The ESG Premium: Investigating the Role of ESG Difference and Momentum in M&A Valuation Tobias Carlryd & Adam Karlsson Hakala Supervisor: Ted Lindblom Master’s thesis in Accounting and Financial Management Spring 2025 Graduate School, School of Business, Economics and Law, University of Gothenburg, Sweden Abstract This thesis investigates how ESG Difference and ESG Momentum influence M&A deal premiums, providing insights into the pre-merger financial implications of sustainability factors in corporate transactions. Using a sample of 158 and 304 transactions, for ESG Difference and ESG Momentum respectively, from 2003 to 2024, multivariate OLS regression models were applied to measure the relationship between ESG variables and deal premiums. The results indicate that larger ESG Difference, particularly environmental differences, have a negative effect on deal premiums, suggesting that larger difference between target and acquirer is perceived as a misalignment risk. In contrast, long-term ESG Momentum is positively associated with higher deal premiums, whereas short-term ESG Momentum had no effect. This indicates that ESG Momentum is only considered by acquirers when it is sustained over longer periods, as it implies a target’s deliberate strategy rather than superficial changes. These findings contribute to the literature by highlighting the importance of relative ESG differences and trajectory in M&A deal valuations. Keywords: Mergers and Acquisitions, ESG, Deal Premium, ESG Difference, ESG Momentum Acknowledgements We would like to express our gratitude to our supervisor, Ted Lindblom, for providing his support, guidance, and insightful feedback throughout the writing process. His expertise and encouragement have been invaluable to the completion of this thesis. In addition, we would like to thank our fellow students for their constructive feedback and engaging discussions during the development of this thesis. Gothenburg, 22nd of May 2025 Table of Contents 1. Introduction........................................................................................................................................................ 1 1.1 Problem Statement...................................................................................................................................... 2 1.2 Purpose........................................................................................................................................................ 3 1.3 Research Questions..................................................................................................................................... 3 2. Theoretical Framework & Literature Review.................................................................................................4 2.1 Theoretical Framework............................................................................................................................... 4 2.1.1 Stakeholder Theory........................................................................................................................... 4 2.1.2 Signaling Theory...............................................................................................................................6 2.1.3 Resource-Based View....................................................................................................................... 7 2.2 ESG & M&A Deal Premiums.....................................................................................................................8 2.3 ESG Difference........................................................................................................................................... 9 2.4 ESG Momentum........................................................................................................................................11 3. Methodology and Data Collection...................................................................................................................13 3.1 Research Strategy......................................................................................................................................13 3.2 Research Design........................................................................................................................................13 3.3 Data Collection..........................................................................................................................................14 3.4 Sample Selection....................................................................................................................................... 14 3.5 Control Variables.......................................................................................................................................17 3.6 Definition of Key Variables.......................................................................................................................18 3.7 OLS Regression.........................................................................................................................................19 3.8 Robustness Tests....................................................................................................................................... 20 3.8.1 Assumptions for Linear Regression................................................................................................20 3.8.2 Homoscedasticity............................................................................................................................ 20 3.8.3 Endogeneity.................................................................................................................................... 21 3.8.4 Multicollinearity..............................................................................................................................22 3.8.5 Multivariate Normality................................................................................................................... 22 3.9 Methodological Limitations...................................................................................................................... 23 3.10 Use of Artificial Intelligence...................................................................................................................23 4. Results................................................................................................................................................................24 4.1 Descriptive Statistics................................................................................................................................. 24 4.1.1 ESG Difference............................................................................................................................... 24 4.1.2 ESG Momentum............................................................................................................................. 25 4.2 ESG Difference Regression Results..........................................................................................................26 4.3 ESG Momentum Regression Results........................................................................................................ 29 4.4 Endogeneity Concerns...............................................................................................................................32 5. Analysis..............................................................................................................................................................34 5.1 ESG Difference Analysis.......................................................................................................................... 34 5.2 ESG Momentum Analysis.........................................................................................................................37 6. Conclusions & Future Research......................................................................................................................40 6.1 Conclusions............................................................................................................................................... 40 6.2 Future Research.........................................................................................................................................41 References............................................................................................................................................................. 42 Appendix............................................................................................................................................................... 47 Appendix 1: Descriptive Statistics.................................................................................................................. 47 Appendix 1.1 Descriptive Statistics ESG Difference.............................................................................. 47 Appendix 1.2 Descriptive Statistics ESG Momentum.............................................................................49 Appendix 2: Robustness Tests.........................................................................................................................50 Appendix 2.1 Breusch-Pagan Homoscedasticity Tests............................................................................50 Appendix 2.2 2SLS Results..................................................................................................................... 51 Appendix 2.3 VIF Results........................................................................................................................53 Appendix 2.4 Shapiro-Wilk W Results....................................................................................................54 Appendix 3: OLS Regressions........................................................................................................................ 55 Appendix 3.1 ESG Difference OLS Regressions.................................................................................... 55 Appendix 3.2 ESG Momentum OLS Regressions...................................................................................56 List of Figures Figure 1: Deal Premiums (%).................................................................................................................................47 Figure 2: ESG Score Comparison.......................................................................................................................... 47 Figure 3: E Score Comparison............................................................................................................................... 47 Figure 4: S Score Comparison................................................................................................................................47 Figure 5: G Score Comparison............................................................................................................................... 48 Figure 6: Market Cap (Target)................................................................................................................................48 Figure 7: MTB (Target).......................................................................................................................................... 48 Figure 8: ROE (Target)...........................................................................................................................................48 Figure 9: Leverage (Target)....................................................................................................................................48 Figure 10: Deal Premiums......................................................................................................................................49 Figure 11: ESG Momentum................................................................................................................................... 49 Figure 12: E Momentum........................................................................................................................................ 49 Figure 13: S Momentum.........................................................................................................................................49 Figure 14: G Momentum........................................................................................................................................ 49 Figure 15: Market Cap (Target)..............................................................................................................................49 Figure 16: MTB (Target)........................................................................................................................................ 50 Figure 17: ROE (Target).........................................................................................................................................50 Figure 18: Leverage (Target)..................................................................................................................................50 Figure 19: Q-Q Plot................................................................................................................................................ 54 Figure 20: KDE Plot...............................................................................................................................................54 List of Tables Table 1: ESG Difference Screening Break-Down..................................................................................................16 Table 2: ESG Momentum Screening Break-Down................................................................................................ 16 Table 3: Descriptive Statistics of ESG Difference Variables................................................................................. 25 Table 4: Descriptive Statistics of ESG Momentum Variables................................................................................26 Table 5: OLS Regression on ESG Difference........................................................................................................ 28 Table 6: OLS Regression on ESG Momentum.......................................................................................................31 Table 7: Breush-Pagan Test Results....................................................................................................................... 50 Table 8: 2SLS ESG Difference...............................................................................................................................51 Table 9: 2SLS ESG Momentum............................................................................................................................. 52 Table 10: VIF Results on ESG Difference............................................................................................................. 53 Table 11: VIF Results on ESG Momentum............................................................................................................53 Table 12: Shapiro-Wilk W Test Results................................................................................................................. 54 Table 13: OLS Regression on ESG Difference (30, 15, and 7 Day Deal Premium)..............................................55 Table 14: OLS Regression on ESG Momentum (30, 15, and 7 Day Deal Premium)............................................ 56 Table 15: OLS Regression on 2-year ESG Momentum (42 and 30 Day Deal Premium)...................................... 57 Table 16: OLS Regression on 2-year ESG Momentum (15 and 7 Day Deal Premium)........................................ 58 Table 17: OLS Regression on 3-year ESG Momentum (42 and 30 Day Deal Premium)...................................... 59 Table 18: OLS Regression on 3-year ESG Momentum (15 and 7 Day Deal Premium)........................................ 60 1. Introduction Mergers and Acquisitions (M&As) plays an integral part in the business landscape, with over 790 000 worldwide transactions, amassing a total value of over 57 trillion USD since 2000 (IMAA, n.d). Hossain (2021) mentioned several reasons to pursue M&As, such as hedging against firm specific risks, enabling changes in corporate structure, and acting as a strategic device to expand businesses to reach higher performance and growth to name a few. However, M&A activities are time-consuming, costly and risky, meaning that it is imperative to understand which factors create value or are important for successful integration between companies in order to avoid big losses. In recent decades, there has been a rising interest in companies' environmental, social, and governmental (ESG) efforts. This trend can be seen not only through increased interest by firms’ stakeholders, but also through efforts of large international bodies such as the sustainable development goals by the United Nations (United Nations, n.d), and the corporate sustainability reporting directive being introduced by the European Union (European Commission, n.d). This development has, according to Tsang et al. (2023), put pressure on companies to demonstrate that they effectively comply with these goals and regulations. As a result, nonfinancial metrics such as ESG scores have become increasingly popular as a tool to measure these efforts. ESG and corporate social responsibility (CSR) are viewed as concepts with vague boundaries and often used interchangeably. In fact, many studies use ESG databases as proxies for corporate sustainability and CSR. Chollet and Sandwidi (2018), Gomes and Marsat (2018), Krishnamurti et al. (2019), and Tampakoudis and Anagnostopoulou (2020) rely on Thomson Reuters Asset4, while Boone and Uysal (2020) and Deng et al. (2013) use the KLD database. Although this thesis focuses on ESG, references to CSR in previous studies are conceptually and empirically aligned with ESG. In the M&A landscape, ESG factors have grown into being a strategic component of deals. Companies have increasingly started to recognize that ESG factors can drive value in ways that go beyond traditional financial metrics (Deloitte, n.d). For example, Arjona et al. (2023) argued that ESG considerations have been emphasized in due diligence processes to identify and mitigate risks that could negatively affect the combined entity’s performance. Acquirers often assess whether targets comply with ESG objectives to avoid potential downgrades of the merged entity’s ESG score or costly integration efforts. Beyond risk mitigation, ESG can 1 create value through operational efficiency, improved access to capital markets, and increased revenues by overcoming regulatory barriers or attracting sustainability-conscious customers. Further, better investor perception and long-term growth potential also plays a significant role. According to Deloitte’s (n.d) trends survey, 74% of firms now evaluate potential deals based on ESG factors, a significant increase from previous years. This trend reflects the growing perception that strong ESG performance not only aligns with societal values but also enhances long-term value creation. 