer and compare the final offer price with target s price 43 days prior to the bid announcement. For deals with missing values or outliers even using this method, we replace the either missing or inaccurate premium information by the recorded premium in the SDC database. By combining these three methods to calculate the deal premium, we try as best as we can to reduce the possibility of missing premiums and inaccurate premium information. The announcement period return is the short run performance of both target s and acquirer s stocks upon the announcement of the deal. We examine 3-day (-1, +1) announcement period abnormal returns. The combined wealth gain is calculated in two forms following Kim et al (1988) and Kale, Kini and Ryan (2003): (1) the weighted average of acquirer s and target s short run performance and (2) the dollar value of the aggregate short run gain or loss of both acquirers and targets. Investigating the results in Table 3 provides a first insight into how the involvement of boutique banks affects the results of the deals. First, we look at how the acquirer s choice of advisors affects deal outcomes (Panel A). We can see that on average, mega advised merger deals last 137 days with an average premium of 49%. Boutique advised deals last 154 days with 51% premium. The difference between mega and boutique on duration is statistically significant, while the difference on premium is not. Another significant difference is the dollar amount combined wealth. Boutique advised deals are value-creating while mega advised deals are value-destroying. However, in terms of percentage gain or loss, there is no significant difference. From the point of view of the targets choice, there is no significant difference in any of the six deal outcome measurements (Panel B). However, it is interesting to note 11

13 that when the target hires a boutique advisor, the choice of using boutique advisors or not by acquirers has an explanatory power for both duration and deal premium. Panel C shows that when the target uses a boutique advisor, it takes a longer time to complete the deal if the acquirer s advisor is also a boutique bank. On average, the duration for boutique advisors on the acquirer side is 157 days compared to 138 days for mega acquirer advisors. However, the acquirer pays a significantly lower premium when using boutique advisors vs. mega advisor: 43.7% versus 52.3%. The comparison indicates that the benefit to an acquirer using a boutique investment bank is most remarkable when the opponent s advisor is a boutique investment bank. This is interesting because it suggests that an acquirer can benefit more than a target from using a boutique advisor. This could be due to the boutique s independence. An advisor s conflicts of interest are more likely to affect the acquirer than the target when the acquirer s advisor has a self interest in pushing through the deal, they might suggest that the acquirer pay a higher premium. It could also be due to the boutiques greater expertise in valuation. The most complex part of a deal is target valuation. Therefore when the target firm hires a boutique advisor, it may signal that the asymmetric information about target s value is larger. This would be exacerbated when the acquirer hires a mega advisor rather than a boutique advisor. Thus, if the mega advisor has greater conflicts of interest or if the boutique has greater skills in target valuation, the acquirer ends up paying a higher premium when hiring a mega advisor. Insert Table 4 Here Because mergers and tender offers are different along many dimensions, we examine them separately. Table 4 shows that merger deals advised by boutique banks on 12

14 the acquirer side last on average 163 days, compared to deals advised by mega, which last 146 days. The difference in the duration is statistically significant. In general, tender offers take a much shorter time to complete than merger deals. In the case of tender offers, boutique advised deals also last longer. On average, it takes 4.5 more days to complete boutique-advised deals than their mega advised counterparts, although the difference is not significant. This analysis shows that the significant difference in duration presented in Table 3 is probably driven by mergers rather than tender offers. The use of boutique advisors for acquirers seems to matter more in mergers than in tender offers, which is reasonable since mergers involve more negotiation, and hence provide more opportunity for boutiques to fully exploit their advantages in being more specialized and focused then mega advisors. In terms of premiums, there is no significant difference regardless of whether the acquirer s advisor is a boutique or not in either mergers or tender offers. There is also no significant difference whether the target s advisor is boutique or not in mergers. However, in tender offers, when the target s advisor is a boutique investment bank, the premium is 71.9% on average, which is higher than mega advised deals average of 58.5% (at 10% significance level). Boutique investment banks seem to have an advantage in advising targets in tender offers. Consistent with previous studies, we also find that the target s announcement period return is much higher on average than acquirer s announcement return. The average target 3-day CAR is 18.3% while the average acquirer CAR is -2.5%. To examine the difference between mega and boutique advisors, we look at mergers and tender offers separately. The average acquirer announcement period return is about -3% 13

