The effect of sustainability on investors’ reaction to M&A announcements

(1)

UNIVERSITY OF AMSTERDAM

The effect of sustainability on investors’ reaction to M&A

announcements

ABSTRACT

In this paper I examine whether the environmental, social and governance (ESG) performance of acquirer firms, measured by the KLD STATS ESG ratings, has an effect on the reaction of acquirers’ and targets’ shareholders to merger and acquisition announcements. Using two different samples of acquirers and targets, I find that the ESG score of the acquirer firms does not influence the cumulative abnormal returns of either the targets’ or the acquirers’ shareholders. These results contradict the stakeholder value-maximization theory and learning effect theory, which state that focusing on the interests of stakeholders by investing in sustainable activities increases shareholders’ wealth, and that targets learn from the ESG practices of the acquirer firm. Thus, it appears that investors of the merging companies do not believe that sustainability will create value for them.

Name: Alessia Zurli

Student number: 11119837 Supervisor: Shivesh Changoer Faculty: Economics & Business Track: Finance & Organization University: University of Amsterdam 12 ECTS

(2)

Statement of Originality

This document is written by Alessia Zurli, who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document are 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)

1. Introduction

The US SIF Foundation reported in 2016 that $8.72 trillion of total assets in the United States were being used on sustainable responsible investing (SRI) strategies. Individuals, companies, and financial institutions often use these strategies to obtain long-term competitive financial returns (US SIF, 2016). Other reasons for why they pursue SRI strategies is to manage risk and to support businesses or products that help promote environmental, social and governance (ESG) practices (US SIF, 2016).

According to the Investment Leaders Group (2014), ESG strategies can create value for investors and their clients in various ways: by increasing returns, achieved through an improved sustainability performance by the firms involved; by lowering the share price volatility, by internalizing costs sustainable firms are protected from earning shocks and market reactions; and by promoting a more stable, long-term shareholder base. ESG strategies can also create non-financial value, by improving the ESG performance of companies (Investment Leaders Group, 2014).

On the other hand, ESG strategies can be value-destroying (Walley & Whitehead, 1994). Walley and Whitehead (1994), for example, state that adopting these strategies creates higher costs and thus reduces the profitability of a company.

According to Deng, Kang, and Low (2013) there are two theories that explain the value creation or destruction of SRI strategies: the stakeholder value maximization theory and the shareholder expense theory. The stakeholder value-maximization theory states that “CSR activities have a positive effect on shareholder wealth because focusing on the interests of other stakeholders increases their willingness to support a firm’s operation, which increases shareholder wealth” (Deng et al., 2013, p. 88). On the other hand, the shareholder expense view suggests that “ managers engage in socially responsible activities to help other stakeholders at the expense of shareholders” (Deng et al., 2013, p. 89). Thus, there is a tradeoff between maximizing shareholder value and maximizing stakeholder value.

Because it is not clear whether sustainable responsible investing creates or destroys value, prior research examines whether effective management of SRI factors has a positive or negative effect on firms’ financial performance. Almost 90% of 2200 studies find a non-negative relation between ESG performance and firm financial performance (Friede, Bush and Bassen, 2015). The other studies find either a negative or a nonsignificant effect of sustainability on firm value.

(4)

Since the findings are mixed, there is still an on-going debate among researchers on whether SRI creates or destroys value for shareholders. I focus on mergers and acquisitions (M&As), because they are one of the largest investments that companies undertake in their life cycle, and thus affects shareholders and stakeholders (Betton, Eckbo & Thorburn, 2008). M&As can thus help to understand the tradeoff mentioned above between shareholders and stakeholders. Moreover, the use of M&As can mitigate the reverse causality problem present in previous studies on the effect of CSR on firm performance, since M&As are unanticipated events (Deng et al., 2013). This allows me to examine abnormal returns.

In this paper I answer the following research question: How does the ESG (environmental, social and governance) performance of acquirer firms affect investors’ reaction to M&A announcements?

To find an answer to this question, I examine the three-day cumulative abnormal returns (CARs) around the announcement day (-1,1), which is defined as day 0. The CARs measure the market reaction to the M&A announcement (Rosen, 2006). In other words, CAR shows how investors react when a firm announces a merger or acquisition (Rosen, 2006). To measure sustainability I use the KLD STATS ESG ratings, which are based on seven different sustainability areas, namely environmental, human rights, diversity, community, employee relations, product and governance.

I find no significant relation between either the acquirers’ or the targets’ shareholders’ returns and the environmental, social and governance performance of acquirers. This finding holds after carrying out four different sensitivity analyses to control for other variables, event periods, deal announcement periods and models. As mentioned before, CAR represents investors’ reaction to M&As (Rosen, 2006). How investors react to such an announcement depends on whether they believe that the M&A will create value for them (Rosen, 2006). For example, a CAR bigger than 0 means that investors believe that the M&A will create value for them. I find insignificant coefficients on ESG score for both acquirer and target samples (Rosen, 2006). Thus, it appears that investors of the merging companies do not believe that sustainability will create value for them.

My research paper builds on work by Deng et al. (2013). They investigate the relation between acquirers’ corporate social responsibility (CSR) and abnormal returns of acquirers’ shareholders around the merger announcement day. In contrast, I examine the abnormal returns of acquirers’ and targets’ shareholders around the merger announcement day. As far as I am aware, my paper is the first to investigate whether a relation between acquirers’ ESG performance and the returns of targets’ shareholders exist. I examine both abnormal returns of

(5)

acquirers’ and targets’ shareholders because, if a learning effect of sustainability takes place, we should also expect higher cumulative abnormal returns for targets’ shareholders (see Section 2.4).

The remainder of the paper is divided into 4 sections. In section 2 I present a literature review around the topic of mergers and ESG performance, and the theory behind the two hypotheses that I test. In section 3 I include a description of the method, the model and the sample. In section 4 I present the results from univariate and multivariate analyses. In section 5 I carry out four sensitivity analyses. In Section 6 I draw conclusions.

2. Literature review

2.1 Mergers and acquisitions

Since 1985, more than 300,000 merger and acquisition deals have been announced, with a value of almost 33,200 billion US dollars ("The Institute for Mergers, Acquisitions and Alliances", 2018). The Deloitte M&A trends report (2017) predicts an increase in the number of M&As in 2018.