1.1 Problem Statement Research and practice in M&A has increasingly integrated ESG considerations into shaping deals and valuation outcomes. Strong ESG scores among the target companies have been linked to better post-merger performance, mitigation of financial risks, and higher deal premiums (Tampakoudis & Anagnostopoulou, 2020; Ozdemir et al., 2022; Zrigui et al., 2024; Chollet & Sandwidi, 2018). Still, most studies focus on the ESG score of the target company in isolation. For example, Tampakoudis and Anagnostopoulou (2020) found that acquiring firms benefits when the target’s ESG score are higher, especially when the target’s ESG practices complement or enhance the acquirer’s profile. Similarly, Ozmedir et al. (2022) demonstrated that higher ESG scores of target firms can be associated with higher deal premiums. Moreover, D’Souza et al. (2024) demonstrated how adverse ESG events prior to acquisition date exacerbate perceived risks and diminish perceived value of the transaction. Despite these insights, little attention has been paid to the effects of differences in ESG scores between the target company and the acquirer on deal premiums. The degree of alignment or misalignment between the companies ESG scores may significantly affect how synergies are perceived, risks are assessed, and future value is projected by acquirers (Boone & Uysal, 2020; Cho et al., 2021; Tampakoudis & Anagnostopoulou, 2020). This raises important questions, such as: What are the implications for deal premiums when the target has a higher ESG score than the acquirer, or vice versa? Exploring these scenarios could provide further insights into the impact ESG has on M&A and deal premiums. While much of the existing literature focuses on static ESG scores, the influence of recent changes on risk perception remains underexplored in the M&A context. ESG Momentum, defined as the rate and direction of change in a company’s ESG performance, provides the dynamic perspective which static ESG scores do not capture. For example, positive ESG Momentum may signal a commitment to improving sustainability practices, resulting in 2 improvements in the positive effects of better ESG scores. Conversely, negative ESG Momentum ought to have the opposite effect. Cauthorn et al. (2023) show the effect of ESG Momentum on investor behaviour and asset pricing, where rating changes affect the perceived risk and credibility of a company. While their study focuses on stock market reactions to the changes in ESG scores, the findings suggest that changes in ESG rating have no financial effect in the short term. However, Berg et al. (2022) found that there are significant stock-price reactions in the medium to long-term time horizon. Nonetheless, the impact of ESG Momentum remains unclear in terms of its effect on M&A deal premiums. In the context of M&A, deal premiums are tied to current and future performance (Ozdemir et al., 2022; D’Souza et al., 2024). If ESG Momentum accurately reflects a company’s sustainability trajectory, it could significantly influence the valuation strategy of targets, where a target with strong positive momentum can be viewed as a strategic asset for long-term growth. Therefore, understanding ESG Momentum provides an additional lens through which to assess target valuation, stakeholder alignment, and post-acquisition synergies. By integrating insights from ESG Differences and ESG Momentum, this thesis aims to explore the extent to which ESG affects M&A deal premiums and its implications for corporate strategy. Moreover, as most prior research in the M&A setting has treated ESG score as a static element, this paper intends to explore how the dynamic nature of ESG rating impacts deal premiums. 1.2 Purpose The purpose of this paper is to investigate how ESG Difference and Momentum influences M&A deal premiums, providing insights into their financial impact and implications for corporate strategy. 1.3 Research Questions In line with the purpose of this study, the following research questions are formulated to guide the empirical analysis: How does the ESG Difference between target and acquirer influence M&A deal premiums? How does target ESG Momentum influence M&A deal premiums? 3 2. Theoretical Framework & Literature Review 2.1 Theoretical Framework This study is built on three key theories to examine the effects of ESG Difference and ESG Momentum on M&A deal premiums: stakeholder theory, signaling theory, and resource-based view. These theories provided the theoretical basis which guided the interpretation of the results. 2.1.1 Stakeholder Theory Freeman (1984) introduced stakeholder theory as an extension of Friedman’s (1970) traditional shareholder theory, which states that corporations have a single responsibility: to maximize shareholder value. Stakeholder theory expanded on this notion by including several stakeholders such as employees, customers, suppliers, communities, and the environment. The core concept of stakeholder theory is that businesses need to take into consideration the interests of all parties affected by corporate decisions rather than focusing solely on the shareholders' financial returns. Donaldson and Preston (1995) further developed stakeholder theory by suggesting that managing stakeholders interests effectively can increase long-term profits. They also suggested three ways to understand stakeholder theory: descriptive, instrumental, and normative. The descriptive view explains how corporations need to engage multiple stakeholders in order to operate, the instrumental view examines the link between stakeholder management and corporate performance, and the normative view explains how corporations have ethical and moral reasons to engage with stakeholders. M&A transactions impact various stakeholder groups, including employees, customers, and local communities, and thus they require evaluation beyond financial synergies. Segal et al. (2020) emphasized that successful M&A outcomes do not only depend on financial and operational integration but also on managing stakeholder relationships effectively in order to reduce resistance and enhance long-term value creation. Therefore, understanding stakeholders expectations and concerns can mitigate post-merger disruptions and contribute to a smoother transition process. Stakeholder theory also plays a crucial role in relation to CSR and ESG considerations. Dmytriyev et al. (2021) showed that businesses that focus on stakeholder-oriented approaches were more likely to engage in socially responsible practices. By integrating ESG considerations into corporate strategy, firms could address stakeholder 4 concerns related to sustainability, ethical governance, and social impact which could increase the perception of corporate legitimacy. When analyzing studies on ESG’s effect on M&A deal premiums, Gomes and Marsat (2018) suggested that acquirers value targets that provide strong, stakeholder-oriented ESG practices as this reduces firm specific risks. Further, Krishnamurti et al. (2019) suggested that CSR-oriented firms use a stakeholder-oriented approach for M&A transactions where balancing financial performance with ethical considerations and long-term sustainability is important. Their argument aligns with both instrumental stakeholder theory as better stakeholder management leads to better financial performance, and normative stakeholder theory as CSR-oriented firms act in a way that considers broader stakeholder interests beyond short-term shareholder gains. Malik and Mamun (2024) found that high-CSR acquirers are willing to pay a premium for high-CSR targets, underscoring that alignment in stakeholder values is prized in M&A transactions. Moreover, from a momentum perspective, when a target's ESG rating improves, as noted by Shanaev and Ghimire (2022), it may indicate that the target is actively engaging with its stakeholders, thereby reducing future risks. Therefore, this positive momentum aligns with stakeholder theory by reflecting a trajectory toward enhanced long-term value creation. In contrast, a decline in ESG performance may trigger concerns among stakeholders, suggesting potential risks that could impact the merged entity’s value. Recent empirical evidence further suggests that the positive effects of stakeholder-oriented practices, such as ESG engagement, may be conditional on a firm's financial strength. Tsai and Wu (2022) found that improvements in CSR activities are associated with enhanced firm value, but this relationship is stronger for firms that are financially healthy. In times of limited financial resources, firms with weak profitability may struggle to translate CSR efforts into value, making stakeholders, and thereby acquirers, less likely to reward these efforts. Similarly, Zrigui et al. (2024) reported that while ESG engagement by targets is associated with lower acquisition premiums overall, this may reflect a perception that ESG initiatives are costly unless they are clearly backed by strong financial structures. Thus, ESG activities may enhance firm attractiveness and value creation in M&A contexts primarily when they are supported by strong profitability metrics. This suggests that stakeholder theory is applicable when the firm is financially stable, whereas the perspective of the shareholder theory is better suited in absence of a strong financial situation. 5 2.1.2 Signaling Theory Signaling theory was introduced by Spence (1973) and provides a framework for understanding how information can be sent out in environments where information asymmetry is present. The theory states that one can use signals to communicate valuable but otherwise unobservable attributes to another party in a credible way. In the context of labor markets, Spence (1973) showed how education could serve as a signal of productivity which in turn helped employers differentiate between high- and low-quality applicants. To ensure the credibility of the signal, there is a signaling cost as only those with the desired trait can afford to send strong signals without experiencing large costs. According to Reuer et al. (2012), signaling theory in the context of M&A transactions can help understand acquisition premiums. As acquirers often deal with information asymmetry when evaluating potential targets, these target firms can use signals to communicate their quality and reduce the perceived risk for acquirers. Beyond traditional financial indicators, firms use ESG factors to signal to external stakeholders. Zerbini (2017) explored how CSR initiatives are used as market signals by helping firms differentiate themselves. By engaging in socially responsible activities, firms can signal their commitment to ethical business practices and long-term sustainability, which may attract investors, consumers, and other key stakeholders. The credibility of these signals depends on the alignment between CSR initiatives and the firm’s core business strategy, as well as the perceived authenticity of these efforts. Similarly, Lee et al. (2022) examined how firms use ESG signals to increase their brands valuation. Their findings suggest that strong ESG performance can serve as a credible signal of financial stability, risk management capabilities, and long-term value creation. The positive association between strong target ESG performance and higher deal premiums, as found by Ozdemir et al. (2022) and Gomes and Marsat (2018), can be interpreted as the target firm sending a signal of high-quality management or ethical practices. These signals reduce uncertainty and enhance the perceived future value of the deal, thereby justifying a higher premium. Further, in the context of signaling theory, ESG Difference could serve as a signal of compatibility or misalignment between the two firms. Cho et al. (2021) found that when a target’s CSR performance exceeds that of the acquirer, the market reacts positively. This could be interpreted as a signal that the target’s superior ESG practices may drive long-term synergies for the acquirer. Contrary, Malik and Mamun (2024) suggested that high-CSR acquirers pay higher premiums for targets with strong ESG performance, 6 suggesting that alignment in ESG performance is a valuable signal in M&A transactions. Lastly, signaling theory is particularly relevant to the concept of ESG Momentum, as changes in ESG scores provide real-time information about a firm’s trajectory. Positive ESG Momentum may signal proactive management and the potential for future value creation, leading to higher premiums. Conversely, negative ESG Momentum may signal future risks, as noted by Shanaev and Ghimire (2022) and Cauthorn et al. (2023). 2.1.3 Resource-Based View The resource-based view (RBV) is a framework introduced by Barney (1991) who suggested that by developing and using valuable, rare, inimitable, and non-substitutable resources, firms can gain as well as sustain competitive advantages. Thus, the RBV shifted the previous focus from industry structure and external competitive forces to the internal capabilities of firms by suggesting that firms with unique and strategically valuable resources can outperform competitors and maintain long-term profitability. Barney (1991) identified four key characteristics that resources must possess to provide sustainable competitive advantages. First, they must be valuable, allowing firms to exploit opportunities or neutralize threats. Second, they must be rare, ensuring that few competitors possess them. Third, they must be inimitable, preventing easy replication. Fourth, they must be non-substitutable, meaning that no similar or equivalent resource can replace them. In relation to M&As, the RBV provides a critical lens through which M&A transactions can be analyzed. According to Mahoney and Pandian (1992), firms engage in M&A to acquire strategic resources that can enhance their competitive advantage. Acquirers often seek to gain access to intangible assets such as brand reputation, proprietary technologies, and managerial expertise, which are difficult to replicate organically. Mahoney and Pandian (1992) also argued that generating excess returns from resources through M&A, and thus creating a competitive advantage, is most likely when resources are complementary and can be effectively combined. From a CSR perspective, the RBV framework has been extended to incorporate sustainability and environmental concerns through the natural-resource-based view (NRBV), as proposed by Hart (1995). The NRBV argues that firms must integrate environmental and social resources into their competitive strategies to achieve long-term sustainability. Unlike traditional RBV, which focuses on economic and technological resources, Hart (1995) emphasized capabilities related to sustainable development, pollution prevention and product stewardship. When revisiting the NRBV, Hart and Dowell (2010) discussed how the 7 framework has expanded to consider how firms develop capabilities that integrate environmental sustainability into their competitive advantage. Further, their work discusses how CSR related resources contribute to competitive advantage by enhancing firm reputation, regulatory compliance, and stakeholder relationships. Thus, Hart and Dowell (2010) emphasized that sustainable resource management is becoming increasingly vital as firms navigate the complexities of climate change, regulatory requirements, and societal expectations. The RBV perspective offers a valuable framework for understanding the impact of ESG Difference and ESG Momentum on M&A deal premiums. Tampakoudis and Anagnostopoulou (2020) highlighted that acquiring a target with superior ESG performance enhances the acquirer’s ESG profile post-merger, leading to increased market valuation. This aligns with the RBV, as firms acquiring ESG-strong targets can integrate these sustainability resources to strengthen their competitive position. Further, Boone and Uysal (2020) found that differences in environmental reputation may reduce perceived synergy, reinforcing the RBV claim that resource alignment is crucial for value creation. From a momentum perspective, D’Souza et al. (2024) argued that poor ESG performance by the acquirer before acquisitions increases deal premiums due to the acquirer’s need to compensate for reputational and operational risks. This reinforces the RBV perspective that ESG capabilities serve as strategic resources, and firms with strong ESG trajectories can leverage them to enhance their market positioning and long-term performance. 2.2 ESG & M&A Deal Premiums Prior studies have examined how companies' ESG-related performance influences M&A deal premiums. Jost et al. (2022) examined if CSR performance among both targets and acquirers affect deal premiums. This was done using a multivariate OLS regression, with results showing that neither acquirer nor target CSR performance had a significant effect on deal premiums. However, the results showed that acquirers with better governance practices paid lower deal premiums, meaning that higher quality of governance impacted deal premiums negatively. These results partly contradict the findings of other studies. Krishnamurti et al. (2019) showed that acquiring firms who have a high CSR score pay lower deal premiums. With regards to targets’ CSR performance, Ozdemir et al. (2022) used OLS regression models which resulted in a significant and positive relationship between target companies 8 CSR performance and deal premiums. Thus indicating that acquirers are willing to pay a premium for good CSR performance by targets. The notion that acquirers see a value in target companies with good CSR performance is also supported by Gomes and Marsat’s (2018) study which showed a positive relationship between the variables. Further, CSR performance was split into environmental and social performance to obtain deeper analysis. Overall, the results remained positive but when controlling for domestic contra cross-border deals, only total CSR performance and environmental performance showed a positive relationship with deal premiums for both deal types. For social performance, its effect was only significantly positive for cross-border deals. However, other studies have also found evidence suggesting that higher target ESG performance results in lower deal premiums. Zrigui et al. (2024) showed that when regressing deal premiums on ESG scores (and control variables), there is a negative relationship between the two. Further, this result was consistent for both environmental and social factors when decomposing the score while the governance factor was insignificant. 