Non-Linear Impact of Environmental, Social, and Governance Scores on Deal Premiums

Abstract

Increasingly, scholars have been researching how ESG ratings appear to impact the returns of a merger as well as the expected synergies of the merger. This paper adds to the literature by using a non-linear method to test the impact that ESG ratings, differences in ESG ratings between the acquirer and the target, and ESG components have on the deal premium. We find overwhelming evidence, using multiple deal premium measurements, of an inverted U-shaped relationship between the target’s ESG scores at the time of the announcement and the deal premium. Moreover, we find some evidence that differences between the ESG scores of the acquirer and the target also impact the deal premium but in a U-shaped relationship. Finally, our results show that the social scores of both the acquirer and the target impact the deal premium, again in an inverted U-shaped manner, as does the governance rating of target, but only in relatively smaller deals.

1. Introduction

Over the past several years, investors have paid more attention to environmental, social, and governance (ESG) scores and social responsibility, especially after the Paris Agreement in 2015 and the global pandemic. With increased reporting of ESG activities and scores, investors have questioned the added value that ESG provides to companies. Put another way, does the investment in ESG activities yield tangible returns, such that increasing ESG scores adds to the market value of a company. While the market value is assessed daily by investors, it is also subject to market sentiment at any particular time. Also, there may be reverse causality between ESG scores and firm value since firms with strong financial performance may invest more in corporate responsibility (Deng et al., 2013). As such, researchers have often found conflicting evidence regarding the impact of ESG activities on firm value.

Another avenue for exploring the value of ESG activities is to assess the value acquirers place on ESG activities when bidding to acquire control of a company. Companies pay a deal premium to acquire another company. This deal premium is a function of the synergies that the acquirer believes stem from combining the companies. Synergies include marketing, operational, management, and financial synergies. The ESG activities of the target, acquirer, or the two combined can be considered part of the management and operational synergies that result from a merger.

ESG scores or activities, however, may already be reflected in the stock price before the merger. For example, high ESG scores may reflect a premium brand. Thus, the deal premium would then need to reflect the additional, projected marketing, management, or operational synergies from ESG that the acquirer believes it is getting from the acquisition.

There are several avenues through which ESG scores could impact deal premiums. First, higher ESG scores may appeal to socially conscious investors. Secondly, ESG scores can indicate the likelihood of environmental or social problems. Furthermore, the debate between the shareholder view of profit maximization versus the stakeholder argument of firms being responsible not only to stockholders but also to customers, suppliers, employees, etc., plays a role as to whether ESG scores impacts deal premiums. Other factors that may affect how or if ESG scores impact deal premiums include the size and location of the merging companies and the industry.

There have been numerous studies (See Feng, 2021; Tampakoudis & Anagnostopoulou, 2020; Huang et al., 2022; Zheng et al., 2023) that have examined the impact of ESG on the post-merger financial performance of the firm or on post-merger ESG performance, but few (See Erben et al., 2022; Malik & Mamun, 2024; Maung et al., 2020; Gomes & Marsat, 2018; de Waal, 2023) have examined the impact of ESG scores on deal premiums. Given the uncertainty around the value of ESG, this paper assesses the impact that ESG scores overall and individually have on the deal premium. The hypothesis is that target ESG scores have a positive impact on the deal premium due to the added value that the scores suggest regarding the management and workforce of the target; however, we expect there to be a diminishing impact of target ESG activities on the deal premium. Similarly, we expect the difference between acquirer and target ESG scores to have a positive impact on the deal premium, but again to have diminishing returns. We also expect the governance of the target to be the primary driver of the ESG–deal premium relationship. Finally, we expect the effect to be prevalent in smaller versus larger deals due to the lower premia that tends to accompany larger deals because of the increased complexity in integrating larger companies (Alexandridis et al., 2013).

2. Literature Review

Early management scholars, such as Friedman (1970), argued that corporations exist solely to maximize profit. As such, expenditures on ESG could be viewed as wasteful and, therefore, as negatively impacting short-term profits. In contrast, Freeman (1984) as well as Porter and Kramer (2006) advocate for a broader vision of corporate responsibility, to include other stakeholders, such as employees, customers, and the community. They suggest that expenditures on ESG are part of a firm’s corporate responsibility, which can positively impact shareholder returns. More significantly, taking into account both shareholder benefits and stakeholder concerns helps sustain long-run economic growth and develop the company’s base of support (this discussion was taken from Zrigui et al., 2024) (Freeman, 1984).

2.1. IIa: Theories Regarding the Impact of ESG on Firm Value and the Deal Premium

The deal premium is the market value plus the additional amount the acquirer is paying to effectuate the deal. The premium is a function of the synergies the acquirer believes it can obtain as well as bargaining between the acquirer and the target. ESG activities can impact the market value of the firm, but that value is ostensibly already in the enterprise value. In evaluating the impact of ESG on the deal premium then we need to consider how these activities may affect the synergies accrued from gaining control of the target as well as the bargaining position of the acquirer. Synergies take many forms, including management and operational synergies.

There are many theories, such as signaling, agency, resource-base, shareholder spending, and too much of a good thing, that connect ESG ratings to firm value. These theories also apply to how ESG impacts the deal premium. Signaling theory (this discussion was taken from Y. Wang & Sonenshine, 2025) suggests that increasing environmental and social disclosures increase the value of the target by enhancing its reputation and reducing policy uncertainty (See Ahsan & Qureshi, 2021; Tang et al., 2024). Strong ESG performance may also reduce financing costs and increase the target’s value to the acquirer (see Richardson & Welker, 2001; Plumlee et al., 2015). Benabou and Tirole (2010) further argue that corporate and social responsibility investment helps establish long-term management perspectives, which will increase the target’s value over time. Finally, Tang et al. (2024) adds that ESG scores are a critical way to measure sustainability risks.

Similarly, the resource-based theory highlights the strategic use of internal resources, such as human capital, a key subcomponent of social responsibility, to gain a competitive advantage (Whelan et al., 2021). This competitive advantage may be a source of bargaining by the target in merger negotiations.

However, Zrigui et al. (2024) countered that ESG activities could have a negative impact on the deal premium aligning with the shareholder spending view, whereby executives pursue ESG activities for personal gains at the expense of shareholders. Similarly, Miralles-Quirós et al. (2018) proposed agency theory whereby managers may engage in ESG activities to enhance their own reputations rather than boosting corporate profits; the result may lead to wasted resources, which could lead to lower deal premiums. Moreover, Pierce and Aguinis (2013) contend that the impacts of ESG initiatives can become negative when they exceed a certain threshold, as additional ESG costs surpass their benefits. They suggest that ESG investments may be characterized by an inverted U-curve in their relationship with firm value and by extension the deal premium, which led them to propose the too much of a good thing theory.

2.2. Empirical Studies Relating ESG Scores to Deal Premiums

The interest in the impact of ESG ratings on value creation from mergers stems in large part from a PwC (2012) study of corporate buyers. This study found that a company’s sustainability performance positively impacts the value of the target company and thereby raises the deal premium. The rationale as to why ESG scores may impact deal premiums has often been attributed to potential risk mitigation. For example, Choi (2015) contends that higher corporate responsibility ratings (CSRs) reduce risk by lowering levels of information asymmetry.

Ung and Urfe’s (2021) results were also similar to the PWC study as they found that the target ESG scores positively impact the deal premium. However, their results showed the effect diminishes when the target is in the upper tercile of analyst coverage, suggesting the effect diminishes among the largest firms. In contrast, Zrigui et al. (2024) found that target ESG scores negatively impact the deal premium. They explain that acquirers with a strong focus on shareholder value may view ESG spending to be excessive and thus offer lower premiums for companies with high levels of ESG activity. Given these and other conflicting results, there has been significant deliberation regarding the impact of ESG on the deal premium and/or abnormal returns resulting from a merger announcement.

Some of these conflicting results also relate to ESG components. For example, Jo and Na (2012) argue that good environmental performance lowers the risks of penalties and concerns over obtaining permits. Furthermore, Renneboog et al. (2008) contend that high levels of environmental performance suggest the firm has quality management particularly relating to risk mitigation. In contrast, Jia et al. (2016) argue that poor environmental scores could actually be seen as an opportunity for an acquirer to obtain a bargain in the merger, as these low ratings likely attract a fair amount of attention, causing the firm to improve their practices. Alternatively, Aktas et al. (2011) suggest that acquiring firms can gain value simply by divesting or reducing the target’s environmental and/or social activities. Irrespective of these explanations, Giese et al. (2019) argue that acquirers are in a better position to assess the accuracy of environmental performance than are investors, which can explain the negative relationship between environmental scores and the premium paid.

Conflicting results have also involved social and governance ratings. For example, Khan et al. (2020) argues that quality governance reduces agency costs as well as the misuse of funds by managers. They explain that part of the value of acquiring a well-managed firm stems from the transferability of management and management practices. Furthermore, Deng et al. (2013) and C. Wang and Xie (2009) argue that a merger creates value when there is a substantial and rising difference in governance practices between the acquirer and target. However, de Waal (2023) found that increasing governance scores have a negative effect on the deal premium, but social ratings have a positive effect.

The one finding where results have been consistent relates to the impact of social scores on the deal premium, with the effect being positive. Zrigui et al. (2024) argue that above-average ESG scores suggest the target firm may have strong community relations, allowing them to more easily recruit, market, and obtain approvals and certifications. As a result, acquirers would likely pay higher merger premiums for targets with higher social scores. Similarly, Kanchel and Lassoued (2022) contend that high or improving social scores suggest rising employee productivity, due to better working conditions.

Other empirical studies regarding the impact of ESG on deal premiums or abnormal returns, as well as an explanation of their results, are presented in

Table 1. Empirical studies covering the effect of ESG on mergers events.

