The effect of the choice to hire an investment bank on the acquirer wealth creation in synergy motivated mergers and acquisitions

(1)

The effect of the choice to hire an investment bank on the acquirer

wealth creation in synergy motivated mergers and acquisitions

Doortje Broersen

Master Thesis Finance 29 November 2014 Abstract

The role of investment banks in mergers and acquisitions is still under research by the existing literature. This thesis contributes to the existing literature by investigating the effect of the choice to hire an investment bank and the investment bank’s compensation contract used on the acquirer wealth creation in synergy motivated mergers and acquisitions. In line with the empirical evidence, the results implicate investment banks do not have a significant effect on the value created in mergers and acquisitions. Neither does the investment bank’s compensation fee selected by the client firm have a significant effect on the acquirer wealth creation in mergers and acquisitions. Therefore, both the results on the role of investment banks and on their compensation fees suggest investment banks do not create added value in the synergy motivated merger and acquisition process.

(2)

Over the past decades, mergers and acquisitions have shown to be highly frequent recurring corporate events within the United States and world economies. These types of corporate events represent massive reallocations of resources within and across industries, making them of great importance for the economy in general. Therefore, mergers and acquisitions have been subject to a large amount of research papers (Andrade, Mitchel & Stafford, 2001). Investment banks are considered to be important intermediaries to facilitate the merger and acquisition process (Servaes & Zenner, 1996). But the role of investment banks in these types of transactions is far less covered in the existing literature (Servaes & Zenner, 1996). For that reason this thesis will focus on the added value for client firms when they choose to engage the services of an investment bank in mergers and acquisitions.

Several theories discussed in previous literature provide motives for firms to engage the services of investment banks. The services of investment banks can be captured in two main functions: investment banks help to lower the transaction costs and the contracting costs of mergers and acquisitions (Servaes & Zenner, 1996). Both these functions should have a value-enhancing role on the shareholder value created in mergers and acquisitions. However, Servaes and Zenner (1996) state the idea about the importance of investment banks has not remained unchallenged, implying the use of investment banks creates less value than expected or can even be value destroying for the shareholder value of the client firm.

This thesis describes two different issues associated with the involvement of investment banks in order to shed light on the concerns about the value created by investment banks. The first concern is related to the commonly used compensation contract for investment banks. In practice, client firms choose investment banks’ compensation fees that are based on deal completion in 80% of the deals (McLaughlin, 1990). This form of compensation fee should create the right incentive for investment banks to act upon firm’s objectives. However, compensation fees contingent on deal outcome are not particularly dependent on service quality. These types of investment bank’s compensation could also create an incentive for investment banks to advise on closing the deal regardless of whether the merger or acquisition creates value for the client firm (McLaughlin, 1990). In other words, the contingent form of payment could, instead of preventing investment banks to act upon their own interest, enhance a potential agency problem between the investment bank and the client firm, leading to lower client firm value (McLaughlin, 1990).

The second concern could provide another explanation for lower shareholder value creation, besides the agency problem between the investment bank and the client firm. Lower

(5)

shareholder value in mergers and acquisitions could be caused by management that is subject to hubris or by a second agency problem; an agency problem between management and shareholders within the firm, rather than by the involvement of the investment bank itself (Wang and Whyte, 2010). Ill-motivated managers tend to engage in value-destroying merger and acquisition activity (Berkovitch and Narayan, 1993). If ill-motivated managers have a higher incentive to hire an investment bank and management motives for mergers and acquisitions are excluded from the analysis, investment banks may seem the cause of lower firm value, while in fact firm management is the problem. In other words, the involvement of investment banks might in essence not be value destroying, but investment banks are more often involved in the value-destroying merger and acquisition activity of firms’ managements. Therefore, looking at the role of investment banks in mergers and acquisitions in which management problems are hardly present or entirely absent should shed some new light on the role of investment banks in mergers and acquisitions.

Empirical evidence on the effect of hiring an investment bank on acquirer wealth creation varies widely (Wang & Whyte, 2010). Among others, Fuller, Netter and Stegemoller (2002) and Servaes and Zenner (1996) argue the wide variation in empirical evidence on the abnormal returns of both target and bidder firms around deal announcement is caused by differences in firm and deal characteristics. In addition, Andrade et al., (2001) state these differences do not only affect acquirer abnormal returns, but also the motivation for the choice to hire an investment bank. Rau (2000), McLaughlin (1992), Kale, Kini and Ryan (2003) and Bowers and Miller (1990) suggest the wide variation in empirical evidence on the abnormal returns is caused by the reputation of the employed investment bank that is affecting the abnormal returns of acquiring firms. Bao and Edmans (2011) find a significant positive relation between the role of investment banks and abnormal returns in their study to the fixed effect of investment banks on merger and acquisitions returns (their research is explained in more detail later in this thesis). In contrast, Wang and Whyte (2010) find, when managerial rights are strong (in case managers are difficult to discipline by shareholders and agency problems are more prone), the use of investment banks increases and abnormal returns are negatively affected. However, when controlling for the interaction between managerial rights and the use of investment banks, the use of investment banks does not significantly contributes to the changes in shareholder value around deal announcement.

To contribute to the existing literature, this thesis tries to find significant empirical evidence on whether investment banks have a value-adding role in synergy motivated mergers and acquisitions. The research focuses on bidding firms, since they are assumed to rely more

(6)

heavily on investment banks’ services, thereby affecting bidder firm value more substantially (Servaes & Zenner, 1996). As mentioned before, there are two concerns associated with the use of investment banks and the value they create for client firms in mergers and acquisitions. This thesis particularly examines if the use of investment banks in mergers and acquisitions creates value in the absence of the second concern, that means if a firm’s management is not subject to hubris or agency problems. Thus in line with Wang and Whyte (2010), this thesis brings the role of investment banks in connection to management motives for mergers and acquisitions. However, where Wang and Whyte (2010) investigate the effects of strong managerial rights on the role of investment banks and the client firm wealth effects created, this thesis particularly looks at whether investment banks have a value enhancing role in mergers and acquisition in case all firms that are subject to hubris or agency problems are excluded from the sample (the Wang and Whyte (2010) paper is discussed in more detail in the theoretical framework). Hereby, avoiding potential differences in wealth effects associated with investment banks that are actually caused by ill-motivated management and enabling to examine the isolated effect of the role of investment banks in mergers and acquisitions that are from nature valuable. The results should indicate if the first concern, a potential agency problem between the investment bank and client firm, leads to lower shareholder value in mergers and acquisitions.