2.3 ESG Difference Tampakoudis and Anagnostopoulou (2020) examined the effects of ESG Differences by focusing on how the target's ESG performance impacts acquirers post merger. Using multivariate regression models, the study found that acquisitions where the target had better ESG performance than the acquirer resulted in improved ESG scores and higher market valuations for the acquirer post merger. However, Hussain and Loureiro (2022) found that when decomposing ESG and focusing on the governance factors, bidders with stronger firm-level governance structures than their targets earned higher returns from the transaction. When focusing solely on environmental reputation, Boone and Uysal (2020) found evidence that acquirers categorized as “Green firms” lost their environmental reputation when the target had a different reputation. Further, different reputations between target and acquirer were shown to lower acquirer returns. The results were concluded to indicate that acquirers account for potential negative spillover effects when engaging with targets whose reputations may conflict with their own. 9 Cho et al. (2021) examined the relationship between CSR differences and market reactions to M&A announcements. Using cumulative abnormal returns (CAR) as their dependent variable, they found that when a target’s CSR performance was stronger than the acquirer’s, the market reacted positively, resulting in higher CARs for the target. This effect was stronger when the acquirer was well-governed, supporting the findings from Hussain and Loureiro (2022). Lastly, Malik and Mamun (2024) analyzed the interaction between acquirers' and targets' CSR performance and its impact on deal premiums. Their study categorized acquirers as “high CSR” or “low CSR” using a dummy variable approach. The results indicated that high CSR acquirers were willing to pay higher premiums for high CSR targets. Additionally, component-level analysis revealed that environmental performance had the strongest effect on acquisition premiums, suggesting that acquirers prioritize environmental performance as it can affect post-merger reputation, which is in line with Boone and Uysal (2020). Other dimensions tested by Malik and Mamun (2024), such as diversity and community-related factors, also showed positive associations with premiums, though employee-related factors had a negative effect. As discussed above, the relationship between ESG Differences and M&A transactions remains an area of active research, with prior studies highlighting the importance of ESG performance in post-merger outcomes. Studies like Tampakoudis and Anagnostopoulou (2020) and Cho et al. (2021) demonstrate that acquirers benefit from acquiring targets with superior ESG performance, which can improve post-merger ESG ratings, market valuation and cumulative abnormal returns. This suggests that firms recognize the strategic value of ESG resources and given these benefits, firms may be willing to pay higher premiums for targets with relatively stronger ESG performance. In contrast, Boone and Uysal (2020) showed that when ESG Differences create reputational risks, acquirers may adjust their payment structure to account for potential spillover effects. These findings imply that differences in ESG scores are not only relevant for post-merger integration but also influence the valuation process at the deal-making stage. Further, while studies like Malik and Mamun (2024) found that acquirers classified as high-CSR firms tend to pay higher deal premiums for targets with high CSR performance, it fails to examine the relative performance between target and acquirer. Therefore, this study hypothesizes that higher ESG Differences between target and acquirer, particularly in environmental and social dimensions, result in higher deal premiums, as acquirers perceive these attributes as valuable post-merger resources. However, as governance quality has been shown to influence acquirer returns negatively when 10 differences exist (Hussain & Loureiro, 2022), this study expects governance differences to have a negative effect on deal premiums. Based on this, our first main hypothesis is as follows: Hypothesis 1: An increase in the ESG Difference between target and acquirer is positively associated with the M&A deal premium. 2.4 ESG Momentum Shanaev and Ghimire (2022) investigated the impact of changes in ESG rating to stock performance from 2016 to 2021. The methodology employed was a calendar-time portfolio approach. They found that negative effects of downgrades were more impactful on ESG leaders than on ESG laggards, indicating that investors punish leaders when they fail to maintain high ESG standards. Additionally, Shanaev and Ghimire (2022) found that a downgrade in ESG rating resulted in negative abnormal returns in the short-term, averaging -1.2% per month. Similarly, upgrades resulted in positive abnormal returns, which were less significant and less consistent than the downgrades. Cauthorn et al. (2023) contested the findings of Shanaev and Ghimire (2022) by arguing that the reliance on calendar-time portfolio methodology introduces biases. Specifically, Cauthorn et al. (2023) mentioned that weights assigned to downgraded stocks were inconsistent, which skewed the results towards statistical significance. They also addressed the fact that ESG ratings are subject to heterogeneity problems, as substantiated by Berg et al. (2022), a limitation Shanaev and Ghimire (2022) did not address. After correcting for the methodological flaws, Cauthorn et al. (2023) concluded that changes in ESG rating have no significant short-term effects on stock performance. Berg et al. (2022) and Galema and Gerritsen (2022) did find that changes in ESG ratings had a significant medium- to long-term impact on the stock price. Although the aforementioned articles focused on the relationship between ESG ratings changes and stock performance, they did not explore how changes in ESG ratings impact M&A deals. D’Souza et al. (2024) offered a partial perspective from an M&A perspective by showing how adverse ESG events among acquiring companies prior to the acquisition increased the transaction price. This occurs because the acquiring company must compensate the target shareholder for the increased operational and reputational risks associated with the adverse ESG event. These results underline the economic cost of recent poor ESG performance and its consequent effect on M&A deals. Conversely, strong ESG performance 11 has several long-term economic benefits, including risk reduction (Chollet & Sandwidi, 2018; Lokuwaduge & Heenetigala, 2016), enhanced stock performance (Berg et al., 2022; Tsai & Wu, 2022), stronger stakeholder trust (Lokuwaduge & Heenetigala, 2016; Tsai & Wu, 2022), and improved deal premiums (Ozdemir et al., 2022; D’Souza et al., 2024). While the findings of D’Souza et al. (2024) underscored the significance of recent adverse ESG events in shaping deal valuation, it focuses predominantly on the acquirer's ESG immediate trajectory before deal completion. Thus, it leaves a gap in understanding how ESG Momentum at the target level influences deal premiums. Further, the study does not examine positive momentum effects on deal premiums. If negative ESG performance increases transaction costs due to heightened risk, then by extension, firms exhibiting positive ESG Momentum should command higher deal premiums. This is particularly relevant for environmental and social momentum, which have been linked to stronger stakeholder trust and financial performance (Berg et al., 2022; Tsai & Wu, 2022). However, governance momentum may have a weaker effect, as Shanaev and Ghimire (2022) indicate that governance structures are more rigid and less likely to generate immediate valuation shifts. Therefore, this study hypothesizes that positive ESG Momentum at the target level will be associated with higher deal premiums, particularly in the environmental and social pillars whereas governance momentum has a weaker or negligible effect. Thus, our second main hypothesis is as follows: Hypothesis 2: An increase in the target’s ESG Momentum is positively associated with the M&A deal premium. 12 3. Methodology and Data Collection 3.1 Research Strategy Considering the stated aim of the thesis and the stated hypotheses, a quantitative research strategy was deemed to be the most suitable approach. Furthermore, the usage of cross-sectional data in this study enabled measurements and statistical analyses, further solidifying the selection of the quantitative research approach, which is in line with Patel and Davidson’s (2019) methodological recommendations. This strategy allows for empirical testing of relationships between ESG factors and M&A deal premiums, ensuring objective and reproducible findings. The chosen research strategy aligns with the positivistic research paradigm, which assumes that reality exists independently of human perception and can be studied through systematic empirical observation. This approach therefore emphasizes objective measurements, hypothesis testing, and the use of empirical evidence as a basis for drawing generalizable conclusions. Moreover, the study follows a deductive approach, in which existing theories are tested using data analysis to validate or refine theoretical assumptions, in line with standard principles of quantitative research methodology (Patel & Davidson, 2019; Bryman et al., 2022). 3.2 Research Design This study aims to examine how ESG Difference and ESG Momentum impact M&A deal premiums. Following prior related studies by Ozdemir et al. (2022), Gomes and Marsat (2018), and Alexandridis et al. (2013), a multivariate regression model was used, with the natural logarithm of deal premiums as the dependent variable. Given that it is assumed that M&A deal premiums are influenced by both deal-specific and firm-specific factors, a multivariate approach is suitable for isolating the effect of our ESG variables while controlling for other determinants. The main model and the three sub-models of this study will all have the same dependent variable and use the same control variables. As this research design tests for linear relationships, potential non-linear relationships were not considered in this study. This assumption aligns with the aforementioned studies we follow. 13 3.3 Data Collection Data was primarily collected from two reputable financial databases. S&P Capital IQ Pro (Capital IQ) as well as London Stock Exchange Group (LSEG) was used to gather financial data on the company level, M&A transactions, and market performance data. The primary source of ESG data is LSEG, previously known as Eikon, as it provides standardized, comparable, and granular ESG data as well as data on the three ESG pillars across firms globally and over time. The ESG scores by LSEG (2024) are calculated using over 870 metrics and cover around 16,000 companies globally. They reflect how well a company performs on environmental, social, and governance issues relative to its industry and country peers. LSEG’s methodology involves scoring across ten ESG categories, which are weighted according to their relevance to each industry. These category scores are combined into the three pillar scores, environmental, social, and governance, which are then aggregated into an overall ESG score ranging from 0 to 100. The scores are updated weekly and aim to provide a transparent, data-driven view of a company’s ESG standing. As share prices of inactive and acquired companies were required to calculate the deal premiums, we had to manually retrieve share prices for four separate dates for each company to be able to calculate the four different deal premiums used. In many cases, currency conversions were also necessary to ensure consistency with the currency used for premium calculations. Furthermore, variables such as ROE, Market Cap, and Leverage were often missing from the LSEG screener, requiring manual collection in addition to the share prices. To ensure data reliability, a cross-referencing process was conducted on a subsample of 60 firms to compare key metrics across the two databases. In a few instances, company balance sheets were consulted directly to obtain missing financial information. 3.4 Sample Selection The sample collected for the testing of our hypotheses consists of M&A transactions that took place between January 1, 2003, and December 31, 2024, covering 22 years of transaction data. This time period was chosen to ensure the largest possible sample, as ESG scores from LSEG date back to 2002. To ensure the availability of complete transaction details, only completed transactions were included in the sample. Transactions categorized as 14 recapitalizations, self-tenders, and repurchases were excluded, as these involve the same firm acting as both the acquirer and target. Since such transactions do not provide insights into how a target firm's ESG performance influences the acquirer’s shareholder value creation, they were omitted. Additionally, spin-offs, minority stake purchases, acquisitions of remaining interest, exchange offers, and privatizations were excluded, as recommended by Alexandridis et al. (2010), to ensure that only transactions representing clear changes in corporate control were analyzed. To further refine the sample, the acquirer was required to own less than 50% of the target before the transaction and more than 50% after its completion, thereby ensuring a change in control. This criterion follows the methodology employed by Jost et al. (2022) and Gomes and Marsat (2018). As acquirers may purchase shares leading up to an acquisition, ownership was assessed six months before completion to ensure compliance with this requirement. In line with prior literature, the minimum transaction value was set at $1 million, the same approach as Alexandridis et al. (2010), ensuring that only economically significant transactions were considered. Additionally, transactions were included only if data were available for deal premiums and all control variables, allowing for the appropriate estimation of our regression models. For Hypothesis 1, which examines ESG Difference, both the target and acquirer needed to be publicly listed to ensure the availability of ESG scores. Since the ESG Difference is calculated as the target’s ESG score minus the acquirer’s ESG score, both firms required ESG scores that predate the announcement of the transaction. The earliest available annual ESG score before the announcement date was used, and data had to be available for the three individual ESG pillars. For Hypothesis 2, which investigates ESG Momentum, only the target was required to be publicly listed, as the hypothesis focuses on changes in the target’s ESG performance. To assess ESG Momentum before the transaction, ESG scores, as well as each separate pillar, had to be available for at least two years prior to the announcement date. This ensured that changes in ESG performance leading up to the acquisition could be accurately measured. As two different databases were used to collect M&A transactions, thorough cross-referencing was conducted to ensure that no duplicate entries were included in the final sample. Additionally, to maximize sample size and data availability, both LSEG and Capital IQ were used to gather stock price data and the control variables included in the regressions. 15 To mitigate the risk of inconsistencies across databases, we manually compared observations by selecting a test sample of 30 transactions from each regression (one for ESG Difference and one for ESG Momentum) where complete data was available in both databases to ensure that the reported numbers were the same. By applying these selection and validation procedures, the sample was structured to provide robust empirical evidence on the relationship between ESG performance and M&A deal premiums while aligning with methodologies established in prior research. Tables 1 and 2 present the screening process for the ESG Difference and ESG Momentum samples, consisting of 158 and 304 observations respectively. The Data Loss column shows how many transactions were excluded at each step. Observations and data loss are initially reported separately for Capital IQ and LSEG to show the filtering for each source. The Data Source column indicates where the data was gathered. In the second row, only LSEG is listed since ESG scores were retrieved solely from LSEG. However, the accompanying numbers for this row still reflect data from both sources to provide a comprehensive view of the screening process. Table 1: ESG Difference Screening Break-Down Table 2: ESG Momentum Screening Break-Down 16 3.5 Control Variables Following prior research by both Alexandridis et al. (2013) and Ozdemir et al. (2022), we included several firm-specific and deal-specific control variables in our OLS regression. The firm-specific control variables include Market-to-Book (MTB), the target’s market value compared to its book value; market capitalization (Market Cap), the natural logarithm of the target’s market capitalization; return on equity (ROE) of the target; and lastly Leverage, the ratio of the target’s long-term debt to its total assets. The deal specific control variables include a dummy variable for Cash, where the dummy code 1 signifies if the transaction is a full cash offer; Hostile, where the dummy code 1 signifies if the acquisition is a hostile offer; Strategic, where the dummy code 1 signifies if the acquisition is strategic; Competing, where the dummy code 1 signifies that there was at least one competing bidder; Tender, where the dummy code 1 signifies that the offer was a tender offer; and lastly Unsolicited, where the dummy code 1 signifies that the acquirer made an unsolicited bid. Moreover, we included an interaction term between ROE and ESG scores. This interaction captures the idea that the effect of a firm's ESG performance on acquisition premiums may depend on its financial strength. This allowed us to test whether the market rewards ESG engagement differently depending on the target’s financial health, as suggested by Tsai and Wu (2022) and Zrigui et al. (2024). To enhance the robustness of our momentum regression and address potential biases, we included a dummy variable that accounts for ESG score methodology changes by the data provider LSEG. The purpose of this dummy is to capture potential distortions in ESG Momentum calculations that could arise if changes in scoring methodology, and not actual changes in firm performance, drive