Author(s)	Results	Explanation
Aktas et al. (2011)	Higher abnormal returns with targets that exhibit better CSR performance.	Acquirers learn from the practices of targets.
Deng et al. (2013)	Higher acquirer abnormal returns correspond to higher acquirer CSR performance	Acquirer’s social performance is the critical factor influencing merger returns and the likelihood that the merger is completed.
Tang et al. (2024)	Greater difference between ESG scores between acquirer and target results in lower abnormal returns.	Larger ESG differences increase the likelihood of merger integration problems and costs while reducing the likelihood of the merger being completed.
Malik and Mamun (2024)	Positive impact of a target’s CSR quality and the acquisition premiums, particularly for large targets and acquirers with high CSR performance.	CSR explains some of the variability in deal premiums.
Maung et al. (2020)	Lower deal premiums result from lower target ESG scores in cross-border mergers.	Deal premiums were lower for firms with ESG incidents reported in the media, particularly if the number of incidents was higher than those reported by the acquirer.
Brunner-Kirchmair and Wagner (2024)	CSR has a negative effect on abnormal returns in Europe.	Suggests “too much of a good thing,” meaning CSR activities may reduce a target’s value.
Erben et al. (2022)	Neither acquirer nor target ESG scores impact deal premiums, but acquirer governance interacting with ESG scores negatively impacts the premiums paid. Suggests a non-linear relationship between ESG and the deal premium.	Stronger governance practices by acquirers allow for a significant investment in CSR activities.
Hussain and Shams (2022)	The higher the bidder’s CSR scores relative to the target’s, the higher the combined cumulative abnormal returns of bidders and targets.	Synergies relate to differences in CSR scores between the acquirer and target
Alexandridis et al. (2022)	Differences in CSR scores resulted in lower announcement event returns.	Lower synergies due to a clash in cultures.
Fairhurst and Greene (2022)	Non-linear U-shaped relationship between CSR and the likelihood of a takeover and abnormal gains in takeovers.	Acquirers seek to gain by taking corrective action among firms with low CSR scores. Also, firms with high CSR scores are takeover targets due to the synergies of combining efforts.
Gomes and Marsat (2018)	Deal premiums are positively impacted by acquirer and target CSR; social scores only cause a premium in cross-border mergers, while environmental performance positively impacts premiums.	Lower information asymmetry.
Godfrey et al. (2009)	Deal premiums are positively impacted by higher target ESG scores.	ESG activities act as insurance in the event of a negative shock as high CSR levels reduce losses.
de Waal (2023)	Target ESG scores do not impact deal premiums. However, social scores have a positive impact, while governance scores have a negative impact.	Differential impact of social and governance scores lead to insignificant impact of ESG on deal premiums.

.

Given the conflicting results regarding how ESG and its components, particularly environmental and governance, impact the deal premium, we analyzed this problem using a non-linear approach. Few studies use a non-linear approach, and none to our knowledge use a non-linear approach to assess how, not only the target and acquirer ESG impact the deal premium, but also how the difference in their ESG scores effects the deal premium, as well as how the results differ for ESG components and between large and small deals.

3. Data and Data Summary

3.1. IIIa. Data Description

The data set consists of approximately 275 mergers1 over 5 years (2020 to 2024) with a deal value greater than USD 1 billion. The study focuses on relatively large deals as ESG data is more available on larger companies. For each deal, we gathered merger-specific data consisting of the deal value, the relative size of the acquirer versus the target based on annual sales, the percentage of the company that was acquired, the percentage of the merger financed with cash versus stock, the presence of a competitive bid, and whether the merger was completed. In addition, we gathered the 1-day, 1-week, and 4-week deal premium. The deal premium is defined as the percentage difference between the deal value and the market value of the target during a set time (t) prior to the merger announcement date (A). The formula for the deal premium is shown in Equation (1) with t referring to either 1 day, 1 week, or 4 weeks.

$${\mathrm{D}\mathrm{e}\mathrm{a}\mathrm{l} \mathrm{p}\mathrm{r}\mathrm{e}\mathrm{m}\mathrm{i}\mathrm{u}\mathrm{m}}_{\mathrm{A}-\mathrm{t}}={\displaystyle \frac{{\mathrm{D}\mathrm{e}\mathrm{a}\mathrm{l} \mathrm{V}\mathrm{a}\mathrm{l}\mathrm{u}\mathrm{e}}_{\mathrm{A}-\mathrm{t}}-\mathrm{M}{\mathrm{a}\mathrm{r}\mathrm{k}\mathrm{e}\mathrm{t} \mathrm{V}\mathrm{a}\mathrm{l}\mathrm{u}\mathrm{e}}_{\mathrm{A}-\mathrm{t}}}{\mathrm{M}{\mathrm{a}\mathrm{r}\mathrm{k}\mathrm{e}\mathrm{t} \mathrm{V}\mathrm{a}\mathrm{l}\mathrm{u}\mathrm{e}}_{\mathrm{A}-\mathrm{t}}}}$$

(1)

Merger information was taken from the LSEG Data and Analytics (formerly Thomson–Refinitiv) database. In doing so, we downloaded all mergers over USD 1 billion in size and filtered this list to ones containing deal premium information. In addition, we eliminated mergers where the acquirer was a special acquisition company (SPAC2) or if the merger entailed an internal purchase of the company’s stock.3

ESG scores were obtained from the MSCI time series database. These scores pertain to the target and acquirer for the month and year of the merger announcement as well as the month and year prior to the merger announcement. ESG scores include the composite weighted score as well as scores for governance, social and environmental4.

ESG components can be further sub-divided into the following sub-categories shown in

Table 2. Listing of ESG Components and Subcomponents.

ESG Components	ESG Subcomponent and Weight
Environmental 1	Climate change Natural capital (natural resource) Pollution and waste (waste management) Environmental opportunities
Social 2	Human capital Product liability Stakeholder opposition Social opportunities
Governance 3	Corporate governance Corporate behavior

¹ Environmental includes 13 issues that are organized into the course subcomponents: climate change, natural capital or resource, pollution and waste, and environmental opportunities. See ESG Ratings Methodology (msci.com) for a list of the 13 items organized into the four subcomponents that comprise the environmental rating. ² Social covers health and safety, human capital development, labor management, and supply chain labor standards, which are issues in the human capital subcomponent. Social also covers consumer financial protection, privacy and data security, product safety and quality, and responsible investment, all of which are part of product liability. Lastly, social covers community relations and controversial sources that are part of stakeholder opposition as well as access to health care and opportunities in nutrition and health that are part of social opportunities. ³ Corporate governance covers pay, ownership and control, and accounting, while corporate behavior includes business ethics and tax transparency.

.

MSCI ratings are determined by analysts who evaluate the level of risk each company faces in the three areas. The information used to arrive at the rating comes from company reports regarding their operations as well as relevant macro-level data. The rating is based on the difference between the best practices in the industry and the company’s governance and environmental and social policies.

Table 3. Summary statistics.

Variable	Observations	Median	Mean	Standard Deviation	Range Min, Max	Mean Small Deals (<3696)	Mean Large Deals (>3696)
Deal premium—1 day	325	30.2	34.7	21.9	0.7, 96.5	35.5	34.3
Deal premium—1 week	331	32.0	36.4	23.0	0.2, 98.2	36.9	35.9
Deal premium—1 month	316	36.0	39.70	23.6	1, 99	39.5	39.2
Deal value (in millions of USD)	329	3696	6834	9853	100, 81,053	1655	12,012
Target-Weighted ESG	331	5.0	5.0	0.95	1.8, 8.0	4.87	5.16
Target Governance	331	5.9	5.6	1.62	0, 8.7	5.66	5.62
Target Environmental	331	5.0	5.0	2.69	0, 10.0	5.07	5.05
Target Social	331	4.6	4.6	1.57	0, 10.0	4.56	4.61
Acquirer-Weighted ESG	310	5.1	5.0	1.10	0.5, 8.0	5.0	5.0
Acquirer Governance	310	5.7	5.4	1.57	0, 8.8	5.4	5.4
Acquirer Environmental	310	5.5	5.5	2.43	0.7, 10.0	5.5	5.5
Acquirer Social	310	4.9	4.9	1.42	0.10	4.9	4.8
Difference Current vs. Prior Target ESG	252	0.3	0.5	0.33	0, 3.4	0.3	0.3
Difference Current vs. Prior Acquirer ESG	248	0.3	0.5	0.80	0, 5.4	0.3	0.3
Difference Current vs. Prior (Acquirer–Target) ESG	252	0.2	0.4	0.68	0, 5.4	0.2	0.2
Relative Sales (Acquirer Sales/Target Sales)	283	3.57	515	4044	0.02, 46,200	584	449
Percent Competitive Bids	283	-	17%	0.36	0.100	14%	18%
Percentage Completed	283	-	88%	-	-	-	-
Percentage Shares Acquired	283	-	87%	0.29	0.100	84%	89%
Percent Cash	283	-	59%	0.44	0.100	70%	68%
Financial	310	-	16%	-	-	-	-
Life Sciences	310	-	16%	-	-	-	-
Technology	310	-	13%	-	-	-	-
Industrial Goods	310	-	11%	-	-	-	-
Services	310	-	18%	-	-	-	-
Retail/Packaged Goods	310	-	7%	-	-	-	-
Energy/Resources	310	-	19%	-	-	-	-

summarizes the merger, company, and ESG information used in this study. The summary data is also segmented by small (below the median deal value) and large (above the median) deals.

From Table 3 we see the average 4-week deal premium is slightly larger than the 1-week premium, which again is slightly larger than the 1-day premium. Furthermore, we see the 1-day deal premium is one percentage point higher in relatively small versus large deals. This difference dissipates when considering the 1-week and 1-month deal premiums.

