The research, led by Xianyou Pan of Shanghai University of Electric Power together with colleagues from Basin Electric Power Cooperative and State Grid Zhoushan Electric Power Supply Company, tackles one of the most consequential puzzles in corporate climate finance. According to the 28th Conference of the Parties, global climate finance flows currently amount to only one-third of what is required to meet temperature control goals, creating what economists call a funding bottleneck. Annual investments of 2.4 trillion dollars are expected to be needed by 2030. Yet corporate willingness and efficiency in green investment remain generally low, because such projects often generate limited short-term economic returns and carry considerable uncertainty. Under resource constraints, rational entrepreneurs who favor certain returns are reluctant to proactively allocate funds to green projects, and each enterprise adopts a wait-and-see approach out of concern that it will bear the costs of the green transformation alone.

Pan and colleagues argue that this dilemma cannot be understood by treating each company as an isolated decision-maker. Corporate green investment, they contend, is embedded within a network of group decision-making comprising enterprises in the same industry, region, or supply chain. Their theoretical framework draws on collective intelligence theory—which holds that when most individuals within a group engage in coordinated action, they can generate substantial collective benefits—and organizational information processing theory, which examines how firms selectively interpret and respond to external signals. The key insight is that the investment behavior of peer enterprises constitutes an external information pool, and whether, how, and to what extent an enterprise utilizes this pool depends on its internal group decision-making mechanisms, including the board’s network for information acquisition, the information-processing structure of senior management, and the transparency of information disclosure.

To test these ideas empirically, the team assembled an unbalanced panel of 821 Chinese listed enterprises from 2010 to 2020, comprising 4,065 observations. Green investment data were manually collected from the annual reports of listed enterprises, screening expenditures directly related to environmental protection in the notes to construction in progress—projects for wastewater treatment, exhaust gas control, dust removal, energy conservation, desulfurization, and denitrification. Green investment efficiency was measured using the Richardson model, a capital allocation framework in which the absolute value of the regression residual captures the gap between actual investment and the economically optimal level conditional on enterprise fundamentals. The primary explanatory variable, Deviation, measures the absolute difference between a company’s green investment and the average green investment of its peer group, defined by industry classification according to the China Securities Regulatory Commission’s standards. The analysis employed fixed-effects models with individual and year fixed effects, alongside system GMM estimators to address dynamic endogeneity.

The benchmark results reveal a striking asymmetry. The coefficient on peer green investment deviation is significantly positive at the 1 percent level (0.0033), supporting the hypothesis that peer effects enhance corporate green investment efficiency. In economic terms, a one-standard-deviation increase in deviation leads to approximately a 0.0039 change in the green investment inefficiency index, corresponding to roughly 26 percent of the mean inefficiency during the sample period—indicating that peer convergence behavior can alleviate about one-quarter of the efficiency deviation. When the team decomposed the effect, they found that negative peer deviation is significantly positive (0.0034, p < 0.01), whereas positive peer deviation is statistically insignificant. In other words, efficiency losses occur only when peer enterprises’ green investment falls below that of the focal enterprise, suggesting the peer effect operates through a directional information-learning mechanism rather than simple overinvestment.

The most provocative finding concerns whom companies choose to imitate. The team classified enterprises into leaders (the top third) and followers (the bottom two-thirds) along two dimensions: the scale of green investment and corporate social responsibility scores. When followers simply replicated the visible green-investment scale of leaders, the coefficient was insignificant, suggesting that such simple replication does not improve efficiency. By contrast, when followers learned from CSR-oriented leaders, the effect was significantly positive (0.5660, p < 0.01), and this promoting effect even exceeded the peer effect within the CSR leader group itself (0.4520, p < 0.01). The interaction term test further indicated that the follower group exhibits significantly higher sensitivity to peer deviation than the leader group. The explanation, the authors argue, is that green-investment-scale leaders may be motivated by non-economic considerations such as policy compliance and image-building; followers who copy the visible scale without discerning the underlying rationale are prone to blind imitation, triggering a bandwagon phenomenon that crowds out genuinely efficient green investments. CSR leaders, however, reflect more robust long-term strategies and stakeholder relationships, and learning from them constitutes rational information learning through which followers can effectively optimize investment allocation.

