A behavioral perspective on investors' reactions to acquisitions announcements : the role of bidding companies' corporate reputation in the assessment of value creation possibilities

(1)

MASTER THESIS:

A Behavioral Perspective On Investors’ Reactions to Acquisitions

Announcements: The Role of Bidding Companies’ Corporate Reputation

In The Assessment of Value Creation Possibilities.

MSc in Business Administration: Strategy Track

Student: Giancarlo Glisenti

Student Number: 11374071

Supervisor: Hesam Fasaei

(2)

ABSTRACT

Drawing on studies on the process perspective of value creation in mergers and acquisitions, on corporate reputation and behavioral studies of management and economics, the aim of this study was to investigate the effects of bidding companies’ reputation on market investors’ assessment of the deal, given the information asymmetry they face towards decision of acquiring of companies’ managements. The assumed effect of reputation was studied both as direct on abnormal returns and indirect on other factors that other behavioral theories highlighted as affecting the perceived uncertainty on the outcomes of the deal: bid premium, payment method adopted and bidding companies’ acquisition experience. The method adopted was a linear regression that did not yield significant results. However, building on previous theories, the study provides a theoretical background to assume that this relationship exists and that adopting different methodologies future research might be able to prove it.

Statement of originality

 

This document is written by Student Giancarlo Glisenti who declares to take full

responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and

that no sources other than those mentioned in the text and its references have

been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision

of completion of the work, not for the contents.

(3)

INTRODUCTION

Mergers and acquisitions (M&As) are important vehicles for firms’ business, product and geographic strategies and have become a major way of enhancing companies’ growth (Hitt et al., 2001). They also have significant effects on firms’ performance (Laamanen & Keil, 2008) and generate long-term consequences for the firm. The most common theoretical rationale behind the decision to undertake M&A activity lies in the interest of acquiring companies of maximizing their shareholder’s wealth (King et al., 2004).

The effects of M&A are usually so profound that they can alter considerably share prices of both targets and acquiring companies. These are affected in different ways and different levels depending on the perception of stockholders of the value that will be created by the deal (Franks et al. 1991). Given that the company will be increasing its market share, shareholders should be optimistic of the returns that will be generated (Gersdorff & Bacon, 2009).

Despite this assumption and despite event-study has become very popular in finance studies since its inception in the 1960s, there is very little research throughout literature that analyzes how investors make decisions in relation to events such as M&A deals. Following the points theorized by the Efficient Market Hypothesis (EMH) (Fama, 1970) that conceptualized the theory of share prices reflecting all available information at any time, traditional financial perspective has used investor reactions as a measure of performance and efficiency benefits of events that have generated them (Zajach & Westphal, 2004). This view has eventually lead to a general overlooking of the behavioral process underlying these reactions.

In opposition to this, Baskin (1988) theorized that investor reactions are a “complex and human subject) far from the idealized vision of them as measurement of deals’ performance promoted by previous studies constructed on the Efficient Market Hypothesis. In line with such belief, scholars from behavioral finance has got to the conclusion that to understand such financial

(5)

phenomena the assumption of rationality leading investor decision has to be relaxed (Schijven & Hitt, 2012). Authors who treated this particular field believe so because while making their investment decisions investors only have part of the information regarding the deal that can be found in annual reports and other media supports that show involved companies’ financials. In fact, even though they might be aware of acquiring and target companies’ capital structure, performance, cash flows and other accounting-related information that may help them in the valuation of the possible synergies, they will not be able to fully understand how effectively management will be able to go through the acquisition process from negotiations to post-merger integration through which the process perspective suggests that synergies are practically generated (Jemison & Sitkin, 1986).

Hitt et al. (2001) also highlighted the relevant asymmetry of information between acquiring companies and investors on that tacit knowledge that can only be acquired by undertaking a deal first-hand considering that it is obtained through sub activities as due diligence and integration. Other studies have promoted the concept that reactions to deal announcement cannot be considered as reliable evaluators of these phenomena since they are not completely understood by external investors and that therefore they may be prone to heuristic biases (Delong & Deyoung, 2007; Oler et al., 2008).

Here I will suppose and try to test that, after a rational evaluation to evaluate how efficiently this process will be conducted by the acquiring company and to which extent synergistic possibilities will be exploited to create value, investors will rely on the acquiring company’s ability of delivering consistent outcomes to their whole stakeholder basis in the past. This kind of consistency should be reflected on the company’s reputation among those stakeholders (Tischer & Hildebrandt, 2014), including shareholders which are the subjects considered. Such assumption is in line with the signaling theory of corporate reputation that suggests that the latter is indeed developed through actions and communications and that rational stakeholders

(6)

will expect the rational company to honor the implicit commitment transmitted by a signal (Caruhana et al. 2006).

It is important to state this assumption because through literature there have been not a general agreement on what reputation entails (as it is explained further on in the “Defining reputation” section), but despite this it has become a relevant concern for companies given the perks it has shown to provide to those who develop a solid one. Given that it is said to add value and increase cash flows (Sheldon, 2011), enhance the growth of sales (Shapiro, 1982), enhance the credibility of advertising (Goldberg & Hartwick, 1990), augment product quality perception and customer loyalty (Weigelt & Camerer, 1998), and attract high-quality job applicants (Fombrun & Shanley, 1990), I assume that investors will also believe that the acquiring company will be able to exploit these competitive advantages on the newly acquired business, reducing their perceived risk and uncertainty regarding the outcomes of the deal and making them more confident that value will be created through synergies

As stated before, this study builds on previous studies of behavioral finance to investigate this specific market phenomena that still lack a relevant degree of explanations.

LITERATURE REVIEW AND HYPOTHESIS

DEVELOPMENT

ACQUISITIONS

There are many reasons for why companies may undertake M&A activity. Of all possible motivations, the most cited one in empirical papers is the identification of synergies that could be created by these corporate restructuring activities (Ferreira et al., 2014). These occur when the profits generated by the combined entity are higher than the sum of the profits that would have been generated by the two firms independently (Brock, 2015). Since there are many

(7)

different types of synergies that companies’ managements may seek to create by undertaking M&A activities, they have to go through a long and thorough research to identify a target that represents a close match with each of the possible objectives that they are seeking to accomplish by means of these corporate restructuring operations (De Pamphilis, 2015). In fact, acquirers may seek to create value by obtaining greater operational efficiency driven by economies of scale (i.e. reductions in purchased materials prices due to an increase in bulk purchases), economies scope (i.e. having a single accounting and HR department supporting multiple product lines), by the combination of complementary technical skills with the target company or simply by increasing their market power being able to boost selling prices above competitive levels (Ferreira et. al., 2014; De Pamphilis, 2015).