15 in merger deals. Although deals advised by boutique advisors on the acquirer side have slightly higher acquirer CAR, the difference is not significant. For tender offers, the average acquirer CAR is 0.1% and 0.2% for mega and boutique acquirer advisor respectively and the difference is not significant. The result is similar when classification is based on the target s advisor. The difference of target CAR between deals advised by boutique or mega is not significant in either merger or tender offers. Furthermore, to understand whether the use of boutique advisors brings any wealth gains to society as a whole, we need to look at the effects on combined wealth. Following Kale, Kini and Ryan (2003), we create two variables CWLTH and CPWLTH to capture the combined wealth gain or loss. CWLTH is the dollar amount of combined wealth generated during announcement period and is defined as follows: CWLTH= TWLTH+AWLTH TWLTH= TMV*TCAR*(1-Bidder Toehold) AWLTH=AMV*ACAR CPWLTH= TMV/(AMV+TMV)*TCAR+AMV/(AMV+TMV)*ACAR where TMV (AMV) is the market value of outstanding shares of the target (acquirer) four weeks before the announcement; TCAR is the target announcement period abnormal return. ACAR is the acquirer announcement period abnormal return. CPWLTH is the weighted average of TCAR and ACAR. When examining the CWLTH, we find some significant results in two settings characterized by acquirer s advisor. In merger deals, the average combined dollar wealth loss is 222 million dollars for deals advised by mega investment banks compared to 74 million combined dollar wealth gain for deals advised by boutique investment banks. The 14

16 difference is statistically significant (p-value=0.014). This result suggests that deals advised by boutique banks on the acquirer side appear to create value. In tender offers, the opposite is observed. Deals advised by boutique are value-destroying while deals advised by mega are value-creating. The difference is significant at 10% level. It seems that in tender offers, boutique investment banks produce more value to targets than acquirers. When they advise targets, the target receives a higher premium on average. In contrast, when they advise the acquirer, the losses to the acquirer are so great that the combined wealth declines. However, if we look at the combined wealth gain in percentage forms, there is no significant difference across any scenario. The market does not seem to react differently upon the announcement of deals advised by mega or boutique firms on either the acquirer side or the target side in both merger and tender offer subsamples. Another variable of interest in Table 4 is the percentage of successful deals. In mergers, deals advised by boutiques have success rates of 92.63% or 94.29% (depending on whether the classification is on acquirer or target side). Deals advised by mega banks only have 89.23% or 87.36% success rates, respectively. It seems that boutique advisors are more likely to get deals through. The general conclusion stemming from Table 4 is that boutique banks appear to be superior advisors in certain cases. Although on average it takes a longer time for deals they advise to be completed, they are more likely to close deals than mega banks are. This higher success rate does not appear to be achieved by sacrificing quality as there is no significant difference in the premium, TCAR, ACAR, CPWLTH in most cases. In some cases, deals advised by boutique banks reflect even better quality measurements. 15

17 For example, when the target is advised by a boutique bank in a tender offer, they receive significantly higher premium. Also, the combined dollar wealth gain generated by merger deals is much higher when the acquirer s advisors are boutique banks than when they are mega banks. 3.2 Multivariate Analysis a) The Determinants of Use of Boutique Advisors From the univariate analysis we can see that the use of boutique advisors is associated with particular deal characteristics. To further investigate the relation between the choice of a boutique bank and deal characteristics, we run several probit regressions in which the dependent variable takes a value of 1 if a boutique advisor is used and 0 otherwise. The results are presented in Table 5. Insert Table 5 Here. First, we examine the acquirer s choice of advisors in mergers and tender offers respectively. All three models we run on mergers show that deal size is a very important determinant of the acquirer s choice of advisor. The larger the dollar size of the deal, the less likely it is an acquirer will choose a boutique investment bank as their advisor. This is consistent with the univariate analysis: larger deals tend to be advised by mega investment banks at least one side. In model 3, the variable HOSTILE is also significant and positive. This means that when the deal is hostile, the acquirer is more likely to choose a boutique advisor, consistent with the analysis in the previous section. Two other significant determinants of an acquirer s choice of advisor are the target s sales growth rate during the prior fiscal year and the target s debt-to-equity ratio 16

