The effects of heterogeneous CSR on corporate stock performance: evidence from COVID-19 pandemic in China

The paper examines whether corporate social responsibility (CSR) mitigated the adverse effects of the COVID-19 pandemic on corporate stock performance (CSP) in China and how heterogeneous CSR—responsive versus strategic—differentially influenced short- and long-term stock outcomes. Motivated by the unprecedented volatility and drawdowns in global equity markets during early 2020 and by the established view of CSR as a risk-mitigation tool, the study investigates CSR’s moderating role on the relationship between pandemic intensity and stock returns. It posits that CSR can cushion firms via reputation, immunisation, and sustainable development channels, yet prior work seldom distinguishes CSR types or assesses their timing (short- vs long-term) effects during COVID-19. The authors hypothesize that overall CSR mitigates the negative impact of COVID-19 on CSP, with responsive CSR affecting short-term returns and strategic CSR affecting long-term return recovery, thereby contributing novel evidence from China’s A-share market during the lockdown period.

The literature underscores that pandemics impose substantial negative shocks on stock markets, with volatility and declines linked to case counts and policy news. CSR has been documented as a risk-management and insurance-like mechanism that can enhance resilience, improve stakeholder relations, reduce financial and operational risks, and support long-term value creation. However, most studies treat CSR as homogeneous. Building on Carroll (1979, 2004), Porter and Kramer (2006, 2011), and subsequent work, the paper differentiates responsive CSR (short-term, compliance- or community-oriented actions such as donations) from strategic CSR (long-term, embedded in business strategy, innovation, and competitive advantage). Prior evidence suggests responsive CSR may yield immediate reputational and stakeholder benefits, while strategic CSR supports sustained performance and adaptability. During COVID-19, firms engaging in responsive actions often saw immediate market recognition, whereas strategic CSR enhanced longer-run resilience and investor confidence. Hypotheses: H1: General CSR activities mitigate the negative impacts of COVID-19 on CSP. H2: In the short term, responsive CSR significantly impacts stock performance (stock returns). H3: In the long term, strategic CSR significantly impacts stock performance (recovery of stock returns).

Data: The study integrates four sources for Chinese A-share listed firms, excluding financials. (1) Weekly COVID-19 case counts from China’s National Health Commission covering 3 Feb 2020 (first trading day post Spring Festival) to 10 Apr 2020 (10 weeks, lockdown window). (2) Stock prices and firm characteristics from CSMAR. (3) Corporate donation data during 3 Feb–10 Apr 2020 from the China Association of Public Companies. (4) CSR ratings from the Rankings CSR Ratings (RKS) database. The initial sample covers 3,680 firms across 48 industries (manufacturing further split into 31 two-digit sub-industries). After removing missing values and winsorizing variables at 1% tails, the main weekly-return regressions include 33,215 firm-week observations. For return recovery analyses, the window is extended to 2019 and the second half of 2020, yielding 33,165 observations after computation-related losses. Empirical model: The authors estimate OLS regressions with robust standard errors clustered at the firm level and include week and industry fixed effects. Core specifications model weekly adjusted stock returns (WkRtn%) and weekly stock return recovery (RtnRecovery) as functions of pandemic intensity (LogCOVID19), CSR measures, and their interactions, along with pre-2020 firm-level controls. Interaction terms capture CSR’s moderating effects on the COVID-19–CSP relationship. Analogous daily return and recovery models are used for robustness. Key variables: - Weekly Stock Return (WkRtn%): Firm’s weekly stock return adjusted by the corresponding market return, from last trading day of week t-1 to last trading day of week t. - Weekly Stock Return Recovery (RtnRecovery): Change in average weekly adjusted return during COVID-19 relative to the following 6 months, scaled by the average weekly return in the 3 months prior to COVID-19. - COVID-19 intensity (LogCOVID19): Log of newly confirmed cases in week t (and daily version for daily models). - CSR measures: CSR_total (fulfilled CSR before/during epidemic), StrategicCSR (indicator that firm disclosed CSR reports prior to the epidemic), ResponsiveCSR (indicator that firm donated during the epidemic). - Controls: Firm age (Age), size (Size), leverage (Leverage), ROA, cash holdings (Cash), market-to-book (MtB), growth (Growth), state ownership (SOE).