We also observe that the average acquirer ESG is slightly higher than the target’s ESG. The same is the case for social and environmental. The opposite is the case for governance where scores for the target are higher than scores for the acquirer. Also, we note that ESG scores and ESG component scores are similar between small and large deals. In addition, we found the average difference between the current and prior-year ESG for the target and acquirer to be 0.5, meaning that their ESG scores were improving from the prior year to the announcement year. Furthermore, we observe that the difference between the acquirer and target ESG (current—prior) was also increasing.

Regarding the deals, we see that roughly 17% of mergers involved competitive bids, whether from other companies or the acquirer proposing multiple times, due to their first offer being rejected. The deal premium is assumed to be higher with a competitive bid than it otherwise would have been. Also, 88% of the deals were completed, with on average 87% of the shares being acquired5. Lastly, on average, cash financed 59% of the deals. We anticipate that the higher the percentage of cash used to finance the deal, the higher the deal premium, as targets will accept a lower premium for a stock-financed deal to participate in the gains and losses from the merger. Finally, we see that the deals were fairly evenly distributed between six to seven industries.

Explanatory Data Analysis

We see in the histograms in Figure 1 that the distribution of target and acquirer ESG is fairly similar, though acquirer ESG is more concentrated between 5 and 6. We also see the difference in ESG is very concentrated between 0 and 1.

Figure 2 below shows the scatter plots using a lowess smoother technique of target ESG, acquirer ESG, the absolute value of the difference in acquirer versus target ESG, and the difference between the current- and prior-year target ESG ratings. In addition, there are lowess smoother plots for target ESG for small and for large deals.

From the top left plot in Figure 2, we see the deal premium appears to increase significantly as the target ESG increases from a very low level (2) to a mid-level (4). From the mid-level ESG throughout the rest of the distribution the 1-day premium appears fairly constant, with a possible hump in the middle of the distribution. Regarding the acquirer’s ESG, we see a slight increase in the deal premium in a small range between roughly 3.5 and 4. Otherwise, the deal premium appears to be relatively unaffected by the acquirer ESG. Similarly, the deal premium seems relatively unaffected by the absolute value of the difference between the acquirer and target ESG as shown in the middle, left plot. However, the deal premium appears to decline slightly relative to the difference between the current and prior year ESG levels, suggesting a slightly lower premium accompanies larger differences in target ESG scores. Finally, we see the target ESG ratings for smaller deals in the bottom left plot are pretty dispersed, while the ratings are in a tight range for larger deals in the bottom right plot. Both small and large deals appear to show a partial, inverted U-shaped relationship with target ESG, as the premium increases substantially at low ESG level, and then flattens out to decline slightly at intermediate ESG ranges.

We next look at the scatter plots for the 1-day premium relative to governance, social, and environmental scores for small and large deals (Figure 3).

In the top row we observe little change in the deal premium for low governance ratings, but then an inverted U-shaped relationship for mid-level and high-level governance ratings. However, for large deals we see an increasing relationship between governance and the deal premium except for the middle section when a hump or small U-shaped relationship appears.

In the middle row we see an increasing relationship between social scores and the deal premium in large deals. In contrast, for small deals this increasing relationship is only found at relatively low governance levels. Finally, for environmental ratings there appear to be higher deal premiums paid for smaller deals but not for larger deals.

4. Hypotheses

Building on economic theory, such as shareholder spending and too much of a good thing that we discussed earlier, this section develops a testable framework to explain the complex relationship between ESG scores and the deal premium.

Hypothesis 1.

We expect the deal premium to increase over a range of relatively low target ESG levels and then to hit a maximum level before declining with increasing target ESG scores. As such, we hypothesize an inverted U-shaped relationship existing between the target’s ESG rating and the deal premium, meaning a positive, significant linear coefficient and a negative, significant quadratic coefficient for the target ESG. The rationale is that,

○

At low ESG levels, target firms will be viewed as riskier due to potential issues relating to poor governance, environmental concerns, or social controversies. As a result, acquirers will offer a relatively low deal premium for companies with these low ESG scores. However, as the ESG scores increase, so too will the deal premiums.

○

At moderate ESG levels, targets are seen as stable but still offer room for improvement, making them the most attractive and synergistic acquisition opportunities. In this range, both acquirers and targets are likely to agree on a higher premium.

○

At high ESG levels, targets may already be highly valued and well-managed, limiting the acquirer’s ability to generate additional value post-acquisition. This condition reduces the acquirer’s willingness to offer a high premium.

This hypothesis is accordance with the-too much of a good thing theory.

Hypothesis 2.

We expect the difference in ESG levels between the acquirer and target at the time of the announcement to have a significant impact on the deal premium. Following Alexandridis et al. (2022), we expect the deal premium to decline as the difference in ESG levels increases due to integration problems that will occur post merger as suggested by these differing ESG scores. However, we expect the relationship to be non-linear and to approximate a U-shaped curve as, once the difference in ESG levels between the acquirer and target increase past a certain threshold, there are opportunities to gain synergies from the merger, and thus the deal premium will reverse and begin to increase. As such, we expect to find a negative, significant linear coefficient and a positive, significant quadratic coefficient for the absolute value in ESG between the acquirer and the target.

Hypothesis 3.

The inverted U-shaped relationship between the target ESG ratings and the deal premium will be found in small versus large deals. Thus, we expect again to find a positive, significant linear coefficient and a negative, significant, quadratic coefficient for the target ESG in small deals. In contrast, we do not expect these coefficients to be significant in large deals. The rationale as noted by Ung and Urfe (2021) is that ESG information is more important where information asymmetry is higher, which is more likely in smaller versus larger deals. Information asymmetries may prevail in smaller deals, because there are likely fewer analysts covering smaller companies. This contention is based in part on Ung and Urfe’s (2021) findings that the influence of ESG on deal premiums diminishes when the target is in the upper third of analyst coverage.

Hypothesis 4.

An inverted U-shaped relationship exists between target social and governance ratings and the deal premium for small deals. As such, for small deals we expect the coefficient for the target ESG to be positive and significant, but the coefficient for the quadratic term to be negative and significant.

○

For social scores, we contend the target’s social scores will impact the deal premium because the acquirer can gain value from increased productivity that stems from higher satisfaction levels among the target’s workforce where social ratings are high. However, the positive influence of social scores will reverse, after reaching a maximum, as similarly to the overall ESG score, the firm with high social scores is likely already highly valued.

○

For governance scores, we believe that target governance scores exhibit an inverted U-shaped relationship with the deal premium because at low governance scores, improvements suggest a more resilient company. However, at higher governance levels, companies may be more difficult to integrate and change into the governance systems of the acquirer. That is certain controls (see Borghesi et al., 2019; van Essen et al., 2013) (e.g., shareholder-centric policies, independent boards, separation of the CEO and Chairman of the Board, etc.) that raise governance scores may present post-merger integration problems and thus reduce the upside potential of the merger.

5. Methodology

To test these hypotheses, we used a series of models to estimate the impact of ESG, ESG differentials, and ESG components on the deal premium. See

Table A1. Correlation matrix.

	Dealvalue	%Cash	Acq ESG	Acq ESG sq.	Tgt ESG	Tgt ESG-sq	Diff ESG	Diff ESG-sq	Diff Yr Sq	Diff Yr ESG sq
Dealvalue	1.0000
%cash	−0.1386	1.0000
Acq ESG	0.0069	0.0491	1.0000
Acq ESG-sq	0.0002	0.0462	0.9628	1.0000
Tgt ESG	0.1233	−0.0341	0.2611	0.2680	1.0000
Tgt ESG sq	0.1664	−0.0082	0.2721	0.2876	0.9610	1.0000
Diff ESG	0.0845	0.0771	−0.1451	−0.0013	−0.3258	−0.2065	1.0000
Diff ESG-sq	0.1118	0.0444	−0.2126	−0.0610	−0.3582	−0.2180	0.9116	1.0000
Diff Yr ESG	0.1324	0.0236	0.0062	0.0095	−0.4172	0.2303	0.4668	0.5307	1.0000
Diff Yr ESG -sq	0.1106	0.0523	0.0320	0.0291	−0.5112	−0.3036	0.5231	0.6339	0.902	1.0000
	Acq diff	Acq diff-sq	Diff (diff)	Diff (diff) sq	SharesAQ	RelSale	Completed	Competitive
Acq diff	1.0000
Acq diff-sq	0.8860	1.0000
Diff (diff)	0.6019	0.6280	1.0000
Diff (diff) sq	0.4717	0.5975	0.5975	1.0000
SharesAQ	0.0420	0.0465	0.0927	0.0689	1.0000
RelSale	−0.0217	0.0002	−0.0070	−0.0185	0.0503	1.0000
Completed	0.0710	0.0566	0.0845	0.0633	−0.0137	0.0438	1.0000
Competitive	−0.0452	0.0277	−0.0334	−0.0057	0.0087	−0.0539	−0.4332	1.0000
	Dealvalue	%cash	AEnv	AEnv-sq	Env	Env-sq	ASoc	Asoc-sq	Soc	Soc sq
Dealvalue	1.0000
% cash	0.1415	1.0000
AEnv	0.1208	0.195	1.0000
AEnv-sq	0.2145	0.113	0.3834	1.0000
Env	0.2109	0.146	0.4078	0.9552	1.0000
Env-sq	0.2145	0.113	0.3834	0.999	0.9552	1.0000
ASoc	−0.0050	−0.034	0.1911	0.0571	0.0839	0.0571	1.0000
Asoc-sq	−0.0091	−0.041	0.1565	0.0420	0.0700	0.0420	0.9641	1.0000
Soc	0.1333	0.060	0.0815	0.1619	0.1477	0.1619	0.2558	0.2301	1.0000
Soc sq	0.1175	0.053	0.0689	0.1754	0.1676	0.1754	0.2286	0.2125	0.9578	1.0000
A gov	−0.0350	−0.038	0.0384	0.0310	0.0658	0.0310	0.0642	0.0648	0.0164	0.0303
	Agov	Agov-sq	sharesAQ	RelSal	Completed	Competitive
Agov-sq	1.0000
sharesAQ	0.0943	1.0000
RelSal	−0.0087	0.0457	1.0000
Completed	−0.0408	0.0013	0.0416	1.0000
Competitive	0.1436	0.0993	−0.0496	−0.444	1.0000

for the correlation matrices that accompany these models. We showed the results visually for the 1-day premium and in table form using the 1-day, 1-week, and 4-week deal premiums.

