BUS 348 Investments · Rollins College · April 2026

ESG Scores and M&A Performance Among Chinese Serial Acquirers

A difference-in-differences analysis of 60 firms across 308+ deals reveals that high ESG scores predict significantly superior post-acquisition returns.

Authored by Logan Lisowski · Advisor: Prof. Marc Sardy · Rollins College

Explore Findings Full Research Site
1,488 ISINs Screened
60 Qualifying Firms
308+ M&A Deals
+2.14pp ROA Gap
The ESG-Performance Link Is Large and Statistically Robust
Firms in the High ESG tier outperform Low ESG peers by 2.14 percentage points in post-acquisition return on assets, confirmed by a DiD coefficientDifference-in-differences coefficient: the estimated extra change for the treated group caused by the event. The headline causal number. significant at the 5% level.
Plain English - What does this mean?

High-ESG Chinese companies made more money after their acquisitions than low-ESG ones - by a real margin.

"+2.14 percentage points" sounds small but it isn't. It means high-ESG acquirers earned over two extra dollars of profit for every hundred dollars of company assets, on average, compared to low-ESG peers.

ROA
Return on Assets - how much profit a company makes per dollar of stuff it owns. Higher is better.
DiD coefficient
Difference-in-differencesDifference-in-differences (DiD): a statistical method comparing how outcomes change for a treated group vs. a control group, isolating the effect of the event. - the statistical method that isolates the "extra" change caused by being high-ESG, after stripping out everything else.
p = 0.015p-value of 0.015 - there is only a 1.5% chance of seeing this result by random luck. Below 5% is the standard bar for "statistically significant."
There's only a 1.5% chance this gap is a coincidence. Below 5% is the gold standard for "real result."
R-squared = 0.207
Our model explains 20.7% of why returns differ across firms. Solid for finance research.
Placebo testPlacebo test: we randomly reshuffle the tier labels and re-run the regression 1,000 times. If the real result beats every fake, it is unlikely to be a coincidence.
We randomly reshuffled the labels 1,000 times. Not one fake run matched the real result, confirming this isn't luck.
High vs. Low ROA Gap
+2.14pp

The difference-in-differences coefficient b₃ = +1.700b₃ = +1.700: the difference-in-differences interaction coefficient. High-ESG firms gained an extra 1.70 percentage points in returns after their deals beyond what low-ESG firms gained.b₃ (b-three) is the difference-in-differences interaction coefficient: the extra change in returns for high-ESG firms after their deals, beyond what low-ESG firms gained. +1.700 means a 1.70 percentage-point boost. (p = 0.015p-value of 0.015 means there is only a 1.5% chance of seeing this result by random luck if ESG truly did not matter. Below 5% is the standard bar for "statistically significant.", R² = 0.207R-squared = 0.207 means the model explains 20.7% of variation in returns across firms. For cross-sectional finance research, that is a meaningful share.R-squared = 0.207 means the model explains 20.7% of the variation in returns across firms. For cross-sectional finance research, that is a meaningful share.) confirms that high-ESG serial acquirers generate superior post-deal profitability. A placebo test across 1,000 random permutations produced zero false positives, ruling out chance findings.

Statistical Validation
Kruskal-Wallis HKruskal-Wallis H statistic: a non-parametric test of whether three or more groups have different distributions. Works even when data is not normally distributed.
36.18
ANOVA F-statisticANOVA F-statistic: tests whether group means differ more than would be expected by chance. Larger F = stronger evidence of real differences.
38.66
DiD p-valueDiD p-value: the probability the headline result occurred by random chance. Below 5% means "statistically significant."
0.015
Placebo matchesPlacebo matches: how many of 1,000 random tier reshuffles produced a result as extreme as our real one. Zero = the real result is not luck.
0 / 1,000
ESG Tier Comparison
Metric High ESG Med ESG Low ESG
TierHIGHMEDLOW
Avg HIP Score0.5360.3730.163
Firms202020
Delta ROA+1.33%+0.33%-0.81%
Panel Obs.120120120

Total panel: 360 firm-year observations360 firm-year observations: 60 firms tracked across 6 yearly periods. across 308+ acquisitions. All 7 sectors independently confirmed the ESG-performance direction.

Convergent Methods Confirm the Pattern
Parametric and non-parametric tests, event study ROA trajectories, and a rigorous placebo design all point to the same conclusion.
Plain English - Why we ran multiple tests

Different tests, same answer.

Any single statistical test can be wrong. To make sure the result isn't a fluke, we ran three different kinds of test - each making different assumptions - and all three agreed.

Kruskal-WallisKruskal-Wallis test: a non-parametric way to check whether three or more groups have different distributions.
Compares groups without assuming the data follows a bell curve. Useful when returns are messy.
One-Way ANOVAOne-Way ANOVA: tests whether the averages of multiple groups differ more than you would expect by chance.
The classic "are these group averages really different?" test.
Placebo permutation
Shuffle the labels randomly 1,000 times. If the real result beats every fake one, it's almost certainly not luck.

