Institutional Cross-Ownership and Corporate Sustainability Performance: Empirical Evidence Based on United Nations SDGs Ratings

Abstract

Corporate sustainable development, as a critical component of Chinese-style modernization, is essential for achieving high-quality economic growth, yet the influence of institutional cross-ownership—a prevalent phenomenon in stock markets—on corporate sustainability performance remains contested. Using a sample of Chinese A-share listed companies from 2012 to 2023, this study innovatively employs micro-level data on the degree of the achievement of the United Nations Sustainable Development Goals (SDGs) to measure corporate sustainability performance and investigate the influence of institutional cross-ownership on corporate sustainability performance. This study presents the following findings: (1) Institutional cross-ownership undermines corporate sustainability performance, a finding that remains robust to a series of endogeneity and robustness tests. (2) Mechanism analysis reveals a triple erosion effect: short-termism driven by institutional investors’ preference for immediate financial returns, market power through cross-ownership that dampens competitive pressures, and reduced green innovation investments that weaken sustainability. (3) This negative effect is more pronounced in firms located in high-productivity regions or central and eastern China, in firms facing lax environmental regulations, and in state-owned enterprises. (4) The impact of cross-ownership on sustainability performance varies across dimensions, with the negative effects concentrated in the economic and social dimensions. This study enriches the literature on the factors influencing corporate sustainability performance, providing new empirical evidence for governments to guide institutional investors in long-term value investment and firms to implement effective sustainable development strategies.

1. Introduction

Promoting corporate sustainable development is crucial for achieving high-quality economic growth. During the 2024 Two Sessions (the National People’s Congress and the Chinese People’s Political Consultative Conference), China emphasized the necessity of establishing a set of sustainable development guidelines, standards, and evaluation mechanisms suitable for the characteristics of domestic enterprises. These measures are designed to incentivize Chinese enterprises to engage more proactively in sustainable development practices, ultimately forming a sustainable development pattern with Chinese characteristics. Current research mainly focuses on the benefits of corporate sustainable development, including increasing firm value and financial performance [1,2], enhancing market advantage [3], and mitigating default risks [4]. However, improving corporate sustainability performance remains a critical yet underexplored imperative, as it is critical for advancing the new quality productivity of firms and ensuring sustained well-being across economic, social, and environmental domains.

The relevant literature explores the driving factors of corporate sustainability performance from the internal firm-level perspective and the external policy-driven perspective. On the one hand, the internal firm-level perspective demonstrates that corporate internal factors significantly enhance corporate sustainability performance, including green technology innovation, digital transformation, and board diversity [5,6,7]. On the other hand, the external policy-driven perspective suggests the government interventions promote corporate sustainable development through regulatory instruments like environmental regulation, low-carbon city pilot policy, green credit policy, environmental tax policy, and carbon emissions trading pilot policy [8,9,10,11,12].

However, most existing studies focus on macro-policies and pay insufficient attention to micro-mechanisms. In fact, micro-policies such as institutional cross-ownership—ownership by institutional investors that simultaneously hold significant stakes (at least 5% ownership) in at least two firms in the same industry—also play an important role in advancing corporate sustainability performance. The rapid development of the market economy increases capital accumulation and facilitates large-scale interest groups. In this context, the phenomenon of institutional cross-ownership has become increasingly frequent [13]. This naturally raises the question in this paper: what role does institutional cross-ownership play in corporate governance? Does it have an impact on corporate sustainability performance? Existing research presents conflicting evidence. On the one hand, institutional cross-shareholders are able to reduce information asymmetry by assigning directors and participating in shareholders’ meetings and therefore contribute to the enhancement of corporate sustainability performance by alleviating financing constraints and improving corporate governance [14,15]. On the other hand, cross-ownership may increase resource sharing and enhance corporate market monopoly position, which may weaken the competition among the affiliated firms and their incentives for sustainable development investment [16,17]. There is a lack of consensus on the role of institutional cross-ownership in influencing corporate sustainability performance.