Following the paper of Berkovitch and Narayan (1993), using a sample existing of mergers and acquisitions in the United States over the time period from 1985 to 2012, this thesis makes a distinction between synergy motivated and hubris or agency motivated mergers and acquisitions. Within the synergy sample, the causal effect between the choice to engage the services of an investment bank and the shareholder wealth created for the client firm is then tested. In addition, to test whether a potential agency problem between the investment bank and the client firm is caused by the investment bank’s compensation contract selected, this thesis also looks at the causal effects of different investment bank’s compensation contracts on the client’s firm value in mergers and acquisitions.

In line with Fuller et al. (2002), Servaes and Zenner (1996) and Andrade et al. (2001), this thesis finds the choice to hire an investment bank has no significant effect on the acquirer wealth creation in mergers and acquisitions. Looking at the results from both the total sample of mergers and acquisitions and the subsample of synergy motivated mergers and acquisitions, differences in acquirer abnormal returns are rather caused by the various firm and deal characteristics than by the use of investment banks. Especially, the payment method used to buy the equity of the target firm seems to be an important determinant of the

(7)

shareholder wealth created in mergers and acquisitions. According to Bao and Edmans (2011) the effects on acquirer abnormal returns caused by deal characteristics should be attributed to the role of investment banks, since deal characteristics are often the responsibility of the advisor in mergers and acquisitions. However, excluding deal characteristics from the analyses provides the same insignificant results related to the role of investment banks in mergers and acquisitions. In addition, the investment bank’s compensation contract selected by the client firm also appears to have no significant effect on the shareholder wealth created. The results suggest firms are able to select the right compensation contract to avoid a potential agency problem between the investment bank and the client firm. Or investment banks do not have a substantial role in the synergy merger and acquisition process, which means they have not enough power to influence the wealth created. This automatically puts less weight on the contract form selected (McLaughlin, 1990). Overall, based on this thesis’ findings there is no reason to believe investment banks have a value-destroying role on acquiring firm values when hubris or agency motivated mergers and acquisitions are excluded from examination. Neither do the results prove investment banks have the value-enhancing role suggested by previous (theoretical) literature in synergy motivated mergers and acquisitions. This means shareholders of acquiring firms should be always asking themselves if their management is hiring an investment bank for the right reasons during the merger and acquisition process.

The remainder of this thesis is structured as follows. Section 2 gives a more extensive discussion on the existing literature and empirical evidence, including theories on the role of investment bank and motivations for mergers and acquisitions and how they can be linked together. Section 3 describes the methodology used, followed by section 4 covering the construction of the dataset. Section 5 presents the findings and section 6 provides insights on the robustness of the results presented in section 5. At last, section 7 concludes on the role of investment banks in mergers and acquisitions.

2. Theoretical framework

2.1 Motivation engaging the service of an investment bank

To gain insights in the variety of value adding roles investment banks might have in the merger and acquisition process, various theories are discussed below. According to the skilled advice hypothesis of Bao and Edmands (2011), certain investment banks have the ability to provide valuable advice to client firms in the merger and acquisition process. In their theoretical approach to the determinants of investment bank employment, Servaes and Zenner

(8)

(1996) explain there are three motives to hire an investment bank that support the skilled advice hypothesis: they can lower information asymmetry, transaction costs and contracting costs. The transaction costs- and contracting costs hypotheses (including the information asymmetry lowering role of investment banks in the transaction cost hypothesis) have proved to be the foundation for a predominant part of the existing literature on the role of investment banks in mergers and acquisitions.

2.1.1 Transaction costs hypothesis

The transaction costs hypothesis suggests investment banks have an advantage in analysing potential mergers and acquisitions compared to individual firms (Bao & Edmands, 2011, Servaes & Zenner, 1996). This advantage comes from their experience, expertise and economies of scale in acquiring information and research (Servaes & Zenner, 1996). Using this advantage, investment banks can lower the cost of target or bidder firms identification, reduce the information asymmetry between participating firms, value the participating firms and put together lower takeover bids or higher premiums (Servaes & Zenner, 1996). Bowers and Miller (1990) support that investment banks can increase the shareholder’s wealth creation in mergers and acquisitions, by identifying (more rapidly) the most suited bidders or targets and by assisting their clients in the valuation of the acquisition premium. Bowers and Miller (1990) and Bao and Edmans (2011)point out investment banks also have a key role in the bargaining process, negotiating the transaction to the most favourable outcome. Kale, Kini and Ryan (2003) believe financial intermediaries perform two distinct roles, also in line with the transaction costs hypotheses: identifying and/or structuring the best mergers and acquisitions and advising on the strategic activities during the transaction. Kale et al. (2003) focus on the fact investment banks can support shareholder wealth creation at the other parties’ cost (for example advising not only on the most beneficial offer for bidders, but also on the deployment of defensive strategies for targets).

Taking their knowledge and expertise in information management, research and takeover strategies together, investment banks are therefore most valuable in complex transactions or in case firms have less experience. Investment banks are also particularly valuable when there is a high degree of information asymmetry (Servaes & Zenner, 1996).

2.1.2 Contracting costs hypothesis

The contracting costs hypothesis posits that investment banks can lower contracting costs by their monitoring and signalling function, because both functionalities reduce agency costs

(9)

within the client firm (Servaes & Zenner, 1996). Financial intermediaries need to sell their products in the market repeatedly to existing or new customers, with as result that investment bank’s reputation is considered a major driver of success (Kale et al., 2003). As seen with IPOs, reputation concerns lead to self-selection by investment banks and monitoring of client firms’ managements to avoid value-destroying transactions (Wang & Whyte, 2010). Therefore, the investment bank’s certification of the value of a merger or acquisition provides reliable information about the quality of the firm and deal (Servaes & Zenner, 1996) and simultaneously motivates management to act upon shareholders objectives (Kale et al., 2003).