$$\begin{gathered} DealPremium={\beta }_o+{\beta }_1{ESG}_{it}+{{\beta }_2{ESG}_{it}}^2+{{\beta }_{3}lDealval}_m+{\%Cash}_{it}{+\beta }_5{\%sharesAcq}_{it} \\ +{\beta }_6{Competitive}_m+{\beta }_7{Completion}_m+{\beta }_{8}Re{lSales}_{it}+{\beta }_{9..14}{Industries}_{it}{+ϵ}_{it} \end{gathered}$$

(2)

$$\begin{gathered} DealPremium={\beta }_o+{\beta }_1{DiffESG}_{it}+{{\beta }_2{DiffESG}_{it}}^2+{{\beta }_{3}lDealval}_m{+\beta }_4{\%Cash}_{it}{+\beta }_5{\%sharesAcq}_{it} \\ +{\beta }_6{Competitive}_m+{\beta }_7{Completion}_m+{\beta }_{8}Re{lSales}_{it}+{\beta }_{9..14}{Industries}_{it}{+ϵ}_{it} \end{gathered}$$

(3)

$$\begin{gathered} DealPremium={\beta }_o+{\beta }_1({Diff\_diff)ESG}_{it}+{{\beta }_2{\left(Diff\_diff\right)ESG}_{it}}^2+{{\beta }_{3}lDealval}_m+{\beta }_4{\%Cash}_{it}{+\beta }_5{\%sharesAcq}_{it} \\ +{\beta }_6{Competitive}_m+{\beta }_7{Completion}_m+{\beta }_{8}Re{lSales}_{it}+{\beta }_{9..14}{Industries}_{it}{+ϵ}_{it} \end{gathered}$$

(4)

In Equations (2)–(4), i refers to the company, t to the month and year, and m refers to the merger. Separate regressions were run using Equation (2) for the current-year acquirer- and target-weighted ESG level for the month and year of the merger announcement. Diff in Equation (3) refers to the absolute value of the difference between the current-year acquirer and target ESG levels. Diff also refers to the difference between the target’s current-year and prior-year ESG level, as well as the acquirer’s current-year and prior-year ESG levels. Separate regressions were run for these three differences. Finally, a regression was run using Equation (4) with the term Diff(diff) referring to the difference in the acquirer–target ESG levels for the current year minus the prior year. We used the squared variable for ESG, and each of the Diff and Diff_diff variables to account for nonlinearities6.

Table 4. Explanation of testable variables.

Equation	Testable Variable(s)
Equation (2) (ESG term)	Target ESG Acquirer ESG
Equation (3) (Diff term)	Absolute value of Acquirer − Target ESG (same period) Absolute value of Acquirer ESG (same period) − Acquirer ESG (prior period) Absolute value of Target ESG (same period) − Target ESG (prior period)
Equation (4) (Diff_diff)	Absolute value of difference in Acquirer ESG − Target ESG (same period versus prior period)

summarizes the terms used.

In addition, we segmented the sample into small (below the medium deal value) and large (above the medium deal value) mergers. We then ran the regressions covered in Equations (2)–(4) for small and large deals to explore the differences.

Finally, we tested the impact of the individual ESG components, environmental (E), social (S), and governance (G), as shown in Equation (5), with Comp. referring to each of the components. Separate regressions were run for E, S, and G to avoid collinearity and allow us to assess the significance of the non-linear effects.

$$\begin{gathered} Deal Premium={\beta }_o+{\beta }_1({Comp.)ESG}_{it}+{{\beta }_2{\left(Comp.\right)ESG}_{it}}^2+{{\beta }_{3}lDealval}_m{+\beta }_4{\%Cash}_{it}{+\beta }_5{\%sharesAcq}_{it} \\ +{\beta }_6{Competitive}_m+{\beta }_7{Completion}_m+{\beta }_{8}Re{lSales}_{it}{+\beta }_{9..14}{Industries}_{it}{+ϵ}_{it} \end{gathered}$$

(5)

To assess the presence of a U-shaped or inverted U-shaped relationship, we followed the methodology outlined in Y. Wang and Sonenshine (2025), which applies the Sasabuchi–Lind–Mehlum (SLM) test as discussed in Lind and Mehlum (2010), Sasabuchi (1980), and Haans et al. (2016). The test evaluates the significance and curvature direction of the quadratic terms and verifies whether the turning point lies within the observed data range. Full details of the procedure can be found in our earlier work.

6. Results

The results covering the impact of the target and acquirer ESG scores on the deal premium (Equation (2)) are shown in Section 6.1. Section 6.2 then follows showing the results covering the difference in acquirer and target ESG scores per Equations (3) and (4). In Section 6.3 we then break down the sample into relatively small and large deals to assess the impact of the factors covered in Equations (2)–(4) on the deal premium. Finally, Section 6.4 covers the impact of ESG components on the deal premium.

6.1. ESG and the Deal Premium

In this sub-section we analyze the impact of the target and acquirer ESG scores on the deal premium, with many of the results shown in Figure 4 below.

The x-axis represents weighted ESG values or the difference in ESG values, and the y-axis represents the 1-day deal premium. The stars under the figures indicate whether the Sasabuchi test statistic is significant, meaning that the slopes of the initial and terminal lines are different, indicating a U or inverted U-shape relationship. The red dashed line shows the ESG or difference in ESG levels associated with the optimal deal premium, while the blue dashed line marks the deal premium at the mean ESG rating.

Table 5. Effect of ESG ratings on deal premiums.

Response Variable 1-Day Deal Premium	Target ESG (1)	Acquirer ESG (2)	Acquirer–Target ESG Difference (3)	Current– Prior Year Target ESG Difference (4)	Current- Prior Year Acquirer ESG Difference (5)	Current-Prior Year Difference in (Acq.—Target) ESG (6)
Log (dealvalue)	−0.02	−0.02	−0.02	0.001	0.05	0.014
	(0.02)	(0.05)	(0.05)	(0.06)	(0.05)	(0.06)
Percent cash	0.004 ***	0.003 ***	0.004 ***	0.004 ***	0.004 ***	0.004 ***
	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)
Shares AcQ	0.001	0.001	0.002	0.000	0.000	0.000
	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)
Relative sales	0.001 *	0.001	0.001 *	0.001	0.001 *	0.001
	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)
completed	0.0332	−0.0202	0.027	0.02	−0.007	0.02
	(0.172)	(0.165)	(0.171)	(0.187)	(0.180)	(0.193)
competitive	0.115	0.064	0.094	0.046	0.044	0.04
	(0.156)	(0.155)	(0.155)	(0.175)	(0.179)	(0.183)
Target ESG	0.812 **
	(0.337)
Target ESG-squared	−0.077 **
	(0.036)
Acquirer ESG		−0.16
		(0.14)
Acquirer ESG-squared		0.017
		(0.02)
Abs Acquirer—Target			−0.16
ESG difference			(0.1)
Abs Acquirer—Target			0.04 *
ESG difference-sq			(0.024)
Current—Prior Target				0.25
ESG difference				(0.24)
Current—Prior Target				−0.14 *
ESG difference-squared				(0.07)
Current—Prior Acq.					0.01
ESG difference					(0.17)
Current—Prior Acq.					0.01
ESG difference-squared					(0.03)
Acquire—Target ESG (Current—Prior)						−0.11 ((0.19)
Acquire—Target ESG (Current—Prior) -squared						0.04 (0.04)
Year effects	Yes	Yes	Yes	Yes	Yes	Yes
Industry effects	Yes	Yes	Yes	Yes	Yes	Yes
Slope—low end	0.86	0.14	−0.15	0.25	−0.05	−0.19
Slope—high end	−0.46	−0.11	0.40	−1.39	2.7	0.51
Sasabuchi test stat.	1.67 **	0.86	1.36 *	01.05	0.30	1.04
Threshold (−β1/(2β2))/within data range	5.27/yes	4.75/yes	1.75/yes	0.88/yes	0.88/yes	1.62/yes
95% Fieller interval for extreme point	4.55; 13.6	-Inf/inf	-inf, 15.9; 3.0, Inf	-Inf, 4.3/1.9, Inf	-Inf/inf	-Inf/inf
Observations	275	279	270	241	235	229
R-squared	0.113	0.124	0.117	0.143	0.114	0.121

Robust standard errors in parentheses *** p < 0.01, ** p < 0.05, * p < 0.1.

covers the regression results from models (2)–(4) with the 1-day deal premium being the response variable.

Table A2. Effect of ESG ratings on Deal premiums (1 and 4-week premiums).