DePamphilis (2009) describes mergers and acquisitions as “the buying, selling, dividing and combining of companies”. Even though the distinction between a merger and an acquisition has become less important in recent years, they are different instruments of corporate takeover (DePamphilis, 2009). An acquisition is when one firm becomes part of another such that, after the deal, the target firm disappears as a legal entity (DePamphilis, 2009). A merger is defined as two firms joining to create a new legal entity (DePamphilis, 2009). M&As are one of the fastest ways for corporate restructuring and expansion (DePamphilis, 2009). In the remainder of this paper I will use interchangeably the terms merger and acquisition to refer to takeovers in general.

One of the reasons why M&As create value for shareholders of the merging firms is that they increase efficiency gains by creating economies of scale, economies of scope and economies of vertical integration (Motis, 2007). Another reason why M&As create value is that they can create synergy gains through the diffusion of know-how and R&D (Motis, 2007). Also, M&As can lead to cost savings through the process of rationalization (which consists of a more optimal reallocation of production across the merging firms), by purchasing power, and by creating internal capital markets (Motis, 2007). Financial cost savings, achieved through tax advantages, cheaper interest rates and diversification, are also another reason why M&As can create value for shareholders of targets and acquirers (Motis, 2007). Moreover, M&As create value for

(6)

shareholders by strengthening the market power of a company, through unilateral effects (by threatening to raise its prices after the merger), collusion, by raising entry barriers, spreading their portfolio and by obtaining a multimarket contact (combining firms active in different markets) (Motis, 2007). Lastly, M&As can create value by having a defensive motive or a disciplinary motive (Motis, 2007).

However, M&As can also destroy value for shareholders of the merging firms. Value destruction occurs when managers seek to maximize their own wage instead of the firm’s value, thus when they are looking to gain at the expense of shareholders (Motis, 2007). One of the reasons why M&As destroy shareholder value is managers engaging in empire building (Motis, 2007). Empire building occurs when managers increase the size of the company, since this might lead to a higher compensation for them. However, empire building can be very costly for the shareholders of both acquirers and targets (Motis, 2007). Another reason why M&As can be value destroying for shareholders is hubris, which occurs when managers believe to be better able to manage other companies (Motis, 2007). Lastly, M&As can destroy value for shareholders if managers seek to construct an optimal portfolio by risk spreading and diversifying (Motis, 2007).

Thus, it is not clear whether M&As create or destroy value for firms’ shareholders.

2.2 Empirical findings of prior research

Numerous studies analyze the effects of M&As on shareholders’ returns, measured by the change in stock prices of acquirer and target firms around the announcement day. To determine whether value creation or destruction occurs, short-term event studies are often used. The idea behind this method is that it compares the returns of a stock around the event date (i.e. the merger announcement day) and the expected return if there had been no event (De Jong, 2007). The difference between the actual return in the event period and the expected return is defined as the abnormal return (De Jong, 2007). In an efficient market, prices adjust quickly after the announcement of a merger, thus easily representing the change in firm value (Jensen & Ruback, 1983).

A large number of studies (see, e.g., Jensen & Ruback, 1983) show that M&As create value for targets’ shareholders. However, there is no real consensus whether M&As create value for acquirer firms, even though it is known that acquirers pay large premiums for target firms (Jensen & Ruback, 1983). For example, Jensen and Ruback (1983) find that mergers and acquisitions are value creating for the shareholders of both acquirer and target firms, though

(7)

most of the gains are collected by target shareholders (Jensen, 2002). However Dodd (1980) finds that shareholders of acquirer firms earn small, but significant negative abnormal returns at the announcement day of the merger. Dodd (1980) argues that the lack of positive returns to the acquirer firms’ shareholders is due to merger gains not being measured accurately.

2.3 ESG performance in mergers and acquisitions

Various studies analyze the effect that M&As have on shareholders’ gains. However, few papers have analyzed the role of sustainability in this context. Moreover, no paper to my knowledge as studied the effect of the ESG performance of acquirer firms on the target firms’ shareholders’ returns.

The paper by Deng et al. (2013) tests the effect of acquirers’ corporate social responsibility (CSR) on acquirers’ shareholders’ returns using a sample of 1556 mergers in the United States between 1992 and 2007. They calculate the cumulative abnormal return (CAR) of acquirers in different event periods around the announcement day. Insignificant results are found for the full sample, so the authors divide this into a subset of 786 firms with a high CSR level and a subset of 770 firms with a low CSR level (Deng et al., 2013). From these subsets Deng et al. (2013) conclude that the CAR of high CSR firms in the period (-1,1) is slightly positive but insignificant, while the CAR of low CSR firms is found to be significant at the five percent significance level. Deng et al. (2013) carry out other tests and find that acquirers with a higher level of corporate social responsibility realize higher announcement abnormal returns. These results suggest that investors of the merging firms believe that sustainability will create value for them.

Atkas, Bodt and Cousin (2011) analyze the effect of the ESG performance of target firms on the returns of both acquirers’ and targets’ shareholders. They test this using a sample of 106 M&A deals between 1997 and 2007 and by calculating the cumulative abnormal returns of acquirers and targets. The authors find that when targets have a high ESG performance, acquirers earn positive and significant cumulative abnormal returns, compared to targets with a low ESG performance. Moreover, the returns of targets’ shareholders do not depend on the targets’ sustainability performance (Atkas et al., 2011). The authors show that this value creation is driven by the acquirer learning from the targets’ sustainable activities. Overall the results suggest that investors of acquirer firms value the acquisition of a high ESG target positively (Atkas et al., 2011).

(8)

2.4 Hypotheses

The general idea I test in this research paper is that shareholders, of both acquirer and target firms, have a positive valuation of sustainability.

Hypothesis 1: The ESG score of acquirer firms has a positive effect on the CARs of acquirers.

Deng et al. (2013) argue that investing in ESG activities can create value for the firms’ shareholders. The reason is that if firms invest, stakeholders of the firm become more willing to support a firm’s operation, because high ESG firms have a higher reputation for committing to their implicit contracts (Deng et al., 2013; Jensen, 2001). Thus, stakeholders of these high ESG firms have a stronger incentive to support a firm’s operation by contributing resources and effort, leading to higher shareholder wealth (Deng et al., 2013; Jensen, 2001).

M&As put at risk the relations between a firm and its stakeholders because often the stakeholders have to renegotiate their contracts with the new merged firm (Deng et al., 2013). For this reason, having a high reputation for committing to implicit contracts, is important for the success of M&As (Deng et al., 2013). Thus, the ESG performance of an acquirer firm has a positive effect on the wealth of its shareholders. In other words, shareholders of a high ESG acquirer benefit more from the merger than shareholders of a low ESG acquirer.