In short, the contracting and previously discussed transaction costs hypotheses explain why investment banks are considered to be important intermediaries. But as mentioned in the introduction, the idea about the importance of investment has not remained unchallenged. Instead of adding value to their client firm’s shareholders, the role of investment banks is also associated with less sufficient value creation then expected or even value destroying activity (Servaes & Zenner, 1996). In other words, there might be a negative relation between the investment bank and the client firm. This can be explained by the potential agency problem between the investment bank and the client firm (McLaughlin, 1990).

2.2 Investment bank contracts and the potential agency problem

To gain a better understanding of the potential agency problem between investment banks and client firms, according to the existing literature, it is essential to look at the motivation for particular investment bank contracts and why the contracts are associated with a potential agency problem (McLaughlin, 1990). Because financial intermediaries need to sell their products in the market repeatedly, it is in the investment bank’s interest to market their (contribution to) projects for client firms as valuable as possible, despite the amount of resources they have spend or the quality of service they have delivered (Chemmanur & Fulghieri, 1994). Moreover, according to Bao and Edmans (2011), investment banks allocate the majority of their human capital to client coverage, with as main responsibility to pitch new projects instead of deal execution. This makes it difficult for firms to determine the credibility of financial intermediaries and distinguish between investment banks that act in good faith and investment banks that act in their own interest instead of client firms’ objectives. This causes an agency problem between the investment bank and client firm. Because investment banks could choose to spend less resources and time on a project of their client firm, but demand a compensation appropriate for a much higher amount of resources and time spend, thereby increasing their payoff (Chemmanur and Fulghieri, 1994). Nevertheless, hiring firms

(10)

can use various compensation fee contracts as a tool to avoid this potential agency problem and secure good quality service from investment banks (McLaughlin, 1992).

2.2.1 Investment bank compensation contracts

If compensation fee contracts have the right structure, they can create the correct incentives for both the client firm and the investment bank. Logically, a firm should ensure the contract creates incentives for the investment bank to act upon firm’s objectives and the investment bank should agree on a contract by which it can optimize its expected payoff and gets recognition for its knowledge and capabilities (McLaughlin, 1990). According to McLaughlin (1990), there are three standard categories of investment bank contract forms: fixed-fee, share-based fee and value-based fee contracts, each encouraging bankers to provide services to client firms in a different manner. Fixed-fee contracts are not related to offer outcome and often used in atypical situations, when investment banker’s role is limited (e.g. if bidders already own a great part of the target’s equity or targets only need an opinion letter) and when there is automatically less room for the investment bank to act upon its own interest (McLaughlin, 1990). Share-based fee contracts are contingent on the deal outcome (the number of shares acquired) and are the usual contract for bidders (McLaughlin, 1992). Value-based fee contracts are also paid on completion of the deal, but the amount is Value-based on the value of the transaction and frequently used by targets (McLaughlin, 1992). In practice, on average 80% of investment banks’ fee compensations are paid contingent on deal outcome (McLaughlin, 1992). This implies that in a predominant part of mergers and acquisitions, contingent compensation contracts are believed to create the right incentives for investment banks and can be an effective solution to the potential agency problem between the investment bank and the client firm.

2.2.2 Potential agency problem investment banks

The downside of compensation fees that are based on takeover success is that deal completion is not necessarily the most optimal outcome for a client firm (e.g. in case of inferior take over terms or a suboptimal offer value) (McLaughlin, 1992). Therefore, by trying to create the right incentive for investment banks to act upon firms’ objectives, potential agency problems between the investment bank and client firm are perhaps rather enlarged instead of resolved (McLaughlin, 1992). In IPOs for example, if the investment bank is engaged with its own capital, it is inclined to provide the best advise in their own interest (Wang & Whyte, 2010). In contrast, with mergers and acquisitions investment bank’s shareholder wealth is not

(11)

directly reliant on a client firm’s performance. In other words, investment banks are encouraged to support deals regardless of the value created by the deal and not incentivised to provide the best possible quality of service (Wang & Whyte, 2010). Especially, when investment banks are not penalized for advice on value-destroying activities, the value created by mergers and acquisitions could be significantly lower (Servaes & Zenner, 1996). Nevertheless, the potential agency problem is mitigated if investment banks have less room to act upon their own interests. Which would also make the choice for the compensation contract less important (McLaughlin, 1990). How the role of investment banks differs among different types of mergers and acquisitions will be discussed on the basis of the different motives of mergers and acquisitions later in this section.

2.3 Motivations mergers and acquisitions

There could be another explanation for the lower shareholder wealth created associated with the role of investment banks in mergers and acquisitions. There could be a second agency problem, between management and shareholders within the firm or management that is subject to hubris, that are affecting a firm’s value, but also the choice to engage the service of an investment bank and the quality delivered by the investment bank. Thereby suggesting in case there are no or hardly any management problems, the effect of the involvement of investment banks on the client firm shareholder value creation should be less severe or even mitigated. This in line with the study of Wang and Whyte (2010) to the relation between managerial rights in acquiring firms and the decision to use an investment bank in mergers and acquisitions. Wang and Whyte (2010) argue firms with strong managerial rights (in which case managers are difficult to discipline by shareholders and agency problems are more prone) are associated with lower firm value and more likely to rely on the services of an investment bank then firms with low managerial rights. Hereby implicating an interaction between management problems, the use of investment banks and negative shareholder wealth effects (the study of Wang and Whyte (2010) is in more detail discussed in 2.5 Empirical evidence). To explain this idea, it is again crucial to look at the motives for mergers and acquisitions in general and subsequently how these various motives can be linked to the choice to hire an investment banks.

To understand the implications for bidder firm values associated with the various mergers and acquisitions motives and how the different motives affect the choice to engage the services of investment banks and the role of investment banks, this section provides a discussion on the different theories on merger and acquisition motives. In accordance to

(12)

Berkovitch and Narayan (1993), there are three major motives for mergers and acquisitions: the synergy, the agency and the hubris motive. This thesis includes a fourth motive, the in the existing literature more recently discussed overvaluation motive (Savor & Lu, 2009).