Notably, the unique institutional background of China’s stock markets may amplify the double-edged sword effects of cross-ownership. On the one hand, the concentrated ownership structure—where central state-owned enterprises account for over 40% of A-share market capitalization—could incentivize institutional investors to form interest coalitions through cross-ownership, thereby reducing their motivation to monitor management. On the other hand, significant regional disparities in environmental law enforcement efficiency imply that institutional investors may acquiesce to corporate cuts in sustainable investments for short-term returns when external regulatory constraints weaken. This institutional configuration of “weak regulation coupled with strong concentration” makes cross-ownership in China more likely to trigger collusion rather than supervision. Furthermore, the 2022 Opinions on Accelerating the Construction of a National Unified Market by the CPC Central Committee and the State Council, which emphasizes anti-monopoly measures and market competition normalization, serves as a targeted response to such institutional risks. Consequently, investigating the mechanisms through which institutional cross-ownership affects corporate sustainability performance holds practical urgency for identifying critical policy intervention points.

Therefore, it is imperative to examine the impact of institutional cross-ownership on corporate sustainability performance, which holds great significance for the implementation of “Chinese-style modernization” by firms. Based on the United Nations Development Goals (SDGs) rating data of Chinese A-share listed companies from 2012 to 2023, this study provides a comprehensive examination of the relationship between institutional cross-ownership and corporate sustainability performance. The findings indicate that institutional cross-ownership significantly reduces corporate sustainability performance, attributed to the short-term orientation of institutional funds, the market power resulting from inter-firm collusion, and the reduced levels of corporate green innovation caused by institutional cross-ownership. This negative effect is particularly significant in firms located in high-productivity areas or central and eastern China, in firms facing lax environmental regulations, and in state-owned enterprises. Furthermore, the impact of cross-ownership on sustainability performance varies across dimensions, with the negative effects primarily observed in the economic and social dimensions.

This study contributes to the literature in two ways. First, it complements existing studies by utilizing SDGs data rather than ESG data to measure corporate sustainability performance. Compared with ESG performance, the SDGs encompass the “5P” dimensions: Planet, People, Peace, Prosperity, and Partnership [18], which not only supplement the ESG evaluation criteria but also align more closely with the values conveyed by the UN 2030 Agenda. Furthermore, the great differences in the selection of ESG contribution indicators and the level of contribution measured by different assessment agencies lead to divergent ESG ratings, causing confusion for investors, companies, and regulators [19,20]. As a comparison, the SDGs scoring system shows distinct advantages since the SDGs are categorized into 17 sub-goals and 169 specific goals that can be monitored and reported, thereby partially alleviating the problem of ESG evaluation criteria.

Secondly, this study enriches the literature on the factors influencing corporate sustainability performance. Considering the profit maximization goal, this study suggests that institutional cross-ownership facilitates inter-firm collusion to increase corporate market power, thereby reducing their incentives to enhance sustainability performance as a competitive advantage. Further analysis uncovers the underlying mechanisms behind this negative impact, driven by three factors: the short-termism of institutional investors, market power, and the decline of green innovation. This study deepens corporate understanding of the intrinsic motivation of sustainability decisions.

The subsequent sections of this study are structured as follows: Section 2 reviews related literature and proposes the hypotheses. Section 3 introduces data sources and empirical methodology. Section 4 presents the empirical findings. Section 5 provides the conclusion of the study.

2. Literature Review and Hypothesis Development

2.1. Institutional Cross-Ownership and Corporate Sustainability Performance

As one of the behavioral subjects of sustainable development, corporate entities serve as the micro-level foundation for China’s pursuit of high-quality economic development. Therefore, it is imperative to comprehensively enhance their sustainability performance. Previous studies on the determinants of corporate sustainability performance primarily focus on two aspects: corporate strategy and corporate governance. From the macro perspective of corporate strategy, corporate green transformation is identified as a critical driver of sustainable development. Empirical research suggests that specific measures, such as the promotion of green innovation [21], the adoption of green-oriented strategies [22], and the implementation of green training programs [23], significantly improve corporate sustainability performance. In addition to internal strategies, external institutional pressures, such as low-carbon city pilot policy [9], green credit policies [10], environmental tax policy [11], and Sci-Tech finance policy [24], are also demonstrated to effectively motivate firms to improve their sustainability performance. Moreover, in the digital age, corporate digital transformation utilizes big data analytics and advanced technologies to substantially enhance sustainability performance via data-driven decisions and enhanced operational efficiency [25,26,27].