Since I only measure the ESG performance of acquirer firms, I will consider only the shareholders and stakeholders of acquirer firms in order to be able to apply the theory of stakeholder value-maximization. Thus, based on this theory, I expect a positive relation between the ESG performance of acquirers and the three-day cumulative abnormal returns of acquirers’ shareholders around the announcement day. Specifically, acquirers with a high ESG performance realize higher merger announcement returns compared to acquirers with a low ESG performance.

Hypothesis 2: The ESG score of acquirer firms has a positive effect on the CARs of targets.

As mentioned in Section 2.2, the paper by Atkas et al. (2011) finds that acquirers learn from the targets’ SRI knowledge (e.g. by pushing the acquirer to meet certain sustainable standards) and that this increases acquirers’ shareholders’ returns. This holds only if SRI practices can be considered value-creating (Atkas et al., 2011).

(9)

I expect that if acquirers have a high ESG performance targets will benefit from the acquirers’ SRI knowledge. Indeed, SRI knowledge can improve employee and customer satisfaction, increase a firm’s reputation with outside parties (allowing the firm to more easily obtain access to capital), signal the quality of management and create new market opportunities (Atkas et al., 2011). I argue that, since it is often the acquirers’ strategy that is continued after the merger, the target firm is probably strongly influenced by the sustainable strategy of acquirers. Therefore, if a learning effect is present in sustainability, I expect a positive relation between the ESG performance of acquirer firms and the three-day cumulative announcement returns of target firms around the merger announcement day. Specifically, targets acquired by firms with a high ESG performance realize higher merger announcement returns than targets acquired by firms with a low ESG performance.

3. Research design

3.1 Method

To test my hypotheses I run the following regressions.

𝐶𝐴𝑅_𝑎𝑡 = 𝛼 + 𝛽₁𝐸𝑆𝐺_𝑡+ 𝛽₂𝑆𝑖𝑧𝑒_𝑡+ 𝛽₃𝑇𝑞_𝑡+ 𝛽₄𝐹𝐶𝐹_𝑡+ 𝛽₅𝐿𝑒𝑣_𝑡+ 𝛽₆𝐴𝐶𝐷_𝑡+ 𝛽₇𝐴𝑆𝐷_𝑡+ + 𝛽8𝐻𝑇𝑡+ 𝜀 (1)

𝐶𝐴𝑅_𝑡𝑡 = 𝛼 + 𝛾₁𝐸𝑆𝐺_𝑡+ 𝛾₂𝑆𝑖𝑧𝑒_𝑡+ 𝛾₃𝑇𝑞_𝑡+ 𝛾₄𝐹𝐶𝐹_𝑡+ 𝛾₅𝐿𝑒𝑣_𝑡+ 𝛾₆𝐴𝐶𝐷_𝑡+ 𝛾₇𝐴𝑆𝐷_𝑡+ 𝛾₈𝐻𝑇_𝑡 + 𝜔 (2),

where CARat are the three-day cumulative abnormal returns of the acquirer firms, based on the

market model and in the event period (-1,1); CARtt are the three-day cumulative abnormal

returns of the target firms, based on the market model and in the event period (-1,1);

Abnormal returns are calculated by taking the difference between the actual returns and the expected stock returns (Jensen, 2002). I calculate the expected stock returns using the market model, consistent with Deng et al. (2013) and Atkas et al. (2011), in order to take into account the effect of market wide events on the returns of the firms in my sample. I estimate the market model parameters using two hundred trading days of returns that end eleven days before the merger announcement day, following the paper by Deng et al. (2013). I sum the daily abnormal stock returns and obtain the cumulative abnormal stock return (CAR) in the event period (-1,1), which represents the three days surrounding the announcement day. Most studies estimate the

(10)

effect of M&A announcements on stock prices through the use of an event study methodology. Specifically, to measure such effect, abnormal stock returns around the announcement day are used (Jensen, 2002).

ESGt is the environmental, social and governance score of the acquirer firms taken from the

KLD STATS ESG database and measured at the end of the fiscal year prior to the acquisition announcement. I measure the ESG performance of the acquirer firms using a score based on the KLD STATS ESG database, an annual data set of positive and negative environmental, social and governance (ESG) performance indicators. This database has been used extensively in prior research on corporate social responsibility (Deng et al., 2013; Jiao, 2010; Waddock & Graves, 1997). In this database, ESG performance indicators are scored using a binary scoring model. If a company meets the assessment criteria of an indicator, then this is identified with ”1”, if that company does not meet the criteria it is identified with a “0”. To arrive at the final rating I sum all strengths and subtract all concerns in seven main different performance areas, namely: environmental, human rights, diversity, community, employee relations, product and governance. This method is consistent with the research by Deng et al. (2013).

To ensure that my analysis does not proxy for other factors that are known to influence merger announcement returns, I follow the papers by Deng et al. (2013) and Masulis, Wang and Xie (2007) and include nine control variables related to bidder characteristics and deal characteristics: Sizet, Tqt, FCFt, Levt, ACDt, ASDt, HTt. All control variables are measured at

the end of the fiscal year prior to the acquisition announcement. Following the paper by Atkas et al. (2011), I do not control for targets’ characteristics, since the authors only control for bidder and deal characteristics although having the targets’ CAR as dependent variable. I extract the data for these companies from the Compustat database.

Sizet is the logarithm of the total assets of the acquirer firm (Deng et al., 2013). Tqt is the

market value of assets divided by the book value of assets (Deng et al., 2013). FCFt is defined

as operating income before depreciation, minus interest expenses, minus income taxes, minus capital expenditure, divided by the book value of assets (Deng et al., 2013). Levt is the book

value of long term debt and short term debt, divided by the market value of assets (Deng et al., 2013). ACDt is a dummy that equals one if the deal was fully financed by cash and zero

otherwise (Deng et al., 2013). ASDt is a dummy that equals one if the deal was fully financed

by stock and zero otherwise (Deng et al., 2013). HTt is a dummy variable that equals one if

both the acquirer and the target firm operate in a high-tech industry, and zero otherwise (Deng et al., 2013).

(11)

I include Size because Moeller, Schlingemann, and Stulz (2004) find that the size of the acquirer firms is negatively correlated with the acquirers’ announcement period CAR, specifically that larger acquirers make acquisitions that generate negative synergies. A reason for this is that it takes more resources to acquire a larger target. Thus I expect the markets’ reaction to be less positive the bigger the size of the acquirer firm.

Following existing literature I include Tq. Prior studies find an unclear effect of acquirers’ Tobin’s q on acquirers’ announcement period CAR. For example, Servaes (1991) reports a positive relation between Tobin’s q and CAR while Moeller et al. (2004) find a negative relation. I include Tobin’s q as control variable since these studies find that it has an effect (positive or negative) on acquirers’ announcement period CAR.