2.3.1 Synergy motive

According to the synergy motive, firm management only engages in mergers and acquisitions to generate growth opportunities and efficiency improvements in the form of synergies, to create economies of scale, market power, improved asset management, better bidder investment opportunities and market discipline (e.g. removal of incompetent target management) (Andrade, et al., 2001 and Dong, Hirshleifer, Richardson & Hong Teoh, 2006). This management objective is in the interest of firm shareholders and in line with the idea mergers and acquisitions are corporate transactions meant to create shareholder value (Berkovitch & Narayanan, 1993). The synergy motive therefore implies mergers and acquisitions activity only occurs if it results in gains to both bidder and target shareholders (Berkovitch & Narayanan, 1993). However, the synergy motive does not provide an explanation for merger and acquisition activity that appears to lack growth opportunities and value creation in the form of synergies.

2.3.2 Agency motive

The agency motive suggests mergers and acquisitions are simply motivated by self-interested management. One manifestation of the agency problem between firm management and shareholder management is illustrated by the free cash flow hypothesis of Jensen (1986). The free cash flow hypothesis posits managers tend to engage in empire building, in other words grow beyond the optimal size of the firm, to increase their management power. Another form of agency motivated mergers and acquisitions are takeovers without apparent synergy or growth opportunities, but that are in the area of expertise of the manager, with the purpose of increasing the firm’s dependence on the manager (Shleifer & Vishny, 2003). Thus in agency motivated mergers and acquisitions managers tend to extract wealth available to shareholders for their own welfare, thereby inducing agency costs that reduce the potential value of mergers and acquisitions. The agency motive therefore provides an explanation for worthless or even value-destroying mergers and acquisitions. When bidder and target gains are negative, takeovers are most likely initiated by self-serving management (Berkovitch & Narayanan, 1993).

(13)

2.3.3 Hubris motive

The hubris motive assumes managers do not act upon their own benefit, but they support mergers and acquisitions because of management’s overbearing presumptions their valuation of the bidder or target firm is correct (Roll, 1986). In this case mergers and acquisitions are motivated by managers’ mistakes, because they are likely to overestimate the value of the deal (or underestimate the value of the deal, but then there would be no transaction). The hubris motive therefore implies potential gains from merger and acquisition synergies are offset by overpayments of bidder management (Berkovitch & Narayanan, 1993) and there are no combined firm shareholder gains (Roll, 1986).

2.3.4 Market-timing motive

The more recent market-timing theory has provided a fourth motive for deals that do not entail any synergies (Savor & Lu, 2009). Following this theory, overvalued bidders use their equity as currency in mergers and acquisitions. Expecting the mispricing of their stock will become public and investors will eventually adjust their beliefs about the firm value downwards, bidders could use mergers and acquisitions as a manner to convert their overvalued equity into (less overvalued) assets at a discount. Thereby creating shareholder value (Savor & Lu, 2009). Because market-timing motivated managers have the same incentive as synergy motivated managers, to create shareholder value, this thesis assumes market-timing motivated mergers and acquisitions have the same implications for the choice to engage the services of investment banks as synergy motivated transactions. Therefore this motive is disregarded from further discussion.

2.3.5 Conclusion motivation mergers and acquisitions

As mentioned before, this thesis predicts investment banks are important financial intermediaries and firms contract investment banks rationally, which does not explain why investment banks are possibly causing lower firm shareholder value in mergers and acquisitions. The different discussed theories on motives for mergers and acquisitions do provide an explanation for lower firm value. If mergers and acquisitions are motivated by synergies, expected total shareholder gains are positive. But if mergers and acquisitions are agency or hubris motivated, this leads to no or even negative shareholder gains. However, the agency or hubris motives do not explain why it seems that firms choosing to hire an investment bank are associated with lower deal outcomes. Therefore it is essential to look at how the choice to engage the services of an investment bank is affected by the different

(14)

motives for mergers and acquisitions. If hubris and self-serving managers, already causing lower firm value, are more likely to hire an investment bank, but management motives are not included in the analysis, this could be a plausible explanation why it seems investment banks are causing lower client firm wealth effects.

2.4 Discussion on the role of investment banks

To show whether lower firm value associated with the role of investment banks is in fact attributable to agency or hubris motivated managers that are more likely to hire an investment bank, motives for merger and acquisitions are linked to the theory on the motivations to use an investment bank in the discussion below.

Firm management has, despite their motive for the deal, justified reasons to hire an investment bank if management believes investment banks are in essence valuable and do fulfil an important role in lowering transaction and contracting costs of the mergers and acquisitions process (McLaughlin, 1990). This means, as long as synergy motivated managers recognize the value of the role of investment banks, there is no reason to believe ill-motivated managers have greater incentives to hire an investment bank than synergy motivated managers (Wang & Whyte, 2010). However, looking at both the lowering transaction and contracting costs function of investment banks from a firm’s, subject to agency or hubris problems, perspective, it might be justified that ill-motivated managers are more likely to hire an investment bank. And in addition, it can show how the misuse of investment banks by management subject to hubris or agency problems can lead to the engagement of investment banks in value-destroying activity.

Whereas it is difficult to define how hubris or agency motivated managers are better served with the lowering transaction costs function of investment banks, certain firm or deal characteristics that increase the need for lowering transaction costs (and thus hiring an investment bank) can be associated with self-serving or hubris managers, as shown below. Looking at the contracting costs hypothesis, the certification role of investment banks certainly can provide greater incentives for self-serving or hubris than for synergy motivated managers to involve an investment bank, as described below.

2.4.1 Lowering transaction costs role in ill-motivated mergers and acquisitions

An example of why ill-motivated managers benefit more from the lowering transaction costs role of investment banks is given by Roll (1986). In his paper to the hubris motive for mergers and acquisitions, he explains in case an industry is characterised by broad-ranging