From the micro perspective of corporate governance, diversified firm executives are proven to improve sustainability performance [28] and management competence development [29]. Empirical evidence shows that firms with higher female participation in senior leadership improve effectiveness in implementing social welfare projects and environmental protection efforts [30]. A recent study further highlights the critical role of women’s education, employment, and rights awareness in advancing sustainable development, underscoring the importance of gender equality in this process [31]. Regarding board characteristics, a notable association exists between board size and corporate sustainability performance [32]. A higher proportion of independent directors strengthens rational decision-making [33], thereby reinforcing corporate commitment to social responsibility implementation. Moreover, existing research indicates a substantial positive correlation between board stock ownership and corporate social responsibility performance (CSR) [34,35]. An ownership-alignment mechanism can help explain this connection: increased board shareholding connects personal career aspirations with business goals, which leads to more dedication to following CSR disclosure rules and sustainability guidelines, ultimately enhancing ESG performance. Current studies confirm the efficacy of corporate governance by institutional cross-shareholders in China [36,37], which raises essential questions: Does institutional cross-ownership significantly affect corporate sustainability performance as a fundamental element of corporate governance? What channels facilitate the manifestation of these effects?

Gao et al. [38] indicate that institutional cross-shareholders enhance corporate sustainability performance through the managerial synergy effect. However, this effect is observed exclusively among long-term investors, who are inclined to advocate for strategic ESG reforms through board engagement. In contrast, Azar et al. [39] point out that institutional cross-shareholders strengthen corporate market monopolistic power, hence undermining sustainability performance, and this negative effect is mostly influenced by funds. This discrepancy may arise from the different participatory governance approaches of the two types of investors: long-term investors generate positive effects through in-depth participation in governance, whereas short-term profit-seeking investors produce negative effects through market power manipulation.

Researchers demonstrate that institutional investors play complex and multifaceted roles in corporate governance [40,41]. Their engagement in governance practices is driven both by a sense of responsibility and ethical considerations, as well as self-interest and external factors. This dual nature results in both positive and negative manifestations of institutional investors’ governance effects. Consequently, institutional cross-holders may simultaneously exert constructive and detrimental influences when participating in corporate governance. Therefore, the participation of institutional cross-shareholders in corporate governance shows the following characteristics: In terms of investment objectives, Azar et al. (2018) [39] reveal that institutional cross-shareholders actively facilitate collusion among portfolio companies by discouraging mutually destructive competitive strategies, thereby enabling the extraction of excess profits. This strategic coordination ultimately fosters market monopolization and distorts pricing mechanisms through suppressed market competition. In corporate operations, based on the commonalities among firms within the same industry, these investors can substantially decrease the costs associated with information gathering and processing, thereby achieving economies of scale in informational governance [42]. Furthermore, He and Huang (2017) [43] show that institutional cross-shareholders leverage inter-industry connections to gain privileged access to informational resources, thereby serving as conduits for cross-organizational knowledge transfer and experiential learning. While their findings establish that such strategic advantages enhance market competitiveness and profitability, the study does not take into consideration how these positive governance externalities might influence corporate sustainability performance.

These characteristics indicate that institutional cross-ownership may exert dual effects on corporate sustainability performance. On the one hand, information sharing and coordinated actions among institutional cross-shareholders may generate risks of collusive fraud. From the perspective of enhancing portfolio value, institutional cross-ownership may facilitate collusion among firms within the same industry to assert pricing power in the product market [39], hence enhancing market competitiveness. From the perspective of their bridging role, institutional cross-shareholders utilize their informational resources to influence the portfolios of holding firms, thus reducing competition among affiliated firms [44]. Consequently, increased corporate power and weakened market competition reduce corporate impetus to adopt sustainability practices for competitive advantage [16,17], ultimately undermining sustainability performance.

On the other hand, institutional cross-ownership creates synergistic governance effects through resource integration and coordinated actions. From the perspective of corporate governance, institutional cross-shareholders facilitate communication between firms to mitigate information asymmetry [45] and reduce information gathering and processing costs, eventually strengthening corporate supervision [37]. This enhanced governance structure promotes long-term value growth and improves sustainability performance [46]. Considering portfolio value maximization, firms neglecting sustainable development may face severe financing constraints and reputational damages, which directly reduce corporate value. Notably, Aouadi and Marsat (2018) [47] posit that superior social performance metrics function as an insurance mechanism, enabling firms to both mitigate the reputational damage from ESG-related controversies and strategically leverage such incidents into value creation opportunities. Their empirical analysis reveals that this dual dynamic not only buffers against stakeholder backlash but also enhances corporate valuation through improved investor perceptions of crisis management capabilities. Therefore, to protect investment interests, institutional cross-shareholders may actively participate in the board decision-making to combat short-termism [37], thereby improving corporate sustainability performance. Therefore, this paper proposes the following hypothesis:

H1a. 