Following existing literature, I include FCF. There is uncertainty regarding the value this coefficient could take in relation to the acquirers’ announcement returns (Masulis et al., 2007). Firms with more free cash flow have more capital and thus might engage in empire building (Masulis et al., 2007). However, higher free cash flows could also signify better firm performance, which could be correlated with higher quality managers, who tend to make better acquisition decisions (Masulis et al., 2007). I include free cash flow as control variable since these studies find that it has an effect (positive or negative) on acquirers’ announcement period CAR.

I include Lev because leverage helps reduce future free cash flows, limit managerial discretion and incentivize managers to improve firm performance, since there is a bigger risk that their firms might fall into financial distress if they have large amounts of debt (Masulis et al., 2007). For these reasons I expect Lev to have a positive effect on acquirers’ announcement period CAR (Masulis et al., 2007).

I include ACD and ASD because during the acquisition the acquirer can choose either cash or stock financing or a combination of the two, and this decision can have significant effects on shareholders’ gains (Masulis et al., 2007). Stock offers take longer time because they have to be approved by the SEC, thus represent a higher cost of transaction (Masulis et al., 2007). Also, the issuance of stock is viewed negatively by the market (Masulis et al., 2007). Thus I expect investors’ reaction to be positively related to mergers financed with cash and negatively related to mergers financed with stocks.

I include HT since acquirers in high tech industries are likely to underestimate the costs and overestimate the synergies of the M&A deal (Masulis et al., 2007). Thus, I expect HT to have a negative effect on investors’ reaction to M&As.

(12)

Hypothesis 1. H0: B1=0 vs H1: B1>0

Hypothesis 2. H0:

γ

1=0 vs H1:

γ

1>0

3.2 Sample

The initial data set includes companies taken from the Thompson One database. M&A deals are selected based on 5 criteria. The firms involved in the takeover are listed firms, are located in the United States and have a deal value bigger than $10 million. The deal is completed and is announced between the 1st of January 2005 and the 31st of December 2015, so in a ten year period. I focus on M&A deals in the United States since the country has had the largest number of deals and the biggest deal values during the years ("The Institute for Mergers, Acquisitions and Alliances", 2018).Also, I focus on the ten- year period from 2005 to 2015 to analyze more recent M&As since other papers on the topic only cover deals up to 2007 (Deng et al., 2013; Atkas et al., 2011).

The selection criteria give 19,846 firms, among which, however, some duplicates are present. The sample is reduced to include only those deals for which acquirers’ ESG ratings are available in the KLD STATS ESG database. Moreover, I take only those firms for which all strengths and concerns within all seven qualitative issue areas are available, and sum (substract) all strengths (concerns). Also, I make sure that data for acquirer firms’ and deal characteristics is available in the Compustat database. In order to test my two different hypotheses I divide the sample into an acquirer sample and a target sample. The first sample includes M&A deals for which acquirers’ CARs are available in the WRDS database. I find 931 that meet the criteria. The second sample includes M&A deals for which targets’ CARs are available in the WRDS database. For this sample I find only 213 deals. This is consistent with the small samples used by Atkas et al. (2011). Furthermore, in order to test the difference between the effect of high and low ESG acquirers on the CAR of shareholders, I divide each sample into two subsamples according to the sample median ESG score of acquirers. This results in a subsample of high ESG acquirers and a subsample of low ESG acquirers.

In Table 1 I present summary statistics both for the full samples of acquirers and targets (Panel A) and the subsamples of high and low ESG acquirers (Panels B and C). In Panel D I test the difference in means between the subsamples of targets and acquirers in panels B and C. I do this for all the variables in the model. Panel D shows that acquirers with a high ESG score have a significantly lower firm size and leverage, and tend to acquire more targets in high-tech

(13)

industries as compared to low ESG acquirers. Also, high ESG acquirers are significantly more likely to finance the deal with stocks and have a significantly higher Tobin’s q, indicating that high ESG acquirers tend to have a higher firm value than low ESG acquirers. Also, targets acquired by a high ESG score acquirer, have a significantly higher Tobin’s q, indicating that targets acquired by a high ESG firm tend to have a higher firm value, as compared to targets acquired by low ESG firm. Moreover, they tend to be acquired by more firms in high-tech industries as compared to targets acquired by low ESG firms.

TABLE 1

Summary statistics

Panel A: Full samples

Variable Acquirer sample (N=931) Target sample (N=213)

Mean Std. Dev. Min Max Mean Std. Dev. Min Max

CAR 0.016 0.073 -0.296 0.860 0.099 0.174 -0.131 0.918

ESG score -0.565 2.163 -9.000 10.000 -0.718 2.731 -9.000 9.000

Firm size 3.187 0.727 1.027 5.876 3.498 0.800 1.610 5.876

Tobin's q 1.836 1.003 0.595 8.445 1.853 1.019 0.617 8.098

Free cash flow 0.052 0.094 -1.102 0.555 0.068 0.072 -0.315 0.453

Leverage 0.151 0.153 0.000 0.892 0.155 0.145 0.000 0.720

All-cash deala _0.047 _0.212 _0.000 _1.000 _0.127 _0.334 _0.000 _1.000

All-stock deala _0.793 _0.406 _0.000 _1.000 _0.357 _0.480 _0.000 _1.000

High techa _0.144 _0.351 _0.000 _1.000 _0.150 _0.358 _0.000 _1.000

Panel B: Target subsamples

Variable High ESG acquirers (N=133): A Low ESG acquirers (N=80): B

CAR 0.111 0.183 -0.131 0.918 0.079 0.158 -0.119 0.634

ESG score 0.759 2.108 -1.000 9.000 -3.175 1.675 -9.000 -2.000

Firm size 3.443 0.802 1.610 5.876 3.589 0.794 1.744 5.481

Tobin's q 1.871 1.035 0.617 8.098 1.823 0.997 0.863 6.717

Free cash flow 0.067 0.080 -0.315 0.453 0.069 0.059 -0.056 0.250

Leverage 0.000 0.153 0.000 0.720 0.171 0.131 0.000 0.535

All-cash deala _0.113 _0.318 _0.000 _1.000 _0.150 _0.359 _0.000 _1.000

All-stock deala _0.368 _0.484 _0.000 _1.000 _0.338 _0.476 _0.000 _1.000

(14)