Descriptive statistics indicate substantial dispersion in returns and recovery across firms during the period (e.g., mean weekly return 0.436% with SD 7.36; mean recovery 4.974% with SD 14.922). Correlations and VIF diagnostics suggest no multicollinearity concerns among main regressors. Main regressions (weekly returns): COVID-19 case intensity significantly depresses weekly returns (e.g., LogCOVID19 coefficient around −0.60, p<0.01). CSR moderates this effect: the interaction CSR×LogCOVID19 is positive and significant (e.g., ≈0.018, p<0.05), indicating that higher CSR attenuates COVID-19’s adverse impact on returns. When decomposing CSR, short-term moderating effects are primarily via responsive CSR, consistent with H2, while strategic CSR’s short-term moderating effect is not robust. Return recovery (weekly): CSR moderates the relationship between pandemic intensity and subsequent recovery; the interaction terms with total CSR are significant and positive. Heterogeneity shows that the long-run moderating effect is driven by strategic CSR (consistent with H3), whereas responsive CSR’s effect on recovery is weak or insignificant. Robustness (daily data): Using daily returns, total CSR and responsive CSR exhibit significantly positive interactions with COVID-19 intensity (e.g., CSR×LogCOVID19day ≈0.061***; ResponsiveCSR×LogCOVID19day ≈0.134***), supporting short-term cushioning. For daily return recovery, both total and strategic CSR show stronger positive moderating effects than responsive CSR (e.g., CSR×LogCOVID19day ≈0.531***; StrategicCSR×LogCOVID19day ≈0.916***; ResponsiveCSR interaction not significant), reinforcing that strategic CSR primarily supports longer-term recovery. Sub-sample heterogeneity: CSR’s moderating effects are more pronounced among non-state-owned enterprises (SOEs show weaker effects), firms with lower financial leverage (responsive CSR in the short term; strategic CSR in recovery), and larger firms (both responsive and strategic CSR influence short-term returns; strategic CSR drives recovery). The paper also notes stronger effects outside medical/Hubei contexts. Overall, findings support H1–H3: CSR mitigates pandemic-induced stock underperformance, with responsive CSR salient in the short term and strategic CSR in the long term.

The results answer the core question by demonstrating that CSR acts as an effective risk-mitigation mechanism during large negative shocks. The moderating effects align with theorized channels: reputational capital and stakeholder support enhance short-run resilience (responsive CSR), while strategic alignment with sustainability, innovation, and long-term value creation supports recovery post-shock (strategic CSR). The stronger effects in non-SOEs, lower-leverage, and larger firms suggest investor expectations and visibility shape CSR’s capital market impact. Collectively, the evidence implies that firms can reduce exposure to systemic shocks by deploying responsive CSR for immediate cushioning and strategic CSR for durable recovery, enriching the CSR–CSP literature with a nuanced, time-differentiated perspective.

The paper contributes by: (1) establishing that CSR attenuates the negative effect of COVID-19 on stock returns and recovery; (2) distinguishing heterogeneous CSR types, showing responsive CSR delivers short-term market support while strategic CSR underpins long-term return recovery; and (3) revealing stronger moderating effects among non-state-owned, lower-leverage, and larger firms. Managerial implications include: treating CSR as a risk-management investment; adopting immediate responsive CSR actions during crises to stabilize returns; committing to strategic CSR to accelerate recovery; and enhancing CSR disclosure to bolster reputational benefits and stakeholder trust. Future research should refine measurement of CSR heterogeneity and further examine the causal roles of strategic versus responsive CSR across different shock types and institutional settings.

The study acknowledges limitations in measuring heterogeneous CSR: responsive CSR is proxied by donations during the epidemic and strategic CSR by pre-epidemic CSR disclosure/engagement, which may not fully capture the constructs’ breadth and depth. Additionally, the generalizability of the moderating roles of strategic and responsive CSR to future shocks and different contexts requires further validation.