Institutional cross-ownership will undermine corporate sustainability performance.

H1b. 

Institutional cross-ownership will enhance corporate sustainability performance.

2.2. Theoretical Path of Institutional Cross-Ownership Influencing Corporate Sustainability Performance

As major participants in the stock market, institutional investors adhere to the principle of profit maximization in their investment decisions and tend to prioritize a company’s short-term financial performance [48]. This narrow perspective may lead to firms that create value through long-term sustainability initiatives, such as fulfilling corporate social responsibility and addressing climate change, being undervalued by the market. To avoid the risk of undervaluation, corporate decision-makers may be forced to sacrifice long-term sustainability goals to meet short-term profit demands. Furthermore, improving sustainability performance requires structural changes, such as increasing environmental investments and strengthening the intensity of long-term investments. However, these types of investments are characterized by high initial costs, long return periods, and significant uncertainty. Under the short-termism orientation of institutional investors, firms may significantly reduce resources in areas such as environmental technology research and development, green production transformation, and nurturing future growth potential due to liquidity constraints and performance assessment pressures.

Specifically, institutional cross-ownership may facilitate tacit collusion through information sharing, weakening market competition intensity, thereby reducing the motivation for firms to use sustainability performance as a differentiated competitive advantage. This mechanism aligns with the theoretical model of Azar and Vives [49], where investor preferences indirectly affect market structure by influencing corporate strategic choices. Moreover, the heterogeneity of different types of institutional investors exacerbates this effect: institutional investors represented by funds, facing frequent short-term performance evaluations and redemption pressures, are more inclined to convey a “short-term profit priority” signal to their portfolio companies. This forces firms to cut back on environmental investments and long-term R&D expenditures, shifting their focus to short-term visible financial indicators. The transmission mechanism of this investment style not only suppresses the capital allocation of firms in areas such as environmental protection technology and low-carbon transformation but also weakens their endogenous motivation to achieve sustainable development through long-term technological accumulation. Therefore, the short-termism of institutional investors not only directly reduces corporate sustainability performance but also indirectly exacerbates the structural contradictions of sustainable development through two paths: suppressing environmental investment and long-term investment. In view of this, the following hypothesis is proposed:

H2. 

The short-termism of institutional investors negatively affects corporate sustainability performance.

Institutional cross-shareholders, as a bridge for communication among firms within the same industry, have significant information advantages. When multiple firms are jointly held by the same group of institutional investors, it becomes easier for these firms to engage in tacit cooperation to maintain price stability or limit production adjustments. This implicit alliance significantly alters the dynamics of market competition. Consequently, the way firms maintain profits gradually shifts from cultivating sustainable competitiveness through technological innovation and environmental enhancements to relying on collective market influence. Specifically, this cooperation influences corporate decision-making through three interrelated mechanisms: First, as the number of institutional investors connected to a firm increases, the capital alliance supporting it encompasses a broader spectrum. This dilution of direct competitive pressure among peer firms results in a diminished urgency for the firm to pursue differentiated advantages via green transformation. Second, if a limited number of large institutional investors hold shares in both upstream and downstream firms in the supply chain, these institutional investors may influence supplier pricing or sales channels through their control of resources, making firms more inclined to use short-term bargaining power instead of investing in long-term environmental governance. Furthermore, firms located at the core of the ownership network, particularly those forming equity connections with banks, investment funds, and industry leaders, frequently obtain priority access to information regarding policy changes or technological trends. This information advantage may instead foster speculative behavior, such as postponing the upgrade of environmental protection equipment in anticipation of government subsidies. The synergistic impact of three key factors—diminished competitive pressure, enhanced resource control, and the exploitation of informational advantages—compels firms to redirect their resource allocation strategies towards sustaining short-term market positions. Therefore, this study proposes the following hypothesis:

H3. 