TABLE 1-Continued Panel C: Acquirer subsamples

Variable High ESG acquirers (N=432): C Low ESG acquirers (N=499): D

CAR 0.017 0.080 -0.296 0.860 0.015 0.054 -0.118 0.260

ESG score 0.349 1.707 -1.000 10.000 -2.850 1.337 -9.000 -2.000

Firm size 3.109 0.716 1.027 5.876 3.383 0.721 1.452 5.481

Tobin's q 1.870 1.046 0.595 8.445 1.750 0.882 0.610 6.717

Free cash flow 0.052 0.097 -1.102 0.555 0.052 0.085 -0.473 0.379

Leverage 0.144 0.153 0.000 0.892 0.170 0.151 0.000 0.694

All-cash deala _0.042 _0.201 _0.000 _1.000 _0.060 _0.238 _0.000 _1.000

All-stock deala _0.811 _0.392 _0.000 _1.000 _0.748 _0.435 _0.000 _1.000

High techa _0.174 _0.380 _0.000 _1.000 _0.068 _0.252 _0.000 _1.000

Panel D: Tests of difference

Variable Test of difference (A-B) Test of difference (C-D)

Mean Mean CAR 0.032 0.002 ESG score 3.934*** 3.199*** Firm size -0.146 -0.274*** Tobin's q 0.048* 0.120*

Free cash flow -0.002 0.000

Leverage -0.171 -0.026**

All-cash deala _-0.037 _-0.018

All-stock deala _0.030 _0.063**

High techa _0.140*** _0.106***

This table presents summary statistics for the full samples of acquirers and targets (Panel A) and for the subsamples of high and low ESG acquirers (Panels B and C). The subsamples are obtained by dividing the full samples of acquirers and targets according to the sample median ESG score of acquirers. Panel D reports the test of difference in means between the subsamples of acquirers and targets in Panels B and C. The samples consist of merger and acquisition deals announced over the period 2005-2015 and extracted from the Thompson One database. The extraction criteria are presented in Section 3.2. All variables are defined in Table A1. CAR is in percent, Firm size is in USD Mil. a _{Denotes a dummy variable. * p < 0.05, ** p < 0.01, *** p < 0.001.}

4. Results

4.1 Univariate analyses

In Table 1, Panel A, I present the mean CARs for the full sample of acquirers and targets. In Panels B and C, I present the mean CARs for the subsamples of high and low ESG acquirers. I divide the acquirers into high and low according to the sample median of the acquirers’ ESG score. Panel D includes the test of difference in mean CARs between the subsamples of high

(15)

and low ESG acquirers. These tests report insignificant mean differences in CARs between the subsamples of high and low ESG acquirers. This result holds for both target and acquirer samples. Overall the results show that acquirers with high or low ESG scores have similar announcement returns. Similarly, targets acquired by firms with high or low ESG scores have similar announcement returns.

In table 2 I present correlation matrices for the full samples of acquirers and targets. Panel A presents the correlation matrix for the full sample of acquirers. Panel B presents the correlation matrix for the full sample of targets. The table shows that the acquirers’ ESG score is not significantly correlated to either the acquirers’ or targets’ announcement period CAR. Since CAR represents investors’ reaction to M&As, these results suggest that the environmental, social and governance performance of acquirers does not affect investors’ reaction to M&A announcements (Rosen, 2006).

4.2 Multivariate analyses

To better understand whether sustainability creates value for shareholders, I carry out two multivariate analyses using either the acquirers’ announcement period CARa (1) or the targets’

announcement period CARt (2) as the dependent variable, and the acquirers’ ESG score as the

main independent variable. I report the results in Table 3. Column 1 presents the estimates of the OLS regression for the full acquirer sample. Column 2 presents the estimates of the OLS regression for the full target sample. I find that both coefficients on the acquirers’ ESG score are not significant at the 5% level. Thus the ESG score of the acquirer firms is not significant in explaining the cumulative abnormal returns of either the acquirer or target firms’ shareholders. These results contradict the stakeholder value-maximization theory, thus focusing on the interests of other stakeholders does not increase shareholder wealth. Also, the results contradict the learning effect theory, thus no learning effect of sustainability takes place, more specifically, targets do not learn from the ESG practices of acquirers.

Overall the results show that the environmental, social and governance performance of acquirers does not affect investors’ reaction to M&A announcements. Moreover, the results show that shareholders of high ESG acquirers do not realize higher announcement returns compared to shareholders of low ESG acquirers. Similarly, shareholders of targets acquired by high ESG firms do not realize higher announcement returns compared to shareholders of targets acquired by low ESG firms. These results confirm the univariate results reported in Tables 1 and 2 and contradict my first and second hypothesis. In conclusion, results indicate that

(16)

TABLE 2

Correlation matrix

Panel A: Acquirer sample

CARa ESG score Firm size Tobin's q Free cash flow Leverage All-cash deal All-stock deal High tech

CARa -

ESG score -0.0151 -

Firm size -0.0876** -0.0025 -

Tobin's q -0.0222 0.1289*** -0.1334*** -

Free cash flow -0.0544 0.1072*** 0.1060*** 0.2663*** -

Leverage 0.0870** -0.1265*** 0.2155*** -0.3836*** -0.1914*** -

All-cash deal -0.0484 -0.0542 0.0888** 0.0223 0.0397 -0.0386 -

All-stock deal -0.0023 0.0441 -0.1775*** -0.0096 -0.0704* -0.0289 -0.4355*** -

High tech -0.0321 0.1737*** -0.1011** 0.1788*** 0.1029** -0.2245*** 0.0240 0.0436 -

Panel B: Target sample

CARt ESG score Firm size Tobin's q Free cash flow Leverage All-cash deal All-stock deal High tech

CARt -

ESG score 0.1115 -

Firm size -0.0263 0.0240 -

Tobin's q -0.0236 0.1280 -0.1700** -

Free cash flow 0.0038 0.1441* 0.0195 0.6024*** -

Leverage -0.0567 -0.1280 0.2808*** -0.3817*** -0.3286*** -

All-cash deal -0.2246*** -0.0964 -0.0224 0.0890 0.0353 -0.1181 -

All-stock deal 0.5488*** 0.0885 0.0868 -0.0433 0.0031 0.1029 -0.2838*** -

High tech 0.1608* 0.1928** -0.1168 0.1240 0.0688 -0.1930** 0.0373 0.1531* -

This table presents correlations for the full samples of acquirers and targets. The samples consist of merger and acquisition deals announced over the period 2005-2015 and extracted from the Thompson One database. The extraction criteria are presented in Section 3.2. All variables are defined in Table A1. CAR is in percent, Firm size is in USD Mil. *, **, and *** denote significance at the 10%, 5%, and 1% level, respectively.

(17)

sustainability is valued neutral by shareholders.