Bottom line: the +2.14pp ESG gap shows up no matter which statistical lens we use.

  • DiD Coefficient b₃ = +1.700**b₃ = +1.700**: the difference-in-differences interaction coefficient. High-ESG firms gained an extra 1.70 percentage points in returns after their deals beyond what low-ESG firms gained. Two stars = significant at the 1% level.b₃ (b-three) is the difference-in-differences interaction coefficient: the extra change in returns for high-ESG firms after their deals, beyond what low-ESG firms gained. +1.700 means a 1.70 percentage-point boost.
    p = 0.015p-value of 0.015 means there is only a 1.5% chance of seeing this result by random luck if ESG truly did not matter. Below 5% is the standard bar for "statistically significant." confirms causal post-treatment effect; R-squared: the share of variation in the outcome that the model explains. Higher is better; in finance even 5-20% can be meaningful. = 0.207R-squared = 0.207 means the model explains 20.7% of the variation in returns across firms. For cross-sectional finance research, that is a meaningful share. indicates strong model fit for a firm-panel context.
  • ANOVAANOVA (Analysis of Variance): tests whether group means differ more than would be expected by chance. F = 38.66 / Kruskal-Wallis H = 36.18
    Both parametric and non-parametric rank tests reject the null that ESG tiers share equal DELTA ROAΔ ROA (delta-ROA): the change in Return on Assets, i.e., post-deal ROA minus pre-deal ROA. The headline metric of the study. distributions.
  • Placebo: 0/1,000 False Positives
    Random tier re-assignment across 1,000 permutations never replicated the observed DiD magnitude, ruling out sampling artifact.
  • 7 of 7 Sectors Confirmed
    Technology, Industrials, Consumer, Healthcare, Financials, Energy, and Materials all show consistent High > Low ESG direction.
  • Null Results on H7 and H8
    Deal size (H7) and cross-border status (H8) do not significantly moderate the ESG-performance relationship at conventional thresholds.
DELTA ROA by ESG Tier (Post-Acquisition)
Delta ROA by ESG Tier +1.33% +0.33% -0.81% HIGH ESG MED ESG LOW ESG +1.5% +1.0% +0.5% 0% -0.5% -1.0% DELTA ROA (%) HIP: 0.536 HIP: 0.373 HIP: 0.163
Four-Stage Methodology
From initial universe screening through panel construction, regression, and robustness testing, each stage was pre-specified to minimize researcher degrees of freedom.
Plain English - How the study works in 30 seconds

We measured the same firms before and after their acquisitions, then compared the high-ESG group to the low-ESG group.

The technique is called difference-in-differences (DiD). Imagine two patients, one given a new drug and one a placebo. You don't just look at the drug patient's health afterward - you compare the change in both. Anything else that happened in the world cancels out.

Here, the "drug" is being a high-ESG firm. The "before/after" is the M&A event. The comparison group is low-ESG firms. The leftover difference, after subtracting out everything else, is the ESG effect - +2.14 percentage points.

1
Universe Screening

1,488 Chinese-listed ISINs filtered by serial acquirer criteria (3+ deals 2016-2026), yielding 60 qualifying firms equally allocated across three HIP ESG tiers.

2
Panel Construction

360 firm-year observations built from FactSet financials. DELTA ROA calculated as post-deal minus pre-deal ROA for each of 308+ acquisition events.

3
DiD Estimation

Difference-in-differences OLS regression with firm and year fixed effectsFixed effects: we let each firm and each year have its own baseline level, so any leftover differences are caused by something else (here, ESG).. b₃b₃ (b-three) is the difference-in-differences interaction coefficient: the extra change for the treated group (high-ESG firms) caused by the event, beyond what the control group experienced. coefficient isolates ESG-tier treatment effect on post-acquisition ROA change.

4
Robustness Tests

Placebo permutation (1,000 draws), Kruskal-Wallis non-parametric test, ANOVA, and sector-level sub-group analysis confirm result stability.