The greater the market power generated by corporate collusion, the poorer the corporate sustainability performance.

3. Methodology and Data

3.1. Sample Selection and Data Sources

This study utilizes annual data from A-share listed firms in China covering the period 2012–2023 as research samples and conducts empirical tests. The comprehensive sustainability ratings and SDGs classifications for these firms are obtained from Robeco, a leading global asset management organization dedicated to SDG-aligned sustainable investing. Since 2010, Robeco has been assessing sustainability performance for over 20,000 publicly traded firms globally, including 1364 Chinese firms across 18 sectors. Data on institutional holdings and financial indicators are obtained from the CSMAR database, a premier source of financial market research data in China. To ensure data quality, we exclude observations with absent or abnormal values from the raw dataset. Furthermore, all financial variables are winsorized at the 1st and 99th percentiles to mitigate the influence of outliers.

3.2. Definition of Variables

3.2.1. Dependent Variable

The dependent variable, corporate sustainability performance (CSP), is quantified by the overall SDG score sourced from Robeco, a leading international asset management organization that evaluates corporate contributions to the UN’s 17 Sustainable Development Goals. The total SDG score is calculated in two steps. First, evaluate corporate performance on each relevant SDG in seven categories: low, medium, and high positive scores (+1, +2, +3) indicate progressive alignment with a specific SDG; neutral (0) reflects no material SDG-related influence; and low, medium, and high negative scores (−1, −2, −3) denote detrimental effects on a specific SDG. Second, calculate the overall SDG score based on the “min–max” principle: if all SDG-specific scores are non-negative, the total score equals the maximum value (ranging from 0 to 3); if any SDG-specific score is negative, the total score equals the minimum negative value (ranging from −1 to −3).

This overall SDG score makes sure that corporate sustainability performance is evaluated through both granular SDG evaluations and a prudent aggregation principle that prioritizes material negative impacts to reduce the risk of greenwashing. Consequently, firms identified as causing considerable damage to environmental goals should be assigned low ratings, while those that substantially advance environmental goals deserve high ratings. Compared with ESG ratings, SDGs scoring shows three notable advantages: (1) target-specific granularity facilitating accurate impact assessment, (2) increased responsiveness to sustainable investment preferences, and (3) structural alignment with the EU Taxonomy for Sustainable Activities and climate mitigation frameworks [50].

3.2.2. Independent Variables

The independent variable is institutional cross-ownership. The institutional investor is limited to those who simultaneously hold more than 5% of shares in two or more firms within the same industry [39]. According to the 2012 edition of the Securities Regulatory Commission (CSRC) Industry Classification Code, we classify non-manufacturing sectors as primary industries and manufacturing industries as secondary industries. Based on this classification and related studies [43], we construct the following two indicators for institutional cross-ownership: (1) Number: We compute the quarterly count of institutional cross-shareholders for each firm, average these figures to obtain annual data, and subsequently increase the annual value by 1 before applying the natural logarithm; (2) Percent: We calculate the quarterly proportion of shares owned by institutional cross-shareholders for each firm and average these proportions to obtain the annual indicator.

3.2.3. Control Variables

Following the prior literature [16,43], we controlled for the following variables: leverage ratio (Lev), fixed asset ratio (Ppe), cash ratio (Cash), return on assets (Roa), firm size (Size), the year since the firm’s initial public offering year (Listdura), the largest shareholder’s ownership (Holder1), total institutional ownership (Total), board size (Drnumber), and number of independent directors (Indnumber).

3.3. Descriptive Statistics

Table 1. Descriptive statistics of main variables.