18 for the prior fiscal year. The larger the changes in sales, the more likely it is for an acquirer to choose a boutique advisor. At the same time, the higher the debt equity ratio, the less likely for an acquirer it is to choose a boutique advisor. When the change in sales is large, it indicates that the target is in the growth stage of the life cycle, which signals the complexity of valuating the target. In this situation, the expertise of a boutique bank is more valuable in arriving at an accurate valuation. Therefore, an acquirer is more likely to choose a boutique bank when target is growing rapidly. On the other hand, a high D/E ratio indicates the target has a greater ability to borrow. This signals that target has less asymmetric information, making the selection of a boutique bank less valuable. In tender offers, deal size is also the significant determinant of advisor. One factor that is not significant in mergers, but significant in tender offer is the degree of relatedness of the industry in which the acquirer and the target are in. If the tender offer is a horizontal one, the acquirer is more likely to choose a boutique bank. Panel B of table 5 repeats the results of target s choice of advisors. As was the case for acquirers, deal size is a highly significant determinant in both mergers and tender offers. The larger the deal is, the less likely it is that a boutique is chosen as the advisor by the target. On the target s side, the medium of exchange affects the decision to hire a boutique advisor. Targets are more likely to have boutique investment banks as advisors when stock is used as the medium of exchange. This is true in both mergers and tender offers. This is consistent with the hypothesis that the boutique bank has more expertise in valuation. When the target shareholders are to be paid with acquirer s stock instead of cash, the asymmetric information problem is more severe and therefore it is harder for the target to accurately evaluate the offer. The target s choice of a boutique bank in this 17

19 circumstance indicates that the boutique bank s expertise in valuation is recognized and valued. b)the Impact of Boutique versus Mega Advisors on Deal Outcomes i) Deal Duration In order to see if the effect of the type of advisor on the duration of the deal is subsumed by other potential variables that were not considered in the univariate analysis, we proceed to analyze the combined effects of several factors in the case of mergers and tender offers. Insert Table 6 Here Table 6 shows that in mergers, relative size, deal size, toehold, the deal being hostile and whether there is a competing bidder are factors that affect the duration of the deal. Most importantly, the use of a boutique bank by either the acquirer or the target significantly elongates the deal s duration. The acquirer s use of boutique is significant at 1% level and target s use of a boutique bank is significant at 10% level. Factors that lengthen the duration are deal size, toehold and competing bid. These factors signal the complexity of deals. In general, the more complicated a deal is, the longer the duration. This is quite intuitive. Factors that shorten the duration are relative size and hostile. The interesting and somewhat surprising finding is that hostile mergers on average last 57.5 days less than friendly mergers. In tender offers, hostile deals on average take 76.7 more days to complete than friendly ones. The significance of the use of boutique by either the acquirer or the target disappears when the analysis is done on tender offers. The only significant factors that 18