Other than operational-level, synergies can also be related to financial matters, in fact Mirvis and Marks (2002) argued that companies involved in M&A activities can notice overhead costs reduction and tax benefits (i.e. tax savings generated by acquiring a company with unused net operating losses and tax credits, and the additional depreciation resulting from the write up of the net acquired assets to the fair market value). Moreover motivations for M&A can be find in the willingness of the acquiring company to realign strategically to changes in its external environment that can be related to their regulatory systems or to disruptive technological innovations that threaten their market share (De Pamphilis, 2015).

Finally, a reason which is less attractive for shareholders can be overconfidence/hubris of acquiring firms’ CEO and management. This is when CEO’s or managers over-estimate their ability of generating returns from the management of the combined entity and may drive them to undertake risky deals that could destroy value (Malmendier & Tate, 2005).

By having a look at all these possible motivations that may drive managements’ decision, it is clear that the first challenge they have to face is the identification of a target matching their needs in the global market and how tough this challenge is. In the following chapters I will

(8)

describe the whole process of acquisitions using evidence from previous literature. In fact, according to scholars who outlined the process perspective on acquisitions, value is created throughout all the steps involved in such processes by the ability of handling strategic decisions of the companies’ managements (Jemison & Sitkin, 1986; Mirvis and Marks, 1992).

MANAGERIAL ISSUES IN ACQUISITIONS PROCESSES

Acquisition planning and target identification

The process perspective on acquisitions was conceptualized in the mid-80s by Jemison and Sitkin (1986). It proposes that acquisitions should be seen as a series of decision processes that have cascading influence throughout all the steps of the acquisition. Scholars supporting this perspective, highlighted how acquisition processes are influenced by individual, organizational and situational characteristics and how the acquisition outcomes strongly depend on those, particularly when to obtain the sought advantages a high degree of communication and collaboration is needed between the two firms. In fact better outcomes are subordinated to the choice of a better target, negotiating a better financial deal and effective sharing of complementary strategic knowledge and skills. Because of all the decision processes they have to go through and because of the unpredictability of decision outcomes, managers have to face a relevant degree of risk that may influence perceptions on decision-related information, evaluation of alternatives and choices made (Pablo et al., 1996). Hence acquisitions strategies are distinguished by a high degree of uncertainty, testified by the high failure rate of these operations (Ravenscraft and Scherer, 1989). This uncertainty stems from process-related challenges as the non-routineness, speed of decision making, restricted use of information, participation and debate that CEO’s and management must face throughout each step of the deal.

(9)

Mirvis and Marks (1992) provided an overview of the main decision-making processes in acquisitions and the managerial skills required to overcome the problems that may occur. Firstly, following previous literature on the topic, they traced success in M&A activity to fit between companies. Further on, they divided fit in three subsections that are financial fit, strategic fit and fit between human organizations. Financial fit is verified by firm’s M&A teams (sometimes with external help) using valuation methods such as DCF or multiple analysis. The more precise the valuations will be the more acquiring companies will be able to avoid massive and value destroying cost cutting lead by undervaluation of cash flows and earnings or unmanageable debt created by an inflatable purchase price. Strategic fit is of course of utmost concern too. It starts with a thorough self-scrutiny with a detailed SWOT (Strengths, Weaknesses, Opportunities, Threats) analysis that serves to identify the sought M&A objectives. These objectives must then be translated into specific strategic and investment criteria with an extensive involvement of managers of different levels. In fact, even though the first screening phase for attractive targets is usually assigned to staff executives or head of corporate development, the final steps of the selection are assigned to line managers who need to test for potential production, marketing or manufacturing synergies. Finally, fit between human organizations regards finding an agreement between the two parts on whose systems will be used and on how to manage departments to exploit the potential synergies.

Through a study on acquisitions of eleven different multinational companies Haspeslagh and Jemison (1991) found that most companies develop systematic processes for selecting and handling acquisition candidates but those are hardly followed because the human factors make each deal unique and require ad hoc managerial behaviors. Mirvis and Marks (2001) went deeper on this topic by analyzing each step that acquiring companies’ top managers go through and how within each step the human side makes it impossible to fully standardize the selection

(10)

and handle process. As stated previously the first step involves deciding an acquisition criteria based on the requirements the target will need to match.

Fogg (1976) identified the main criteria that can be adopted in this phase. The most common are related to financial performance and in particular they set minimum ROI, sales and profit growth of the target and on the maximum cost that the deal could have. Then he highlighted that there can be product considerations regarding, for example, how long product line should be and what is the quality demanded by the market and other regarding distribution system such as what is minimum to access the market or provide basis for growth and especially how to make it fit with the acquiring existing distribution channels. The list goes on with target’s customer base considerations, size, measured with total sales or total assets, and competences or skills to be found among target’s employees. Finally, Fogg argued that investment criteria may be also linked to which kind of manufacturing facilities the acquiring company is looking for (i.e. what excess capacity may be needed), to which unions affiliations they consider acceptable and to legal considerations (i.e. performing what it is considered a good acquisition that may induce anti-trust action). In this sense, with a survey proposed to partners from four different countries of a consultancy company that had adviced acquiring companies Very and Schweiger (2001), identified 55 unique problems that were associated with each stage of the acquisition. The main problems in this first phase where all related to the collection of reliable information on the target, on its market and even on its country legal situation. Table 1 shows the problems that were mentioned by the respondents in relation to deal planning.

Table 1-Issues in deal planning and candidates selection (Very and Schweiger, 2001) Identifying and selecting

acquisition candidates: Finding candidates with a strong strategic fit Difficulties in identifying candidates in new countries Limited number of potential

(11)

The second phase according to Mirvis and Marks (2001) consists in thorough analysis and screening of the possible candidates that match the established criteria. They argue that in this phase managers need to have developed skills to get input from people with different backgrounds and specialties on strengths and weaknesses of each possible candidate together with reconciling any possible conflict between these subjects and getting them working cooperatively. Other than checking for strategic and cultural match with the criteria previously selected, this also includes understanding the reasons why sellers want to sell (i.e. Are they driven by a business opportunity or by personal motives, does senior board want to stay on board after the sale? Do the buyers want the seller to stay and how will be the cooperation managed) and also if management teams of lower levels share mindsets and will be able to cooperate effectively. Table 2 from Very and Schweiger (2001) study summarize management issues involved in this phase.