Response Variables 1-Week & 4-Week Deal Premium	Lpremium_ 1-Week (1)	Lpremium_ 1-Week (2)	Lpremium_ 1-Week (3)	Lpremium_ 1-Week (4)	Lpremium_ 4-Weeks (5)	Lpremium_ 4-Weeks (6)	Lpremium_ 4-Weeks (7)	Lpremium_ 4-Weeks (8)
Log deal value	0.005	−0.011	0.017	0.008	0.006	−0.004	0.153	−0.002
	(0.05)	(0.05)	(0.05)	(0.05)	(0.04)	(0.04)	(1.31)	(0.04)
Percent cash	0.004 ***	0.004 ***	0.004 ***	0.004 ***	0.005 ***	0.005 ***	0.121 ***	0.005 ***
	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.03)	(0.00)
Shares AQ	0.001	0.002	0.001	0.000	0.000	0.001	−0.039	−0.001
	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.07)	(0.00)
Relative sales	0.001 *	0.000	0.000	0.001 **	0.001 **	0.001 **	0.000	0.001 ***
	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)
completed	0.202	0.229	0.306	0.296	0.182	0.206	5.956	0.311 *
	(0.18)	(0.19)	(0.21)	(0.22)	(0.15)	(0.16)	(4.94)	(0.18)
competitive	0.318 **	0.331 **	0.293 *	0.332 **	0.123	0.148	8.124 *	0.108
	(0.13)	(0.14)	(0.16)	(0.17)	(0.11)	(0.12)	(4.69)	(0.14)
Target ESG	0.788 ***				0.874 ***
	(0.28)				(0.26)
Target ESG-sq	−0.081 ***				−0.093 ***
	(0.03)				(0.03)
Acquirer—Target		−0.183 *				−0.14
ESG difference		(0.11)				(0.10)
Acquirer—Target		0.0428 *				0.0409 **
ESG difference-sq		(0.02)				(0.02)
Current-Prior Target ESG diff			0.29 (0.23)				0.72 (7.11)
Current-Prior Target ESG diff-sq			−0.181 ** (0.09)				−1.98 (2.08)
Acq.–Target ESG (Current—Prior)				−0.156 (0.19)				−0.348 ** (0.15)
Acq—Target ESG (Current—Prior) -sq				0.040 (0.04)				0.083 *** (0.04)
Year effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Slope—low end	29.2	−4.1	−0.18	−2.4	29.1	−0.14	−1.13	−8.7
Slope—high end	−19.4	6.3	0.35	5.2	−20.9	0.36	−17.84	14.2
Sasabuchi test stat.	4.05 ***	1.07	1.70 **	0.45	4.01 ***	1.45 *	-	1.42 *
Threshold (−β1/(2β2))/within data range	4.81/yes 4.3/5.3	2.53/yes -Inf./Inf.	2.13/yes -Inf./Inf.	1.83/yes -Inf./Inf.	4.65/yes 4.1/5.2	1.71/yes −5.55/2.44	−0.39/no -Inf, 2.3./1.3, Inf.	2.21/yes -Inf./Inf.
Observations	275	279	270	241	235	229	237	215
R-squared	0.108	0.095	0.097	0.129	0.102	0.107	0.127	0.167

Robust standard errors in parentheses *** p < 0.01, ** p < 0.05, * p < 0.1.

7 covers these results using the 1-week and 1-month deal premium.

From the results in Figure 4 and Table 5 (Column 1), it is clear that target ESG impacts deal premiums in an inverted U-shaped relationship. We found the coefficients for the linear and quadratic terms to be significant, and the Sasabuchi test statistic to be significant. The coefficients for target ESG and ESG-squared terms are also highly significant using the 1- and 4-week premiums, as shown in Table A2. These results suggest that acquirers are willing to pay higher premiums as the target’s weighted ESG score increases, but only up to a certain threshold—approximately a score of 5—beyond which the premium begins to decline, as shown in Figure 4, top-left plot. Moreover, it appears from Figure 4 that the ESG level associated with the optimal deal premium is very close to the mean deal premium, suggesting that acquirers are paying near the optimal amount in deal premium to consummate their mergers. Finally, we see in Table 5 (Column 2) and Figure 4 (right-side figure) that the acquirer ESG scores do not have a significant impact on the deal premium.

6.2. Differences in ESG (Acquirer vs. Target and Current vs. Prior Period) Scores and the Deal Premium

Next, we turn to the difference in the acquirer and target ESG levels at the time of the announcement as well as the difference in the current- and prior-year ESG levels. These results are shown visually below in Figure 5 as well as in

Table 6. Effect of ESG ratings on deal premium by deal size.

Response Variable 1-Day Deal Premium	Small Deals Target ESG (1)	Large Deals/ Target ESG (2)	Small Deals/ Acq-Targ ESG (3)	Large Deals/ Acq-Targ ESG (4)	Small Deals/ Current-Prior Year Acquirer ESG (5)	Large Deals/ Current-Prior Year Acquirer ESG (6)	Small Deals/ Acquire—Target ESG (Current–Prior Year) (7)	Large Deals/ Acquire—Target ESG (Current–Prior Year) (8)
Log (dealvalue)	−5.383	−3.123	−4.692	−2.316	−5.056	−2.335	−3.517	−2.381
	(3.34)	(2.56)	(3.29)	(2.60)	(3.49)	(2.82)	(3.74)	(2.79)
Percent cash	0.0503	0.118 ***	0.064	0.109 ***	0.089 *	0.093 **	0.096 *	0.094 **
	(0.04)	(0.03)	(0.05)	(0.04)	(0.05)	(0.04)	(0.05)	(0.04)
Shares AcQ	−0.021	−0.014	−0.004	−0.010	−0.050	−0.075	−0.044	−0.072
	(0.09)	(0.07)	(0.10)	(0.07)	(0.09)	(0.08)	(0.10)	(0.09)
Relative sales	0.000	0.001 **	0.000	0.001 **	0.000	0.000	0.001 **	0.001 **
	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)
Completed	−0.339	4.895	3.183	4.732	2.829	4.932	3.499	4.579
	(7.89)	(5.19)	(8.25)	(5.19)	(8.04)	(5.83)	(8.30)	(5.91)
competitive	−1.809	11.93 **	0.0619	12.30 **	−0.759	13.83 **	−1.287	14.20 **
	(6.28)	(4.95)	(6.95)	(4.90)	(7.51)	(5.55)	(7.24)	(5.88)
Target ESG	29.63 **	32.74 ***
	(11.86)	(9.832)
Target ESG2	−2.845 **	−3.197 ***
	(1.236)	(1.039)
Acquirer—Targ			−10.95	−0.034
ESG difference			(8.575)	(4.271)
Acquirer—Targ			3.240	0.135
ESG difference-sq			(2.792)	(0.793)
Current—Prior					−13.11	3.602
Acq ESG diff					(10.38)	(6.41)
Current- Prior					5.384 *	−0.662
Acq ESG diff-sq					(3.17)	(1.24)
Acq—Target ESG							−32.77 ***	3.365
(Current—Prior)							(11.72)	(6.921)
Acq—Target							14.49 **	−0.647
(Current—Prior)-sq							(5.568)	(1.382)
Year effects	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Industry effects	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Constant	7.101	26.93	68.19 **	37.47	77.84 **	44.54	55.68 *	44.17
	(36.37)	(27.21)	(28.84)	(23.84)	(31.16)	(27.09)	(30.72)	(27.02)
Slope—low end	29.6	32.7	−11.24	−1.35	−13.4	3.6	−32.5	−2.1
Slope—high end	−15.8	−18.4	26.83	3.53	47.4	−4.1	136.5	4.7
Sasabuchi test stat.	1.92 **	2.53 ***	1.09	0.31	1.26 *	0.53	2.49 **	0.30
Threshold (−β1/(2β2))/within data range	5.2/yes	5.15	1.84/yes	1.72/no	1.22/yes	2.72/yes	1.13/yes	1.78/yes
95% Fieller interval for extreme point	4.5/8.5	4.4/6.4	-Inf/Inf	-Inf/Inf	-Inf/Inf	-Inf/Inf	0.78/1.94	-Inf/Inf
Observations	128	156	120	150	111	128	106	127
R-squared	0.196	0.284	0.173	0.252	0.162	0.189	0.182	0.184

Robust standard errors in parentheses—*** p < 0.01, ** p < 0.05, * p < 0.1.

, Columns (3) through (6).

From Figure 5 and Table 5 (Columns 3 and 4), we find some evidence of a non-linear relationship between the acquirer–target ESG difference and the deal premium, indicated by the significant quadratic term and the Sasabuchi test statistic being marginally significant at the 10% level. These coefficients are also significant using the 1- and 4-week premiums, as shown in Table A2, Columns (2), (3), and (6). Interestingly, the absolute value of the difference in the acquirer versus target ESG levels appears to have a U-shaped relationship with the deal premium8. At lower levels of ESG difference, increasing divergence between the acquirer and target is associated with a decrease in the deal premium. Deal premiums then hit a minimum at a threshold point and then start to increase with an increase in the difference in the ratings. The rationale may be that increasing differences in ESG levels between the target and acquirer raise integration costs or present integration barriers. As a result acquirers do not project large synergies due to these differences. However, in rare cases with extremely large ESG differences, some acquirers may perceive opportunities to drive improvements in governance or employee engagement, potentially viewing the target as a turnaround candidate. This large difference could justify a higher premium, although such deals may also carry higher integration risk. Further research is needed to better understand whether such premiums reflect expected synergies, signaling effects, or potential underestimation of integration risk. In Figure 5, the blue line is left of the red line, suggesting that the deal premium paid for the mean difference in ESG is higher than the optimal deal premium paid for a larger difference in ESG levels.

With regard to the control variables, we observe that the greater the percentage of cash in financing the deal, the higher the deal premium paid. This result fits with the finance literature as targets will accept a lower premium for stock transactions in order to gain some stock in the new company and share in the gains and risks. Turning to other deal characteristics, we see the coefficient for competitive is surprisingly not significant in Table 5. However, this coefficient is significant in Table 6 for large deals (columns 2, 4 and 6) but not small deals. This result suggests that the impact of competitive bids on deal premiums may be conditional on deal size, and such heterogeneity would be obscured in pooled regressions.

In addition, from Table 5, we see that the coefficient for relative sales is positive and significant in three of the regressions. This result provides limited evidence that deal premiums paid to consummate a merger of relatively similar-sized companies are smaller than the deal premiums paid when the acquirer and target are relatively dissimilar as measured by annual sales. This result may be driven in part by mergers in the pharmaceutical industry, where massive pharmaceutical companies, such as Pfizer or Sanofi, pay a very large premium to acquire a startup or relatively new company that has made a major drug discovery.