20 affect deal duration in tender offers are deal size, the medium of payment and the deal attitude. This may not be surprising since larger deals, hostile deals and/or deals paid with stock are more complex and therefore may elongate the duration. Our analysis indicates that, for merger deals, it takes longer time to complete deals advised by boutique investment banks than those by mega investment banks, even after controlling for other deal characteristics. This is true on both acquirer and target sides. One possible explanation is that the boutique banks tend to be smaller and have fewer resources; therefore it takes them longer to complete the deals However, it might also because boutique banks spend more time doing due diligence. In order to see whether our result is conflated with boutiques being smaller, we control for the rank of investment banks in the robustness check section. The results still hold when the advisors rank based on market share is taken into account. ii) Deal Premium When examining the effects of using boutique banks on the quality of transactions, we recognize that the use of a boutique bank is endogenously determined by deal or firm characteristics. This is confirmed by the results in the previous section on the choice of boutique advisor. There are several deal characteristics strongly related to the use of boutique advisors. This endogenous selection process therefore has the potential to bias estimates of the impact of using boutique advisors on the merger outcomes that we evaluate. To control for the potential self-selection bias, we employ both OLS regressions and a two-equation treatment procedure. Insert Table 7 Here. 19

21 The independent variables in the OLS regressions are dummy variables for boutique advisors to the acquirer, to the target, and other deal characteristics. We find that relative size, toehold, competition, acquirer s market to book ratio, target sales growth, target debt equity ratio significantly affect deal premium in mergers, while only toehold significantly affects deal premium in tender offers. It seems that neither acquirer s choice nor target s choice on boutique vs. mega affects deal premium. However, this OLS regression does not consider the endogeneity problem of the choice of boutique advisors by the merging firms. We therefore use a two-stage treatment procedure to address this endogeneity problem. In the first step, we estimate the choice of boutique advisor when deal characteristics differ. The effect of boutiques on premium is then examined in the second step, where we estimate the deal premium using deal characteristics and boutique dummies. In merger deals, both the acquirer s choice and the target s choice of boutique bank significantly affects deal premiums. The acquirer s boutique advisor is negatively related to the deal premium, which is beneficial to the acquirer. In tender offer deals, neither the acquirer s nor the target s boutique advisor is significantly affects the deal premium. This result is consistent with our univariate analysis. Our multivariate analysis suggests that boutique advisors have a bargaining advantage in mergers, and that the advantage seems to be more on the acquirer side than on the target side. The advisors performance in terms of deal pricing is comparable to the performance of mega advisors in tender offers. iii) Announcement Period Returns 20

22 We also use both the OLS regressions and two-stage treatment procedures in analyzing announcement period returns to control for the endogeneity problem. When we first examine how a boutique advisor affects either the acquirer s or the target s short-run performance CARs 5 (-1,+1) in mergers and tender offers using OLS regressions, we find that acquirer s choice of boutique is not significantly related to acquirer s short run announcement returns. Instead, it is positively related to target s short run performance in merger deals, as shown in model 1 of Panel A of Table 8. Insert Table 8 Here However, when we take into account the endogeneity problem and use a twostage treatment procedure, some interesting results emerge. In the case of mergers, the use of a boutique bank by the acquirer is significantly and positively related to the acquirer s announcement abnormal returns and has no significant effect on target s short run abnormal returns. On the other hand, the use of boutique by the acquirer has no effect on either the acquirer s performance or the target s performance in tender offers. The use of a boutique advisor by the target also seems to benefit acquirer shareholders, as it is positively related to acquirer 3-day CAR. Overall, the use of the boutique advisor, either by the acquirer or by the target is welcomed by the market since the market react positively to the merger announcement. The benefit of using a boutique advisor is most valuable for the acquirer s shareholders because the possible conflicts of interest of mega banks are more likely to affect the acquirer s shareholders than to the target s shareholders since conflicts of interest might induce advisors on the acquirer side to suggest higher premium or push through bad deals. 5 When calculating CARs, the market model is estimated using continuously compounded returns over the 200-day period ending 50 days before the initial acquisition announcement. The CRSP value weighted index is used as a market proxy. 21