(15)

information asymmetries, there is more room for hubris motivated mergers and acquisitions. This is because due to the lack in available information, managers rely more heavily on their personal valuations increasing the probability they overestimate the value of the deal. But also, in line with the lowering transaction costs hypothesis, the higher the information asymmetry, the more valuable the experience, expertise and economies of scale in acquiring information of investment banks (Servaes & Zenner, 1996). Both these findings suggests, hubris motivated managers rely more on the services of investment banks and firms would in essence benefit more from the services of the investment bank, except that management engages in value destroying activity. Another example is given by Wang and Whyte (2010). As stated previously in this thesis, firms subjected to severe agency problems are more likely to pursue empire building. However, these firms are also associated with lower firm performance and therefore are expected to experience more resistance from shareholders against take over deals. If there is more shareholder resistance, management experiences more difficulties closing the deal. Since investment banks have better knowledge and capabilities for deal completion, agency motivated managers have a greater incentive to hire an investment bank, because investment banks can help them to handle the shareholder resistance and complete the value-destroying deal (Wang & Whyte, 2010). A third example is given by Jensen (1986). He states self-serving managers do not only engage in mergers and acquisitions more often, they tend to engage more in diversifying take over programs. The problem with diversifying mergers and acquisitions is that they are generally more complex and firms cannot rely on their own knowledge and experience in the industry (Servaes & Zenner, 1996). Again, this assumingly creates an interest for investment banks because of their lowering transaction costs function and self-serving management benefit in essence more of the involvement of an investment bank (Servaes & Zenner, 1996). To conclude, these three examples provide reasons why investment banks are in essence more valuable to management subject to hubris or agency problems and how investment banks support their value destroying activity, that lead to lower firm shareholder value. However, it is still possible managers choose to hire investment banks to facilitate the lowering transaction costs function even if hubris or agency problems are less severe (Wang & Whyte, 2003). Hence, whether firm and deal characteristics of ill-motivated mergers and acquisitions could validate the more frequent expected choice for investment banks is questionable. Therefore it is important to look at the contracting costs function of investment banks.

(16)

2.4.2 Lowering contracting costs role in ill-motivated mergers and acquisitions

From a shareholder point of view, if a firm’s management is more likely to be subject to hubris or agency problems, shareholders have reason to rely more heavenly on the support of investment banks (Servaes & Zenner, 1996). Because the monitoring and signalling role of investment banks can be used to monitor and control a client firm’s management (Wang & Whyte, 2010). That would mean, the role of investment banks should be value enhancing for the client firm. However, from a self-serving manager point of view, if managers believe they receive personal benefits from the investment bank or think investment banks support deal completion, which in turn regularly leads to higher bonuses, they have a higher tendency to hire an investment bank (Rau, 2000). But more important, managers subject to hubris or self-interested managers can use investment banks to justify their behaviour, thereby creating public confidence (McLaughlin, 1992). In other words, the fairness opinion of a financial intermediary could be used as a protection against shareholders’ legal actions instead of an incentive for managers to act upon shareholders’ objectives (Servaes & Zenner, 1996). And although investment banks are dependent on their reputation, they could decide to collaborate with self–interested managers, because of the likelihood agency motivated managers will provide more deals in the future (Wang & Whyte, 2010). Or investment banks could ignore reputation concerns in case ill-motivated managers select investment bank compensation contracts that are in line with their own interest, but (unintentionally) also provide a greater pay off for investment banks (McLaughlin, 1992). Therefore, the certification role of investment banks leads in essence to a greater stimulus for shareholders of agency or hubris infected firms to hire an investment bank, which would only create shareholder wealth. But because ill-motivated firm managers themselves can benefit more from the choice to engage the services of an investment bank (McLaughlin, 1992), again investment banks are believed to support the value-destroying activity of ill-motivated management, leading to lower shareholder firm value.

2.4.3 Conclusion discussion on the role of investment banks

To come to an end, the previously discussed lowering transaction costs hypothesis does not particularly proves hubris or self-serving managers are more likely to involve an investment bank. Even when management problems are less severe, firms still choose to hire an investment bank for their lowering transaction costs function (Wang & Whyte, 2003). However, looking at the contracting costs hypotheses, shareholders of firms that are subject to hubris or agency problems show they have a greater incentive to hire an investment bank

(17)

(Servaes & Zenner, 1996). In addition, ill-motivated management also profits from the lowering contracting function of investment banks, providing several benefits that are assumingly less valuable for synergy motivated managers (McLaughlin, 1992, Rau, 2000 and Servaes & Zenner, 1996). This implies investment banks are expected to be more often associated with value destroying merger and acquisition activity, since in essence they provide more benefits management subject to hubris or agency problems. In other words, ill-motivated management misuse the lowering transaction and contracting costs function of investment banks to complete deals in their own interest instead of shareholders’ interest. Thus looking at the role of investment banks in mergers and acquisitions, not controlling for the motives for mergers and acquisitions can result in negative associations with investment banks. In addition, to come back at the agency problem between investment banks and client firms, since investment banks have a more important role in ill-motivated mergers and acquisitions, there is more room for investment banks to act upon their own interest (McLaughlin, 1990). Thereby implying, the potential agency problem between investment banks and client firms is higher in hubris of agency motivated mergers and acquisitions, leading to lower shareholder wealth creation in these types of transactions. For that reason, this thesis looks at the effect of the choice to engage the services of an investment bank if hubris or agency motivated mergers and acquisitions are excluded from the analyses. In order to see if investment banks are in essence valuable if management have the right motives for mergers and acquisitions.

2.5 Empirical evidence

To position the theories and discussions about the role of investment banks in mergers and acquisitions in the existing literature, the related empirical evidence on the subject is discussed below. The existing literature looks at the client firm’s performance around merger and acquisition announcement, because changes in a client firm’s abnormal stock returns, should provide information about market’s perception on the shareholder wealth creation in a takeover deal, including how the choice for investment bank employment is received (Servaes & Zenner, 1996).

Empirical evidence is widely diverse regarding the effect of hiring an investment bank on the abnormal returns of the client firm (Wang & Whyte, 2010). Among others, Fuller et al. (2002) illustrate empirical evidence on the abnormal returns of both target and bidder firms around merger and acquisition announcement varies tremendously, because differences in firm and deal characteristics explain a great deal, if not everything, of the abnormal returns