Table 2- Issues in deal structuring and valuations (Very and Schweiger, 2001)

Before the closing: Evaluating strategic fit Transparency on M&A facts

Reliability of financials

Finding information on the reliability of the country

economicand political situation Poor quality of order backlog Tax laws

Over-estimation of synergies Country restrictions

Finding ethical problems Currency control regulations Suspecting environmental problems Keep the acquisition confidential Trustworthiness regarding target's

accounting principles

Competitive bidding for the target

Difficulties in pricing Legal and tax effective structuring of the deal Difficulties in valuation Valuing some assets

(12)

Making the deal: Negotiations process

In Walsh (1989) words the nature of a negotiation affects the future relationship between two parties. In the context of mergers and acquisitions problems in the negotiations may affect the willingness of cooperating of the two companies once they become one single entity. Previous literature has identified several success factors in this phase of the acquisition. A first factor affecting negotiations that has been identified is attitude (Frensch, 2006). This relates both to the acquirer and the target and it indicates whether the acquisition is hostile or friendly. In a multiple case study on 24 acquisitions Hitt et al. (1998) found out that friendliness has a positive effect on acquisition success. In fact, all of the 12 successful acquisitions that he found from his sample were friendly acquisitions (i.e. some acquirers offered the target to maintain some independence while achieving synergies or some acquirers got into the deal as white knights to help the target not being acquired by another undesired bidder). Similarly even Haleblian and Finkelstein (1999) conducted a study on 449 acquisitions and found out that whether the deal is friendly or hostile has also effects on the subsequent stock market returns of the acquirer. Friendliness in a deal seems to affect variables in the later stages of the acquisition as ingroup-outgroup reactions or “we against them” attitudes that may hinder cooperation, therefore complicating the processes of generation of synergies (Frensch, 2006). Another factor affecting negotiations identified in previous literature is the method of payment. The established equity price of the deal can be paid in three different methods: cash, stocks, a combination of cash and stocks. Previous studies on this aspect of the deal have reached different results. Haleblian and Finkelstein (1999) and Hayward and Hambrick (1997) tested the independent effect of the payment method and did not find any relationship between the payment method and the stock market returns, while Sirower (1997) find out that cash payment had more chances of resulting in positive stock performance. Despite these controversial

(13)

studies on the topic, issuing stock suffers of an adverse selection process since the acquirer would only decide to use this method if he felt the stock was overvalued, moreover acquiring shareholders take on the entire risk that the expected synergy embedded in the premium paid will not materialize (Rappaport and Sirower, 1999). This would send a strong negative signal to the market, negatively affecting stock performance (Frensch, 2006).

Finally Frensch (2006) cited as another factor involved in the negotiation phase, the premium paid. As it is defined by Sirower (1998) and Hayward and Hambrick (1997) premium is the overpayment over market value or stand-alone value of an acquisition target’s share. In his work Sirower (1997) argued that the payment of a large premium in most cases is driven by an overvaluation of synergies. He tested the effect of the premium on 28 shareholder performance measurement and found out that it had a strong negative effect. Hayward and Hambrick (1997) and Roll (1986) instead tested and verified that

a high premium can reflect CEO’s hubris and overconfidence that may drive them into deals that will likely destroy value.

In Table 3 from Very and Schweiger (2001) are summarized the main problems of this stage that they identified from their survey.

Table 3- Issues in negotiations (Very and Schweiger, 2001) Negotiation with target and other

stakeholders: Price negotiation Ethical problems

Obtaining warranties and

representation Negotiation about employment Getting the customer list

included in the contracy Respect of anti-trust law Negotiating envieonmental

(14)

Managing the post-merger situation

A relevant part of literature has been concerned with post-acquisition process. The objective of these studies has been to understand the impact of managerial actions concerning this phase of the acquisition on value-creation (Birkinshaw et al., 2000). The assumption on which these studies are based is that all value creation actually takes place after the acquisition (Haspeslagh and Jemison, 1991). Birkinshaw et al. (2000) observed that the process of value creation in this phase involves both task integration and human integration. CEOs and managers should work to achieve both effectively because focusing on just one of the two may hinder the success of the acquisition. Putting all efforts on task integration may drive to creating operational synergies but also employee lack of motivation, while focusing on human integration may prevent the realization of the expected synergies. “Good managerial” efforts should therefore focus on reconcile strategies to achieve both types of integration. The authors stated that task integration achievement was affected by the choice of integration mechanisms (staff meetings, joint R&D meetings, cultural awareness seminar ecc.), the degree of task specialization during integration (i.e. the extent to which R&D units of both entities were directed to more specialized and clear areas of the responsibility following the acquisition), to the degree of operating autonomy assigned to target units once acquired and by the level of inter-unit communication that the management is able to enhance (i.e. frequency of contact between acquiring and acquired companies’ employees). Regarding human integration they argued that is affected by prior experience with acquisitions of the acquiring company, by the visibility and ability of exercising leadership even on the new entity of acquiring companies’ leaders, by the amount and effectiveness of communication in the new entity, by the retention and voluntary walking away of acquired employees, by how much employees’ tasks are changed by the new organization and finally by the differences between the two companies’ cultures.

(15)

Frensch (2006) did a similar analysis of the post-acquisition processes. He also identified several kinds of variables that may affect the success of the acquisition and on which managerial efforts should focus. One of this factors are the environmental variables and in particular labor unions. Along with Bryson (2003) study on human resource management in the integration processes, he argued that they play a significant role in employee attitude to the new situation. If the relationship with them is managed effectively they can help to get employees on board, while in the opposite case they may have a negative effect on employee’s commitment to the new working conditions.

Just like Birkinshaw et al. (2000), Frensch (2006) also assess the cultural difference factor as potential detriemental variable on the outcomes of an acquisition. Even though considerations on this topic should begin in the acquisition planning and target identification with cultural due diligence (i.e. visit and tours in the partnering firm, surveys etc.), sometimes fit between the two companies’ cultures may be partially overlooked in favor of strategic and financial fit (Mirvis and Marks, 1992). Therefore in the post-acquisition situation managers have to work on target’s current culture to prevent clashing situations in the new entity. Buono and Bowditch (1989) specified three levels of integration that managers could aim to achieve: cultural pluralism, cultural blending and cultural takeover. Going for cultural pluralism means essentially recognizing the differences between companies’ cultures and organizing in a way that they can coexist independently. Cultural blending refers to a situation in which the new company’s culture inherits traits from both the partnering companies. Lastly, cultural takeover is when the acquiring company forces the adoption its own culture erasing the other one. This shows Buono and Bowditch (1989) assumption that management can act to reconcile the differences in mindsets and beliefs of the merged company, for example by hiring new employees, by enhancing socialization by the old ones and the new ones and also by removing those that do not support changes that they want to enact. Again, Table 4 from Very and

(16)

Schweiger (2001) shows what problems were faced by their respondents in the post-acquisition situation.