In column 1, the coefficients on Size and Lev are both significant at the 1% level, the coefficients on all other variables are insignificant. The coefficient on Size is negatively significant. This result is in line with the research by Moeller et al. (2004), who find that the size of the acquirer firm is negatively correlated with the acquirer's announcement period CAR. The coefficient on Lev is positively significant. This finding confirms the results of Masulis et al. (2007), who state that leverage incentivizes managers to improve firm performance. In column 2 all except for the coefficient on all-stock deal dummy are insignificant. The positive coefficient on all-stock deal dummy contradicts the research by Masulis et al. (2007).

TABLE 3

Regressions of CARs (-1,1) for acquirers (1) and targets (2) on explanatory variables

(1) (2)

Variable CARa CARt

ESG score 0.000 0.003 (0.02) (0.68) Firm size -0.011** -0.010 (-3.20) (-0.72) Tobin's q 0.001 -0.009 (0.37) (-0.66)

Free cash flow -0.020 -0.027

(-0.74) (-0.15) Leverage 0.050** -0.138 (2.84) (-1.74) All-cash deala _-0.018 _-0.043 (-1.43) (-1.37) All-stock deala _-0.008 _0.191*** (-1.16) (8.60) High techa _-0.004 _0.027 (-0.49) (0.91) Constant 0.051*** 0.107 (3.47) (1.97) Sample size 931 213

(18)

TABLE 3-Continued

(1) (2)

Variable CARa CARt

R2 _0.023 _0.330

This table presents multivariate regression analyses of announcement abnormal returns. The regressions are based on a sample of M&A deals announced in the period 2005-2015 and extracted from the Thompson One database. The extraction criteria are presented in Section 3.2. In the acquirer sample, the cumulative abnormal return of the acquirer firm CARa is the dependent variable. In the target sample, the cumulative abnormal return

of the target firm CARt is the dependent variable. Both CARs are based on the market model and estimated

around the announcement day of the M&A deal. The market model parameters are estimated using two hundred trading days of return data ending 11 days before the merger announcement. The method of calculation of the CARs is described in Section 3.1. ESG score is the main variable of interest and is extracted from the KLD ESG STATS database. The method of calculation of ESG score is described in Section 3.1. All variables are defined in Table A1. CAR is in percent, Firm size is in USD Mil. a _{Denotes a dummy variable. The numbers in}

parentheses are t statistics. * p < 0.05, ** p < 0.01, *** p < 0.001.

5. Sensitivity analyses

I replicate the analyses using four different regressions in order to determine if controlling for an additional variable, changing the event period, the model, or excluding a time period, impacts the CAR of shareholders.

In Table A2 in the Appendix I carry out a regression similar to the one in section 4.2 but include an additional control variable “Related”. Related deals generate higher abnormal returns for shareholders (Betton et al, 2008; Atkas et al., 2011). Eckbo (1983) finds that horizontal mergers generate positive abnormal returns for shareholders of acquirers and targets because they increase the probability of successful collusion among rival producers, which reduces the number of independent producers in the industry and thus reduces monitoring costs. Related is a dummy variable that equals one if the acquirer and the target are in the same industry and zero otherwise (Atkas et al., 2011). I identify related firms using four-digit SIC codes from the Thompson One database.

In Table A3, I carry out a similar regression as in section 4.2 but exclude the merger deals occurred during the years affected by the financial crisis (from 2007 to 2009 included) to analyze whether the crisis might have had a negative effect on the announcement returns of merger deals.

In Table A4, I carry out a similar regression as in section 4.2 but extend the event period to (-10, 10). I conduct this analysis because it is possible that before the official announcement day information is released, making it hard to understand which date represents the correct event date. For example, Oler, Harrison, and Allen (2007) analyze the effect of using different event windows in an event study methodology of a large sample of acquisitions. They argue that when information revealed to the market is complex, the market’s initial response after the

(19)

event might be incomplete and might not capture the economic impact of the event. Specifically, they find different returns in the short and long-window, indicating that the market did not have enough information to make a complete assessment of the event. Thus, to account for these issues, I analyze a longer event window, both before and after the announcement day. In Table A5, I carry out a similar regression as in section 4.2 but I use a different model to calculate the expected stock returns, the Fama- French Three Factor Model. I conduct this test, because the Three Factor Model takes a different approach to explain market pricing (Fama & Kenneth, 1993). It assumes that when investors make decisions they take into consideration more than one factor, specifically: market, size and value (Fama & Kenneth, 1993). Thus, I control also for these factors when running the regression of the ESG score on the cumulative abnormal returns of firms.

In all three sensitivity analyses I arrive at the same conclusions as in the main analysis in Section 4.2. Indeed, I find a negative significant coefficient on firm size and a positive significant coefficient on leverage for column 1. I find insignificant coefficients on all other variables. Also, I find the same positive and significant coefficient on all-stock deal dummy in column 2. Most importantly I find an insignificant coefficient on ESG score in both columns, suggesting that the environmental, social and governance score of acquirers does not influence the cumulative abnormal return of either the targets’ or the acquirers’ shareholders. Thus, the ESG performance of acquirers does not affect investors’ reaction to M&A announcements.

6. Conclusion

In this paper I analyze how the environmental, social and governance (ESG) performance of acquirer firms in M&A deals, announced in the US in the period from 2005 to 2015, affects investors’ reaction to M&A announcements.

In order to answer my research question I analyze M&As using an event study methodology, where cumulative abnormal returns are used as a measure of firms’ financial performance. To measure sustainability I use the KLD STATS ESG ratings, which is based on seven different sustainability areas, namely environmental, human rights, diversity, community, employee relations, product and governance. I run a regression where I include the ESG score of acquirer firms as explanatory variable and the cumulative abnormal return of acquirers around the announcement day as dependent variable, controlling for other variables. I also run a second regression of the acquirers’ ESG score on the cumulative abnormal returns of targets.

(20)

I expect to find positive and significant returns for both firms’ shareholders, according to the theories of stakeholder value-maximization and learning effect. The results, however, show that the ESG score of acquirers does not influence the cumulative abnormal returns of either the targets’ or the acquirers’ shareholders. Specifically, the results show that shareholders of high ESG acquirer firms do not realize higher announcement returns compared to shareholders of low ESG acquirer firms. Similarly, targets’ shareholders, acquired by a high ESG firm, do not realize higher announcement returns compared to targets acquired by a low ESG firm. These results suggest that the ESG performance of acquirers does not affect investors’ reaction to M&A announcements.

These findings are in line with the research carried out by McWilliams and Siegel (1999) and Halbritter and Dorfleitner (2015). The authors find that CSR has a neutral impact on the financial performance of a firm.