Variables	N	Mean	Std	Min	Median	Max
CSP	4113	0.509	1.748	−3.000	1.000	3.000
Number	11,124	0.197	0.377	0.000	0.000	1.914
Percent	11,124	0.086	0.155	0.000	0.000	0.899
Lev	14,613	0.436	0.200	0.010	0.456	0.997
Ppe	14,613	0.213	0.166	0.00003	0.157	0.876
Cash	14,611	0.871	2.180	0.0004	0.396	129.310
Roa	14,613	0.054	0.074	−2.646	0.045	0.863
Size	14,613	22.965	1.327	17.813	23.499	28.697
Listdura	14,613	19.132	6.110	2.000	19.000	64.000
Holder1	14,613	34.073	16.367	2.230	31.575	89.990
Total	14,613	51.940	23.440	0.162	54.415	99.970
Drnumber	11,124	8.694	1.8142	4.000	9.000	18.000
Indnumber	11,124	3.266	0.663	1.000	3.000	8.000

presents the descriptive statistics for all the variables. The mean value of corporate sustainability performance (CSP) is 0.509, the standard deviation is 1.748, and the minimum and maximum values are −3.000 and 3.000, respectively. This finding suggests that the sustainability performance of the sample firms is relatively low and there are significant differences among these firms. For institutional cross-ownership (Number), the minimum value of 0 indicates no institutional cross-shareholders, whereas the maximum value of 1.914 implies that as many as six institutional investors may engage in cross-holding shares within a firm.

3.4. Model Construction

To investigate the effect of institutional cross-ownership on corporate sustainability performance, we constructed the following baseline model:

$${CSP}_{i,t}=\alpha +\beta X_{i,t}+\mathsf{\partial }{Controls}_{i,t}+{\gamma }_i+{\mu }_t+{\theta }_k+{\epsilon }_{i,t},$$

(1)

where i and t represent firm and year, respectively. ${CSP}_{i,t}$ represents the sustainability performance of firm i; $X_{i,t}$ represents the indicator of institutional cross-ownership, including Number and Percent; ${Controls}_{i,t}$ is the control variables of this study. Regarding model selection, the result of the Hausman test shows a chi-square statistic of 87.64 with a p-value of 0.000, which rejects the null hypothesis that “the random-effects model is preferred”. Therefore, this study adopts the fixed-effects model for regression analyses. Additionally, the model controls for the firm ${\gamma }_i$, year ${\mu }_t$, and industry ${\theta }_k$ fixed effects, and ${\epsilon }_{i,t}$ is the random disturbance term. This study focuses on the regression coefficient β; if it is significantly negative, then H1a is confirmed. And if β is significantly positive, it indicates that H1b is confirmed.

4. Empirical Results

4.1. Benchmark Regression

Table 2. Effects of institutional cross-ownership on sustainability performance: baseline results.

	(1)	(2)	(3)	(4)
	CSP	CSP	CSP	CSP
Number	−0.433 ***	−0.439 ***
	(−4.545)	(−4.562)
Percent			−1.019 ***	−1.017 ***
			(−4.250)	(−4.231)
Lev	0.086	0.086	0.089	0.088
	(0.410)	(0.406)	(0.426)	(0.415)
Ppe	0.155	0.159	0.183	0.183
	(0.558)	(0.573)	(0.660)	(0.659)
Cash	0.034	0.037	0.039	0.039
	(1.405)	(1.449)	(1.597)	(1.518)
Roa	−0.210	−0.207	−0.237	−0.235
	(−0.875)	(−0.856)	(−0.986)	(−0.971)
Size	0.179 ***	0.178 ***	0.186 ***	0.185 ***
	(2.891)	(2.862)	(3.006)	(2.987)
Listdura	−0.016	−0.016	−0.017	−0.017
	(−0.995)	(−0.975)	(−1.083)	(−1.079)
Holder1	0.003	0.003	0.007 **	0.007 **
	(0.931)	(0.865)	(2.039)	(1.976)
Total	0.001	0.001	−0.000	−0.000
	(0.269)	(0.259)	(−0.045)	(−0.049)
Drnumber	−0.006	−0.006	−0.010	−0.009
	(−0.357)	(−0.345)	(−0.594)	(−0.562)
Indnumber	0.024	0.023	0.031	0.030
	(0.582)	(0.561)	(0.769)	(0.749)
Year FE	Yes	Yes	Yes	Yes
Firm FE	Yes	Yes	Yes	Yes
Industry FE	NO	Yes	NO	Yes
N	3375	3375	3375	3375
R2	0.089	0.089	0.087	0.088

Note: Standard errors are clustered at the firm level. The t-statistics are reported in parentheses. *** and ** indicate two-tailed tests of significance at the 1% and 5% levels, respectively.

reports the baseline results of the effect of institutional cross-ownership on corporate sustainability performance. Columns (1)–(4) present the regression results based on two measures of institutional cross-ownership. The results in columns (1) and (3) account for the firm and year fixed effects, whereas those in columns (2) and (4) account for firm, industry, and year fixed effects. The coefficients of both measures are negative and statistically significant at the 1% level. These results indicate that institutional cross-ownership is negatively associated with corporate sustainability performance, which supports the hypothesis H1a proposed in Section 2.