Table 4- Issues In the post-acquisition situation (Very and Schweiger, 2001)

After the closing integration:

Dealing with poor quality of target

management Compensation of management Integration of

multinatonal targets Keeping management in place after the close Managing firms'

culturla differences Combining different organizational structures Maintaining market share Integrating management to international group view Controlling

expenditures Integration planning and execution

Dealing with unions

Difficulties in implementing

sophisticated procedures and techniques

In this work they also pointed out that retention of target’s management often becomes a success factor especially in the case of cross-border mergers since time expenditures of learning processes of new functions, markets and customers may become more costly than coping with the cultural divergences. Finally they summarized the solutions that may be adopted by acquiring companies’ management teams to overcome these issues in intercultural management and workshops to help employees to familiarize, frequent visits by acquiring operational and top managers to the target in order to communicate the strategic intent and explain the advantages of working for the new entity (i.e. better salaries and pension plans) and hiring local human resource with the precise task of supporting the cultural integration. Across all of this process stock market investors will face a lack of information on a relevant part of the decisions taken by the acquiring company (Myers & Majluf, 1984) increasing their risk of adverse selection. In this work, I will test if the consideration they have about the acquiring company, represented by the company’s reputation, will affect their perception that

(17)

during the process right decisions will be taken eventually making it successful and making their investment profitable.

DEFINING REPUTATION

“It takes 20 years to build a reputation and 5 minutes to ruin it.” (Warren Buffet, 1995:109). This quote gives an idea of how much good reputation is a difficult and volatile achievement to obtain. Since it is a “soft concept” throughout organizational studies it has been sometimes overlooked since most of them were preferably directed to assess more hard-edged, day-to-day corporational urgencies. Despite this, a relevant body of research from the late 80s has highlighted how good reputations have strategic value for firms that possess them (Dierckx & Cool 1989; Rumelt, 1997; Weigelt & Camerer, 1988). According to the resource-based view, possessing assets that are valuable and rare can bring superior returns to firms and if they are also inimitable, as reputation is, they can also become a source of sustained competitive advantage (Barney, 1991). Tischer and Hildebrant (2011) go even further, stating that intangible assets are actually the major drivers of sustained competitive advantage, since they are the less likely to be neutralize by competitors compared to tangible assets and are in general not tradable in factor markets.

For this reason during the first decade of 2000s things changed and a remarkable trend of studies have been thoroughly dedicated to organizational reputation, directing attention on developing a satisfying definition of it and, later on, on understanding and measuring the impact it has on corporations. In an attempt to inspire and guide management researchers by clarifying what organizational reputation is Lange et al. (2011) synthesized the outcomes of researches of the precedent decade and identified three general group definitions that have been given about it and those are: being known, being known for something or generalized favorability of the organization. Between all the papers regarding organizational reputation,

(18)

under the “Being known” category of definitions there are Bromley (2000) that states that it is the way key stakeholders or other interested parties actually conceptualize an organization, Scott and Walsham (2005) that adds that the expectations of key stakeholders must coincide with reputation claim, Spence (1974) interpreting reputation as the outcome of a competitive process in which firms signal key characteristics to constituents to maximize their social status and several others. For the “Being known for something” category, among others, Benjamin & Podolny (1999) can be mentioned, in fact it stated that a corporation’s reputation comes from factors as previous quality investments and also history and affiliations over time, also Mahon (2002) that highlights how reputation is anchored to individual and organizational based past actions and recent strings of behavior and Pfarrer, Pollock and Rindova (2010) that claim that from an economic perspective the term “reputation” refers to perceptions of a particular attribute which is the ability to deliver quality products to customers. Finally, reputation as the evaluation of a firm by its stakeholder in terms of affect, esteem and knowledge (Deephouse, 2000) and reputation as observers’ collective judgements of a corporation based on assessments of the financial, social, and environmental impacts attributed to the corporation over time (Barnett, Jernier, Lafferty; 2006) are two of several definitions grouped under the last cathegory: “Generalized favorability”.

Authors highlighted how reputation affects firms’ performance and shareholder value (Raithel, Schwaiger; 2015) through different factors as customer retention (Nguyen, Leblanc; 2001), customer loyalty (Bontis, Bart, Wakefield; 2007), investor loyalty, (Helm, 2007) investor reactions (Pfarrer, Pollock, Rindova; 2010) and brand performance (Lai et al., 2010). In case of good reputation a firm may benefit of a value increase in the financial market and, by charging premium prices to customers for products and services, can obtain an advantageous and more profitable competitive position in its industry. These different aspects are prime for a company’s success making all studies regarding reputation a remarkably interesting matter

(19)

for them. In addition to the effects that are induced by reputation, it is also fascinating and very useful for companies to try and understand from where it is generated and how it can be enhanced.

Some studies have also demonstrated how internal individuals’ behavior plays an important role in stakeholder’s consideration of the company, firstly when these come from CEOs or top management’s members but also from all other employees.

In her study on Nakra (2000) finally adds that corporate stakeholders, including customers are taking renewed interest in business goals and objectives, in corporate competitive strategy and advantage, and the company’s ability to survive and thrive in the high-change global market. Summarizing, this work is based on the concept of reputation as “a relative stable, aggregated and indirectly suggestible perception within multiple stakeholder groups based on a company’s past actions and future prospects in comparison to some reference” (Tischer & Hildebrandt; 2011; Walker, 2010).

The aim of this work is to analyze whether such collective perception may play a relevant role on stock returns of an acquiring company. To be more specific, the purpose will be understanding if investors see as more attractive deals involving better reputable bidding companies. In fact, since reputation is built along time through good management of assets and, through demonstration of successful managing skills (Nakra, 2000), it seems reasonable to think that investors will believe that the parent company’s management will be able to manage successfully issues arisen throughout the whole process of the deal as previously defined: planning and target identification, deal definition and post-acquisition situation. Reputation is therefore assumed to reduce the uncertainty and risk perceived by financial market actors due to limited information (Myers & Maluf, 1984) in relation to the value creation possibilities in acquisitions (Petty & Cacioppo, 1986).