year-on-year score variations. Although LSEG provides limited transparency regarding whether methodology updates are applied retroactively to historical scores, prior research by Berg et al. (2021) finds evidence that historical ESG scores in the Refinitiv dataset have been systematically rewritten following methodology changes. If scores are fully updated retroactively, momentum measures would, in theory, remain valid. However, given the uncertainty and the known ongoing adjustments to past scores, we adopted a precautionary approach. The dummy variable takes a value of 1 for any observation where the calculation of ESG Momentum spans a period that includes a known methodology change, and 0 otherwise. This design ensures that we can control for artificial changes in ESG Momentum potentially 17 caused by adjustments in the rating methodology rather than real changes in ESG performance. Moreover, since the methodology updates were spread across different years and not isolated to a single economic event (such as COVID-19), the dummy is unlikely to merely capture broad economic shifts. Instead, it specifically aims to isolate the methodological effects on ESG Momentum, preserving the validity of our results. After the completion of the data sets, it was noted that some dummy variables had few observations. These were Hostile, which had zero and three observations in each respective data set, Strategic, which had eight and eleven observations in each respective data set, and Competing, which had nine and 22 observations in each respective data set. Hostile is thus removed as a variable, whilst the other variables are kept. 3.6 Definition of Key Variables To measure the sustainability alignment between the acquiring and target firms, ESG Difference is defined as the difference in the most recent annual ESG scores at the time of deal announcement. Specifically, the following equation is used: 𝐸𝑆𝐺 = 𝐸𝑆𝐺 − 𝐸𝑆𝐺 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 𝑇𝑎𝑟𝑔𝑒𝑡 𝐴𝑐𝑞𝑢𝑖𝑟𝑒𝑟 To capture recent developments in the target’s ESG performance, ESG Momentum is defined as the relative change in ESG score over the year preceding the M&A. Following the approach of Magnani et al. (2024), the percentage change is calculated using the following expression: 𝐸𝑆𝐺 − 𝐸𝑆𝐺 𝐸𝑆𝐺 = 𝑡 𝑡−1𝐸𝑆𝐺 × 100 𝑀𝑜𝑚𝑒𝑛𝑡𝑢𝑚 𝑡−1 where t denotes the most recent available annual ESG score prior to the M&A, and t-1 refers to the score one year earlier. Deal premium represents the premium paid by the acquirer over the target’s pre-announcement share price. This study adopts the definition used by Ozdemir et al. (2022), applying the natural logarithm to reduce skewness in the distribution. The main deal premium is calculated as specified below: 𝐷𝑒𝑎𝑙 𝑝𝑟𝑒𝑚𝑖𝑢𝑚 = 𝐴𝑐𝑞𝑢𝑖𝑠𝑖𝑡𝑖𝑜𝑛 𝑝𝑟𝑖𝑐𝑒 𝑝𝑒𝑟 𝑠ℎ𝑎𝑟𝑒 − 𝑇𝑎𝑟𝑔𝑒𝑡'𝑠 𝑠𝑡𝑜𝑐𝑘 𝑝𝑟𝑖𝑐𝑒 42 𝑑𝑎𝑦𝑠 𝑏𝑒𝑓𝑜𝑟𝑒 𝑎𝑛𝑛𝑜𝑢𝑛𝑐𝑒𝑚𝑒𝑛𝑡 𝑇𝑎𝑟𝑔𝑒𝑡'𝑠 𝑠𝑡𝑜𝑐𝑘 𝑝𝑟𝑖𝑐𝑒 42 𝑑𝑎𝑦𝑠 𝑏𝑒𝑓𝑜𝑟𝑒 𝑎𝑛𝑛𝑜𝑢𝑛𝑐𝑒𝑚𝑒𝑛𝑡 18 3.7 OLS Regression As ESG Difference required both target and acquirer to be public companies as well as having available annual ESG data prior to the M&A, the amount of available data was significantly less for this variable than for ESG Momentum. To avoid unnecessary elimination of data, this study employed two main OLS regression on ESG scores in general. Furthermore, additional OLS regressions were conducted on each pillar of the ESG scores for both momentum and difference in both main OLS regressions. Lastly, the significance level employed in this study is 5%, although significance levels of 10% and 1% are also noted in the tables and discussed. Model 1: ESG Difference 𝑙𝑛(𝐷𝑒𝑎𝑙 𝑝𝑟𝑒𝑚𝑖𝑢𝑚) = β + β 𝐸𝑆𝐺 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 + β 𝑀𝑇𝐵 + β 𝑀𝑎𝑟𝑘𝑒𝑡 𝐶𝑎𝑝 0 1 𝑖 2 𝑖 3 𝑖 + β 𝑅𝑂𝐸 + β 𝑅𝑂𝐸 × 𝐸𝑆𝐺 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 + β 𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒 + β 𝐶𝑎𝑠ℎ + β 𝑆𝑡𝑟𝑎𝑡𝑒𝑔𝑖𝑐 4 𝑖 5 𝑖 𝑖 6 𝑖 7 𝑖 8 𝑖 + β 𝐶𝑜𝑚𝑝𝑒𝑡𝑖𝑛𝑔 + β 𝑇𝑒𝑛𝑑𝑒𝑟 + β 𝑈𝑛𝑠𝑜𝑙𝑖𝑐𝑖𝑡𝑒𝑑 + ϵ 9 𝑖 10 𝑖 11 𝑖 𝑖 Model 2: ESG Momentum 𝑙𝑛(𝐷𝑒𝑎𝑙 𝑝𝑟𝑒𝑚𝑖𝑢𝑚) = β + β 𝐸𝑆𝐺 𝑀𝑜𝑚𝑒𝑛𝑡𝑢𝑚 + β 𝑀𝑇𝐵 + β 𝑀𝑎𝑟𝑘𝑒𝑡 𝐶𝑎𝑝 + β 𝑅𝑂𝐸 0 1 𝑖 2 𝑖 3 𝑖 4 𝑖 + β 𝑅𝑂𝐸 × 𝐸𝑆𝐺 𝑀𝑜𝑚𝑒𝑛𝑡𝑢𝑚 + β 𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒 + β 𝐶𝑎𝑠ℎ + β 𝑆𝑡𝑟𝑎𝑡𝑒𝑔𝑖𝑐 5 𝑖 𝑖 6 𝑖 7 𝑖 8 𝑖 + β 𝐶𝑜𝑚𝑝𝑒𝑡𝑖𝑛𝑔 + β 𝑇𝑒𝑛𝑑𝑒𝑟 + β 𝑈𝑛𝑠𝑜𝑙𝑖𝑐𝑖𝑡𝑒𝑑 + β 𝑀𝑒𝑡ℎ𝑜𝑑𝑜𝑙𝑜𝑔𝑦 𝐶ℎ𝑎𝑛𝑔𝑒 + ϵ 9 𝑖 10 𝑖 11 𝑖 12 𝑖 𝑖 Each model had three sub-models, one for each ESG pillar, along with the aggregated ESG score. These were denominated as “E Difference”, “S Difference”, and “G Difference”, which refers to the difference between the target’s and the acquirer’s score in each pillar of the ESG score. Similarly, “E Momentum”, “S Momentum”, and “G Momentum” refers to the target's momentum in each ESG pillar. Furthermore, each sub-model had an interaction variable between the relevant ESG pillar and ROE. 19 3.8 Robustness Tests To ensure that the results are robust and trustworthy, we used various robustness checks. Firstly, alternative time periods for the calculation of deal premiums ensures that the regression holds regardless of timeframe. Similar to Ozdemir et al. (2022), we used target’s share prices from 30, 15, and 7 days before announcement day to complement the main timeframe of 42 days. Secondly, by fulfilling the assumptions for linear regression, and all its relevant statistical diagnostics, we ensured that our results are robust and trustworthy. The robustness tests were applied on both aggregate ESG as well as the disaggregated ESG data. 3.8.1 Assumptions for Linear Regression To ensure that the results from the study’s main models were trustworthy and reliable, the model employed needed to fulfill the following conditions (Brooks, 2019). (1) The errors have a mean of zero. (2) Homoscedasticity: The variance of the error term is constant. (3) No autocorrelation: The error terms are linearly independent of one another. (4) No endogeneity: There is no correlation between the error term and the corresponding x variate. In addition to these requirements, there was a requirement of little or no presence of multicollinearity between the independent variables in the multivariate regression as well as normality in the residuals in the multivariate model (Brooks, 2019). As the data in this study had no time component, except the target’s ESG Momentum which is observed only once, there was no need to account for the presence of autocorrelation. 3.8.2 Homoscedasticity To test for the assumption of homoscedasticity, the Breusch-Pagan test was used. The results revealed that the data is heteroscedastic. For the difference data, the chi-squared statistics were 250.11 for ESG, 135.4 for E, 254.74 for S, and 229.42 for G, with all p-values being 0.000. Similarly, for the momentum data, the chi-squared values were 72.44 for ESG, 71.34 for E, 65.15 for S, and 63.45 for G, with p-values of 0.000. See Table 7 in Appendix 2.1. Since the p-values were all below the 5% significance level, the null hypothesis of homoscedasticity was rejected, indicating the presence of heteroscedasticity in the residuals. 20 This suggests that the variance of the errors was not constant across observations, which violates one of the key assumptions of classical linear regression. To account for this, robust standard errors were employed in the main regressions. 3.8.3 Endogeneity To address potential endogeneity concerns, a two-stage least squares (2SLS) model was used. As this study draws on the methodology of Ozdemir et al. (2022), similar reasoning on sources of endogeneity and which instrument to use was employed. We predicted that positive pre-acquisition ESG Momentum and ESG Difference of a target firm leads to higher deal premiums. In other words, the causation is drawn from the ESG Momentum and ESG Difference, not the deal premiums. Therefore, reverse causality, i.e that high deal premiums lead to better ESG performance, was considered not a likely potential cause of endogeneity. Likewise, inadequate measurements of primary constructs was not considered a source of endogeneity as we followed previous studies which use widely accepted operationalization to construct our primary independent and dependent variables. Moreover, as the ESG data was collected from LSEG, a trusted third-party database, there was likely no measurement errors in these primary constructs either. The last source of endogeneity is omitted variable bias, which Ozdemir et al. (2022) argue might be a source of endogeneity in similar research designs. If ESG data is endogenous, then the OLS estimates might also be biased. When testing for endogeneity, this thesis followed Ozdemir et al. (2022) by using lagged ESG performance variables as instruments. Specifically, we used lagged ESG Difference as the instrument for the ESG Difference variable, and lagged ESG Momentum as the instrument for the ESG Momentum variable. This choice was based on the two fundamental requirements for a valid instrument: relevance and exogeneity. Regarding instrument relevance, an instrument must be strongly correlated with the endogenous regressor. In our context, lagged ESG Difference was expected to be strongly correlated with ESG Difference, since ESG performance and corporate sustainability strategies typically evolve gradually over time. Likewise, this was expected to hold for ESG Momentum as well, as changes in ESG performance often occur in sustained trends. Furthermore, the instrument must satisfy the exogeneity assumption, meaning it must be uncorrelated with the error term in the second stage. This means that it should not directly affect the dependent variable (deal premium) other than through its effect on ESG Difference 21 or Momentum. Following the reasoning by Ozdemir et al. (2022), lagged ESG performance was likely to be correlated with ESG performance while not being so with the current year deal premium. Therefore, this thesis made the assumption that this holds for ESG Difference and ESG Momentum also. The lagged ESG Difference reflects relative ESG performance from a previous period and is unlikely to directly influence the deal premium set at the time of the acquisition announcement. Likewise, lagged momentum, which captures past changes in ESG performance, predates the transaction and should not have an effect on the deal premium. While Ozdemir et al. (2022) also used industry-average CSR performance as a second instrument, we excluded this in our analysis. In the context of ESG Difference, introducing an industry average may effectively amount to reintroducing the same difference as both firms are often from the same industry. Thus, inclusion would risk redundancy and a violation of the exogeneity condition. This could lead to correlation between the instrument and the error term. Further, industry-country average ESG performance was also considered, as proposed by Gomes and Marsat (2018). However, since our data had observations from relatively small countries with few public companies in specific sectors, there were not enough observations to obtain reliable average ESG performance in some specific industry-country pairs. As a result, the lack of data coverage could have introduced noise and bias rather than mitigating endogeneity. Therefore, we only used the lagged ESG Difference and momentum as our instruments. 3.8.4 Multicollinearity The Variance Inflation Factors, or VIF test, was used to check for multicollinearity. As is common in academia, the cut-off level was a VIF test score at ten (Ozdemir et al., 2022; Gomes & Marsat, 2018). None of the variables were close to exceeding the cut-off level, as can be seen in Tables 10 and 11 in Appendix 2.3, indicating no signs of multicollinearity. Therefore, none of the chosen variables had to be discarded. 3.8.5 Multivariate Normality To ensure normality in the residuals in the multivariate model, the Shapiro-Wilk W test was employed. The results from the Shapiro-Wilk tests consistently indicated a significant deviation from normality in all cases. The test statistics ranged between 0.746 and 0.796, with 22 associated z-values above seven and all p-values equal to 0.000, leading to a rejection of the null hypothesis of normally distributed residuals at the 1% significance level. These statistical results are further supported visually with the Q-Q plots and kernel density estimates in Figures 19 and 20 as well as Table 12 in Appendix 2.4. The Q-Q plots demonstrate curvature at the ends, indicative of fat tails or skewness, while the kernel density plots show peaked and asymmetric distributions that diverge from the bell-shaped normal curve, specifically with larger tails on the left. Given this evidence of non-normality, all regressions in this study were estimated using robust standard errors. 3.9 Methodological Limitations A methodological limitation, as emphasized by Cauthorn et al. (2023), is the heterogeneous nature of ESG scores. They argued that since there are different ESG scores from different organizations who use different methodologies to calculate their scores, the results of a statistical analysis might be dependent on which ESG database is used. As we only had access to LSEG, this limitation was considered in the conclusions. Further, another potential limitation of this study comes from the use of multiple data providers for collecting stock price data and control variables. While this approach allowed us to expand the sample size, it introduced the risk of inconsistencies due to differences in how financial metrics are reported or calculated across databases. To mitigate this risk, we conducted a consistency check on a subsample of 30 observations from each regression model, confirming that the calculated values aligned across both databases. Despite this, the use of multiple sources may still have introduced measurement variability that could affect the comparability of some variables. 3.10 Use of Artificial Intelligence Artificial Intelligence (AI), more specifically ChatGPT, has only been used to generate code in Stata, give structural suggestions and clarity improvements in accordance with academic integrity guidelines. AI has not been used for analytical reasoning nor empirical analysis. This has solely been done by the authors. 23 4. Results 4.1 Descriptive Statistics 4.1.1 ESG Difference Table 3 presents the descriptive statistics for the difference variables used in the analysis of the first dataset. ESG Difference has a mean of -16.545 indicating that target ESG scores tend to be lower than the respective acquirer's ESG score. The wide range between the minimum and maximum observations (from -84.623 to 40.165) highlights large variation across the observed transactions, which is further substantiated by the standard deviation of 23.979. The three pillars of ESG show similar patterns, with negative means, high standard deviations, and consistently higher scores among the acquirers. This discrepancy in ESG scores between acquirer and target, along with the firm-specific variables, is further visualized in Figures 1 through 9 in Appendix 1.1. Firm characteristics such as MTB and Market Cap show considerable spread. For example, the MTB has a mean of 3.99 and a standard deviation of 5.537, suggesting notable differences in growth expectations or valuation among firms. Market Cap is similarly dispersed, with relatively high upper bounds at 11.016, suggesting the presence of a few large firms in the sample. ROE and Leverage have means of 0.027 and 0.177 respectively, with limited spread, indicating generally low profitability and moderate levels of debt usage across the sample. Finally, the deal premium variables measured at 42, 30, 15, and 7 days before the deal announcement show decreasing means as the window shortens, with values ranging from 0.255 to 0.306. The distribution of these premiums also displays moderate variability, as reflected by their respective standard deviations. 24 Table 3: Descriptive Statistics of ESG Difference Variables N Mean SD Min p25 Median p75 Max ESG Difference 158 -16.545 23.979 -84.623 -32.311 -13.29 0 40.165 E Difference 158 -24.92 32.145 -96.369 -45.034 -20.89 -1.005 60.553 S Difference 158 -17.36 25.990 -89.443 -32.152 -11.269 0 39.505 G Difference 158 -7.647 28.900 -80.394 -26.095 -6.102 9.362 66 Market-to-Book 158 3.99 5.537 .2 .94 1.889 4.601 31.226 Market Cap 158 7.735 1.558 3.988 6.844 7.916 8.617 11.016 ROE 158 .027 0.316 -1.562 .002 .079 .137 .868 Leverage 158 .177 0.166 .001 .048 .119 .28 .764 Deal Premium 42 158 .306 0.388 -.944 .098 .269 .534 1.567 Deal Premium 30 158 .289 0.361 -.946 .089 .266 .493 1.484 Deal Premium 15 158 .268 0.334 -.948 .078 .249 .457 .92 Deal Premium 7 158 .255 0.327 -.948 .084 .238 .419 1.026 Note: Table 3 displays the descriptive statistics for the variables in the ESG Difference analysis. The descriptive statistics consists of the mean, standard deviation, min, max, median, 25th and 75th percentile values, and the count for each variable 4.1.2 ESG Momentum Table 4 displays the descriptive statistics for the momentum variables used in the second dataset. The ESG Momentum values across three measurement periods show increasing means, from 11.465 in ESG Momentum 1, to 23.603 in ESG Momentum 2, and 31.756 in ESG Momentum 3, suggesting a general upward trend in ESG performance over time for the sample firms. This is accompanied by an increase in standard deviation, indicating growing variability in firm-level ESG changes across the longer time horizons. Minimum and maximum values across these variables further highlight the wide range in ESG performance trajectories. The ESG Momentum scores, along with the firm-specific variables, is further visualized in Figures 10 through 18 in Appendix 1.2. When broken down by ESG pillars, the E, S, and G Momentum variables reveal similar upward trends and large variability. Notably, the E Momentum 3 and S Momentum 3 variables reach maximum values exceeding 550 and 625, respectively, with standard deviations also increasing across time windows, reflecting considerable heterogeneity in firm behavior. Median values remain positive for all subcomponents, indicating that the majority of firms experience improvements in their ESG scores over time. Firm characteristics such as MTB and Market Cap maintain relatively stable values compared to the difference data. MTB has a mean of 3.279 and a standard deviation of 4.997, while Market Cap has a similar level of variability, suggesting a sample comprising both large and small firms. Profitability and financial structure, represented by ROE and Leverage, show 25 mean values of 0.015 and 0.217, respectively. The low ROE indicates generally modest profitability across firms, while Leverage appears moderate on average. Deal Premiums measured at 42, 30, 15, and 7 days before the deal announcement reveal a decreasing pattern over time windows, with mean values ranging