23 iv) Combined Wealth Gain There are two ways to look at the combined wealth gain. Panel A of table 9 reports results on combined dollar wealth gain, and Panel B reports the weighted average of CARs of the merging firms. Insert Table 9 Here The use of a boutique bank by the target is associated with reduction on the combined wealth of 335 million dollars in merger deals. However, there is no significant difference of mega and boutique in the weighted average of short run abnormal returns of acquirer and target. Therefore, whether target s use of boutique is associated with valuedestroying deals or not is still arguable. On the contrary, the acquirer s use of boutique in merger deals has no significant effect on the combined dollar wealth, but it is associated with higher weighted average of CARs (the coefficient is.012 and it is significant at 10% level). Therefore, there is some evidence for value-creation of deals advised by boutique on the acquirer side. Again, the positive effect of using a boutique advisor on combined wealth is more notable in mergers. In tender deals, the use of boutique by either the acquirer or the target has no significant effect on combined wealth in either form Robustness Check We have mentioned that boutique banks are relatively smaller firms than the mega banks and boutique banks usually have smaller market shares than mega banks. Previous literature finds that the rank of an investment bank based on market share affects certain aspects of deal outcomes. It is not clear whether our results are driven by the fact that boutique banks are usually small firms and have less market shares. In order to avoid the 22

24 result being contaminated by this confounding factor, we run the robustness check by including the rank of the investment banks as an additional control. We employ two methods to account for the market share effect. First, following Rau (2000), we calculate the average yearly rank of each investment bank in this sample, according to the annual rank of each investment bank on the basis of the total value of transactions advised during the year from SDC s advisor league tables. If a bank is not listed as having advised any acquisitions during the year, a rank of one plus the number of investment banks that participated in the market that year is assigned. Then we take the average rank across the years for each investment bank and use it as the advisor ranking variable. For deals that have multiple advisors, we use the lead advisor s rank. We calculate average rank for each deal for both the acquirer and target advisors. We repeat all the multivariate analysis by including this control variable. Our findings about boutique investment bank in previous sections do not change. The second method to control for the rank of investment bank is to examine a sub-sample of deals advised by top 30 investment banks ranked annually on the basis of transaction value. In this sub-sample, all the advisors are comparable in terms of size and advisor ranking, and hence the confounding effect is largely reduced. Univariate analysis of this sub-sample indicates that the percentage of deals advised by boutique banks is lower than the whole sample, but main results from multivariate analysis still hold. Therefore, we can conclude that our findings about boutique advisors are robust. 4. Conclusion 23

25 In this article, we investigate whether boutique investment banks or mega investment banks are better advisors in mergers and acquisitions. Boutique investment banks have the expertise and independence, but they tend to be smaller and are definitely less diversified. Mega investment banks have resources and diversity, but suffer from potential conflicts of interest. Therefore, which one is the better advisor is subject to empirical examination. Our empirical analysis suggests that boutique investment banks are more likely to be chosen as the advisors when the values of deals are smaller and when the deal attitude is hostile. These are the factors affecting the choice of using a boutique vs. mega advisor. Once the endogenous choice problem is accounted for, the difference that a boutique advisor makes in deal outcomes is more noteworthy in mergers than in tender offers. It is also more remarkable on the acquirer side than on the target side. On average, it takes a longer time to complete deals advised by boutique advisors, but the success rate is higher. There is some evidence that the use of boutique advisors by the acquirer can reduce the premium, improve the acquirer s short term performance and create social welfare in merger deals. There are also situations where mega and boutique advisors are comparable in their service, especially in tender offers. Generally speaking, boutique investment banks are slightly better at getting deals done without sacrificing its client s benefit or the social welfare, while issues such as conflicts of interest embedded in mega investment banks do not seem to be a big concern to the market. Our findings also suggest that boutique advisors expertise in valuation is more appreciated than their independence by both the client and the market. They are better advisors in certain deals due to their specialization and expertise in M&A advisory. 24

26 However, due to the nature of the boutique, there is a limitation on what kind of deals that these niche players can exploit the most of their advantage. The diversity and the broad spectrum of complexity of deals in the market determine that both boutique and mega advisors will have their own supporters and beneficiaries. The balance of their market share might shift from time to time, but it is likely to remain fairly stable. 25