(20)

A BEHAVIORAL PERSPECTIVE ON INVESTORS’ REACTIONS

Traditional financial economic reasoning has been developed on the efficient market hypothesis (EMH) getting to the conclusion that stock prices catch all the effects of information available to market investors (Fama, 1970). EMH is based on three main assumptions. Firstly, investors are assumed to value stocks rationally based on all information they have. The second assumption states that to the extent that they do not behave like this, irrational decisions are assumed to be random therefore neutralizing each other’s effect. Finally, the third assumption argues that, to the extent that investors act irrationally in a nonrandom way, the impact on the market of those who value stocks rationally will be nullified. Since it is impossible to test whether share prices reflect or not the actual value of a stock research should focus on understanding what underlies investors decisions. The underlying reasoning of this proposition comes from the conclusion of scholars from both behavioral finance, strategic management and economy sociology who found evidence that capital markets are not as efficient as previous theories have assumed. In this study we want to analyze the reactions linked to acquisitions’ announcements that have been the most widely researched topic of event studies (Zollo and Meier,2008) and through which investors can show their favor or disfavor toward the deal by driving the acquirer’s share price up or down once the information is disclosed (De Pamphilis, 2015). These reactions represent the general re-assessment of the restructured bidding company based on the new expectations of value creation of investors and not an actual measure of the value that will be created as the EMH in its strong form predicts (Grinblatt & Tittman, 2002). Therefore, similarly to Schijven and Hitt (2012) we will try to develop an argument that valuations of the value created by the deal will be based on management’s perceptions and that the overall valuation of the convenience of the deal made by the acquiring company will be trusted more the better the company is reputable.

(21)

Myers and Majluf (1984) stated that there is always an informational asymmetry related to investment decisions of a company between market investors and the latter, both on the focal target and on the acquirer itself. Studies related to decision-making theory (Cyert & March, 1963; March & Simon, 1958) argue that just like all other decision making under uncertainty, investors’ reactions on the stock market after acquisition announcements originate from a process of “problemistic search” where the problem is related to an informational disadvantage on the deal conditions and on managerial strategies of the acquiring company. This lack of information makes them boundedly rational (Kock, 2005) and since they need to react as quick as possible they have to satisfice (Simon, 1945) rather than maximize their research for information (Schijven & Hitt, 2012; Farrelly, 1980; Sapra and Zak, 2008). The theoretical reasoning behind this study is based on the assumption that this situation of bounded rationality will push investors to rely on a certain extent on their sentiment and trust on the bidding company rather than on a thorough estimation of the value that will be created (Rosen, 2006) and that this sentiment and trust will be driven by the bidders’ reputation (Mui et al., 2002). Such conceptualization builds on research from social psychology which suggests that in decision-making processes perceptions that will drive the final decision are built through two different route: central route and peripheral route (Petty & Cacioppo, 1986; Petty & Wegener, 1999). The first is a careful analysis of all information to get to a conclusion, while the latter is just relying on perceptions of those that have necessary information that is not publicly available, assuming those are right (Chaiken, 1987). Furthermore, social psychology has established that the degree on which one’s perceptions rely on an external source’s perceptions is affected by how expert and reliable this source is considered to be, both characteristics on which a company’s reputation is built. In the words of Petty et al. (2009): “when a source is perceived to be both expert and reliable, individuals can be reasonably confident of the accuracy of their attitudes by merely accepting the position advocated”. Linking these

(22)

behavioral assumption of investors relying on acquirer’s management perceptions with Caruhana et al. (2006) argumentation that reputation is a collective representation of a company’s past actions and results and outlines the ability and consistency of the firm in delivering value to its stakeholders, I argue that to assess the possibilities of value creation investors will rely more on management decision of investing in a new company the more such company has proven to be successful and to deliver positive outcomes in the past. Therefore the first hypothesis of this study will be:

H1: Market investors’ reactions on stocks of acquiring companies are positively affected by the level of reputation the latter have.

The first hypothesis developed is that investor reactions are based to a certain extent on sentiment regarding management perceptions of the deal. In fact despite being boundedly rational, there are some characteristics of deals that stand out immediately to this kind of stakeholders when they have access to public information and that affect the degree to which they trust that the deal will generate value and therefore positively affect share prices. Previous studies has shown that the variance of these variables affect the degree of risk perceived by investors on the possibility of gaining positive returns by purchasing bidding companies’ shares.

Firstly, previous literature has taught us that value created in acquisitions is equal to the difference between synergies created and the premium paid on the target share price (V=S-P)(Bruner, 2004; Sirower, 2007). Schijven and Hitt (2012) has highlighted how traditional financial perspective, based on the efficient market hypothesis, has implicitly pointed out that investors will evaluate the synergies (S) that will be created by the combination of the businesses and then that they will subtract the premium offered (P) to obtain an approximation

(23)

of the value (V) that will be created. Secondly, research from the past has provided evidence that managers and CEOs are susceptible to hubris and overconfidence and this could lead to overestimate their ability to manage the new assets acquired and therefore the synergies that they could create from the new entity (Hayward and Hambrick, 1997). Lastly, since most managers aim at managing large firms (Jensen, 1989) and usually get higher retributions for managing larger companies (Hambrick and Finkelstein, 1995), the acquisition strategies may be driven by ‘empire building’ motivations rather than value creation for shareholders. Either of these motivation could lead to overpayment of the selected target.

Given the conceptualization of a bidding company’s reputation as a measurement for investors of its ability of managing the acquisition process exploiting the possibilities of value creation, a higher level of reputation will eventually affect the degree of trust on the valuation made by the acquiring company on the targeted company and eventually on the premium offered for its shares.

H2: The negative effect of bid premium on market investors’ reactions to acquisition announcements is negatively moderated by the level of reputation of the bidding company.

Another signal that investors could get from publicly available information that could alter the degree of perceived uncertainty on the value creation possibilities of the deal, is the payment method adopted. The purchase in these corporate restructuring events can be done essentially in three methods (Travlos, 2017; Frensch 2006). Firstly the target could be bought using cash only, which will require the acquiring company to have a strong cash flow availability, secondly the acquisition could be done exchanging bidders’ stocks for target stocks and finally the method adopted could involve a part of the total amount paid with cash and another part paid using stock. Using stock as a method of payment has a signaling effect on external

(24)

stakeholders that the firm believes to have an overvalued stock (Frensch, 2006; Travlos; 1987). Additionally, it involves sharing risk with target firms’ investors signaling that bidding firm’s management is not fully confident of the synergistic potential of the deal (Hansen; 1987; Schijven & Hitt, 2012). Therefore the extent to which stock will be used to purchase the target will have a negative effect on trust of investors on getting positive returns from the purchase of bidding companies stocks. Again I argue that the degree of reputation these companies benefit of will mitigate the perceived uncertainty of investors driven by the signaling effect previously mentioned.

H3: The negative effect of the extent to which stock payment is used for the purchase of the target company and investors’ reactions to acquisition announcements is negatively moderated by the level of reputation of the bidding company.