from 0.286 (42-day) to 0.236 (7-day). This pattern suggests that price reactions are more pronounced over longer pre-announcement periods, though the spread in values indicates notable firm-specific variation. Table 4: Descriptive Statistics of ESG Momentum Variables N Mean SD Min p25 Median p75 Max ESG Momentum 1Y 304 11.465 27.196 -31.161 -4.465 4.6 19.864 106.93 ESG Momentum 2Y 249 23.603 47.209 -46.501 -3.273 9.117 40.389 218.09 ESG Momentum 3Y 203 31.756 48.820 -38.207 .91 20.002 54.186 178.683 E Momentum 1Y 249 24.477 82.962 -46.925 -4.295 3.118 20.186 528.983 E Momentum 2Y 207 39.042 92.442 -92.814 -7.048 6.303 43.441 379.601 E Momentum 3Y 162 56.222 122.261 -92.749 -6.457 11.262 56.732 556.098 S Momentum 1Y 304 13.85 38.948 -33.875 -5.631 2.268 17.886 185.494 S Momentum 2Y 249 32.459 71.189 -42.233 -6.637 7.881 41.905 321.582 S Momentum 3Y 203 48.939 112.565 -46.21 -5.979 19.694 59.356 625.006 G Momentum 1Y 304 12.503 40.007 -49.58 -10.392 3.581 27.744 138.409 G Momentum 2Y 249 30.205 86.997 -65.612 -9.577 7.518 43.322 421.067 G Momentum 3Y 203 39.974 103.733 -58.351 -10.219 12.259 47.344 533.06 Market-to-Book 304 3.279 4.997 .2 .811 1.471 3.289 31.226 Market Cap 304 7.37 1.605 3.614 6.413 7.505 8.389 10.841 ROE 304 .015 0.289 -1.038 -.036 .063 .133 .823 Leverage 304 .217 0.181 .001 .06 .169 .342 .764 Deal Premium 42 304 .286 0.449 -.895 .049 .233 .467 2.259 Deal Premium 30 304 .292 0.447 -.897 .064 .231 .476 2.259 Deal Premium 15 304 .251 0.359 -.894 .053 .215 .452 1.246 Deal Premium 7 304 .236 0.355 -.878 .065 .2 .4 1.308 Note: Table 4 displays the descriptive statistics for the variables in the ESG Momentum analysis. The descriptive statistics consists of the mean, standard deviation, min, max, median, 25th and 75th percentile values, and the count for each variable 4.2 ESG Difference Regression Results Table 5 displays the results of the ESG Difference regression with a 42 day deal premium. Table 13 in Appendix 3.1 displays the results for ESG Difference with the 30, 15, and 7 day deal premiums. Beginning with the overall ESG Difference variable, it is statistically insignificant (p > 0.1), suggesting that there is no observable systematic relationship between ESG Difference and deal premiums in this sample. Further, this result remains consistent across the alternative premium windows, reinforcing the conclusion that overall ESG Difference does not appear to be a decisive factor in determining acquisition premiums. In 26 contrast, the environmental pillar, E Difference, shows a negative and statistically significant coefficient of -0.00304 (p < 0.05). This indicates that, all else equal, for every one-point increase in the environmental score gap, the premium decreases by approximately 0.3 percentage points on average. Importantly, this result is consistent and statistically significant across all additional premium windows, supporting the robustness of the finding. However, both the S Difference and G Difference variables are statistically insignificant across all deal premium windows. Thus, differences in social or governance quality between the acquirer and target do not appear to systematically impact the premium offered. Among the interaction effects, the interaction between E Difference and ROE stands out as it is positive and statistically significant (coefficient = 0.0200, p < 0.05). Specifically, this indicates that the more profitable the target is, the weaker the negative effect of the environmental difference becomes. This interaction effect remains statistically significant across the alternative premium windows as well, adding further credibility to the robustness of this dynamic. However, the remaining interaction terms (ESG, S, and G Difference with ROE) do not reach statistical significance. This implies that the moderating role of ROE is unique to the environmental pillar in this sample, and does not extend to other ESG dimensions. Looking at the firm-specific control variables, the results show ROE as a significant and positive predictor of deal premiums in the environmental difference model, with a coefficient of 1.219 (p < 0.05). This result indicates that more profitable target firms tend to command higher acquisition premiums, a finding that is consistent across the different model specifications. In contrast, the MTB ratio, Market Cap, and Leverage show small and statistically insignificant coefficients in all models, suggesting that valuation discrepancies between market and book values, as well as firms’ size, and leverage ratios, are not systematically priced into the acquisition premium in our sample. Turning to the deal-specific control variables, most exhibit limited explanatory power. The dummies for Cash, Tender, and Unsolicited are all statistically insignificant in the 42-day model. Unsolicited approaches marginal statistical significance in shorter window regressions (particularly the 15-day model, where the coefficient is 0.115), suggesting that acquirers may offer slightly higher premiums in unsolicited deals. However, this finding is not consistent across the different deal windows or ESG pillars and is only statistically significant at the 10% level. Further, the strategic- and competing offer dummies are not statistically significant. This may be due to the small number of strategic and competing deals in the dataset, which reduces statistical power. The overall fit of the models, R-squared, ranges from 27 0.124 to 0.2462. The E Difference model has the highest R-squared at 0.246, explaining nearly 25% of the variation in the 42-day model. The F-statistics range between 0.689 and 1.363, indicating that the models do not explain a significant portion of the variation. Table 5: OLS Regression on ESG Difference (1) (2) (3) (4) ESG E S G ESG Difference -0.00157 (0.00161) E Difference -0.00304** (0.00136) S Difference -0.00182 (0.00153) G Difference 0.00139 (0.00115) Return on Equity 0.706 1.219** 0.649 0.224 (0.520) (0.579) (0.472) (0.278) ESG x ROE 0.0131 (0.00968) E x ROE 0.0200** (0.00881) S x ROE 0.0113 (0.00803) G x ROE -0.00587 (0.00750) Market-to-Book 0.00640 0.00816 0.00831 -0.000574 (0.0104) (0.00991) (0.0115) (0.00846) Market Cap -0.00233 -0.0510 -0.0104 0.0543 (0.0665) (0.0691) (0.0676) (0.0522) Leverage 0.0437 -0.0199 0.00502 0.139 (0.184) (0.167) (0.187) (0.233) Cash 0.0874 0.0436 0.0906 0.108 (0.0810) (0.0727) (0.0852) (0.0875) Strategic -0.359 -0.313 -0.373 -0.413 (0.391) (0.390) (0.387) (0.391) Competing 0.175 0.118 0.217 0.183 (0.140) (0.107) (0.165) (0.149) Tender -0.0232 -0.0448 -0.0440 -0.0626 (0.0972) (0.0934) (0.0976) (0.0882) Unsolicited 0.102 0.110 0.0876 0.0860 (0.0709) (0.0666) (0.0693) (0.0672) Constant 0.145 0.0677 0.152 0.172 (0.350) (0.322) (0.333) (0.344) Observations 158 158 158 158 R-Squared 0.136 0.246 0.139 0.124 F-Statistic 0.806 1.363 0.880 0.689 Note: Table 5 displays the main OLS regression for the ESG Difference at a 42 day deal premium. This table presents the coefficients and standard errors from four independent regressions. (1) represents the regression with the aggregated ESG Difference along with the control variables. (2), (3), and (4) have Environmental, Social, and Governance Differences, along with the same control variables. Standard errors in parentheses * p < 0.10, ** p < 0.05, *** p < 0.01 28 4.3 ESG Momentum Regression Results Table 6 displays the results of the main ESG Momentum regression using the 42-day deal premium. The regression results for the 30-, 15-, and 7-day deal premiums are displayed in Table 14 in Appendix 3.2. Starting with the 1-year momentum regressions, the results show that none of the momentum variables, ESG, environmental, social, nor governance, are statistically significant in the 42-day model. The coefficients are generally small in scale, and their standard errors are large relative to the point estimates. The R-squared for the models ranges from 0.1247 to 0.1683, suggesting limited explanatory power. The F-statistics ranges between 1.360 and 1.447, implying that the models do not explain a significant portion of the variation in deal premiums. Notably, these patterns hold across the alternative windows of 30, 15, and 7 days, where the significance and magnitude of the momentum variables remain unchanged, reinforcing the result of no robust relationship at the 1-year period. Turning to the 2-year momentum regressions, a more nuanced picture emerges. The coefficient on overall ESG Momentum remains statistically insignificant in the 42-day model, but two of the individual ESG pillars show statistically significant effects. Specifically, environmental momentum has a positive and statistically significant coefficient (0.000559, p < 0.05), as well as governance momentum (0.000477, p < 0.05). These findings suggest that improvements in a firm’s environmental and governance performance over a 2-year period may be associated with higher deal premiums. These effects appear economically small in magnitude but are consistent across the all premium windows, where both environmental and governance momentum remain positive and significant at the 5% level. The R-squared values for the 42-day models in this group range from 0.0359 to 0.0720, indicating a decrease in explanatory power relative to the 1-year momentum regressions. The F-statistics ranges between 1.50 and 2.23 which are higher than those from the 1-year momentum regressions. However, it still indicates that the models do not explain a significant portion of the variation in deal premiums. While the overall model fit remains limited, the results for environmental and governance momentum are consistent and statistically robust. The results for the 2-year momentum regressions are displayed in Tables 15 and 16 in Appendix 3.2. The 3-year momentum regressions further reinforce the positive association between environmental momentum and deal premiums. In the 42-day window, the coefficient on environmental momentum is 0.000390 and significant at the 5% level. However, neither the ESG, social nor the governance momentum variables are significant in this model. The 29 statistical significance of environmental momentum is robust across the shorter time windows (30, 15, and 7 days), with consistent positive coefficients in the range of 0.000321 to 0.000390 and p-values below 0.05. The R-squared values in the 3-year models remain in line with the previous results, generally falling between 0.03 and 0.11 across different premium windows and specifications, confirming the modest fit of the models. The F-statistics for the 3-year momentum model ranges from 1.50 to 2.56, indicating that the models do not jointly explain a significant portion of the impact on deal premiums. The results for the 3-year momentum regressions are displayed in Tables 17 and 18 in Appendix 3.2. The interaction terms between ROE and the main ESG Momentum variables are statistically insignificant in the 42-day models. However, the interaction effect between environmental momentum and ROE in the 2-year regression had weak statistical significance. Specifically, in the 42-day deal premium model, the interaction term E × ROE has a coefficient of -0.00321 and is marginally significant at the 10% level (p < 0.1). Similarly, in the 7-day premium regression, the interaction term is -0.00241 and also marginally significant at the 10% level. While these results do not meet the 5% threshold, they could suggest that the positive effect of environmental momentum on deal premiums may be moderated by the target firm's profitability. More precisely, for firms with higher ROE, the marginal impact of environmental improvements on deal premium appears to be reduced. Although this pattern does not appear consistently across all models, its marginal statistical significance in two different premium windows lends possibility to the interaction effect and could suggest a potentially interesting dynamic between financial and sustainability performance in M&A contexts. The control variables across all momentum models show some consistent patterns. ROE itself is generally positive but mostly insignificant, except in a few cases such as the 2-year and 3-year Environmental momentum regressions, where ROE becomes significant at the 10% level. MTB and Market Cap show inconsistent signs and are statistically insignificant in most regressions. Leverage consistently shows positive but insignificant effects. Among the deal characteristics, cash offers are positively associated with premiums and are statistically significant at the 10% level in several models, especially in shorter time windows. Competing bids also show some positive and significant effects in the 3-year regressions. Other dummy variables, such as Strategic, Tender, and Unsolicited, display no consistent pattern of significance across specifications. Nonetheless, Competing and Strategic had limited observations, which limits their explanatory power. 30 Table 6: OLS Regression on ESG Momentum (1) (2) (3) (4) ESG E S G ESG Momentum (1-Year) -0.00101 (0.00126) Environmental Momentum (1-Year) 0.000386 (0.000415) Social Momentum (1-Year) 0.000211 (0.000452) Governance Momentum (1-Year) -0.000826 (0.000853) Return on Equity 0.0576 0.230 0.0609 0.00577 (0.185) (0.198) (0.183) (0.170) ESG × ROE -0.00496 (0.00456) E × ROE -0.00251 (0.00207) S × ROE -0.00250 (0.00279) G × ROE -0.00131 (0.00285) Market-to-Book -0.00580 0.0000198 -0.00541 -0.00720 (0.00676) (0.00939) (0.00656) (0.00701) Market Cap 0.0438 0.00836 0.0411 0.0521 (0.0390) (0.0429) (0.0372) (0.0398) Leverage 0.0623 0.0437 0.0716 0.0715 (0.123) (0.129) (0.122) (0.119) Cash 0.100 0.120 0.0910 0.102 (0.0673) (0.0736) (0.0679) (0.0679) Strategic -0.0416 -0.0682 -0.0326 -0.0465 (0.0723) (0.103) (0.0689) (0.0718) Competing 0.100 0.141 0.0849 0.0905 (0.0745) (0.108) (0.0746) (0.0710) Tender 0.0373 0.000334 0.0500 0.0425 (0.0512) (0.0548) (0.0531) (0.0496) Unsolicited 0.0931* 0.0810 0.0888* 0.0896* (0.0495) (0.0525) (0.0482) (0.0490) Method Change (1Y) -0.0287 -0.0116 -0.0268 -0.0267 (0.0685) (0.0830) (0.0676) (0.0673) Constant 0.236 0.151 0.226 0.218 (0.213) (0.237) (0.213) (0.214) Observations 304 249 304 304 R-Squared 0.0654 0.0708 0.0605 0.0605 F-Statistic 1.447 1.360 1.428 1.424 Note: Table 6 displays the main OLS regression for the ESG Momentum at a 42 day deal premium. This table presents the coefficients and standard errors from four independent regressions. (1) represents the regression with the aggregated ESG Momentum along with the control variables. (2), (3), and (4) have Environmental, Social, and Governance Momentums, along with the same control variables. Standard errors in parentheses * p < 0.10, ** p < 0.05, *** p < 0.01 31 4.4 Endogeneity Concerns The results of the 2SLS regressions for ESG Difference and Momentum are respectively displayed in Tables 8 and 9 in Appendix 2.2. Beginning with the ESG Difference variables, the 2SLS results largely confirm the direction and significance of the OLS estimates. The coefficient on environmental difference remains negative and statistically significant at the 1% level (-0.004, p < 0.01), reinforcing the finding that larger environmental gaps between target and acquirer are associated with lower acquisition premiums. This strengthens the credibility of the earlier OLS results by addressing potential reverse causality or omitted variable bias. Notably, the magnitude of the coefficient is slightly larger in the 2SLS model, suggesting that the OLS estimate may be weakened due to measurement error or other sources of endogeneity. The overall ESG Difference becomes statistically significant in the 2SLS model (-0.004, p < 0.05), unlike in the OLS where it was insignificant. This shift implies that once endogeneity is addressed, ESG Difference may exert a modest negative influence on deal premiums. For the social and governance differences, the 2SLS coefficients remain statistically insignificant, consistent with the OLS findings. This suggests that for these two ESG pillars, endogeneity concerns are unlikely to be the main reason behind the lack of association with premiums. The interaction terms between ESG dimensions and ROE were also tested in the 2SLS setting. As in the OLS models, the interaction between environmental difference and ROE remains positive and strongly significant (0.027, p < 0.01), while other interaction terms remain statistically insignificant, echoing earlier results. In the case of ESG Momentum, the 2SLS regressions generally reinforce the earlier OLS findings that ESG Momentum variables do not show a consistent nor statistically significant relationship with acquisition premiums. Across all four ESG dimensions the coefficients remain statistically insignificant in the 2SLS regressions, mirroring the OLS outcomes in sign and significance. Even though the instruments used are statistically strong in the first stage, the absence of significant second-stage effects suggests that short-term changes prior to the acquisition in ESG performance are not priced into M&A premiums in this sample. This alignment between OLS and 2SLS results supports the robustness of the initial finding, that ESG Momentum over a one-year period does not appear to play a meaningful role in deal valuation, even after addressing endogeneity concerns. 32 One caveat arises in the case of the E Momentum instrument, as it does not pass conventional thresholds for instrument strength. As a result, the 2SLS estimates for 1-year environmental momentum are interpreted with caution. Overall, the broader 2SLS results reinforce the credibility of the main OLS findings: ESG Difference, particularly in the environmental pillar, appears robustly associated with acquisition premiums, while short-term ESG Momentum changes do not exhibit a meaningful impact on deal valuation, even after addressing endogeneity concerns. 