However, my results contradict the findings by Deng et al. (2013), who find a positive significant coefficient of CSR on the CARs of acquirers. The different results of my analysis could be due to omitted variable bias, since I do not include all control variables used by Deng et al. (2013). Also, my measure of ESG performance does not acknowledge that the number of strengths and concerns used to calculate the ESG score of acquirers may change every year. The adjusted CSR score used by Deng et al. (2013) gives equal weight to all seven dimensions of the KLD STATS ESG database.

Another limitation of my research paper is the small number of observations in my sample of M&A deals, especially in the target sample. Since I need to measure the sustainability performance of acquirers, I reject all firms that do not have a score available in the KLD STATS ESG database. Moreover, the database covers only 3000 of the largest US companies by market capitalization.

Thus a suggestion for future research is to create a more comprehensive list of firms with sustainability ratings. Another suggestion for future research is to look at the long-term consequences of sustainability. Indeed, Deng et al. (2013) found that the market does not fully value the benefits of sustainability immediately. This could be another reason why I find that the environmental, social and governance performance of acquirers does not influence the cumulative abnormal returns of shareholders. Future research on the topic should look into the firms’ returns over a longer period of time to fully understand the value of sustainability for shareholders.

(21)

References

Aktas, N., Bodt, E. D., Cousin, J. G. (2011). Do financial markets care about SRI? Evidence from mergers and acquisitions. Journal of Banking & Finance, 35(7), 1753-1761. https://doi.org/10.1016/j.jbankfin.2010.12.006

Asquith, P., Bruner, R. F., & Mullins, D. W. (1983). The gains to bidding firms from mergers. Journal of Financial Economics, 11(1-4), 121-39. http://dx.doi.org/10.1016/0304-405x(83)90007-7

Betton, S., Eckbo, B. E., & Thorburn, K. (2008). Corporate takeovers. Journal of Financial Economics, 91(2), 158-178. http://dx.doi.org/10.1016/j.jfineco.2008.02.004

Coase, R. (1937). The Nature of the Firm. Economica, 4(16), 386. http://dx.doi.org/10.2307/2626876

De Jong, F. (2007). Event Studies Methodology [pdf]. Retrieved from

https://blackboard.uva.nl/bbcswebdav/pid-6793728-dt-content-rid-5623450_1/courses/2410E001.6013B0326Y.S12.1.2013/De%20Jong%20%282007%29%2 0-%20Event%20Study%20Methodology.pdf

Deloitte, M. (2017). 2018 M&A trends annual report. Retrieved from

https://www2.deloitte.com/us/en/pages/mergers-and-acquisitions/articles/matrendsreport.html

Deng, X., Kang, J., & Low, B. S. (2013). Corporate social responsibility and stakeholder value maximization: Evidence from mergers. Journal of Financial Economics, 110(1), 87-109. https://doi.org/10.1016/j.jfineco.2013.04.014

DePamphilis, D. M. (2008). Mergers, acquisitions, and other restructuring activities (4th edition). London: Elsevier.

Dodd, P. (1980). Merger proposals, managerial discretion and stockholder wealth. Journal of Financial Economics, 8, 105-138.

Fama, E., & Kenneth, F. (1993). Common risk factors in the returns on bonds and stocks. Journal of Financial Economics, 33, 3-53.

Fatemi, A., Glaum, M., & Kaiser, S. (2017). ESG performance and firm value: The moderating role of disclosure. Global Finance Journal. http://dx.doi.org/10.1016/j.gfj.2017.03.001 Firth, M. (1980). Takeovers, shareholder returns and the theory of the firm. Quarterly Journal

(22)

Friede, G., Busch, T., & Bassen, A. (2015). ESG and financial performance: aggregated evidence from more than 2000 empirical studies. Journal Of Sustainable Finance & Investment, 5(4), 210-233. http://dx.doi.org/10.1080/20430795.2015.1118917

Halbritter, G., Dorfleitner, G. (2015). The wages of social responsibility-where are they? A critical review of ESG investing. Journal of Financial Economics, 26, 25-35.

Institute for Mergers, Acquisitions and Alliances (2018). M&A Statistics - Worldwide, Regions, Industries & Countries. Retrieved from https://imaa-institute.org/mergers-and-acquisitions-statistics/

Investment Leaders Group (2014). The value of responsible investment. University of Cambridge Institute for Sustainability Leadership.

Jensen, M. (2001). Value maximization, stakeholder theory, and the corporate objective function. Journal Of Applied Corporate Finance, 14(3), 8-21. http://dx.doi.org/10.1111/j.1745-6622.2001.tb00434.x

Jensen, M. (2002). Value maximization, stakeholder theory, and the corporate objective function. Business Ethics Quarterly, 12, 235-256.

Jensen, M., Ruback, R. (1983). The market for corporate control: The scientific evidence. Journal of Financial Economics, 11, 5-50.

Jiao, Y. (2010). Stakeholder welfare and firm value. Journal of Banking and Finance, 34, 2549–2561.

Masulis, R. W., Wang, C., & Xie, F. (2007). Corporate governance and acquirer returns. The Journal of Finance, 62(4), 1851-1889.

McWilliams, A., Siegel, D., Teoh, S. H. (1999). Issues in the use of the event study methodology: a critical analysis of corporate social responsibility studies. Organizational Research Methods, 2, 350-372.

Moeller, S., Schlingemann, F., & Stulz, R. (2004). Firm size and the gains from acquisitions.

Journal Of Financial Economics, 73(2), 201-228.

http://dx.doi.org/10.1016/j.jfineco.2003.07.002

Motis, J. (2007). Mergers and Acquisitions Motives. Toulouse School of Economics – EHESS

(GREMAQ) and University of Crete. Retrieved from

https://economics.soc.uoc.gr/wpa/docs/paper2mottis.pdf

Oler, K. D., Harrison, S. J., & Allen, R. M. (2008). The danger of misinterpreting short event- window study findings in strategic management research: An empirical illustration using horizontal acquisitions. Strategic Organization, 6, 151-184.

(23)

Rosen, J. R. (2006). Merger momentum and investor sentiment: the stock market reaction to merger announcements. The Journal of Business, 79(2), 987-1017.

Servaes, H. (1991). Tobin’s q and the gains from takeover. Journal of Finance, 46, 406-419. US SIF. (2016). Report on US Sustainable, Responsible and Impact Investing Trends.

Retrieved from https://www.ussif.org/content.asp?contentid=82

Waddock, S., & Graves, S. (1997). The corporate social performance-financial performance link. Strategic Management Journal, 18(4), 303-319.