Finally, another predictor identified by previous literature that have an effect on how investors will assess the investment is previous experience of the acquiring company with this restructuring activity. Scholars have been increasingly realizing that learning from prior experience may be crucial to perform better in acquisitions and other strategic activities (Barkema and Schijven, 2008). Even though execution of due diligence, negotiation, financing and integration, being all complex operations (Hitt et al., 2001), need to be customized to the specific deal considered (Haspeslagh and Jemison, 1991), companies can benefit of previous experience through learning. The specific term for the influence of prior event on performance is transfer effect and comes from psychology literature and positive transfer refers to improving in the performance of a previously performed task by learning from both positive and negative previous experience (Haspeslagh & Jemison, 1991). Previous experience can enhance

(25)

performance in particular by enriching acquiring company’s understanding of target selection, deal negotiation and post-acquisition integration (Schijven & Hitt, 2012).

Given this possible signaling effect of the degree of experience in acquisitions of the bidding companies and given the conceptualization of reputation as the perception of stakeholders on companies based on management’s ability of delivering consistent positive outcomes in the past on which this study is built, I argue that a high reputation will affect the degree on which investors will trust the bidding company to be able to efficiently exploit insight obtained from previous deals, learning from previous managerial mistakes and successes (Haleblain and Finkelstein, 1999). Further on, another possible explanation for why a high reputation might have a positive effect on how investors assess the degree of experience in acquisitions could be that this stakeholders’ perception is built through success in past actions and strategies that include the acquisition activity. If this kind of restructuring strategy had been proven to be highly unsuccessful or to destroy value it would be hard or even impossible for the company to benefit of a high reputation (Benjamin and Podolny, 1999)

Hypothesis 4: The positive relationship between the level of acquiring companies’ previous experience in acquisition sand investors’ reactions to acquisition announcements is positively moderated by the level of reputation of the bidding company..

The topic addressed has not been significantly considered in all previous research. In fact, lots of studies have been directed to the economic outcomes linked to reputation, and most of these studies have proved the enhancing effect of it. For example, Pfarrer, Pollock and Rindova (2010) discovered positive relations between firms’ high reputation and earning surprises and investor reactions.

(26)

Roberts and Dowling (2002) found a bilateral, positive and dynamic relationship with superior performance and several other studies have highlighted the favorable impact of reputation on customer retention (e.g. Bontis, Booker and Serenko, 2007; Nguyen and Leblanc, 2001; Andreassen and Lindestad, 1998) and positive word of mouth behaviour (e.g.,; Fombrun and van Riel, 1997). All of these results provide a solid basis to deduce that investors, when deciding whether to invest or not in a company that has just announced an acquisition, which would mean accepting a certain degree of uncertainty on their returns on the investment, will take in consideration acquirer’s reputation as a way to reasonably forecast how successful the acquisition will be and whether the acquiring company will likely be able to handle the acquisition process and create value. If the hypothesis will be tested, it will be proven the positive effect of reputation investors’ assessment of the possible valuable created by the deal.

METHODS

The established aim of this work was to understand whether the acquirer’s reputation has effects on post-announcement investor reactions. Since reputation is an intangible asset, the first issue that arose while undertaking this analysis was to find an available satisfactory measurement for it.

Throughout literature there have been different ways of measuring the effects of reputation (mostly quantitative) and Walker (2010) gives a useful review of the most relevant ones. He used a sample of 54 articles to give a review of corporate reputation including a section on reputation measurement.

Throughout such sample several different methods have been used for measuring reputation in order to run these tests. Deephouse and Carter (2005) used an asset quality ratio by a third-party rating agency and a content analysis of media data, Fang (2005) used market share as an

(27)

indicator of reputation, Rao (2004) winning contents, Cable and Graham (2000) and Turban and Greening (1997) adopted ranking by students while Rindova et al. (2005) ranking by recruiters.

The measurement chosen for this work will be the most commonly adopted according to Walker’s study (around 40% of all the quantitative studies considered): Fortune’s “America’s Most Admired Companies” (FMAC) list. This method appears to be the most substantial since it outranks the other methods overcoming two of the major common limitations that affect the effectiveness of the other methods that are hardly retrievable and provide a view of reputation from only one aspect of it In fact companies’s final FMAC ratio is calculated on 7 corporate attributes: Innovation, People management, Use of corporate assets, Social Responsibility, Quality of management, Financial soundness, Long-term investment value, Quality of products/services and Global competitiveness. This choice is also in line with Tischer an Hildebrandt (2011) that sustained that reputation effect could be measured using information contained in public rankings having found evidence on financial performance.

Additionally, of the 54 papers include in Walker’s (2010) sample, 24 were empirical papers of which just 2 used qualitative methods as measurements. Of the remaining 23, some methods included path analysis (Rindova et al., 2005), developing and testing economic formulas specific for the study (Chu and Chu, 1994; Fang, 2005) factor analysis (eg, Fryxell and Wang, 1994 ; Zyglidopoulos, 2005) raw percentile (Dranove and Shanley, 1995) chi-square tests (Fryxell and Wang, 1994) and z-scores (Davies et al., 2001). Although, the most used method was a regression technique which will also be adopted for this work.

Once the measurement was established, the next step was to identify a sample of deals involving bidding North American companies that appeared in the FMAC list and therefore had a reputation ratio calculated at the beginning of the year of in which the acquisition was announced. The most consistent time series of ratios that was publicly available of the internet

(28)

was between 2006 and 2010. In this time window I identified 648 companies that had been evaluated on their corporate reputation level, most of which appeared on a regular basis in the list given its relative stability (Tischer and Hildebrandt; 2011). The next step in the selection of acceptable deals for the study was to look in ZEPHYR’s database for all of those done by these companies in the 5 years considered in which bidding company’s annual reputational ratio was available. The last limitation in the selection process was represented by the fact that just a few deals had a disclosed figure on the bidding premium which was necessary for the study. This process lead to a final sample of analyzable 150 deals. Such number respected Green’s (1991) rule of thumb condition for acceptable size of the sample both for testing the overall model (minimum number of observations=50+8*(number of predictors)) and the individual predictors included in it (minimum number of observations=104+(number of predictors)).

The main hypothesis of my study is that due to an incomplete availability of information on acquisition deals, on the companies involved and on the synergies that are likely to be created, a part of investors’ trust on the acquirer’s management decision of acquiring and on its belief of being able to generate synergies will be affected by the bidding company’s reputation, and this will be reflected on their willingness of investing affecting stock prices. To verify this relationship, I set a regression between bidding company’s last available annual Fortune’s reputation ratio and the Cumulative Abnormal Returns (CAR) generated around the announcement date. The formula to calculate the latter was:

∑ 𝐶𝐴𝑅 =  (𝑅𝑖, 𝑡 – [𝐸(𝑅𝑖, 𝑡)]

2

𝑡=−2

where Ri,t are the actual daily returns per day on the bidder’s stock and E(Ri,t) are the expected

(29)

(CAPM) (Shah & Arora, 2012). I adopted a five days window on the announcement (-2; +2) to allow for some leakage of information prior to announcement and to allow for the news of the announcement to spread (Rosen, 2006; Schijven & Hitt, 2012). Additionally, such a short time window was adopted to avoid abnormal returns generated by other confounding events (McWilliams & Siegel, 1997).