33 5. Analysis 5.1 ESG Difference Analysis The first hypothesis of this study proposed that greater ESG Differences between target and acquirer would be positively associated with deal premiums. However, the empirical findings reject this hypothesis. While ESG Difference was insignificant in the OLS models, it became significantly negative in the 2SLS model. Further, E Difference was consistently negative and statistically significant for both the OLS and 2SLS model, while social and governance differences remained insignificant. These results indicate that when the target’s ESG score, particularly in the environmental dimension, exceeds the acquirer’s, the deal premium decreases. Conversely, this means that when the acquirer has a higher ESG score than the target, premiums increase. This directional nature of the variable means that the interpretation can be analyzed from both perspectives. Thus, the analysis that follows discusses the implications of both scenarios to capture the implications of directionality in ESG Differences and provide a more complete interpretation of the regression results. The findings seem to stand in contrast to those of Boone and Uysal (2020), who showed that environmentally strong acquirers experience reputational losses and weaker post-merger returns when acquiring environmentally weaker targets. In our case, however, such deals are associated with higher premiums, suggesting that acquirers do not perceive the target’s weaker environmental performance as a reputational risk at the negotiation stage. One potential explanation for this divergence is the timing of the measurements. Boone and Uysal’s (2020) study focused on short-term market reactions using CARs calculated within a narrow event window around the announcement. In contrast, our study captures the negotiated premium before the deal. Thus, while stakeholders may initially respond negatively to this type of environmental misalignment, acquirers may see longer-term opportunities by for example improving the target’s ESG practices post-acquisition. This ties into the instrumental view of stakeholder theory described by Donaldson and Preston (1995), which suggests that firms can improve financial performance by managing stakeholder relationships. While Boone and Uysal (2020) showed that acquiring targets with weaker ESG performance can cause short-term concerns, our findings suggest that acquirers may have a long-term perspective. Specifically, they may anticipate long-term value from improving the target’s ESG practices after the acquisition and thus strengthen relationships with both the 34 target’s and their own stakeholders. This proactive stakeholder management aligns with the instrumental view, as it links ESG upgrading with improved corporate performance. This could justify the higher premiums observed in our results despite the initial reputational risks. Hussain and Loureiro (2022) found evidence which suggested that bidders earn higher post merger returns when they themselves have higher governance than targets. Their reasoning was that acquirers can effectively influence the target's governance performance in a positive direction after the acquisition. While our governance variable was insignificant, a similar reasoning could explain the results with respect to ESG and environmental performance. Acquirers may feel confident in being able to influence their targets positively in the long run, and thus are willing to pay a higher premium. However, the results by Boone and Uysal (2020) could indicate that stakeholders are more cautious when it comes to environmental influence between acquirer and target. This concern with integration and stakeholder response also ties into stakeholder theory. Segal et al. (2020) highlighted that successful M&A outcomes depend not only on operational and financial integration, but also on effectively managing stakeholders’ expectations. If ESG Differences are high and the acquirer lacks alignment with the target’s stakeholders, the perceived risk of reputational or integration issues increases. In such cases, acquirers may hedge against this risk by offering lower premiums. Therefore, by extension, when the acquirer is stronger and sees the opportunity to raise the target’s ESG profile, stakeholder concerns may be reduced, leading to higher premiums. Krishnamurti et al. (2019) also argued that CSR-oriented acquirers adopt a stakeholder-oriented approach to M&A, reinforcing that acquirers may not simply reward ESG strength, but rather look for manageable integration to deal with stakeholder trust. A more nuanced perspective emerges when considering the interaction between environmental difference and the target’s financial strength. While we find that higher environmental scores in the target relative to the acquirer are generally associated with lower premiums, this negative relationship is moderated by the target’s financial performance. Specifically, the significant positive interaction between environmental difference and the target’s ROE suggests that financially strong targets receive less of a discount for having stronger environmental scores than the acquirer. In other words, acquirers may be more willing to overlook ESG misalignment when the target has the financial capacity to support integration and sustain ESG initiatives. This finding complements the argument by Tampakoudis and Anagnostopoulou (2020) that the benefits of acquiring ESG-strong targets depend on successful integration. Our results suggest that financial strength may be a critical 35 enabler of such integration. Similarly, Tsai and Wu (2022) emphasize that CSR initiatives are more likely to yield value in firms with robust financial foundations, and Zrigui et al. (2024) note that ESG engagement is costly and requires sufficient financial resources. Taken together, these findings highlight that the perceived risks of ESG misalignment can be offset by strong financials. Acquirers may therefore view financially sound targets as better positioned to realize ESG-related synergies, even when the targets’ environmental scores exceed those of the acquirer. Similarly to Tampakoudis and Anagnostopoulou (2020), Cho et al. (2021) also found evidence of positive market reactions when targets have stronger ESG performance. However, their study focused on targets’ short-term CARs before the acquisition, showing results that these increased for targets with better ESG performance, although this does not necessarily translate into higher premiums from the acquirer. As discussed above, our findings suggest that acquirers may perceive integration challenges. According to signaling theory, as described by Reuer et al. (2012), ESG performance can help reduce information asymmetry by signaling firm quality. Yet, as Zerbini (2017) argues, the effectiveness of these signals depends on their credibility and compatibility with the firm’s core business strategy. If the target’s ESG strategy diverges too far from the acquirer’s, the signal may be seen as an obstacle by the acquirer. This in turn may lead to lower premiums despite the positive market reactions. This highlights the important distinction that while external stakeholders and markets may respond positively to strong ESG signals from the target, the acquirer may see obstacles instead. As emphasized by Gomes and Marsat (2018), acquirers have deeper operational insights as they conduct thorough due diligence and are more focused on firm specific risks. Thus, they can assess whether ESG strengths are transferable and value-adding in the specific deal context, or if there may be anticipated costs or challenges in leveraging the targets’ ESG strength. If the result is the latter, the acquirer may assign a lower premium to such targets to avoid overpaying, explaining the divergence between public market reactions and negotiated deal premiums. When comparing our results to Malik and Mamun (2024), an interesting distinction emerges. While their study finds that high-CSR acquirers pay higher premiums for high-CSR targets, it does not account for the degree or direction of ESG alignment. Therefore, our results extend this by demonstrating that directionality matters. It is not only the acquirer’s and target’s ESG strengths that is rewarded, but rather the acquirer’s ability to steer ESG integration which may only be possible if the target is over a certain threshold. This aligns well with the RBV, 36 particularly Mahoney and Pandian (1992) who implied that integration ability is as important as resource complementarity in value creation. From this perspective, high-CSR targets may only be attractive at a premium when they fall within a range that acquirers believe they can manage or enhance. The findings by Malik and Mamun (2024) combined with the insights from Mahoney and Pandian (1992) and our results highlight the importance of both resource complementarity and integration capability. Further, as discussed before, Tampakoudis and Anagnostopoulou (2020) support this view by emphasizing that the benefits of ESG-related acquisitions are dependent on the success of integration efforts. Thus, reinforcing the RBV idea that synergies are only realized when capabilities can be effectively combined. 5.2 ESG Momentum Analysis The second hypothesis of this study suggested that positive ESG Momentum at the target level would be associated with higher M&A deal premiums. This expectation was grounded in stakeholder theory, signaling theory, and the RBV. Contrary to our second hypothesis, the empirical results provide partial evidence for this hypothesis. Beginning with the 1-year momentum regressions, no statistically significant relationships were observed between overall ESG Momentum, environmental momentum, social momentum, or governance momentum and M&A deal premiums. The coefficients were small and statistically insignificant across all premium windows, and this pattern persisted in the 2SLS regressions, reinforcing the conclusion that short-term changes in ESG performance prior to acquisition are not systematically priced into deal premiums. Our findings contrast with the results of Shanaev and Ghimire (2022), who found significant short-term effects of ESG rating changes on stock performance. Instead, the lack of short-term significance in our analysis is more consistent with the analysis presented by Cauthorn et al. (2023), who found no short-term effects from changes in ESG ratings. While Shanaev and Ghimire (2022) and Cauthorn et al. (2023) examined stock price reactions, this study focuses on deal premiums in M&A transactions, making the outcomes comparable, although they relate to different financial outcomes. Several factors may explain the lack of significance. As shown in the descriptive statistics, the median ESG Momentum across the sample is positive, indicating that a majority of firms improve their ESG scores over time. If ESG Momentum had a uniform and linear impact on deal premiums, it would be expected that deal premiums would be systematically higher each year across the board, which is not observed. This suggests that minor ESG improvements, 37 although common, may not be sufficiently meaningful or credible to influence acquirers’ valuation decisions. It is plausible that only substantial or high-quality short-term ESG improvements act as effective signals of enhanced firm value, which implies the existence of a threshold effect that is not currently captured. Second, the measurement of ESG Momentum as a percentage change introduces potential distortions, particularly for firms with very low initial ESG scores. For those firms, even slight absolute improvements can translate into very large percentage increases, potentially overstating the perceived ESG Momentum. When examining longer momentum horizons, a more nuanced picture emerges. Environmental and governance momentum showed positive and statistically significant relationships with deal premiums over the 2-year window, and environmental momentum remained significant at the 3-year window. These findings suggest that sustained ESG improvements, particularly environmental improvements, are recognized by acquirers and factored into deal valuations. These results are consistent with the NRBV, which highlights the strategic value of hard-to-replicate capabilities like environmental stewardship (Hart, 1995; Hart & Dowell, 2010), and with signaling theory, where credible, sustained ESG improvements act as signals of proactive and reliable management (Spence, 1973; Zerbini, 2017). Furthermore, our results are consistent with longer-term economic benefits such as improved deal premiums (Ozdemir et al., 2022; D’Souza et al., 2024), and possibly improved risk reduction (Chollet & Sandwidi, 2018; Shanaev & Ghimire, 2022), and stronger stakeholder trust (Lokuwaduge & Heenetigala, 2016; Tsai & Wu, 2022). Social momentum did not have significant effects at any period, suggesting that social factors may be less important to acquirers or slower to translate into measurable firm value. This finding is partially consistent with the stakeholder theory (Freeman, 1984; Donaldson & Preston, 1995), which recognizes that corporate responsibility includes, beyond maximizing shareholder value, the wellbeing of employees, customers, suppliers, communities, and the environment. Malik and Mamun (2024) further observe that while social and community-related CSR aspects can positively influence premiums, their effects are generally weaker compared to environmental aspects. This could be an explanation to why there is no significance in social momentum while there is significant results in environmental momentum. Similarly, governance momentum's significance at the 2-year horizon but not at the 3-year horizon indicates that the valuation impact of governance improvements may be more fragile or dependent on timing. This is in line with the findings by Hussain and Loureiro (2022), which emphasized the long-term importance of governance 38 alignment for post-merger success. It also aligns with Shanaev and Ghimire (2022), who claimed that governance structures are relatively rigid and slow to evolve, which reflects the slower pace at which governance quality is perceived in M&A transactions. These results raise the question of why ESG Momentum is insignificant over a one-year period, but becomes significant over two- and three-year periods. A potential explanation could be that the contrast between the non-significant short-term momentum results and the significant longer-term momentum effects imply that ESG improvements might need time to be perceived as credible and strategically valuable by acquirers, as argued by Cauthorn et al. (2023) and confirmed by both Berg et al. (2022) and Galema and Gerritsen (2022). In the short term, changes in ESG performance can be viewed as volatile, superficial, or subject to reporting adjustments, which limits their influence on deal valuations. In contrast, sustained improvements over two or three years may signal strategic commitment to ESG principles, aligning more closely with the RBV and signaling frameworks. From a signaling perspective, acquirers may interpret consistent, multi-year ESG improvements as costly and credible signals of superior management quality (Spence, 1973; Zerbini, 2017). From the perspective of RBV, sustained environmental progress can be seen as an indicator of durable, competitive capabilities (Hart, 1995; Hart & Dowell, 2010). Thus, the results imply that time plays a critical role in transforming ESG Momentum from a weak signal into a valuable strategic asset recognized in M&A pricing. Further, as suggested by Gomes and Marsat (2018), stakeholder-oriented ESG practices may reduce firm specific risks. Nonetheless, as our results indicate, this only becomes significant if it is a sustained practice in the prospective company, further supporting the signaling perspective by Spence (1973) and Zerbini (2017). The interaction variable between ROE and ESG Momentum revealed that the positive impact of the target firm’s environmental momentum might be moderated by its profitability. The negative interaction term suggests that for highly profitable firms, the marginal benefit of ESG environmental improvements on deal premiums diminishes. This aligns with signaling theory, indicating that strong financial performance may already serve as a sufficient signal of firm quality, reducing the incremental signaling value of ESG improvements. This finding is consistent with Tsai and Wu (2022), who argue that the financial benefits of CSR improvements are more pronounced for firms with stronger financial health, but may taper off when profitability is already high. Similarly, Zrigui et al. (2024) suggest that while ESG engagement can create value, its impact on acquisition premiums may be perceived as costly or redundant unless backed by financial strength within the company. 39 6. Conclusions & Future Research 6.1 Conclusions This study set out to investigate how ESG Difference and ESG Momentum influence M&A deal premiums, providing insights into their financial impact and strategic implications. Specifically, we sought to answer two research questions: (1) How does the ESG Difference between target and acquirer influence M&A deal premiums? and (2) How does target ESG Momentum influence M&A deal premiums. The empirical findings reveal important distinctions between ESG Difference and ESG Momentum. ESG Difference, particularly in the environmental dimension, was found to be negatively associated with acquisition premiums, suggesting that acquirers discount deals where targets have stronger ESG profiles than themselves while rewarding deals where their own ESG score exceeds the target’s. This indicates that upward ESG gaps may introduce friction, whereas downward ESG gaps may be perceived as opportunities for improvement or value extraction. In contrast, positive ESG Momentum at the target level over longer periods was positively associated with deal premiums, indicating that sustained improvements in ESG performance are recognized and valued by acquirers. These opposing effects can be understood through the perspectives of strategic fit, stakeholder expectations, and integration challenges. From an ESG Difference perspective, even if the target’s ESG practices are strong, the acquirer may perceive this as a source of uncertainty regarding stakeholder expectations or when it comes to successful integration. Conversely, when the acquirer has a stronger ESG profile, the integration may appear more manageable, and the acquirer may anticipate adding value through its superior ESG capabilities. This asymmetric dynamic reflects concerns around stakeholder management, signaling theory, and the challenges of integrating resources. ESG Momentum reflects a different signal. Positive and sustained ESG improvements, especially over two to three years, may signal to acquirers that the target has a proactive and adaptive management capable of continuous improvement and development. Rather than introducing misalignment risks, positive momentum suggests dynamism and resilience, qualities that are attractive in a M&A context and which might enable integration. Moreover, because momentum reflects changes within the target rather than gaps between firms, it does not necessarily threaten the acquirer's existing ESG positioning or stakeholder relationships. 