(18)

captured around announcement dates. Servaes and Zenner (1996) suggest that, even controlling for firm and deal characteristics and the determinants of the choice to employ an investment bank, the role of investment banks does not significantly affect abnormal returns of target or bidder firms. Neither does the investment bank’s reputation. Rau (2000) shows investment bank quality has a negative effect on the client firm’s abnormal returns in mergers and acquisitions. McLaughlin (1992) suggest in line with Fuller et al. (2002) this negative effect of good reputation banks on the client firm’s abnormal returns is rather caused by a higher degree of complexity of deals in which, in general, better reputation banks are involved. In contrast to Rau (2000), Kale et al. (2003) provide evidence that, in case both the bidder and target hire a more reputable investment bank, the combined shareholder value created increases in a successful deal and if the bidder includes a relative more reputable investment bank, the wealth gains accrue to the bidder firm. In addition, Bowers and Miller (1990) state when either the bidder or the target firm hires a more reputable investment bank, combined acquirer and target abnormal returns are higher than when neither employs a high quality investment bank. Bao and Edmans (2011) recognize finding the right measure for advisors’ ability or quality is challenging and might be a reason for insignificant results about the role of investment banks in mergers and acquisitions. They therefore use a different approach and apply a fixed effect methodology in their study to 143 banks that advised on at least 10 mergers and acquisitions over the period from 1980-2007. This means, instead of looking at the effect of the choice to assign an investment bank in individual deals, they analyse the average abnormal returns to all deals per investment bank and the average residual of abnormal returns by controlling for fixed time effects, acquirer characteristics and acquirer fixed effects. Thereby examining whether advisors exhibit differences in client firm value in the first place, without having to specify an appropriate proxy for investment bank ability or quality. Bao and Edmans (2011) find investment bank’s fixed effects have a significant positive effect on announcement returns in mergers and acquisitions. As mentioned before, Wang and Whyte (2010) havemade an attempt to find evidence on the role of management in mergers and acquisitions and how management affects the choice to use an investment bank and the wealth created in the mergers and acquisitions. Their research of 10767 mergers and acquisitions in the United States from 1990 to 2002 is based on two analyses; they first examine the causal relation between a higher degree of managerial rights and the probability to use an investment bank using a Probit choice model. They subsequently analyse the causal relation between the use of an investment bank and the wealth effects around deal announcement, using an instrumental variable regression model. To test managerial rights the

(19)

Governance Index (GI) dataset of Gompers, Ishii, Metrick (2003) is used. The GI is a number ranging from 1 to 24 that reflects the amount of antitakeover provisions adopted by firms. Wang and Whyte (2010) argue that the higher is the GI number, the stronger the manager rights are in the given firm. The firms are then, based on their GI ranking, divided into five groups and the two analyses are performed on all five groups separately. Wang and Whyte (2010) find, when managerial rights are strong, the use of investment banks increases and abnormal returns are negatively affected. However, when controlling for the interaction between managerial rights and the use of investment banks, the use of investment banks shows no significant contribution to the changes in shareholder value around deal announcement. In addition, they find when firms include a more reputational bank the negative wealth effects associated with the use of investment banks are mitigated (Wang & Whyte, 2010).

2.6 Overall conclusion theoretical framework and contribution thesis

The role of investment banks in mergers and acquisitions is still under research in the existing literature and the relatively small amount of empirical evidence on the effect of investment bank choice on firm performance is characterised by a wide variation in results (Wang & Whyte, 2010). Previous papers on the subject do provide several theories on motives for firms to engage the services of an investment bank that can be captured in two main functions: the lowering transaction and the lowering contracting costs functions of investment banks (Servaes & Zenner, 1996). Despite these apparent shareholder value enhancing functions, the contribution of investment banks has also been associated with lower client firm value, which has resulted in ambiguities about the value investment banks bring to the merger and acquisition process (Servaes & Zenner, 1996). McLaughlin (1990, 1992) suggests investment bank compensation contracts are a reason to believe investment banks’ involvement in mergers and acquisitions could lead to lower firm value. In a predominant part of the deals compensation fees are contingent on deal outcome. The concern with compensation fees based on deal completion is that they, instead of preventing, possibly induce a potential agency problem between the investment bank and the client firm (Servaes & Zenner, 1996). This thesis contributes to the existing literature by questioning the potential agency problem between an investment bank and client firm that could arise when hiring an investment bank. Wang and Whyte (2010) find there is an interaction between the role of investment banks, managerial rights and lower firm value in mergers and acquisitions. But when controlling for the interaction between managerial rights and the use of investment banks, the role of

(20)

investment banks in mergers and acquisitions does not have a significant effect on shareholder wealth around deal announcement. This thesis tries to follow up on these findings, by looking at the role of investment banks in mergers and acquisitions in case all firms that are subject to hubris or agency problems are excluded from the sample. Because excluding hubris or agency motivated mergers and acquisitions from the sample enables to separate the possible negative effects on firms’ shareholder wealth creation caused by potential agency problems between the firm and the investment bank from the negative effects caused by ill-motivated management that (mis)use the role of investment banks in mergers and acquisitions. The answer to the question regarding the role of investment banks is of great significance for future employment of investment banks in mergers and acquisitions and in corporate events in general.

3. Methodology and hypotheses

3.1 Objectives and hypotheses

The aim of this thesis is to investigate if the choice to engage the services of an investment bank in the merger and acquisition process affects the client firm’s wealth created, if client firms are not subject to agency or hubris problems. As previously discussed, to incentivise the investment bank to act upon a firm’s objectives the investment bank gets, in a predominant part of the deals, rewarded based on deal completion (Servaes & Zenner, 1996). However, instead of aligning the interest of the investment bank with that of the client firm, this form of compensation could also enlarge a potential agency problem between the investment bank and the client firm (Servaes & Zenner, 1996). In addition, inferior value creation or even value-destroying in mergers and acquisitions could also be caused by agency or hubris motivated management, that have a better incentive to hire an investment bank to get support for their value-destroying activity. This would also leave more room for investment banks to act upon their own interest. To test whether there is any reason to believe the questions about the role of investment banks are even justified it is fundamental to this research to test whether the choice to hire an investment bank is leading to differences in firm value at all when both synergy and hubris or agency motivated mergers and acquisitions are included. As mentioned before, the focus of this research is on bidder firms, assuming acquiring firms rely more heavenly on investment banks’ services throughout the merger and acquisition process and the use of investment banks is expected to have a more substantial effect on acquirer shareholders’ value creation (Servaes & Zenner, 1996). This leads to the following hypothesis:

(21)

H1: The choice to hire an investment bank has on average a value destroying effect on acquirer shareholder wealth in all mergers and acquisitions