The study also identifies the conditions under which peer effects translate into efficiency gains, examining three sequential stages of information learning. Information transparency was proxied by the KV index, a market-based measure of corporate disclosure quality in which a smaller value indicates more comprehensive disclosure. Information acquisition capability was measured by the betweenness centrality of independent directors within the director network, capturing the structural-hole positions through which directors access heterogeneous information. Information analysis capability was proxied by the number of executives with environmental backgrounds, identified by resume keywords such as ecology, low carbon, sustainability, energy conservation, and clean energy. All three interaction terms were significant in the expected direction: higher transparency, stronger acquisition capability, and greater analytical capability each amplified the positive impact of peer effects on green investment efficiency. Notably, the main effect of information analysis capability was positive and significant, while transparency and acquisition capability exerted only indirect influence—consistent with the theoretical expectation that environmentally versed executives can directly optimize green investment decisions while also strengthening the interpretation of peer signals.

The team conducted extensive robustness checks to distinguish the information-learning mechanism from regulatory imitation. The interaction term between industry deviation and the intensity of environmental regulation was significantly positive (4.739, p < 0.01), and remained so when regulation was lagged by one period (2.989, p < 0.01), ruling out reverse causality. The authors reason that if the peer effect were solely attributable to passive imitation by enterprises seeking to comply with regulatory requirements, environmental regulation should not reinforce the efficiency-enhancing effect of the peer effect. Cross-validation using a VRS-DEA output-oriented model that incorporates both environmental and economic performance dimensions confirmed the benchmark findings (0.0161, p < 0.05), although a pure green DEA model relying solely on a single environmental output dimension failed to achieve statistical significance, suggesting that a single environmental indicator is insufficient to capture the economic-environmental trade-offs inherent in green investment efficiency. After controlling for geographical-level peer effects, the industry deviation coefficient remained significant while the geographical coefficient did not, indicating that technological trajectories and institutional pressures among enterprises in the same industry are the dominant channels through which peer effects operate.

The findings carry pointed implications for both policymakers and corporate managers. From the government perspective, the authors recommend leveraging the guiding role of leading enterprises by establishing model and pilot enterprises to stimulate other companies in the same industry to actively participate in green investment, while strengthening oversight mechanisms involving the government and the public to prevent the compliant bandwagon phenomenon. From the corporate governance perspective, enterprises should enhance their information coordination capabilities through disclosure, acquisition, and analysis, fostering rational information learning behaviors and preventing blind following of trends. The authors also acknowledge limitations: the Deviation variable cannot fully distinguish whether convergence stems from active information learning or passive institutional imitation, the reflection problem and common-shock problem inherent in peer-effect research are difficult to fully resolve with existing methods, and peer relationships defined by industry classification fail to capture supply chain linkages and interlocking director networks. Future research leveraging exogenous policy shocks, such as environmental regulations or green finance pilot programs rolled out in batches, would help disentangle these mechanisms.

As climate finance flows remain far short of what is needed, the study offers a sobering message: the aggregate quality of green capital allocation may depend less on how much companies invest and more on how they learn from one another. In a group decision-making environment with excessive information noise, managers can easily form a false consensus and trigger the bandwagon phenomenon, leading to resource misallocation and overcapacity. But when enterprises possess the internal information-processing capabilities to observe, acquire, and analyze peer signals rationally, the collective intelligence of the market can compensate for the shortcomings of policy pressures and incentives, amplifying investment variability and steering green capital toward genuinely efficient uses. The difference, it turns out, is not imitation itself but the sophistication of the learner.

Subject of Research: Enhancement mechanism of corporate green investment efficiency from the perspective of group decision making

Article Title: Enhancement mechanism of corporate green investment efficiency from the perspective of group decision making

Article References: Pan, X., Wu, X., Ma, J., Lu, H., & Liu, B. (2026). Enhancement mechanism of corporate green investment efficiency from the perspective of group decision making. Frontiers in Environmental Economics, 5, Article 1876957. https://doi.org/10.3389/frevc.2026.1876957

Image Credits: AI Generated

DOI: 10.3389/frevc.2026.1876957

Keywords: Enhancement, mechanism, corporate, green, investment, efficiency, perspective, group, decision, making, scientific research