40 Instead, it offers the potential to import a valuable trajectory of growth and improvement, consistent with signaling theory and the RBV's emphasis on dynamic capabilities. These findings carry important implications for dealmakers. Acquirers should carefully assess ESG differences early in the M&A process to anticipate potential integration challenges or stakeholder misalignment. Meanwhile, targets with a track record of ESG improvements may be better positioned to command higher premiums by communicating their progress as a signal of management quality and strategic value. Nonetheless, it is important to acknowledge a potential methodological limitation. Previous research has highlighted the heterogeneity of ESG scores across different rating providers (Cauthorn et al., 2023), raising concerns about the consistency and comparability of ESG metrics between sources. This study relied on ESG data from a single provider due to access constraints, which may influence the generalizability of the results. To conclude, the results show that ESG factors are priced in M&A transactions not merely based on their absolute levels, but based on their strategic alignment and trajectory. Acquirers reward firms that demonstrate credible, sustained ESG improvements, but penalize firms where target ESG profiles exceed acquirers, fearing integration risks. 6.2 Future Research While this study provides new insights into the relationship between ESG Difference, ESG Momentum, and M&A deal premiums, there are several avenues for future research that could expand and deepen the understanding of these dynamics. First, the reliance on a single ESG database presents a potential limitation. Previous research has emphasized the heterogeneity of ESG scores across different providers (Cauthorn et al., 2023), suggesting that results can vary depending on the database used. Future studies could strengthen the robustness and generalizability of our findings by validating our results using multiple ESG data providers. Second, as this study is limited by existing data coverage, ongoing improvements in ESG reporting, data collection, and database expansion may allow future researchers to access a broader range of M&A transactions. A larger dataset could enable subgroup analyses, particularly across different industries, regions, and timeframes. Finally, while this study examined ESG Momentum in general, future research could explore the existence of a significant threshold or cut-off level of ESG Momentum beyond which the effect on M&A premiums becomes economically meaningful in the short-term. 41 References Alexandridis, G., Fuller, K-P., Terhaar, L., & Travlos, N-G. (2013). Deal size, acquisition premia and shareholder gains. Journal of Corporate Finance, 20, 1-13. https://doi.org/10.1016/j.jcorpfin.2012.10.006 Arjona, L., Herfurth, J., Nielsen, J., Barrett, C., Cornet, G., & Degen, D. (2023). Capitalizing on ESG synergies in M&A. Boston Consulting Group. https://www.bcg.com/publications/2023/capitalizing-on-esg-synergies-in-mergers-and-acquis itions Barney, J. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17(1), 99-120. https://doi.org/10.1177/014920639101700108 Berg, F., Fabisik, K., & Sautner, Z. (2021). Is history repeating itself? The (un) predictable past of ESG ratings. SSRN Electronic Journal, 1-59. https://dx.doi.org/10.2139/ssrn.3722087 Berg, F., Heeb, F., & Kölbel, J. (2022). The Economic Impact of ESG Ratings. SAFE Working Paper No. 439. http://dx.doi.org/10.2139/ssrn.4088545 Boone, A., & Uysal, V. B. (2020). Reputational concerns in the market for corporate control. Journal of Corporate Finance, 61, 101399. https://doi.org/10.1016/j.jcorpfin.2018.08.010 Bryman, A., Bell, E., & Harley, B. (2022). Företagsekonomiska forskningsmetoder. Liber Cauthorn, T., Dumrose, M., Eckert, J., Klein, C., & Zwergel, B. (2023). Rating changes revisited: New evidence on short-term ESG momentum. Financial Research Letters, 54, 103703. https://doi.org/10.1016/j.frl.2023.103703 Cho, K., Han, S. H., Kim, H. J., & Kim, S. (2021). The valuation effects of corporate social responsibility on mergers and acquisitions: Evidence from U.S. target firms. Corporate Social Responsibility and Environmental Management, 28(1), 378-388. https://doi.org/10.1002/csr.2055 Chollet, P., & Sandwidi, B. W. (2018). CSR engagement and financial risk: A virtuous circle? International evidence. Global Finance Journal, 38, 65-81. https://doi.org/10.1016/j.gfj.2018.03.004 42 Deloitte. (n.d.). The increasingly vital role of ESG in M&A. Deloitte. https://www2.deloitte.com/us/en/pages/mergers-and-acquisitions/articles/role-of-esg-in-deals. html Dmytriyev, S. D., Freeman, R. E., & Hörisch, J. (2021). The Relationship between Stakeholder Theory and Corporate Social Responsibility: Differences, Similarities, and Implications for Social Issues in Management. Journal of Management Studies, 58(6), 1441-1470. https://doi.org/10.1111/joms.12684 Donaldson, T., & Preston, L. E. (1995). The Stakeholder Theory of the Corporation: Concepts, Evidence, and Implications. The Academy of Management Review, 20(1), 65-91. https://doi.org/10.2307/258887 D’Souza, R., Ho, C. Y., & Yang, J. W. (2024). The cost of corporate social irresponsibility for acquirers. Journal of Banking and Finance, 162. https://doi.org/10.1016/j.jbankfin.2024.107132 European Commission. (n.d.). Corporate sustainability reporting. European Commission. https://finance.ec.europa.eu/capital-markets-union-and-financial-markets/company-reporting- and-auditing/company-reporting/corporate-sustainability-reporting_en Freeman, E. R. (1984). Strategic Management: A Stakeholder Approach. Cambridge University Press. Friedman, M. (1970). The Social Responsibility of Business Is to Increase Its Profits. Corporate Ethics and Corporate Governance, 173-178. https://doi.org/10.1007/978-3-540-70818-6_14 Galema, R., & Gerritsen, D. (2022). ESG Rating Score Revisions and Stock Returns. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.4218969 Gomes, M., & Marsat, S. (2018). Does CSR impact premiums in M&A transactions?. Finance Research Letters, 26, 71-80. https://doi.org/10.1016/j.frl.2017.12.005 Hart, S. L. (1995). A natural-resource-based view of the firm. The Academy of Management Review, 20(4), 986-1014. https://doi.org/10.2307/258963 43 Hart, S. L., & Dowell, G. (2010). Invited editorial: A natural-resource-based view of the firm: Fifteen years after. Journal of Management, 37(5), 1464-1479. https://doi.org/10.1177/0149206310390219 Hossain, M. S. (2021). Merger & acquisitions (M&As) as an important strategic vehicle in business: Thematic areas, research avenues, and possible suggestions. Journal of Economics and Business, 116, 106004. https://doi.org/10.1016/j.jeconbus.2020.106004 Hussain, T., & Loureiro, G. (2022). Portability of firm corporate governance in mergers and acquisitions. Research in International Business and Finance, 63, 10177. https://doi.org/10.1016/j.ribaf.2022.101777 IMAA. (n.d.). M&A Statistics. https://imaa-institute.org/mergers-and-acquisitions-statistics/ Jost, S., Erben, S., Ottenstein, P., & Zülch, H. (2022). Does corporate social responsibility impact mergers & acquisition premia? New international evidence. Finance Research Letters, 46, 102237. https://doi.org/10.1016/j.frl.2021.102237 Krishnamurti, C., Shams, S., Pensiero, D., & Velayutham, E. (2019). Socially responsible firms and mergers and acquisitions performance: Australian evidence. Pacific-Basin Finance Journal, 57, 101193. https://doi.org/10.1016/j.pacfin.2019.101193 Lee, M. T., Raschke, R. L., & Krishen, A. S. (2022). Signaling green! firm ESG signals in an interconnected environment that promote brand valuation. Journal of Business Research, 138, 1-11. https://doi.org/10.1016/j.jbusres.2021.08.061 Lokuwaduge, C. S. D. S., & Heenetigala, H. (2016). Integrating Environmental, Social and Governance (ESG) Disclosure for a Sustainable Development: An Australian Study. Business Strategy and the Environment, 26(4), 438-450. https://doi-org.ezproxy.ub.gu.se/10.1002/bse.1927 LSEG. (2024). Environmental, Social and Governance scores from LSEG - October 2024. https://www.lseg.com/content/dam/data-analytics/en_us/documents/methodology/lseg-esg-sc ores-methodology.pdf 44 Magnani, M., Guidolin, M., & Berk, I. (2024). Strong vs. stable: the impact of ESG ratings momentum and their volatility on the cost of equity capital. Journal of Asset Management, 25, 666-699. https://doi.org/10.1057/s41260-024-00377-w Mahoney, J. T., & Pandian, J. R. (1992). The resource-based view within the conversation of strategic management. Strategic Management Journal, 13(5), 363-380. https://doi.org/10.1002/smj.4250130505 Malik, M., & Mamun, M. A. (2024). Impact of target firm’s social performance on acquisition premiums. Journal of Contemporary Accounting & Economics, 20(2). https://doi.org/10.1016/j.jcae.2024.100417 Ozdemir, O., Binesh, F., & Erkmen, E. (2022). The effect of target’s CSR performance on M&A deal premiums: a case for service firms. Review of Managerial Science, 16, 1001-1034. https://doi.org/10.1007/s11846-021-00471-y Patel, R., & Davidson, B. (2019). Forskningsmetodikens grunder: Att planera, genomföra och rapportera en undersökning (Fifth edition). Studentlitteratur. Reuer, J. J., Tong, T. W., & Wu, C. (2012). A Signaling Theory Of Acquisition Premiums: Evidence From IPO Targets. Academy of Management Journal, 55, 667-683. http://dx.doi.org/10.5465/amj.2010.0259 Segal, S., Guthrie, J., & Dumay, J. (2020). Stakeholder and merger and acquisition research: a structured literature review. Accounting & Finance, 61(2), 2935-2964. https://doi.org/10.1111/acfi.12688 Shanaev, S., & Ghimire, B. (2022). When ESG meets AAA: the effect of ESG rating changes on stock returns. Financial Research Letters, 46, 102302. https://doi.org/10.1016/j.frl.2021.102302 Spence, M. (1973). Job Market Signaling. The Quarterly Journal of Economics, 87(3), 355-374. https://doi.org/10.2307/1882010 45 Tampakoudis, I., & Anagnostopoulou, E. (2020). The effect of mergers and acquisitions on environmental, social and governance performance and market value: Evidence from EU acquirers. Business Strategy and the Environment, 29(5), 1865-1875. https://doi-org.ezproxy.ub.gu.se/10.1002/bse.2475 Tsai, H-J., & Wu, Y. (2022). Changes in Corporate Social Responsibility and Stock Performance. Journal of Business Ethics, 178(3), 735-755. https://doi.org/10.1007/s10551-021-04772-w Tsang, A., Frost, T., & Cao, H. (2023). Environmental, social, and governance (ESG) disclosure: A literature review. The British Accounting Review, 55(1), 101149. https://doi.org/10.1016/j.bar.2022.101149 United Nations. (n.d.). The 17 goals. United Nations Department of Economic and Social Affairs Sustainable Development. https://sdgs.un.org/goals Zerbini, F. (2017). CSR Initiatives as Market Signals: A Review and Research Agenda. Journal of Business Ethics, 146, 1-23. https://doi.org/10.1007/s10551-015-2922-8 Zrigui, M., Khanchel, I., & Lassoued, N. (2024). Does environmental, social and governance performance affect acquisitions premium? Review of International Business and Strategy, 34(4), 469-494 https://doi.org/10.1108/RIBS-07-2023-0076 46 Appendix Appendix 1: Descriptive Statistics Appendix 1.1 Descriptive Statistics ESG Difference Figure 1: Deal Premiums (%) Figure 2: ESG Score Comparison Figure 3: E Score Comparison Figure 4: S Score Comparison 47 Figure 5: G Score Comparison Figure 6: Market Cap (Target) Figure 7: MTB (Target) Figure 8: ROE (Target) Figure 9: Leverage (Target) 48 Appendix 1.2 Descriptive Statistics ESG Momentum Figure 10: Deal Premiums Figure 11: ESG Momentum Figure 12: E Momentum Figure 13: S Momentum Figure 14: G Momentum Figure 15: Market Cap (Target) 49 Figure 16: MTB (Target) Figure 17: ROE (Target) Figure 18: Leverage (Target) Appendix 2: Robustness Tests Appendix 2.1 Breusch-Pagan Homoscedasticity Tests Table 7: Breush-Pagan Test Results Difference Momentum ESG E S G ESG E S G chi2 250.11 135.4 254.74 229.42 72.44 71.34 65.15 63.45 Prob > chi2 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Note: Table 7 displays the Breush-Pagan test results for ESG Difference and Momentum at a 42 day deal premium, where a significant test result indicated the presence of heteroscedastic residuals. Standard errors in parentheses * p < 0.10, ** p < 0.05, *** p < 0.01 50 Appendix 2.2 2SLS Results Table 8: 2SLS ESG Difference (1) (2) (3) (4) (5) (6) (7) (8) ESG 1st ESG 2nd E 1st E 2nd S 1st S 2nd G 1st G 2nd ESG Diff Lag 0.824*** (0.052) E Diff Lag 0.856*** (0.050) S Diff Lag 0.818*** (0.063) G Diff Lag 0.770*** (0.056) ESG Difference -0.004** (0.002) E Difference -0.004*** (0.001) S Difference -0.003 (0.002) G Difference 0.000 (0.002) ESG x ROE -0.042 0.017 (0.109) (0.010) E x ROE 0.220** 0.027*** (0.100) (0.007) S x ROE -0.109 0.013 (0.151) (0.008) G x ROE -0.178 -0.007 (0.221) (0.008) Return on Equity 1.197 0.984 13.454*** 1.707*** -0.794 0.842 -3.242 0.332 (4.018) (0.625) (4.407) (0.500) (5.571) (0.561) (6.559) (0.373) Market-to-Book 0.024 -0.005 0.062 -0.001 -0.003 -0.005 -0.150 -0.013 (0.261) (0.016) (0.382) (0.014) (0.274) (0.016) (0.475) (0.016) Market Cap -1.857 -0.027 -2.975 -0.095 -1.107 -0.013 -1.518 0.057 (1.518) (0.072) (2.087) (0.071) (1.883) (0.073) (2.453) (0.066) Leverage -10.065* 0.116 -8.006 0.068 -9.249 0.079 -16.846 0.297 (5.600) (0.274) (5.926) (0.210) (6.590) (0.283) (10.253) (0.334) Cash -0.861 0.119 -0.692 0.047 -3.406 0.134 2.124 0.172** (2.108) (0.080) (2.841) (0.070) (2.560) (0.086) (3.507) (0.087) Strategic 9.686** 0.092 7.642 0.132 3.554 0.050 17.816*** 0.024 (4.643) (0.128) (6.261) (0.113) (6.374) (0.122) (6.017) (0.117) Competing -1.413 0.184 -5.793 0.064 3.524 0.264 -4.977 0.223 (3.776) (0.190) (3.554) (0.137) (4.618) (0.228) (5.603) (0.194) Tender 6.023** 0.063 3.894 0.030 5.541* 0.029 8.676** -0.004 (2.351) (0.108) (3.120) (0.095) (2.837) (0.111) (4.110) (0.084) Unsolicited 2.825 0.082 1.146 0.097 -0.109 0.058 5.157 0.072 (2.307) (0.072) (2.726) (0.067) (2.846) (0.068) (3.384) (0.070) Constant -8.075 0.040 -20.989*** -0.015 -6.679 0.047 -1.518 0.014 (5.263) (0.442) (7.928) (0.375) (6.801) (0.434) (7.836) (0.465) Observations 115 115 115 115 115 115 115 115 Note: Table 8 displays the 2SLS regression for the ESG Difference at a 42 day deal premium. This table presents the coefficients and standard errors from four 2SLS regressions. (1) represents the regression with the aggregated ESG Difference and the instrument variable, along with the control variables. (2), (3), and (4) have Environmental, Social, and Governance Difference and their respective instrument variables (IV), along with the same control variables. Standard errors in parentheses * p < 0.10, ** p < 0.05, *** p < 0.01 51 Table 9: 2SLS ESG Momentum (1) (2) (3) (4) (5) (6) (7) (8) ESG 1st ESG 2nd E 1st E 2nd S 1st S 2nd G 1st G 2nd ESG Momentum IV -0.038** (0.017) E Momentum IV -0.029 (0.043) S Momentum IV -0.060*** (0.022) G Momentum IV -0.022*** (0.008) ESG Momentum (1-Year) -0.001 (0.005) Environmental Momentum (1-Year) -0.016 (0.023) Social Momentum (1-Year) 0.004 (0.003) Governance Momentum (1-Year) -0.004 (0.003) ESG × ROE -0.269 -0.006 (0.427) (0.005) E × ROE 1.981** 0.029 (0.914) (0.051) S × ROE -0.579* -0.001 (0.349) (0.003) G × ROE -0.463 -0.003 (0.375) (0.003) Return on Equity 0.437 0.083 -16.205 0.050 2.781 0.092 -2.735 0.007 (5.684) (0.200) (22.665) (0.556) (6.320) (0.199) (8.301) (0.190) Market-to-Book -0.158 -0.006 -0.479 -0.012 -0.010 -0.006 -1.107** -0.010 (0.566) (0.007) (1.558) (0.022) (0.921) (0.008) (0.528) (0.008) Market Cap 1.592 0.054 6.434 0.133 0.588 0.049 6.049** 0.080* (2.092) (0.041) (6.078) (0.144) (3.224) (0.040) (2.602) (0.046) Leverage 2.709 0.121 -30.564 -0.421 -7.027 0.160 33.231** 0.213 (10.456) (0.137) (18.787) (0.802) (15.301) (0.158) (13.680) (0.144) Cash 5.972 0.126* 17.369* 0.426 3.126 0.099 9.371 0.149* (4.360) (0.075) (9.923) (0.354) (6.494) (0.080) (5.958) (0.081) Strategic 4.058 -0.063 9.519 0.015 4.708 -0.083 -3.491 -0.083 (8.534) (0.087) (17.168) (0.381) (11.091) (0.074) (7.679) (0.095) Competing 7.771 0.145 50.106 1.043 6.843 0.106 12.298 0.176* (8.439) (0.101) (46.672) (1.069) (8.239) (0.105) (12.817) (0.098) Tender -5.896* 0.036 -19.730 -0.331 -4.844 0.075 -7.515 0.022 (3.524) (0.055) (12.179) (0.393) (4.912) (0.059) (5.532) (0.055) Unsolicited 7.203* 0.078 -2.841 0.002 12.899** 0.030 9.092* 0.095 (4.042) (0.068) (7.000) (0.127) (6.386) (0.068) (5.420) (0.067) Method Change (1Y) -6.762 -0.058 -5.084 -0.106 -10.016 -0.022 -4.171 -0.061 (5.960) (0.094) (10.648) (0.210) (8.707) (0.094) (7.822) (0.089) Constant 10.894 0.182 -8.592 -0.061 30.908* 0.091 -5.304 0.132 (10.570) (0.244) (45.990) (0.795) (16.280) (0.247) (13.432) (0.255) Observations 249 249 204 204 249 249 249 249 Note: Table 9 displays the 2SLS regression for the ESG Momentum at a 42 day deal premium. This table presents the coefficients and standard errors from four 2SLS regressions. (1) represents the regression with the aggregated ESG Momentum and the instrument variable, along with the control variables. (2), (3), and (4) have Environmental, Social, and Governance Momentum and their respective instrument variables (IV), along with the same control variables. Standard errors in parentheses * p < 0.10, ** p < 0.05, *** p < 0.01 52 Appendix 2.3 VIF Results Table 10: VIF Results on ESG Difference Table 11: VIF Results on ESG Momentum VIF Difference ESG E S G VIF Momentum ESG E S G ESG Difference 1.28 - - - ESG Momentum 1.03 - - - E Difference - 1.32 - - E Momentum - 1.47 - - S Difference - - 1.25 - S Momentum - - 1.06 - G Difference - - - 1.15 G Momentum - - - 1.07 Return on Equity 3.92 3.87 3.27 1.51 Return on Equity 1.37 1.28 1.44 1.21 ESG x ROE 3.43 - - - ESG x ROE 1.20 - - - E x ROE - 3.46 - - E x ROE - 1.55 - - S x ROE - - 2.83 - S x ROE - - 1.30 - G x ROE - - - 1.15 G x ROE - - - 1.08 Market-to-Book Ratio 2.74 2.66 2.84 2.63 Market-to-Book Ratio 2.68 2.38 2.68 2.70 Market Cap 6.79 6.80 6.78 6.19 Market Cap 4.98 5.52 5.00 4.97 Equity 6.63 6.69 6.55 6.06 Equity 4.85 5.07 4.84 4.86 Leverage Ratio 1.08 1.06 1.10 1.06 Leverage Ratio 1.07 1.09 1.08 1.08 Cash 1.34 1.37 1.39 1.34 Cash 1.19 1.29 1.19 1.21 Strategic 1.13 1.12 1.12 1.14 Strategic 1.08 1.06 1.08 1.09 Competing 1.07 1.07 1.08 1.07 Competing 1.07 1.11 1.06 1.07 Tender 1.28 1.28 1.28 1.32 Tender 1.14 1.18 1.13 1.14 Unsolicited 1.22 1.22 1.24 1.21 Unsolicited 1.12 1.10 1.14 1.12 Methodology Change (1Y) 1.66 1.62 1.65 1.66 Methodology Change (2Y) 2.30 2.28 2.30 2.32 Methodology Change (3Y) 1.70 1.68 1.70 1.70 Note: Tables 10 and 11 displays the VIF results for ESG Difference (Table 10) and ESG Momentum (Table 11) with the 42 day deal premium as the dependent variable. The threshold where multicollinearity is suspected is at a value of 10. 