Finally, we showed in Table 5 that industry effects were included, but we did not provide details regarding the coefficients due to space constraints. Also, we did not find any of the coefficients for the industries to be significant in any of the regressions in Table 5. However, when we segmented the mergers into relatively small and large deals in Table 6, we did find the coefficients for resources, life sciences, and industrial to be positive and significant in most of the regressions, suggesting higher deal premiums in relatively smaller deals in these industries. Also, the coefficients for resources and retail/packaged goods were negative and significant for larger deals in each of the regressions in Table 6. We also generally had the same industry findings when we regressed the ESG components in

Table 7. Effect of ESG component ratings on deal premium.

Response Variable 1-Day Deal Premium	Small Deals Target Env (1)	Large Deals Target Env (2)	Small Deals Acq Social (3)	Large Deals Acq Social (4)	Small Deals Target Social (5)	Large Deals Target Social (6)	Small Deals Acq Gov (7)	Large Deals Acq Gov (8)	Small Deals Target Gov (9)	Large Deals Target Gov (10)
ldealvalue	−5.722 *	−1.842	−5.310 *	−2.264	−5.469 *	−3.309	−5.056	−2.66	−6.327 *	−2.455
	(3.17)	(2.62)	(2.97)	(2.710)	(3.22)	(2.61)	(3.49)	(2.57)	(3.20)	(2.61)
percentcash	0.0272	0.115 ***	0.0795 *	0.108 ***	0.0498	0.109 ***	0.081 *	0.110 ***	0.0295	0.108 ***
	(0.05)	(0.04)	(0.04)	(0.0387)	(0.05)	(0.04)	(0.05)	(0.04)	(0.04)	(0.04)
Relative sales	0.000	0.001 **	0.000	0.001 **	0.000	0.000	−11.67 *	0.000	0.000	0.001 **
	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(6.25)	(0.00)	(0.00)	(0.00)
completed	2.475	5.239	3.828	4.368	2.69	5.282	4.310	5.245	0.89	4.973
	(7.58)	(5.04)	(7.66)	(5.01)	(7.96)	(5.48)	(7.44)	(5.51)	(7.43)	(5.26)
competitive	−0.562	12.23 **	−0.336	12.24 **	0.527	12.93 ***	0.000	13.44 ***	0.0801	12.61 **
	(6.23)	(4.78)	(6.61)	(4.91)	6.23)	(4.84)	(0.00)	(4.93)	(6.05)	(5.17)
sharesAQ	−0.004	−0.009	−0.009	−0.011	−0.014	−0.018	2.83	−0.032	0.008	−0.014
	(0.09)	(0.08)	(0.09)	(0.07)	(0.09)	(0.07)	(8.04)	(0.07)	(0.09)	(0.07)
Target Environ.	5.460 *	−2.161
	(3.166)	(2.702)
Target Environ-sq	−0.333	0.158
	(0.321)	(0.248)
Acq Social			7.531	5.291 *
			(7.706)	(3.038)
Acq Social-squared			−0.747	−0.399
			(0.797)	(0.324)
Target Social					−0.030	8.67 ***
					(4.166)	(3.10)
Target Social-sq					0.104	−0.722 *
					(0.448)	(0.401)
Acq Gover							8.030	5.226 *
							(5.271)	(3.084)
Acq Gover-squared							−0.787	−0.604 *
							(0.552)	(0.368)
Targ Gover									21.32 **	−0.441
									(8.493)	(3.523)
Targ Gover-sq									−1.896 **	0.110
									(0.800)	(0.393)
Year effects	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Industry effects	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Slope—low end	5.46	−2.63	8.55	5.3	0.1	8.0	7.9	5.22	21.3	1.75
Slope—high end	−1.18	1.62	−8.00	2.7	2.16	−5.8	−6.4	−5.40	−11.6	−0.51
Sasabuchi test stat.	0.34	0.66	1.09	0.71	-	1.11	1.21	1.48 *	2.05 **	0.13
Threshold (−β1/(2β2))/within data range	7.94/no	6.18/no	5.17/yes	6.62/yes	−0.58/no	6.0/yes	5.1/yes	4.3/yes	5.61/yes	6.72/no
95% Fieller interval for extreme point	-Inf 5.3/1.53inf	-Inf/Inf	-Inf/Inf	0.33/19.1	-Inf/Inf	-Inf/-inf	-Inf/Inf	−0.49/6.77	7.66/8.19	-Inf/Inf
Constant	55.15 **	37.63	44.02	32.93	66.23 **	30.12	81.78 **	35.72	45.85	33.44
	−27.4	−24.54	−30.43	−25.34	−28.97	−25.06	−31.97	−25.85	(29.22)	−23.91
Observations	128	156	125	154	128	156	112	154	128	156
R-squared	0.205	0.256	0.172	0.264	0.163	0.285	0.161	0.166	0.151	0.153

Robust standard errors in parentheses *** p < 0.01, ** p < 0.05, * p < 0.1.

against the deal premium.

6.3. Analysis of Small Versus Large Deals

Next we examine differences in the impact of ESG on the deal premium by deal size. We see in Table 6 the linear and quadratic coefficients along with the Sasabuchi test statistic are significant for target ESG in small and large deals. These coefficients were also significant, as shown in Columns (1), (2), (5), and (6) in

Table A4. Effect of ESG Component ratings on Deal premium (1 week premium).

Response Variable 1-Week Deal Premium	Small Deals Target Env (1)	Large Deals Target Env (2)	Small Deals Acq Social (3)	Large Deals Acq Social (4)	Small Deals Target Social (5)	Large Deals Target Social (6)	Small Deals Acq Gov (7)	Large Deals Acq Gov (8)	Small Deals Target Gov (9)	Large Deals Target Gov (10)
Log deal value	−4.351	−2.609	−5.475	−2.895	−4.095	−3.314	−5.235	−2.926	−5.072	−2.828
	(4.002)	(2.911)	(3.889)	(2.780)	(3.983)	(2.752)	(3.833)	(2.696)	(4.176)	(2.78)
Percent cash	0.0346	0.148 ***	0.0673	0.147 ***	0.0547	0.155 ***	0.0658	0.150 ***	0.0397	0.145 ***
	(0.0511)	(0.0424)	(0.0493)	(0.0415)	(0.0501)	(0.0425)	(0.049)	(0.041)	(0.049)	(0.041)
Relative sales	7.65 × 10−5	0.002 *	0.001	0.002 *	0.001	0.001	0.000	0.000	0.000	0.001 **
	(0.001)	(0.001)	((0.001)	(0.001	(0.001)	(0.001)	(0.00)	(0.00)	(0.00)	(0.00)
completed	4.616	5.366	5.254	3.961	5.425	5.955	−0.056	0.022	−0.056	0.022
	(8.027)	(5.559)	(8.148)	(5.470)	(8.387)	(5.667)	(0.119)	(0.075)	(0.119)	(0.075)
competitive	0.814	15.0 ***	1.486	15.44 ***	2.767	16.06 ***	3.482	4.816	3.482	4.816
	(6.854)	(5.069)	(7.65)	(5.093)	(6.845)	(5.029)	(8.012)	(5.732)	(8.012)	(5.732)
Shares AQ	−0.067	0.026	−0.042	0.011	−0.082	0.013	1.914	14.77 ***	1.914	14.77 ***
	(0.113)	(0.077)	(0.110)	(0.074)	(0.116)	(0.070)	(6.88)	(0.07)	(0.09)	(0.07)
Environmental	6.246 **
	(3.113)
Environmental	−0.497
	(0.319)
Acq Social			8.925	7.263 **
			(7.843)	(3.546)
Acq Social-sq			−0.830	−0.687 **
			(0.803)	(0.346)
Target Social					−0.218	10.74 **
					(4.835)	(4.709)
Target Social-sq					−0.0249	−1.095 **
					(0.502)	(0.539)
Acq Gover							4.213	6.254
							(5.379)	(4.257)
Acq Gover-sq							−0.584	−0.758 *
							(0.573)	(0.450)
Targ Gover									14.63 *	0.523
									(8.127)	(3.772)
Targ Gover-sq									−1.315 *	−0.0156
									(0.763)	(0.429)
Year effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Slope—low end	6.2	−2.0	8.9	7.3	−0.2	10.7	4.2	6.2	14.6	0.5
Slope—high end	−3.7	1.5	−7.6	−6.5	−0.7	−11.1	−6.1	−7.1	−8.2	0.3
Sasabuchi test stat.	1.04	0.59	0.90	1.71 **	-	1.96 **	0.78	1.87 **	1.52 *	-
Threshold (−β1/(2β2))/within data range	6.29/no	5.75/no	5.37/yes	5.28/yes	−4.37/no	4.90/yes	3.6/yes	4.1/yes	5.6/yes	16.7/no
95% Fieller interval for extreme point	-Inf, 4.3/-0.37, Inf	-Inf/Inf	-Inf/Inf.	1.9/94.4	-Inf/Inf.	3.9/10.2	-Inf/Inf.	1.3/36	-Inf/Inf.	-Inf/Inf.
Year effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Constant	54.04	38.68	50.67	23.02	69.18 *	18.67	66.45 *	32.15	36.41	35.05
	(33.62)	(26.25)	(37.35)	(25.93)	(35.24)	(27.14)	(34.59)	(27.03)	(32.43)	(25.14)
Observations	128	151	125	150	128	151	125	150	128	151
R-squared	0.123	0.197	0.119	0.207	0.093	0.224	0.121	0.211	0.108	0.193

Robust standard errors in parentheses *** p < 0.01, ** p < 0.05, * p < 0.1.

when using the 1- and 4-week premiums. These findings suggest that the inverted U-shaped relationship between ESG and the deal premium is similar in small and large deals. Moreover, the consistent results across 1-day, 1-week, and 4-week premiums suggest that the inverted U-shaped relationship is robust and not sensitive to the time window, reinforcing the reliability of our findings.