Data on stocks was retrieved from the Center for Research in Security Prices (CRSP) section of Wharton Research Data Services’ (WRDS) database. The first step was to compute the beta of acquirer’s stock of each deal. To do so I adopted Yahoo finance’s method that involves a calculation of the volatility of the stock by comparing its monthly returns with the S&P 500 index, which was used as benchmark, for the 36 months preceding the announcement date of the deal. Practically, I obtained the final result by applying a SLOPE formula on Microsoft Excel between the monthly returns of the bidding companies’ stocks and of the S&P 500 index for the selected time period. As soon as I finished this process I could go on calculating the expected returns for the stock by multiplying the beta I had calculated by the returns on the S&P 500 index for the five days window over the announcement date. I then subtracted the daily expected return to the actual daily return on the stocks obtaining daily abnormal returns. Finally I summed these together to calculate the cumulative abnormal returns over the deal. This method is usually adopted in event studies because it allows to isolate to a relevant extent the unanticipated returns on the stock generated by a major event as an acquisition from the normal returns that were anticipated by a statistical or economic model (Reinganum, 1985). In my model I then added a moderating variable based on the payment method adopted. Three are the possible methods of payment (stocks, cash, combination of stock and cash), where stock payment is seen as a sign of perceived risk related to the deal by bidder’s company management (Schijven & Hitt, 2012). Following Hayward and Hambrick (1997) I will adopt a three point

(30)

ordinal scale where 1 will be assigned to cash payments, 2 to combinations of cash and stock and 3 for only stock payments.

Since previous literature has taught us that value created in an acquisition is the difference between synergies created and premium paid (V = S – P)(Bruner, 2004; Sirower, 1997), I argue that the higher the premium offered on target’s shares, the less acquirer’s reputation will be effective on stock returns since the chances that the acquisition will create value would be lower. Therefore even in this case, I ran a regression, this time using premium percentage as a moderator.

To test the moderation effect of acquisition experience on my model I needed a satisfactory quantification of the company’s previous experience. Following Bruton, Oviatt and White (1994) and Fowler and Shmidt (1989) I adopted the sum of the acquisition deals made by the company in the 4 years preceding the deal at hand. Such short period should include all the deals involving the acquiring company that investors would consider while assessing managerial experience in acquisitions of the company’s management. Again, all data on bidding companies’ previous acquisitions was retrieved by ZEPHYR database.

Finally, to isolate investors the effect of other predictors that may not be affected by reputation from the abnormal returns I inserted three control variables that previous studies have highlighted as relevant in the decision process of investors on acquisition deals. Firstly I wanted to control for the effects of relative size of the deal compared to the acquiring company. It is obvious that relatively smaller deals will affect in a different way a company’s future performance compared to bigger ones and this will therefore be reflected to some extent on the willingness to invest of market actors (Moeller et al., 2004; Rehm et al. 2012). The variable adopted is a ratio of the total deal value (retrieved from ZEPHYR database) and total assets of the bidding company (from YCHARTS.com and companies’ annual reports). Secondly, to test whether bidding companies’ previous financial performance affected investors evaluation of

(31)

the deal I used last publicly available ROA ratios for bidding companies, in line with existing work from Zollo and Singh (2004) and Barkema and Schijven (2008). Also Schijven and Hitt (2012) adopted this key financial statistic supporting the theory that a company’s strong performance could induce management to undertake risky deals only driven by the confidence they have obtained from previous successes. Lastly, I controlled for another key financial statistic that investors can retrieve from publicly available statements: the debt-to-equity ratio. In fact acquirer’s capital structure plays an important role in stimulating organizational efficiency (Jensen, 1986). A high acquirer’s debt compared to its equity would mean that it has little cash flows available for financing new projects (Schijven and Hitt, 2012). High debt also makes the company more exposed to bankruptcy risk and probably less keen on undertaking risky acquisition deals (Harford, 1999).

The selected ratio was calculated by dividing Total Liabilities with Total Stockholder’s Equity. The analysis was ran using a multiple hierarchical regression where first the effects of the control variables were tested, and secondly the effect of reputation and its hypothesized moderators. The final model included all the variables and the equation representing it was:

𝒚 = 𝒃𝒐 + 𝒃𝟏𝒙𝟏 + 𝒃𝟐𝒙𝟐 + 𝒃𝟑𝒙𝟑 + 𝒃𝟒𝒙𝟒 + 𝒃𝟓𝒙𝟓 + 𝒃𝟔𝒙𝟔 +

(32)

Table 5 – Description of the variables in the equation

Linear coefficient Variable (equation) Variable (statistical model)

/ y abnormal returns (ret)

b1 x1 reputation (rep)

b2 x2 premium offered (prm)

b3 x3 payment method (p_mtd)

b4 x4 acquisition experience (exp)

b5 x5 relative size (size)

b6 x6 performance (roa)

b7 x7 capital structure (d_e)

b8 (x1*x2) Interaction term between prm and rep

b9 (x1*x3) interaction term between p_mtd and rep

b10 (x1*x4) interaction term between exp and ret

The following models show a graphical representation of the moderating effect of reputation in the hypothesis. Model (1) conceptual model for each specific hypothesis; Model (2) statistical model for all the hypothesis.

Model (1) H2: H3: Reputation Premium offered Abnormal Returns Payment Method Abnormal Returns Reputation

(33)

H4:

Model (2)

ANALYSIS

Verifying Regression Assumptions

The first step of my analysis was to recode the only categorical variable present in the model: payment method. Using Microsoft Excel’s IF function I transformed this data in values referring to the three-pointed ordinal scale described in the methods section so that it was ready to be used in SPSS.