53 Appendix 2.4 Shapiro-Wilk W Results The Shapiro-Wilk W test was performed on all four models on ESG Difference, and all four on ESG Momentum. The results were similar across all eight tests. Therefore, only the visual plots for ESG Difference are shown. Figure 19: Q-Q Plot Figure 20: KDE Plot Table 12: Shapiro-Wilk W Test Results Difference Momentum ESG E S G ESG E S G Observations 158 158 158 158 304 249 304 304 W 0.74651 0.78258 0.76105 0.76108 0.79094 0.79682 0.78637 0.78551 V 30.840 26.451 29.071 29.067 45.066 36.720 46.051 46.236 z 7.794 7.445 7.660 7.660 8.945 8.382 8.995 9.005 Prob > z 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 54 Appendix 3: OLS Regressions Appendix 3.1 ESG Difference OLS Regressions Table 13: OLS Regression on ESG Difference (30, 15, and 7 Day Deal Premium) (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) ESG_30 E_30 S_30 G_30 ESG_15 E_15 S_15 G_15 ESG_7 E_7 S_7 G_7 ESG -0.00137 -0.00151 -0.00115 Difference (0.00157) (0.00149) (0.00143) E Difference -0.00287** -0.00292** -0.00256** (0.00136) (0.00132) (0.00128) S Difference -0.00147 -0.00169 -0.00129 (0.00154) (0.00143) (0.00138) G Difference 0.00131 0.00127 0.00123 (0.00111) (0.00107) (0.00105) ROE 0.700 1.208** 0.645 0.276 0.798 1.222** 0.769* 0.355 0.762 1.169** 0.730* 0.364 (0.515) (0.585) (0.468) (0.276) (0.493) (0.573) (0.442) (0.264) (0.474) (0.572) (0.430) (0.257) ESG x ROE 0.0110 0.0125 0.0113 (0.00947) (0.00911) (0.00868) E x ROE 0.0184** 0.0175** 0.0164* (0.00885) (0.00869) (0.00865) S x ROE 0.00926 0.0117 0.0104 (0.00785) (0.00750) (0.00719) G x ROE -0.00635 -0.00425 -0.00369 (0.00746) (0.00704) (0.00662) MTB 0.00682 0.00892 0.00824 0.000612 0.00998 0.0114 0.0123 0.00359 0.0105 0.0119 0.0125 0.00482 (0.0104) (0.00979) (0.0114) (0.00833) (0.0106) (0.0103) (0.0115) (0.00876) (0.0102) (0.0101) (0.0112) (0.00852) Market Cap -0.0141 -0.0630 -0.0193 0.0356 -0.0326 -0.0752 -0.0428 0.0206 -0.0310 -0.0719 -0.0400 0.0159 (0.0651) (0.0680) (0.0663) (0.0502) (0.0643) (0.0677) (0.0654) (0.0505) (0.0612) (0.0662) (0.0628) (0.0483) Leverage -0.0109 -0.0754 -0.0397 0.0728 0.0318 -0.0223 -0.00817 0.121 0.101 0.0469 0.0683 0.179 (0.176) (0.158) (0.179) (0.226) (0.172) (0.157) (0.175) (0.221) (0.175) (0.162) (0.177) (0.218) Cash 0.0989 0.0567 0.102 0.118 0.0997 0.0602 0.104 0.117 0.119 0.0822 0.125 0.133 (0.0790) (0.0712) (0.0839) (0.0866) (0.0756) (0.0673) (0.0793) (0.0841) (0.0728) (0.0648) (0.0765) (0.0812) Strategic -0.371 -0.327 -0.384 -0.420 -0.328 -0.289 -0.340 -0.379 -0.365 -0.327 -0.375 -0.411 (0.394) (0.393) (0.390) (0.395) (0.403) (0.401) (0.399) (0.403) (0.394) (0.393) (0.390) (0.395) Competing 0.180 0.127 0.214 0.187 0.141 0.0910 0.183 0.147 0.143 0.0962 0.179 0.149 (0.137) (0.103) (0.159) (0.146) (0.0937) (0.0861) (0.119) (0.105) (0.0885) (0.0818) (0.112) (0.101) Tender -0.0342 -0.0532 -0.0516 -0.0729 -0.0194 -0.0396 -0.0401 -0.0519 -0.0489 -0.0670 -0.0676 -0.0779 (0.0972) (0.0942) (0.0976) (0.0878) (0.0940) (0.0919) (0.0939) (0.0860) (0.0903) (0.0886) (0.0904) (0.0839) Unsolicited 0.101 0.109* 0.0895 0.0862 0.115* 0.120* 0.103 0.102 0.0956 0.100 0.0852 0.0828 (0.0683) (0.0642) (0.0668) (0.0644) (0.0672) (0.0635) (0.0657) (0.0642) (0.0652) (0.0624) (0.0642) (0.0619) Constant 0.170 0.0968 0.177 0.193 0.147 0.0763 0.157 0.176 0.137 0.0695 0.146 0.161 (0.348) (0.321) (0.333) (0.338) (0.337) (0.312) (0.319) (0.331) (0.319) (0.296) (0.303) (0.313) Observations 158 158 158 158 158 158 158 158 158 158 158 158 R-Squared 0.144 0.246 0.145 0.140 0.163 0.252 0.169 0.147 0.173 0.254 0.176 0.160 F-Statistic 0.833 1.295 0.907 0.736 0.916 1.177 1.050 0.787 0.789 1.103 0.920 0.832 Note: Table 13 displays the main OLS regression for the ESG Difference at 30, 15, and 7 day deal premium. This table presents the coefficients and standard errors from four independent regressions. Columns (1) to (4) represent the 30 day deal premium regression with the aggregated ESG Difference, disaggregated ESG pillars, along with the control variables. Similarly, columns (5) to (8) represent the 15 day deal premium regression, and columns (9) to (12) represent the 7 day deal premium regression. * p < 0.10, ** p < 0.05, *** p < 0.01 55 Appendix 3.2 ESG Momentum OLS Regressions Table 14: OLS Regression on ESG Momentum (30, 15, and 7 Day Deal Premium) (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) ESG_30 E_30 S_30 G_30 ESG_15 E_15 S_15 G_15 ESG_7 E_7 S_7 G_7 ESG -0.000942 -0.000623 -0.000662 Momentum (0.00126) (0.00124) (0.00116) E Momentum 0.000341 0.000366 0.000399 (0.000404) (0.000352) (0.000338) S Momentum 0.000255 0.000294 0.000201 (0.000432) (0.000405) (0.000374) G Momentum -0.000671 -0.000292 -0.000354 (0.000867) (0.000844) (0.000799) ROE 0.100 0.236 0.105 0.0476 0.170 0.299* 0.152 0.132 0.165 0.267 0.149 0.123 (0.178) (0.196) (0.177) (0.165) (0.163) (0.178) (0.163) (0.147) (0.155) (0.171) (0.153) (0.140) ESG × ROE -0.00509 -0.00506 -0.00506 (0.00454) (0.00385) (0.00378) E × ROE -0.00220 -0.00166 -0.00215 (0.00206) (0.00178) (0.00167) S × ROE -0.00264 -0.00165 -0.00175 (0.00273) (0.00230) (0.00227) G × ROE -0.00135 -0.00303 -0.00251 (0.00283) (0.00241) (0.00224) MTB -0.00759 -0.00139 -0.00718 -0.00883 -0.00127 0.00538 -0.00119 -0.00231 0.000161 0.00537 0.000266 -0.000885 (0.00645) (0.00913) (0.00630) (0.00679) (0.00585) (0.00845) (0.00586) (0.00599) (0.00543) (0.00798) (0.00538) (0.00557) Market Cap 0.0385 0.00393 0.0357 0.0462 0.00201 -0.0346 0.00226 0.00697 -0.00269 -0.0333 -0.00256 0.00289 (0.0379) (0.0426) (0.0362) (0.0390) (0.0326) (0.0370) (0.0318) (0.0325) (0.0311) (0.0352) (0.0302) (0.0312) Leverage 0.0989 0.0703 0.109 0.105 0.133 0.150 0.138 0.129 0.125 0.126 0.128 0.122 (0.121) (0.128) (0.121) (0.117) (0.111) (0.116) (0.111) (0.108) (0.108) (0.114) (0.108) (0.106) Cash 0.117* 0.133* 0.108 0.118* 0.116* 0.129* 0.108* 0.119* 0.123** 0.140** 0.116* 0.125** (0.0667) (0.0730) (0.0674) (0.0673) (0.0634) (0.0704) (0.0641) (0.0635) (0.0603) (0.0677) (0.0610) (0.0606) Strategic -0.0314 -0.0503 -0.0223 -0.0345 0.0259 -0.00626 0.0326 0.0247 0.00137 -0.0273 0.00834 -0.00000444 (0.0635) (0.0853) (0.0615) (0.0633) (0.0655) (0.0890) (0.0633) (0.0651) (0.0632) (0.0841) (0.0608) (0.0624) Competing 0.0903 0.137 0.0750 0.0799 0.0709 0.0745 0.0550 0.0679 0.0442 0.0363 0.0289 0.0390 (0.0724) (0.102) (0.0719) (0.0689) (0.0509) (0.0716) (0.0500) (0.0507) (0.0490) (0.0685) (0.0482) (0.0485) Tender 0.0335 0.00707 0.0465 0.0390 0.0353 0.0122 0.0473 0.0362 0.0209 -0.00304 0.0324 0.0232 (0.0511) (0.0549) (0.0530) (0.0497) (0.0476) (0.0521) (0.0493) (0.0468) (0.0452) (0.0501) (0.0468) (0.0446) Unsolicited 0.0926* 0.0900* 0.0883* 0.0883* 0.0839* 0.0758 0.0785* 0.0779* 0.0728 0.0705 0.0683 0.0673 (0.0483) (0.0517) (0.0470) (0.0478) (0.0475) (0.0511) (0.0460) (0.0468) (0.0453) (0.0493) (0.0438) (0.0446) Method -0.0290 -0.00829 -0.0271 -0.0273 -0.0536 -0.0522 -0.0529 -0.0504 -0.0387 -0.0304 -0.0381 -0.0361 Change (1Y) (0.0671) (0.0830) (0.0661) (0.0663) (0.0620) (0.0762) (0.0612) (0.0618) (0.0606) (0.0744) (0.0596) (0.0601) Constant 0.308 0.204 0.299 0.290 0.235 0.171 0.222 0.219 0.249 0.185 0.239 0.233 (0.210) (0.233) (0.209) (0.210) (0.199) (0.223) (0.199) (0.198) (0.192) (0.215) (0.192) (0.191) Observations 304 249 304 304 304 249 304 304 304 249 304 304 R-Squared 0.075 0.078 0.071 0.068 0.069 0.086 0.062 0.065 0.071 0.083 0.064 0.065 F-Statistic 1.886 1.711 1.901 1.877 1.522 1.473 1.508 1.529 1.689 1.655 1.662 1.668 Note: Table 14 displays the main OLS regression for the ESG Momentum at 30, 15, and 7 day deal premium. This table presents the coefficients and standard errors from four independent regressions. Columns (1) to (4) represent the 30 day deal premium regression with the aggregated ESG Momentum, disaggregated ESG pillars, along with the control variables. Similarly, columns (5) to (8) represent the 15 day deal premium regression, and columns (9) to (12) represent the 7 day deal premium regression. * p < 0.10, ** p < 0.05, *** p < 0.01 56 Table 15: OLS Regression on 2-year ESG Momentum (42 and 30 Day Deal Premium) (1) (2) (3) (4) (5) (6) (7) (8) ESG_42 E_42 S_42 G_42 ESG_30 E_30 S_30 G_30 ESG 0.000533 0.000527 Momentum (0.000532) (0.000557) (2-Year) Environmental 0.000559** 0.000551** Momentum (0.000258) (0.000254) (2-Year) Social -0.000163 -0.000163 Momentum (0.000315) (0.000316) (2-Year) Governance 0.000477** 0.000453** Momentum (0.000226) (0.000215) (2-Year) ROE 0.0815 0.310 0.0936 0.0236 0.131 0.318 0.146 0.0719 (0.201) (0.228) (0.205) (0.185) (0.193) (0.221) (0.198) (0.179) ESG × ROE -0.00554 -0.00568 (0.00468) (0.00468) E × ROE -0.00321* -0.00284 (0.00177) (0.00179) S × ROE -0.00297 -0.00317 (0.00305) (0.00301) G × ROE -0.00122 -0.00131 (0.00309) (0.00305) Market-to-Book -0.00577 -0.00382 -0.00616 -0.00590 -0.00869 -0.00461 -0.00906 -0.00888 (0.00765) (0.00991) (0.00753) (0.00772) (0.00733) (0.00949) (0.00722) (0.00744) Market Cap 0.0534 0.0283 0.0515 0.0563 0.0510 0.0231 0.0488 0.0542 (0.0411) (0.0456) (0.0398) (0.0405) (0.0398) (0.0455) (0.0385) (0.0394) Leverage 0.120 0.0917 0.125 0.0974 0.164 0.106 0.170 0.142 (0.143) (0.141) (0.140) (0.136) (0.139) (0.140) (0.138) (0.131) Cash 0.112 0.138* 0.111 0.104 0.133* 0.144* 0.132* 0.126 (0.0796) (0.0832) (0.0794) (0.0812) (0.0791) (0.0844) (0.0789) (0.0808) Strategic -0.0687 -0.125 -0.0666 -0.0660 -0.0676 -0.109 -0.0653 -0.0651 (0.0830) (0.110) (0.0857) (0.0829) (0.0724) (0.0916) (0.0748) (0.0720) Competing 0.137 0.231 0.122 0.136 0.128 0.221* 0.113 0.127 (0.104) (0.144) (0.0975) (0.0991) (0.102) (0.133) (0.0959) (0.0975) Tender 0.0471 -0.00986 0.0557 0.0526 0.0384 0.00263 0.0474 0.0435 (0.0587) (0.0575) (0.0587) (0.0551) (0.0588) (0.0576) (0.0589) (0.0554) Unsolicited 0.0602 0.0495 0.0751 0.0561 0.0612 0.0569 0.0763 0.0574 (0.0580) (0.0617) (0.0585) (0.0571) (0.0567) (0.0590) (0.0573) (0.0561) Method Change 0.0302 -0.0422 0.0427 0.0302 0.0202 -0,0287 0.0329 0.0201 (2Y) (0.0875) (0.0886) (0.0871) (0.0877) (0.0851) (0,0842) (0.0847) (0.0853) Constant 0.153 0.0841 0.198 0.149 0.238 0.173 0.283 0.234 (0.252) (0.284) (0.255) (0.253) (0.249) (0.286) (0.251) (0.250) Observations 249 204 249 249 249 204 249 249 R-Squared 0.088 0.122 0.083 0.084 0.098 0.119 0.094 0.093 F-Statistic 1.61 1.65 1.62 1.80 1.95 1.95 2.00 2.23 Note: Table 15 displays the OLS regression for the 2-year ESG Momentum at 42 and 30 day deal premium. This table presents the coefficients and standard errors from eight independent regressions. Columns (1) to (4) represent the 42 day deal premium regression with the aggregated ESG Momentum, disaggregated ESG pillars, along with the control variables. Similarly, columns (5) to (8) represent the 30 day deal premium regression,. Standard errors in parentheses. * p < 0.10, ** p < 0.05, *** p < 0.01 57 Table 16: OLS Regression on 2-year ESG Momentum (15 and 7 Day Deal Premium) (1) (2) (3) (4) (5) (6) (7) (8) ESG 15 E_15 S_15 G 15 ESG_7 E_7 S_7 G_7 ESG 0.000260 0.000204 Momentum (0.000506) (0.000507) (2-Year) Environmental 0.000547** 0.000549** Momentum (0.000237) (0.000238) (2-Year) Social -0.000190 -0.000157 Momentum (0.000293) (0.000287) (2-Year) Governance 0.000356* 0.000222 Momentum (0.000205) (0.000198) (2-Year) ROE 0.205 0.387* 0.195 0.158 0.207 0.360* 0.198 0.157 (0.176) (0.206) (0.182) (0.159) (0.168) (0.198) (0.172) (0.153) ESG × ROE -0.00568 -0.00560 (0.00398) (0.00395) E × ROE -0.00193 -0.00241* (0.00149) (0.00144) S × ROE -0.00218 -0.00225 (0.00252) (0.00248) G × ROE -0.00326 -0.00265 (0.00258) (0.00242) Market-to-Book -0.00209 0.00269 -0.00250 -0.00241 -0.000582 0.00261 -0.000935 -0.00106 (0.00672) (0.00871) (0.00673) (0.00659) (0.00628) (0.00825) (0.00623) (0.00617) Market Cap 0.0113 -0.0248 0.0122 0.0140 0.00558 -0.0215 0.00655 0.00943 (0.0333) (0.0376) (0.0327) (0.0319) (0.0318) (0.0355) (0.0313) (0.0308) Leverage 0.201 0.209* 0.201 0.178 0.163 0.153 0.163 0.143 (0.128) (0.126) (0.126) (0.121) (0.126) (0.122) (0.123) (0.120) Cash 0.137* 0.147* 0.134* 0.136* 0.148** 0.156** 0.144** 0.147** (0.0748) (0.0783) (0.0744) (0.0758) (0.0711) (0.0754) (0.0707) (0.0726) Strategic -0.00290 -0.0586 -0.00203 -0.000282 -0.0280 -0.0802 -0.0269 -0.0266 (0.0776) (0.0979) (0.0793) (0.0766) (0.0740) (0.0927) (0.0753) (0.0731) Competing 0.0927 0.150 0.0774 0.102 0.0559 0.0950 0.0412 0.0593 (0.0708) (0.0938) (0.0655) (0.0711) (0.0679) (0.0901) (0.0647) (0.0676) Tender 0.0380 0.00101 0.0475 0.0363 0.0260 -0.00781 0.0360 0.0259 (0.0535) (0.0531) (0.0536) (0.0509) (0.0505) (0.0510) (0.0505) (0.0485) Unsolicited 0.0573 0.0359 0.0671 0.0509 0.0438 0.0280 0.0518 0.0389 (0.0559) (0.0595) (0.0564) (0.0551) (0.0531) (0.0571) (0.0537) (0.0527) Method Change 0.0409 -0.0183 0.0509 0.0346 0.0190 -0.0362 0.0282 0.0147 (2Y) (0.0790) (0.0792) (0.0788) (0.0796) (0.0758) (0.0760) (0.0754) (0.0762) Constant 0.170 0.122 0.203 0.153 0.199 0.131 0.228 0.185 (0.237) (0.268) (0.239) (0.237) (0.229) (0.258) (0.231) (0.229) Observations 249 204 249 249 249 204 249 249 R-Squared 0.091 0.128 0.083 0.091 0.093 0.128 0.085 0.087 F-Statistic 1.50 1.68 1.60 1.71 1.68 1.85 1.69 1.82 Note: Table 16 displays the OLS regression for the 2-year ESG Momentum at 15 and 7 day deal premium. This table presents the coefficients and standard errors from eight independent regressions. Columns (1) to (4) represent the 15 day deal premium regression with the aggregated ESG Momentum, disaggregated ESG pillars, along with the control variables. Similarly, columns (5) to (8) represent the 7 day deal premium regression,. Standard errors in parentheses. * p < 0.10, ** p < 0.05, *** p < 0.01 58 Table 17: OLS Regression on 3-year ESG Momentum (42 and 30 Day Deal Premium) (1) (2) (3) (4) (5) (6) (7) (8) ESG 42 E_42 S_42 G 42 ESG_30 E_30 S_30 G_30 ESG 0.000364 0.000446 Momentum (0.000525) (0.000557) (3-Year) Environmental 0.000390** 0.000377** Momentum (0.000177) (0.000182) (3-Year) Social -0.000164 -0.000145 Momentum (0.000209) (0.000212) (3-Year) Governance -0.000170 -0.000122 Momentum (0.000180) (0.000189) (3-Year) ROE 0.203 0.536* 0.307 0.117 0.224 0.538* 0.329 0.148 (0.278) (0.302) (0.321) (0.249) (0.273) (0.298) (0.318) (0.244) ESG × ROE -0.00645 -0.00577 (0.00588) (0.00597) E × ROE -0.00439 -0.00399 (0.00335) (0.00338) S × ROE -0.00492 -0.00469 (0.00409) (0.00412) G × ROE 0.000972 0.00131 (0.00350) (0.00350) Market-to-Book -0.00588 -0.0117 -0.00554 -0.00486 -0.00962 -0.0128 -0.00943 -0.00884 (0.0104) (0.0110) (0.0101) (0.0103) (0.00994) (0.0101) (0.00960) (0.00999) Market Cap 0.0623 0.0617 0.0537 0.0712 0.0651 0.0599 0.0566 0.0732 (0.0448) (0.0461) (0.0425) (0.0465) (0.0438) (0.0456) (0.0416) (0.0455) Leverage 0.191 0.233* 0.185 0.187 0.258* 0.254* 0.251 0.254* (0.150) (0.139) (0.154) (0.147) (0.150) (0.137) (0.156) (0.146) Cash 0.0943 0.138* 0.0970 0.0901 0.109 0.133 0.113 0.106 (0.0813) (0.0829) (0.0797) (0.0832) (0.0799) (0.0824) (0.0786) (0.0820) Strategic -0.0793 -0.141 -0.0762 -0.0891 -0.0766 -0.133 -0.0743 -0.0860 (0.0950) (0.116) (0.0999) (0.0977) (0.0827) (0.0980) (0.0880) (0.0851) Competing 0.234* 0.170 0.240** 0.219** 0.196 0.128 0.202* 0.181* (0.123) (0.114) (0.120) (0.109) (0.123) (0.0962) (0.121) (0.110) Tender -0.000688 -0.0234 0.00741 0.00789 -0.0122 -0.0116 -0.00540 -0.00446 (0.0631) (0.0642) (0.0616) (0.0599) (0.0640) (0.0655) (0.0627) (0.0609) Unsolicited 0.00331 -0.0144 0.0159 0.00213 0.0000861 -0.0130 0.0133 0.000545 (0.0609) (0.0667) (0.0598) (0.0610) (0.0596) (0.0629) (0.0586) (0.0600) Method Change -0.163* -0.0551 -0.175* -0.170* -0.141 -0.0634 -0.154 -0.148 (3Y) (0.0982) (0.103) (0.0977) (0.101) (0.0960) (0.0767) (0.0958) (0.0987) Constant 0.221 0.143 0.274 0.221 0.313 0.241 0.365 0.316 (0.257) (0.313) (0.261) (0.257) (0.245) (0.303) (0.248) (0.246) Observations 203 160 203 203 203 160 203 203 R-Squared 0.119 0.185 0.125 0.107 0.128 0.181 0.134 0.116 F-Statistic 1.69 2.00 1.83 1.76 1.97 2.56 2.15 2.02 Note: Table 17 displays the OLS regression for the 3-year ESG Momentum at 42 and 30 day deal premium. This table presents the coefficients and standard errors from eight independent regressions. Columns (1) to (4) represent the 42 day deal premium regression with the aggregated ESG Momentum, disaggregated ESG pillars, along with the control variables. Similarly, columns (5) to (8) represent the 15 day deal premium regression,. Standard errors in parentheses. * p < 0.10, ** p < 0.05, *** p < 0.01 59 Table 18: OLS Regression on 3-year ESG Momentum (15 and 7 Day Deal Premium) (1) (2) (3) (4) (5) (6) (7) (8) ESG 15 E_15 S_15 G 15 ESG_7 E_7 S_7 G_7 ESG 0.000235 0.000138 Momentum (0.000512) (0.000494) (3-Year) Environmental 0.000387** 0.000321* Momentum (0.000193) (0.000178) (3-Year) Social -0.000216 -0.000222 Momentum (0.000189) (0.000179) (3-Year) Governance -0.000161 -0.000211 Momentum (0.000177) (0.000165) (3-Year) ROE 0.305 0.530* 0.373 0.225 0.301 0.508* 0.367 0.220 (0.246) (0.293) (0.293) (0.218) (0.233) (0.279) (0.277) (0.209) ESG × ROE -0.00590 -0.00597 (0.00492) (0.00484) E × ROE -0.00280 -0.00326 (0.00273) (0.00257) S × ROE -0.00377 -0.00375 (0.00350) (0.00341) G × ROE -0.00152 -0.00121 (0.00296) (0.00277) Market-to-Book -0.000336 -0.00429 -0.000098 0.000560 0.000773 -0.00319 0.00110 0.00184 (0.00904) (0.00965) (0.00883) (0.00875) (0.00861) (0.00927) (0.00843) (0.00836) Market Cap 0.0206 0.00813 0.0152 0.0276 0.0159 0.00967 0.0105 0.0233 (0.0351) (0.0370) (0.0346) (0.0352) (0.0340) (0.0364) (0.0337) (0.0345) Leverage 0.238* 0.293** 0.229 0.227* 0.176 0.220* 0.168 0.167 (0.138) (0.136) (0.143) (0.133) (0.131) (0.129) (0.136) (0.128) Cash 0.114 0.143* 0.116 0.116 0.126* 0.156** 0.127* 0.126* (0.0761) (0.0810) (0.0746) (0.0772) (0.0733) (0.0780) (0.0718) (0.0741) Strategic -0.0128 -0.0797 -0.0120 -0.0215 -0.0321 -0.0895 -0.0304 -0.0405 (0.0871) (0.105) (0.0924) (0.0899) (0.0823) (0.0984) (0.0867) (0.0851) Competing 0.139* 0.135 0.147* 0.133* 0.107 0.0744 0.116 0.101 (0.0818) (0.0916) (0.0779) (0.0758) (0.0808) (0.0994) (0.0779) (0.0754) Tender -0.00382 -0.0173 0.00405 -0.000581 -0.0109 -0.0276 -0.00243 -0.00596 (0.0598) (0.0620) (0.0585) (0.0574) (0.0567) (0.0590) (0.0555) (0.0544) Unsolicited 0.00194 -0.0269 0.0139 0.000566 -0.00623 -0.0362 0.00450 -0.00922 (0.0587) (0.0645) (0.0581) (0.0591) (0.0557) (0.0613) (0.0551) (0.0560) Method Change -0.120 -0.0393 -0.130 -0.125 -0.113 -0.0373 -0.121 -0.118 (3Y) (0.0939) (0.0991) (0.0936) (0.0958) (0.0899) (0.0953) (0.0897) (0.0918) Constant 0.221 0.215 0.270 0.214 0.221 0.215 0.270 0.214 (0.234) (0.297) (0.237) (0.235) (0.234) (0.297) (0.237) (0.235) Observations 203 160 203 203 203 160 203 203 R-Squared 0.115 0.173 0.117 0.103 0.116 0.172 0.118 0.105 F-Statistic 1.50 1.80 1.87 1.53 1.54 1.74 1.86 1.61 Note: Table 18 displays the OLS regression for the 3-year ESG Momentum at 15 and 7 day deal premium. This table presents the coefficients and standard errors from eight independent regressions. Columns (1) to (4) represent the 15 day deal premium regression with the aggregated ESG Momentum, disaggregated ESG pillars, along with the control variables. Similarly, columns (5) to (8) represent the 7 day deal premium regression,. Standard errors in parentheses. * p < 0.10, ** p < 0.05, *** p < 0.01 60