However, the coefficients for the difference in ESG between the target and acquirer is not significant when the deals are segmented into small and large deals. In addition, the coefficients for the linear and quadratic terms for the difference in ESG using the 1-week and 4-week premiums are also not significant. This segmentation then casts further doubt on the findings that the absolute value of the difference between the acquirer and target ESG has a significant impact on the deal premium.

Finally, the segmentation into large versus small deals appeared to make a difference when examining the impact of the difference in acquirer versus target ESG from the current versus the prior year. Here, we find that for smaller deals the linear and quadratic coefficients for the one-year difference between the acquirer and target ESG are highly significant (see Column 7, Table 6). This result indicates again that in relatively smaller deals, the one-year difference in acquirer and target ESG levels follows a U-shaped relationship with deal premiums increasing after hitting a minimum. We found similar results using the 4-week premium as shown in Column (8),

Table A3. Effect of ESG Ratings on Deal premium by Deal Size (1- and 4-week premiums).

	Small Deals	Large Deals	Small Deals	Large Deals	Small Deals	Large Deals	Small Deals	Large Deals
Response Variables 1-Week & 4-Week Deal Premiums	Premium_1-Week (1)	Premium_1-Week (2)	Premium_1-Week (3)	Premium_1-Week (4)	Premium_4-Weeks (5)	Premium_4-Weeks (6)	Premium_4-Weeks (7)	Premium_4-Weeks (8)
Log deal value	−3.463	−3.054	−3.595	−3.758	−3.548	−4.483	−3.540	−5.805 *
	(3.878)	(2.925)	(4.713)	(2.972)	(5.107)	(3.067)	(5.898)	(3.008)
Percent cash	0.0639	0.155 ***	0.0781	0.130 ***	0.108 *	0.169 ***	0.131 *	0.149 ***
	(0.0502)	(0.0421)	(0.0555)	(0.0467)	(0.0628)	(0.0449)	(0.0697)	(0.0490)
Shares AQ	−0.082	0.001	−0.091	−0.044	0.101	0.001	0.122	−0.054
	(0.115)	(0.073)	(0.127)	(0.089)	(0.132)	(0.081)	(0.150)	(0.095)
Relative sales	0.001	0.001	0.001	0.001 **	0.001	0.001 ***	0.001	0.001 ***
	(0.001)	((0.001)	(0.001)	(0.0001))	(0.001)	(0.0001)	(0.001)	(0.001)
completed	0.981	5.102	5.517	6.898	3.034	8.942	6.467	12.43 *
	(8.465)	(5.705)	(9.151)	(6.740)	(8.729)	(5.763)	(8.855)	(6.887)
competitive	−0.001	14.23 ***	1.720	17.73 ***	−7.802	14.21 ***	−9.600	17.01 ***
	(6.891)	(5.198)	(8.765)	(6.180)	(5.179)	(5.083)	(6.685)	(6.063)
ESG	32.10 **	30.91 ***			25.77 *	34.42 ***
	(12.53)	(8.334)			(13.38)	(8.478)
ESG-squared	−3.414 ***	−3.035 ***			−2.936 **	−3.430 ***
	(1.276)	(0.864)			(1.381)	(0.852)
Acquire—Target ESG (Current—Prior)			−18.81 (12.79)	2.158 (6.867)			−31.51 ** (15.70)	−2.815 (6.747)
Acquire—Target ESG (Current—Prior) -sq			6.483 (6.079)	−0.263 (1.361)			11.14 (8.091)	0.736 (1.255)
Slope—low end	32.1	30.9	−18.8	2.1	−31.5	−2.8	25.7	34.4
Slope—high end	−22.5	−17.4	56.9	−0.9	98.5	5.7	−21.2	−20.0
Sasabuchi test stat.	2.6 ***	2.9 ***	0.9	0.1	1.22	0.4	1.9 **	3.5 ***
Threshold (−β1/(2β2))/within data range	4.7/yes	5.09/yes	1.45/yes	4.1/no	1.41/yes	1.91/yes	4.39/yes	5.07/yes
95% Fieller interval for extreme point	3.6/5.4	4.4/6.2	-Inf/Inf	-Inf/Inf	-Inf, 0.7/-0.04, Inf.	-- /Inf	1.6/5.3	4.4/5.8
Year effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Constant	−5.888	−35.93	64.70 *	49.82 *	1.699	−31.08	61.41	67.88 **
	(42.65)	(29.72)	(38.20)	(28.47)	(47.53)	(30.70)	(48.87)	(28.19)
Observations	128	151	106	123	115	151	96	123
R-squared	0.128	0.223	0.125	0.229	0.116	0.242	0.142	0.265

Robust standard errors in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

; however, the coefficient for this variable was not quite significant using the 1-week deal premium as shown in Column (4), Table A2.

6.4. ESG Components, E, S, and G, and the Deal Premium

Turning to the individual ESG components, we find the coefficients for the social rating for both the acquirer and target to be significant in large deals. This result applies when using the 1-day and 1- and 4-week premiums as shown in Columns (4) and (6) in Table 7 and in Columns (4) and (6) in Table A4 and

Table A5. Effect of ESG component ratings on deal premium (4-week premium).

Response Variable 4-Week Deal Premium	Small Deals Target Env (1)	Large Deals Target Env (2)	Small Deals Acq Social (3)	Large Deals Acq Social (4)	Small Deals Target Social (5)	Large Deals Target Social (6)	Small Deals Acq Gov (7)	Large Deals Acq Gov (8)	Small Deals Target Gov (9)	Large Deals Target Gov (10)
Log deal value	−4.259	−3.723	−4.516	−4.623	−4.075	−4.963 *	−4.573	−4.418	−4.886	−4.488
	(5.096)	(3.095)	(4.870)	(2.924)	(5.138)	(2.931)	(4.685)	(2.895)	(5.392)	(2.898)
Percent cash	0.0877	0.165 ***	0.118 **	0.157 ***	0.0855	0.167 ***	0.108 *	0.161 ***	0.0908	0.157 ***
	(0.0646)	(0.0448)	(0.0581)	(0.0443)	(0.0643)	(0.0453)	(0.061)	(0.044)	(0.061)	(0.044)
Relative sales	0.000	0.001 **	0.000	0.001 **	0.000	0.000	0.000	0.000	0.000	0.001 **
	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)
completed	5.848	8.983 *	4.153	7.573	6.499	9.727	5.723	8.667	4.800	8.712
	(8.162)	(5.414)	(7.836)	(5.531)	(8.410)	(6.035)	(8.376)	(6.125)	(8.349)	(5.759)
competitive	−7.228	14.82 ***	−8.640	15.30 ***	−5.854	16.06 ***	−8.053	15.49 ***	−7.134	15.02 ***
	(5.14)	(4.95)	(5.722)	(5.142)	(5.33)	(5.355)	(6.24)	(5.486)	(5.410)	(5.275)
Shares AQ	0.108	0.026	0.124	0.014	0.107	0.013	0.132	0.0107	0.108	0.021
	(0.131)	(0.0827)	(0.123)	(0.0839)	(0.132)	(0.0745)	(0.126)	(0.077)	(0.140)	(0.081)
Environmental	3.382
	(4.142)
Environmental	−0.322
	(0.404)
Acq Social			13.88 *	8.247 ***
			(8.293)	(3.136)
Acq Social-sq			−1.122	−0.813 ***
			(0.818)	(0.296)
Target Social					3.529	11.31 **
					(6.472)	(5.015)
Target Social -sq					−0.428	−1.149 **
					(0.669)	(0.574)
Acq Gover								5.714
								(4.133)
Acq Gover-sq								−0.745 *
								(0.437)
Targ Gover									5.965	1.919
									(12.74)	(3.616)
Targ Gover-sq									−0.429	−0.160
									(1.207)	(0.407)
Year effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Slope—low end	3.4	−2.7	13.8	8.2	3.5	11.1	5.0	4.2	6.0	1.9
Slope—high end	−3.1	1.2	−8.6	−8.0	−5.0	−11.6	−6.7	−4.0	−1.5	−0.9
Sasabuchi test stat.	0.72	.52	1.0	2.44 ***	0.55	2.26 ***	0.56	1.06	0.18	0.23
Threshold (−β1/(2β2))/within data range	5.2/no	6.9/no	6.2	5.1/yes	4.1/no	4.9/yes	0.56/yes	3.7/yes	4.5	6.0/no
95% Fieller interval for extreme point	-Inf/Inf.	-Inf/Inf.	-Inf/Inf.	3.1/7.5	-Inf/Inf.	3.9/7.9	-Inf/Inf.	-Inf/Inf.	-Inf/Inf.	-Inf/Inf.
Year effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Constant	51.44	50.20 *	23.98	36.22	49.02	31.60	56.58	45.32	43.21	46.87 *
	(42.58)	(27.02)	(46.03)	(26.50)	(45.77)	(28.91)	(41.92)	(29.02)	(43.56)	(25.46)
Observations	115	151	113	150	115	151	113	150	115	151
R-squared	0.099	0.217	0.129	0.226	0.097	0.240	0.115	0.215	0.098	0.211

Robust standard errors in parentheses *** p < 0.01, ** p < 0.05, * p < 0.1.

, respectively. The same applies to the coefficient for the quadratic terms for the social rating for the acquirer and target suggesting an inverted U-shaped relationship with the deal premium as shown in Figure 6 (top row). The Sasabuchi test statistic, however, is not significant when using the 1-day deal premium. However, the test statistic is significant to the five percent level when using the one-week and four-week deal premiums as shown in Columns (3) through (6) in Table A4 and Table A5.