I then went on checking for the assumptions that characterize the regression model. First of all I checked for the normality distribution of the variables. To do so I ran a frequency test to check for skewness and kurtosis of the data distribution. Rep had a skewness statistic of -0.928 and a showed a leptokurtic distribution (kurtosis = 1.840>1) meaning that its shape was pointy with

Abnormal Returns (y) Independent Variable (X) Moderator (M) Interaction Effect (X*M) Abnormal Returns Acquisition Exp. Reputation

(34)

most of the values in the tails. Prm had a highly positive skewness equal to 5.472 meaning that most of the values were distributed toward the left end of the distribution and had a relevant leptokurtic distribution that was easily recognizable by the sharply pointed distribution of its histogram. P_mtd statistics for skewness and kurtosis were

acceptable values adopting the rule of thumb (acceptable skewness and kurtosis values between -1 and +1) in fact they were respectively -0.893 and -0.772. Exp had an extremely positive skewness of 2.113 and a platykurtic distribution meaning it is flatter than the normal. Among the control variables skewness was extremely positive. In fact roa had a skewness statistic of 6.012, d_e had one of 12.215 and size of 4.11. Kurtosis statistics were extremely high for these three variables, respectively 44.919, 9.472 and 20.712. Skewness for the dependent variable ret was almost insignificant with a statistic of 0.266 meaning that all the values were

symmetrically distributed around the mean. Despite this, it had a leptokurtic distribution with a statistic of 6.102.

To try and reduce this values of skewness I ran an outliers check. I checked for frequencies and using SPSS I identified one case for rep, three cases for exp, two cases for roa, two cases for prm, one case for d_e and four for ret that had a z score higher than 3 or lower than -3. To

eliminate those cases and try to approximate the distributions to normality I adopted a winsorizing (Winsor, 1941) technique which consists in changing the values of outliers with the nearest value having an acceptable z-score within the +/-3 range. This would have not relevantly affect the results because the outliers changed were less than 5% of the sample’s observations for each variable (Winsor, 1941; Dixon 1950).

After this I checked again frequencies and found out that skewness values had changed for all of the variables. In fact all the variables were much more distributed around their mean as it could be seen from their respective histograms that showed that observations were more centered to the mean. The new distribution for Roa and d_e showed the most relevant skewness

(35)

variation from the extremely positive skewness statistics they had before the outliers check. In fact these values for their new distributions were respectively 0.825 and 2.766. Such figures showed how roa had now a moderately positive skewness, in the range of the generally adopted rule of thumb for normality tests (between -1 and +1). Regarding d_e, the approximation to normal distribution after the winsorizing process was clearly not precise enough to obtain a skewness statistic within this range. The same process lead to a new skewness statistic for rep of -0.721 showing a still moderate level of skewness. Distribution of prm, exp and size also benefited from the winsorizing and their new skewness values were respectively 1.234, 1.598 and 2.309. These values still showed a substantial level of positive skewness that is not acceptable according to the adopted rule of thumb. To achieve distributions approximated to normality I therefore applied some transformations to those variables that still did not satisfy this assumption of linear regression. For size the best transformation between the power functions that can be used for this purpose was a square root transformation that lead to a skewness value for the new variable of 1.121, just outside the boundaries of the acceptable rule of thumb. I applied this function on the new observations obtained and the new distribution had a skewness statistic of 0.283 and therefore sufficiently approximated the normal distribution. I proceeded recoding the transformed variable as size1. Also exp was transformed using a square root function. The new variable was again recoded as exp1 and showed a moderate positive skewness with an acceptable statistic equal to 0.272. For the remaining variables prm and d_e the square root transformation was initially inapplicable since some observations had negative values and properties of the square root function would not allow it. For this reason the transformation was done adding a constant equal to the absolute value of the minimum negative observation of each variable to all of the observations. The constants added were equal to 22.44 for prm and 12.42 for d_e. Once transformed the variables showed skewness statistics respectively equal to 0.241 and -0.014. Therefore after these

(36)

transformations, all of the variables adhered to the normality assumption according to the adopted rule of thumb.

Table 6 summarizes all the transformations that were done to the skewed variables of the study. Table 6: Skewed variables transformation

Variable Transformation Recoded Variable

prm [prm+|min(prm)|] prm1

size size size 1

d_e [d_e+|min(d_e)|] d_e1

exp exp exp1

After dealing with the normality distribution of the variables I went on computing their standard means and standard deviations. The mean value for reputation (rep) was 6.5393 while the standard deviation was 0.92535. The mean for the bid premium (prm1) was 7.035539 with a standard deviation of 1.7814596. The mean of the variable regarding payment method (p_mtd) was 2,42 and its standard deviation was 0.780. The mean for the variable acquisition experience (exp1) was 2.126737 with a standard deviation of 1.4023908. The mean for the dependent variable of the study abnormal returns (ret) was -0.968 with a standard deviation of 5.89917. In line with these statistics by checking frequencies I found that around 59 percent of the bidding companies faced negative abnormal returns on their shares around the acquisitions’ announcement dates. The means for the control variables measuring acquiring firm’s performance (roa) and capital structure (d_e1) and the relative size of the deal (size1) were respectively 6.9168, 3.899701, and 1.659586. Finally the standard deviations for these variables were respectively 7.633262, 0.6101997 and 0.7042905.

(37)

In Table 7 means and standard deviations for each variable are summarized. In addition the table shows the correlation matrix that includes Pearson correlation levels between each one of them. The linear effect on one another was generally between small and medium (Cohen 1988, 1992). Looking at theses coefficients the assumption of no perfect or close to perfect multicollinearity was also respected. Even if with substantially low effects, correlations were significant between roa and all of the other variables in the model including the dependent variable ret. The correlations on the dependent variable were only significant for roa and p_mtd showing that probably these were the two predictors that had more chance to have an

effect on it. The main dependent variable of the study rep showed a low and non-significant correlation with dependent variable showing that probably its effect on it in the final model will not have been significant. However the weak relationships, highlighted by the low Pearson correlation coefficients did not provide the basis to suppose that these variables had relevant mediating effects on one another and therefore affect the following regression analysis. Further I tested for another of the assumptions related to linear regression that is related to absence of relevant heteroscedasticity along the observations of the sample, which means that the error terms do not have a constant variance. I used a scatterplot (Appendix 1) for the Table 7: Means, Std. Deviations, Correlation

Matrix

Variables MEAN ST. DEV. 1 2 3 4 5 6 7 8

1. Relative Size 1.659586 0.7042905 1 2. Return on Assets 6.9168 7.633262 0.129 1 3. Debt-to-Equity 3.899701 0.6101997 -0.183* -0.184* 1 4. Abnormal returns -0.968 5.89917 0.061 0.189* -0.005 1 5. Reputation 6.5393 0.92535 -0.083 0.210* -0.086 0.093 1 6. Premium 7.035539 1.7814596 -0.264** -0.321** 0.066 -0.146 0.033 1 7. Payment method 2.126737 0.780 -0.224** 0.278** -0.177* 0.198* 0.260** 0.157 1 8. Acquisition Experience 2.126737 1.4023908 -0.151 0.191* 0.046 0.009 0.001 0.195* 0.032 1

**.Correlation is significant at the 0,01 level (2-tailed)