If lower firm value associated with the choice of investment banks is actually caused by hubris or agency motivated managers that have a greater incentive to hire an investment bank, this should automatically mean there would be no apparent negative effect of the choice of investment banks among synergy motivated mergers and acquisitions. In case mergers and acquisitions are synergy driven, expected total gains are positive and there is no reason to involve an intermediary if the potential agency problem between the firm and investment bank out weights the lowering transaction and contracting costs functions of investment banks. This leads to the second hypothesis:

H2: The choice to hire an investment bank has a value enhancing effect on acquirer shareholder wealth in the subsample of synergy motivated mergers and acquisitions

If in the synergy motivated subsample of transactions, the choice to engage the services of investment banks is nonetheless causing lower firm performance, this could indicate the concerns about potential investment bank agency problems are justified. Since the type of compensation fee used are believed to cause the potential agency problems with investment banks (McLaughlin, 1990,1992), lower firm value would indicate client firms fail to select the right investment bank compensation contract. However, this thesis predicts firms should be able to select a contract form that provides the right incentive for investment banks to act upon a client firms’ objectives (McLaughlin, 1990). Which means firms select compensation fees contingent on deal outcome if they believe they create the right incentive for investment banks to act upon firm’s objectives. And firms fixed-fee contracts otherwise. This leads to the third hypothesis:

H3: The choice between investment banks’ fee compensations based on deal completion and fixed-fee investment bank’s compensations has no effect on bidder firm value in the subsample of synergy motivated mergers and acquisitions

In case, compensation fees based on deal completion have a negative effect on acquirer shareholder wealth creation in the subsample of synergy motivated mergers and acquisitions

(22)

and fixed compensation fees do not, this could mean client firms do fail to select the right contracts. In other words, the so often used compensation contracts based on deal completion are then assumed to rather induce an agency problem between investment banks and client firms instead of preventing investment banks to act upon their own interest.

3.2 Methodology

To understand how the three hypotheses are tested, the model and its various aspects are described below.

3.2.1 Methodology

To test the causal effect between acquirer shareholder wealth creation and the choice to hire an investment bank, this thesis research employs a regression analysis, based on the traditional short-term event study approach of Andrade et al. (2001) (discussed in more detail below). With the acquirer shareholder wealth creation measured by announcement period cumulative abnormal returns (CARs) as the response variable and the choice to use an investment bank as explanatory variable. To remain consistent with the existing literature, the model does not make a distinction between specific functions performed by the investment banks (Wang & Whyte, 2010). Recognizing a potential endogeneity issue between a firm’s abnormal returns and the employment of an investment bank, this research applies the instrumental variable (IV) approach instead of performing a standard ordinary least squares (OLS) regression (discussed in more detail below). The regression model is as follows:

𝐶𝐴𝑅_! = 𝛽_!+ 𝛽_!𝐼𝑉𝐼𝑛𝑣𝑒𝑠𝑡𝑚𝑒𝑛𝑡 𝐵𝑎𝑛𝑘_! + 𝛽_!𝐶𝑜𝑛𝑡𝑟𝑜𝑙_!+ 𝜀_! (1)

With ‘Control’ reflecting the control variables included in the regression (discussed in more detail below). To test the causal effect between acquirer wealth creation and the investment bank’s compensation fee used, this study employs a similar research analysis as before, but then with the type of compensation fee as response variable. Because this thesis assumes the compensation fee variable is endogenous in the regression, in this case an OLS regression suffices. Leading to the following regression model:

(23)

Using the above described methodology; this thesis tries to investigate the effect of the choice to hire an investment bank without looking at investment bank’s fixed effects (as with Bao and Edmans, 2011), but treating mergers and acquisitions as individual events. Because this enables to include a higher number of mergers and acquisitions in the sample, since it also includes deals in which acquiring firms did not hire an investment bank or hired infrequent advisors.

3.2.2 Short-term event studies and abnormal returns

The event study approach analyses cumulative abnormal returns (CARs) surrounding the announcement of a corporate event, using different event windows. Following Brown and Warner’s (1985) event study methodology, abnormal returns for this thesis’ model can be retrieved by using modified market model (Fuller et al., 2002):

𝐴𝑅_! = 𝑟_! − 𝑟_!

With ri is return to firm i and rm is the value-weighted market index return from CRSP (Fuller

et al., 2002) (to test the robustness of the market adjusted model, the robustness analyses at the end of this thesis also contains a standard market model event study to generate the CARs surrounding the deal announcements). According to Andrade et al. (2001) most reliable evidence on firm value is attainted by looking at the average short-term abnormal returns around the announcement of the corporate event. This is based on the assumption that in perfect efficient capital markets, information about firms gets quickly incorporated in the respective firms’ stock prices, implying market’s perception on expected value changes is directly reflected in the firm’s stock prices. Therefore, a frequently applied event window is the (-1,+1) window, that is from one day before to one day after the announcement (Andrade et al., 2001). In addition, to ensure all wealth effects associated with the takeover deal are captured in the short-term even study, this research includes a (-10,+10) window event study, that is from ten days before to ten days after the announcement.

Several long-term event studies cast doubt on whether short-time stock prices in the proximity of the announcement impound all the information publicly available and believe that investors fail to assess the full wealth effects generated by mergers and acquisitions in a short time period (Andrade et al., 2001). However, long-term event studies come at the cost of introducing much greater noise and errors (Bhagat, Dong, Hirshleifer & Noah, 2005), hence this research is based on the short-term event study following Andrade et al. (2001).