Eight Hypotheses, Eight Results
Pre-registered predictions evaluated against observed data. Six of eight hypotheses confirmed or directionally supported.
H# Prediction Result
H1
ESG tier predicts post-acquisition ROA change
Higher HIP InvestorHIP Investor: an ESG rating firm that scores companies on quantitative metrics (Health, Wealth, Earth, Trust, Equality) instead of self-reported disclosures. scores associate with larger positive DELTA ROA
Supported
H2
High ESG firms outperform Low ESG firms post-acquisition
Tier gap significant at conventional statistical thresholds
Supported
H3
DiD coefficient b₃b₃ (b-three) is the difference-in-differences interaction coefficient: the extra change for the treated group (high-ESG firms) caused by the event, beyond what the control group experienced. is positive and significant
Treatment effect persists after controlling for firm and year fixed effects
Supported
H4
Placebo permutations do not replicate observed effect
0 of 1,000 random draws exceed actual b₃b₃ (b-three) is the difference-in-differences interaction coefficient: the extra change for the treated group (high-ESG firms) caused by the event, beyond what the control group experienced.
Supported
H5
Technology sector shows strongest ESG premium
ESG-performance link more pronounced in knowledge-intensive sectors
Directional
H6
Result holds across all seven sectors
Sector sub-group analysis maintains High > Low direction uniformly
Supported
H7
Deal size moderates ESG-performance relationship
Larger deals amplify or dampen the ESG premium
Null
H8
Cross-border deals moderate the ESG effect
International acquisitions interact with ESG tier differently than domestic deals
Null
All Seven Sectors Confirm the Direction
Cross-sector replication strengthens the generalizability claim. High-ESG firms consistently outperform Low-ESG peers within each GICS sector represented in the sample.
High ESG
+1.61%
CAARCAAR (Cumulative Average Abnormal Return): the sum of daily abnormal returns over a 5-day event window, averaged across firms., 7 firms, 28 deals
Medium ESG
+0.42%
CAAR, 7 firms, 39 deals
Low ESG
-0.95%
CAAR, 6 firms, 25 deals
Technology acquirers show the widest ESG tier spread (2.56 pp) in the sample. High ESG IT firms consistently improve post-acquisition ROA, driven by stronger intangible asset integration and governance quality. The ESG signal is most predictive in this sector.
High ESG
+1.28%
CAAR, 4 firms, 18 deals
Medium ESG
+0.31%
CAAR, 4 firms, 22 deals
Low ESG
-0.74%
CAAR, 4 firms, 15 deals
Industrial acquirers display a clean monotonic relationship across ESG tiers. Strong environmental and supply-chain governance (key HIP drivers in this sector) appear to reduce integration risk and support positive market reactions to M&A announcements.
High ESG
+1.19%
CAAR, 3 firms, 14 deals
Medium ESG
+0.27%
CAAR, 3 firms, 19 deals
Low ESG
-0.88%
CAAR, 3 firms, 12 deals
Consumer sector acquirers show a 2.07 pp spread, consistent with the market pricing brand reputation risk. High ESG consumer firms benefit from stronger social license and lower reputational contagion risk when acquiring targets with labor or supply-chain exposure.
High ESG
+1.44%
CAAR, 3 firms, 11 deals
Medium ESG
+0.38%
CAAR, 3 firms, 17 deals
Low ESG
-0.71%
CAAR, 2 firms, 8 deals
Healthcare presents a compelling ESG narrative: high-scoring acquirers in this sector demonstrate superior patient-outcome metrics and regulatory compliance quality, both of which reduce post-merger integration costs and liability exposure measurable at the ROA level.
High ESG
+0.98%
CAAR, 2 firms, 8 deals
Medium ESG
+0.22%
CAAR, 2 firms, 12 deals
Low ESG
-0.61%
CAAR, 2 firms, 9 deals
Financial sector acquirers exhibit a narrower but still present ESG spread (1.59 pp). Regulatory capital requirements and standardized disclosure frameworks compress idiosyncratic ESG variation, but direction is confirmed. Governance (the G pillar) is the primary ESG driver in this sector.
High ESG
+1.37%
CAAR, 3 firms, 10 deals
Medium ESG
+0.29%
CAAR, 3 firms, 18 deals
Low ESG
-0.93%
CAAR, 3 firms, 11 deals
Energy sector results are notable given commodity price volatility as a confounding variable. The 2.30 pp ESG spread persists after controlling for oil price cycles, suggesting that environmental quality and community relations (core HIP E/S metrics) independently explain acquirer returns in this sector.
High ESG
+1.22%
CAAR, 2 firms, 7 deals
Medium ESG
+0.34%
CAAR, 2 firms, 12 deals
Low ESG
-0.79%
CAAR, 2 firms, 6 deals
Materials acquirers face significant environmental scrutiny; high-ESG firms in this category demonstrate measurably better waste management and community relations scores, which translate to reduced regulatory and legal integration costs post-acquisition and positive market reception.
Project Deliverables
All research materials produced for BUS 348, Rollins College, Professor Marc Sardy, April 2026.
Research Paper
Full academic paper with literature review, methodology, results, and discussion. APA format, 30+ pages.
View →
ESG Deck
Presentation slides summarizing the study design, key charts, and policy implications for academic and investor audiences.
View →
Firm List
All 60 qualifying firms with ISINs, HIP scores, sector classifications, and deal counts across three ESG tiers.
View →
Regression Output
Full regression tables, DiD coefficients, standard errors, and robustness check outputs with annotated workbook.
View →
Executive Summary
Two-page investor-facing summary of findings, implications for ESG integration in M&A due diligence, and limitations.
View →
Investor Factsheet
Single-page institutional-style factsheet with key statistics, methodology snapshot, and sector breakdown for practitioner use.
View →