Walley, N., & Whitehead, B. (1994). It's not easy being green. Harvard Business review, 72(3), 46-52.

(24)

Appendix

TABLE A1

Variable definitions

Variable Definitions All cash deal

(dummy) One if the deal is purely financed by cash and zero otherwise. All stock deal

(dummy) One if the deal is purely financed by stocks and zero otherwise.

CARa The three-day cumulative abnormal returns of the acquirer firm, based on the market model and in the event period (-1,1). CARt The three-day cumulative abnormal returns of the target firm, based on the market model and in the event period (-1,1).

ESG score The sum (subtraction) of all the strengths (concerns) in seven different performance areas of the KLD STATS database: environmental, human rights, diversity, community, employee relations, product and governance.

Firm size The log of the book value of total assets of the acquirer.

Free cash flow Operating income before depreciation minus interest expenses minus income taxes minus capital expenditures divided by the book value of total assets.

High tech (dummy)

One if both the acquirer and the target operate in high-tech industries and zero otherwise. Leverage The book value of long term debt and short term debt divided by the market value of assets. Related

(dummy)

One if the acquirer and the target are in the same industry and zero otherwise. Tobin's q The market value of assets divided by the book value of assets.

(25)

TABLE A2

Sensitivity analysis 1: Related (dummy) added

(1) (2)

Variable CARa CARt

ESG score 0.000 0.003 (-0.05) (0.70) Firm size -0.011** _-0.012 (-3.18) (-0.88) Tobin's q 0.001 -0.005 (0.40) (-0.41)

(-0.76) (-0.39) Leverage 0.052** _-0.118 (2.92) (-1.48) All-cash deala _0.018 _-0.043 (-1.42) (-1.35) All-stock deala _-0.009 _0.177*** (-1.28) (7.59) High techa _-0.002 _0.037 (-0.23) (1.23) Relateda _-0.005 _-0.077 (-0.83) (-1.86) Constant 0.054*** _0.182** (3.56) (2.70) Sample size 931 213 R2 _0.024 _0.341

around the announcement day of the M&A deal. The market model parameters are estimated using two hundred trading days of return data ending 11 days before the merger announcement. The method of calculation of the CARs is described in Section 3.1. ESG score is the main variable of interest and is extracted from the KLD STATS ESG database. The method of calculation of ESG score is described in Section 3.1. All variables are defined in Table A1. CAR is in percent, Firm size is in USD Mil. a _{Denotes a dummy variable. The numbers in}

(26)

TABLE A3

Sensitivity analysis 2: Merger deals announced in 2007-2009 excluded

(1) (2)

Variable CARa CARt

ESG score 0.000 -0.004 (-0.26) (-0.61) Firm size -0.009* -0.012 (-2.23) (-0.59) Tobin's q 0.004 0.018 (1.35) (0.99)

(-0.36) (-1.20) Leverage 0.057** -0.108 (2.92) (-1.11) All-cash deala _-0.017 _-0.077 (-1.19) (-1.68) All-stock deala _-0.005 _0.160*** (-0.71) (5.06) High techa _-0.001 _0.047 (-0.12) (0.91) Constant 0.087 0.031 (1.92) (1.11) Sample size 779 87 R2 _0.020 _0.326

This table presents multivariate regression analyses of announcement abnormal returns. The regressions are based on a sample of M&A deals announced in the periods 2005-2006 and 2010-2015 and extracted from the Thompson One database. The extraction criteria are presented in Section 3.2. In the acquirer sample, the cumulative abnormal return of the acquirer firm CARa is the dependent variable. In the target sample, the

cumulative abnormal return of the target firm CARt is the dependent variable. Both CARs are based on the

market model and estimated around the announcement day of the M&A deal. The market model parameters are estimated using two hundred trading days of return data ending 11 days before the merger announcement. The method of calculation of the CARs is described in Section 3.1. ESG score is the main variable of interest and is extracted from the KLD STATS ESG database. The method of calculation of ESG score is described in Section 3.1. All variables are defined in Table A1. CAR is in percent, Firm size is in USD Mil. a _{Denotes a}

(27)

TABLE A4

Sensitivity analysis 3: Event period (-10,10)

(1) (2)

Variable CARa CARt

ESG score 0.000 0.003 (0.18) (0.68) Firm size -0.012* -0.006 (-2.18) (-0.39) Tobin's q -0.001 -0.016 (-0.14) (-1.05)

Free cash flow 0.023 -0.144

(0.55) (-0.71) Leverage 0.070* -0.103 (2.56) (-1.14) All-cash deala _-0.031 _-0.049 (-1.57) (-1.34) All-stock deala _-0.006 _0.212*** (-0.61) (8.54) High techa _0.014 _0.049 (1.25) (1.47) Constant 0.047* 0.100 (2.08) (1.64) Sample size 931 220 R2 _0.015 _0.331

around the announcement day of the M&A deal. The market model parameters are estimated using two hundred trading days of return data ending 20 days before the merger announcement. The method of calculation of the CARs is described in Section 3.1. ESG score is the main variable of interest and is extracted from the KLD STATS ESG database. The method of calculation of ESG score is described in Section 3.1. All variables are defined in Table A1. CAR is in percent, Firm size is in USD Mil. a _{Denotes a dummy variable. The numbers in}

(28)

TABLE A5

Sensitivity analysis 4: Fama-French Three Factor Model used as estimation model

(1) (2)

Variable CARa CARt

ESG score 0.000 0.002 (-0.06) (0.66) Firm size -0.011** _-0.010 (-3.10) (-0.79) Tobin's q 0.001 -0.006 (0.18) (-0.48)

(-0.70) (-0.45) Leverage 0.046** -0.122 (2.66) (-1.56) All-cash deala _-0.021 _-0.045 (-1.71) (-1.43) All-stock deala _-0.007 _0.195*** (-1.09) (9.01) High techa _-0.004 _0.017 (-0.50) (0.59) Constant 0.050*** 0.107* (3.49) (2.00) Sample size 931 221 R2 _0.023 _0.337

of the target firm CARt is the dependent variable. Both CARs are based on the Fama-French Three Factor

Model and estimated around the announcement day of the M&A deal. The market model parameters are estimated using two hundred trading days of return data ending 11 days before the merger announcement. The method of calculation of the CARs is described in Section 3.1. ESG score is the main variable of interest and is extracted from the KLD STATS ESG database. The method of calculation of ESG score is described in Section 3.1. All variables are defined in Table A1. CAR is in percent, Firm size is in USD Mil. a _{Denotes a}