These results indicate that higher premiums can be paid with higher social scores both from the acquirer and target, likely due to the expected productivity improvements that the acquirer believes stems from greater employee satisfaction. This finding is particularly salient for larger transactions, perhaps because there are greater opportunities to benefit from the improvements in productivity and overall social conditions in larger deals. The inverted U-shaped relationship that we found can be explained by the too-much of a good thing theory, meaning that above a certain, maximum social score, additional benefits from the social score, perhaps in investments in human capital or shareholder activism, have a negative impact on the synergies from a deal.

Turning to governance we see in Column (9) in Table 7, the linear and quadratic coefficient for the target’s governance score is significant when using the 1-day premium in smaller deals. The Sasabuchi test statistic is also significant for the governance score providing evidence of the inverted U-shaped relationship, as shown in Figure 6 above. Our results suggest that acquirers will pay more for improved governance for the target, up to a certain threshold. Moreover, we see in Figure 6 that on average the target’s governance levels are at or near the governance level associated with the optimal deal premium. The implication is that acquirers believe there are synergies that stem from acquiring a firm with relatively low governance ratings, perhaps because acquirers believe they can improve governance and thus increase value.

Moreover, we see that the linear and quadratic governance coefficients for the acquirer are weakly significant for large deals as shown in Figure 6 and Table 7, Column (8). This finding is consistent9 when using the 1- and 4-week deal premium, as shown in Column (8) in Table A4 and Table A5. Again, this result suggests a negative relationship between the deal premium and governance ratings, when these ratings are above the average level. Finally, we see the acquirer governance level as shown by the blue line is further right of the red line, suggesting that acquirers that are highly rated for governance will pay lower premiums in larger deals.

Finally, we see the coefficient for the target’s environmental rating to be positive and weakly significant for small deals but not for large deals. Also, the coefficient for the quadratic environmental term is not significant, nor is the Sasabuchi test statistic. This result provides limited evidence that better environmental ratings lead to higher deal premiums in smaller deals and that the relationship is linear. This result was similar when using the 1-week premium but not the 4-week premium, as shown in Table A5, which again suggests that caution should be used when making inferences regarding the significance of the environmental rating.

6.5. Robustness Check for Endogeneity

To ensure the robustness of our findings, we tested for endogeneity under the notion that companies may try to improve their ESG scores or components of their ESG score in order to obtain a higher deal premium. To test for this potential source of endogeneity, we followed Zrigui et al. (2024) and performed a series of two stage least square regressions. In the first stage, we estimated the impact of industry, year, region and the interaction of these variables (industry-year10, region-industry, and region-year pairs) on target ESG. As such, the three pairs acted as instrument variables. The results of the endogeneity tests are shown in

Table 8. Two-Stage Test for Endogeneity.

	1st Stage		1st Stage		1st Stage
	Estimating		Estimating		Estimating
	ESG	1-Day Premium	ESG	1-Day Premium	ESG	1-Day Premium
ESG		11.12		36.59 *		21.88 ***
		(16.88)		(21.46)		(8.392)
ESG-squared		−2.505 ***		−2.555 ***		−2.493 ***
		(0.728)		(0.729)		(0.721)
Twenty0	−0.611 *	−5.968	−0.519 ***	7.259	−1.986 ***	0.0623
	(0.340)	(9.195)	(0.164)	(10.93)	(0.533)	(5.295)
Twenty1	−0.437 **	1.151	−0.451 ***	11.59	−0.365 **	5.936
	(0.174)	(7.818)	(0.174)	(9.898)	(0.169)	(4.932)
Twenty2	−0.379 **	4.078	−0.325 *	12.67	−0.344 **	7.853
	(0.179)	(6.993)	(0.176)	(7.818)	(0.169)	(5.168)
Twenty3	−0.477 **	5.959	−0.476 **	17.25	−0.236	11.59 **
	(0.188)	(8.948)	(0.185)	(10.56)	(0.173)	(5.360)
Industry	Not Sig	Not Sig	Not Sig	Not Sig	Not Sig	Not Sig
Residual (Year–Industry)		14.66 (15.25)
Year–Industry	Not Sig
	(0.433)
US–Finance			−0.849 ***	4.154
			(0.252)	(18.30)
US–Other Industries			Not Sig
Residual (Region–Industry)				−10.73 (20.10)
Regions					Not Sig
Twenty0—US					1.541 ***	8.782
					(0.536)	(23.91)
Twenty0—EU					1.479 ***	16.98
					(0.559)	(26.15)
Twenty0—Asia					1.164 *	25.15
					(0.637)	(25.21)
Residual (Year–Region)						3.794 (5.993)
F-test	1.35		2.01 **		4.18 ***
Observations	336	276	336	276	378	275
R-squared	0.064	0.141	0.093	0.141	0.070	0.153

Standard errors in parentheses *** p < 0.01, ** p < 0.05, * p < 0.1.

below.

In Table 8, columns (1), (3), and (5), the results from the first stage estimation are shown. In columns (3) we see the coefficient for the US-finance pair was significant. Moreover, in column (5) we see each of the year-region pairs were significant. Also, the models in columns11 (3) and (5) passed the F-test, allowing us to conclude the instruments are relevant.

Then, we see in columns (4) and (6) that the coefficients for the estimated residuals from the first stage regression were not significant in estimating the deal premium, nor were the coefficients for the instrument variables. As such, we can conclude the instruments are exogenous and that our results are robust relative to endogeneity concerns.

7. Discussion

In this section we discuss how our results compare to those from similar studies, which were highlighted in our literature review.

7.1. Results Regarding Target ESG and Difference in ESG

Our results regarding the inverted U-shaped relationship between the deal premium and the target ESG ratings fits somewhat with the findings of Brunner-Kirchmair and Wagner (2024), who used the too much of a good thing theory to explain the negative impact of the target ESG level on the stock market performance during the merger announcement window. However, we found an inverted U-shaped relationship meaning that the effect differs depending on whether the target’s ESG level is above or below a certain ESG level in the distribution. Our findings of an insignificant relationship between the acquirer ESG and the deal premium, however, conflict with Zrigui et al. (2024) who showed that the ESG score of acquirers has a negative impact on the premium paid, which they explain is due to ethical CEOs paying a lower premium.

Furthermore, our results are broadly consistent with those of Hussain and Shams (2022), who observed that differences in CSRs between acquirers and targets positively affect abnormal returns around merger announcements. However, unlike our study, they did not explore potential non-linear effects.

7.2. Results Regarding Governance and Social Scores

Our findings that social ratings have an inverted U-shaped relationship with the deal premium in large deals match somewhat with the results of Deng et al. (2013) and de Waal (2023), except these authors also do not test for non-linear relationships nor do they distinguish between relatively large and small deals. Moreover, our findings conflict somewhat with Krishnamurti et al. (2019) who found that socially responsible firms are more likely to pay a lower premium as they contend that more ethical CEOs will not pursue self-aggrandizing deals and overpay. Similarly, Zrigui et al. (2024), found social scores to have a negative impact on the deal premium in alignment with the shareholder spending theory. However, as we have often mentioned, these scholars only used a linear approach to test the effect.

Finally, our results regarding governance conflict with those of Zrigui et al. (2024), who found that the target’s governance scores did not influence the deal premium. Instead, we found the target governance scores to have an inverted U-shaped relationship with the deal premium in relatively small deals. Interestingly, we found the coefficients (both linear and quadratic term) for target governance to be insignificant when using the 4-week premium as shown in Table A4, Column (9), which is what Zrigui et al. (2024) used.

8. Conclusions

This study assesses the impact of ESG ratings, including the ratings of the target, the acquirer, the difference between the two, and the difference between the current- and prior-year ESG ratings, on the deal premium. We examined both the linear and non-linear effects using the 1-day, 1-week, and 4-week deal premiums. We also segmented the sample into relatively large and small deals. We found overwhelming evidence supporting the first hypothesis that the target ESG ratings impacted the deal premium in an inverted U-shaped relationship with the deal premium being an increasing function of the target’s ESG rating up to a certain point and then switching to a decreasing function. Our findings were consistent for relatively large and small deals and also using the 1- and 4-week deal premiums. The rationale for these results is the too much of a good thing theory.

We also found limited evidence to support the second hypothesis that the difference in the acquirer and target ESG levels had a significant impact on the deal premium, characterized by a U-shaped relationship with premiums initially falling as the difference increases and then later rising. This finding indicates that greater differences in ESG scores up to a certain point imply integration problems. However, above a certain level, increasing differences suggest opportunities to gain synergies by correcting the large differences between the companies This finding again was consistent using each of the three deal premiums.

Regarding the components, we found that social ratings for both the acquirer and target impacted the deal premium, particularly in large deals. Again, we found this relationship to be characterized by an inverted U shape. The explanation for this relationship is that the social score is a proxy for employee satisfaction and thus synergies from improved productivity. It seems that these synergies are more expected in larger deals, though we did find some evidence of this relationship in relatively smaller deals as well. Finally, we found limited evidence that the inverted U-shaped relationship exists with the acquirer’s governance rating in large deals and the target’s governance rating in small deals. These results indicate the potential synergies and, therefore, value that can be garnered from improving governance. However, this relationship only occurs at lower governance levels. The findings regarding governance and social ratings provide evidence in support of Hypothesis 4.

When considering all of these findings, one can readily see that the ESG ratings impact the deal premium, and that the non-linear model best explains the relationship. This study delved into many different segments to explore this relationship. It is left to other research to use our techniques to explore other aspects of mergers, such as the likelihood of a merger occurring or post-merger performance. Furthermore, our methodology could be used to explore differences by segment, such as geographic region or industry. Finally, case studies can be used to explore the non-linear relationships that we have found in this study.