(24)

3.2.3 IV regression

It is very likely several factors that affect the choice to hire an investment bank also affect market’s reaction around announcement, which leads to the explanatory variable, the use of an investment bank, being correlated with the error term (Wang & Whyte, 2010). This can be illustrated using the previous discussed interaction between management, the choice to hire an investment bank and firm value as an example. If the choice to hire an investment bank is (partially) driven by ill-motivated management, such management is likely to have an influence on stock prices as well (Wang & Whyte, 2010). And even though the group of control variables included in the model is meant to avoid omitted variable bias, they may not capture all aspects of the endogeneity issue. Therefore this research applies the IV approach. A valid instrumental variable must satisfy two conditions: the variable must be correlated with the endogenous variable (the choice to use an investment bank) and the variable itself should be uncorrelated with the market reaction in the second stage of the analysis (Wang & Whyte, 2010). Following Wang and Whyte (2010), this thesis identifies two possible instrumental variables: the percentage of the choice to use investment banks in previous deals (if a high percentage of firms has used an investment bank in previous deals, the respective firm is also more likely to use an investment bank) and if the respective firm has used an investment bank before (the use of an investment bank in a previous deal is expected to increase the probability to use one in the following deal). However, both these instrumental variables have shown to be irrelevant explaining the choice to engage the services of an investment bank in the sample of this thesis. Using the above described criteria, an alternative instrumental variable is identified: the compensation fee variable. The binary compensation fee variable, equal to one if the investment bank’s compensation fee is fixed and zero if the investment bank’s compensation fee is not fixed, satisfies both the conditions for a valid instrument. Naturally, the variable is correlated with the choice to use an investment bank, since the compensation fee is based on the appointment of an investment bank. Whether, the compensation fee variable is exogenous is less straight forward. However, this thesis assumes the compensation fee is selected rationally and based on creating the right incentive for the investment bank, the investment bank’s ability and the preferences of the investment bank itself, rather than on market perception on the deal or other factors that could be omitted from the regression (McLaughlin, 1992). The validity of the instrument is supported by the the exogeneity test of the statistical program used to perform the analyses. However, because missing values are automatically removed from the study, using the compensation fee variable as the instrumental variable would lead to a sample in which only

(25)

mergers and acquisitions that involve an investment bank are included (only for those observations the compensation fee provides a value, one or zero). To avoid excluding observations on mergers and acquisitions that do not involve an investment bank, the compensation fee dummy variable is created, equal to one if the compensation fee variable displays to be missing and zero if not. Thereby assuming if the compensation fee variable is missing, firms chose not to hire an investment bank. This compensation fee dummy variable has proved to be both relevant and exogenous in this thesis’ model.

3.2.4 Compensation fee

As discussed before, there are three standard categories in investment bank contract forms: fixed-fee, share-based fee and value-based fee contracts, each encouraging bankers to provide services to client firms in a different manner. Fixed-fee contracts are not related to offer outcome, in contrast, both share-based and value-based fees are paid based on completion of the merger and acquisition deal (McLaughlin, 1992). To test whether firms induce an agency problem between the investment bank and client firm by selecting a contingent compensation contract instead of avoiding one (thereby affecting shareholder wealth creation negatively), the compensation fee variable is included as the response variable in regression two. The compensation fee variable makes a distinction between fixed and variable based fee contracts and equals one if the investment bank’s compensation fee is fixed and zero if the investment bank’s compensation is not.

3.2.5 Control variables

Empirical evidence on abnormal returns around merger and acquisition announcement varies tremendously, because differences in firm and deal characteristics explain a great deal, if not everything, of the abnormal returns captured around announcement dates, besides the value effect of the corporate event itself (Fuller et al., 2002). In addition, these firm and deal characteristics also explain a great deal of the motivation to hire an investment bank. Therefore, to be able to isolate the effect of the choice to use an investment bank on acquirer abnormal returns, it is essential to include an extensive set of control variables in this research.

One of the factors driving abnormal returns that needs to be captured in the set of control variables is information asymmetry. Especially for bidding shareholders abnormal returns can be negative if there is uncertainty about the potential value created by a merger or acquisition and these negative returns increase with the dispersion of beliefs regarding the

(26)

beneficial synergies of the deal (Hackbarth & Morellec, 2008). In line with the effect of information asymmetry, is the effect of diversifying offers on abnormal returns. Diversifying

offers provide negative combined bidder-targets abnormal returns, suggesting that investors

seem to believe cross-industry transactions are an indication of the absence of good investment opportunities in a bidder’s own industry or that managers do not act in the interest of shareholders. Reflecting more about the stand-alone value of the bidder firm then market’s perception of the benefits of the deal (Fuller et al., 2002). Based on the research of Servaes and Zenner (1996), two proxies are used to capture information asymmetry and diversification: industry relatedness and the number of industries in which the target firm operates. Mergers and acquisitions are classified industry related when the first two SIC-digits are the same for the primary SIC codes of the acquirer and target (Andrade et al., 2001). The number of different SIC codes associated with the target reflects the number of industries in which the target operates (Servaes & Zenner, 1996). Competition among similar firms also decreases the abnormal returns to bidding shareholders. If there is competition for the acquisition of the target, the bargaining power of bidders decreases, which gives the target an advantageous position and the possibility to demand a higher premium, leading to lower acquirer abnormal returns (Morellec & Zhdanov, 2005). According to Andrade et al. (2001) the number of bidders per deal can be used as proxy for competition. The nature of the deal is assumed to be another driver of the abnormal returns around the announcement date. Hostile bids are interpreted as an indication that the bidder has by itself strong cash flow prospects and that managers act in the interest of the bidder shareholders or of both the shareholders of the combined firm. This is shown by higher bidder and bidder-target combined abnormal returns (Fuller et al., 2002). Therefore, a binary variable is included that equals one if the bid is hostile and zero if the bid is friendly (Servaes & Zenner, 1996). When making a distinction between public versus private firms, bidders earn positive abnormal returns in case they merge with or acquire a private firm and negative abnormal returns in case of a public target (Fuller et al., 2002). However, private acquisitions will be entirely removed from the sample, since there is little data and research on these types of deals and moreover, both bidder and target stock prices are needed for the study (explained in more detail in the data section) (Fuller et al., 2002). Regarding the size of firms, mergers and acquisitions consisting of a smaller target and a larger bidder show lower fractions of combined abnormal returns relative to the combined value than similar-sized firms (Fuller et al., 2002 and McLaughlin, 1992). To measure the effect of the size of both the target and the acquirer firm on bidder firm’s abnormal returns, two variables are included in the regression. The logarithm of the total

The effect of the choice to hire an investment bank on the acquirer wealth creation in synergy motivated mergers and acquisitions

The effect of the choice to hire an investment bank on the acquirer

wealth creation in synergy motivated mergers and acquisitions

Table of Contents

1. Introduction ...4

2. Theoretical framework ...7

3. Methodology and hypotheses ...20

4. Data ...28

5. Empirical results ...30

6. Robustness checks ...44

7. Conclusions ...49

References ...51

Appendices ...53