27 Appendix: List of Variables Used in Tests 1. Boutique Dummy: whether the firm is advised by boutique investment banks. 2. Deal size: log of the transaction value reported by SDC. 3. Relative size: the ratio of acquirer size relative to target size. 4. Acquirer Size: the market value of equity of acquirer measured at two months prior to acquisition announcement. 5. Acq. M/B (Target M/B): the acquirer s (target s) market-to-book ratio, measured as the ratio of year-end market value of common stock to the book value of equity for the prior fiscal year (COMPUSTST items 24*25/60). 6. Same_Industry: a dummy variable equals to 1 if the acquirer and target have the same first 2-digits SIC. 7. Target ROE: the ratio of earnings to average equity for the prior fiscal year (COMPUSTAT items 20/(60+60(t-1)). 8. Target sales growth: the proportional change in sales over the prior fiscal year. 9. Target D/E: the ratio of debt to equity of the target for the prior fiscal year (COMPUSTAT items 6/60) ; 10. Stock: is a dummy variable that takes value of 1 if at least 50% of transaction is paid by stock, 0 otherwise; 11. Hostile: is a dummy variable indicating deal attitude, 0 if friendly, and 1 if hostile; 12. Tender: is a dummy variable indicating that deals are identified as a tender offer by SDC; 13. Toehold: is the fraction of the target s stock held by the acquirer prior to merger announcement; deals in which the toehold is larger than 50% are dropped; 14. Competition: is a dummy variable equal to 1 if there is at least one competing bidder for the same target after the deal announcement and before the deal completion/termination; 15. Premium: is the percentage difference between the offer price and target share price four weeks prior to the announcement date. 16. Complete: is a dummy indicating whether the deal is completed or withdrawn. 17. ACAR: acquirer s announcement period (-1,+1) abnormal return. 18. TCAR: target s announcement period (-1,+1) abnormal return 19. CWLTH: Combined wealth in dollar amount. It is the market value of target times TCAR plus market value of acquirer times ACAR. 20. CPWLTH: Weighted average of ACAR and TCAR. The weight is the market value of the merging firms. 26

28 References 1. Allen, Linda, Julapa Jagtiani, Stavros Peristiani, and Anthony Saunders, The Role of Bank Advisors in Mergers and Acquisitions, Journal of Money, Credit, and Banking 36, Bao, Jack, and Alex Edmans, How should Acquirers Select Investment Banks Advisors? Working paper, MIT. 3. Bowers and Robert Miller, Choice of Investment Banker and Shareholders' Wealth of Firms Involved in Acquisitions, Financial Management, Hunter, William C., Jagtiani, Julapa, An Analysis of Advisor Choice, Fees, and Effort in Mergers and Acquisitions, Review of Financial Economics, 12, Hunter, William C., Walker, Mary B., An Empirical Examination of Investment Banking Merger Fee Contracts, Southern Economic Journal, 56, Kale, Jayant, Omesh Kini, and Harley Ryan, Jr., Financial Advisors and Shareholder Wealth Gains in Corporate Takeovers, Journal of Financial and Quantitative Analysis 38, McLaughlin, R., Investment-banking Contracts in Tender Offers: An Empirical Analysis, Journal of Financial Economics 28, McLaughlin, R., Does the Form of Compensation Matter? Investment Banker Fee Contracts in Tender Offers, Journal of Financial Economics 32, Michael, A., Shaked, I., and Lee, Y., An Evaluation of Investment Banker Acquisition Advice: The Shareholders Perspective, Financial Management 20, Officer, M.S., Termination Fees in Mergers and Acquisitions, Journal of Financial Economics, 69, Rau, P., Investment Bank Market Share, Contingent Fee Payments, and the Performance of Acquiring Firms, Journal of Financial Economics 56, Servaes, Henri, and Marc Zenner, The Role of Investment Banks in Acquisitions, Review of Financial Studies 9,