4.2. The Endogeneity Tests

Although the baseline model includes firm, year, and industry fixed effects to partially address endogeneity concerns, other potential factors may still contribute to the endogeneity problem. First, reverse causality could arise from institutional investors’ strategic selection behavior. Given their sophisticated investment strategies and privileged information advantage, these investors may systematically target firms with lower sustainability performance to maximize short-term returns rather than actively reducing corporate sustainability through post-investment interventions. Second, omitted variable bias may exist if unobservable factors simultaneously influence institutional investment decisions and corporate sustainability performance. Third, sample selection bias might occur in the Robeco ratings due to differences in rated firms’ size and industry characteristics.

4.2.1. Lagged Independent Variables

To address potential endogeneity concerns, we implement four checks. First, we re-estimate the baseline model by using independent variables lagged by one period (L. Number, L. Percent), respectively. As presented in columns (1) and (2) of

Table 3. Results of endogeneity tests: Lagged variables, IV approach, Heckman two-stage regression, and PSM method.

	Lagged Variables		Instrumental Variable (IV) Approach				Heckman Two-Stage Regression		Propensity Score Matching
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)
	CSP	CSP	Number	CSP	Percent	CSP	CSP	CSP	CSP	CSP
L. Number	−0.644 ***
	(−5.685)
L. Percent		−0.831 ***
		(−3.038)
Merge_IV			0.355 ***		0.297 ***
			(4.039)		(3.874)
In300			0.033 ***		0.025 ***
			(5.012)		(4.356)
Out300			−0.019 **		−0.013 **
			(−2.370)		(−2.531)
$\widehat{Number}$				−0.154 ***
				(−4.039)
$\widehat{Percent}$						−0.275 ***
						(−3.272)
Number							−0.425 ***		−0.551 ***
							(−4.350)		(−4.893)
Percent								−0.978 ***		−1.216 ***
								(−3.918)		(−5.281)
IMR							−0.156	−0.128
							(−0.972)	(−0.752)
Weak Instrument (F-Stats)			31.223		33.150
Controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Firm/Year/Industry Fixed Effects	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	3332	3332	3375	3375	3375	3375	3365	3365	514	514
R2	0.081	0.082	0.109	0.235	0.118	0.242	0.080	0.017	0.063	0.091

Note: Standard errors are clustered at the firm level. The t-statistics are reported in parentheses. *** and ** indicate two-tailed tests of significance at the 1% and 5% levels, respectively.

, the results indicate that L. Number and L. Percent are significantly negative. These findings are consistent with the results in Table 2.

4.2.2. Instrumental Variable (IV) Approach

Then, we employ an instrumental variable (IV) approach with two IVs. Following the approach of He and Huang (2017) [43], we employ institutional mergers (Merge_IV) as the first instrumental variable. Merge_IV is set to 1 if a firm experiences an increase in cross-ownership due to mergers between institutional investors in year t-1, and 0 otherwise. This design satisfies the two key requirements for valid instruments: (1) Mergers between institutional investors often lead to portfolio consolidation, which may result in the merged institution holding shares in multiple firms; (2) Institutional mergers are typically driven by regulatory changes or strategic considerations at the institutional level, rather than the sustainability performance of individual firms. In addition, according to Gao (2019) [38], we construct the second IV using the adjustments to the CSI 300 Index constituents. In300(Out300) equals 1 if a firm is newly included (excluded) from the CSI 300 Index in year t-1, and 0 otherwise. On the one hand, the higher liquidity and larger market capitalization of CSI 300 constituents make them more attractive to institutional investors, creating a correlation between index inclusion/exclusion status and institutional cross-ownership. On the other hand, changes in index composition are driven by exchange rules rather than corporate initiatives, which reduces direct association with corporate sustainability performance.

Therefore, we conduct the 2SLS regression analyses to reexamine the impact of institutional cross-ownership on corporate sustainability performance. The results are shown in columns (3)–(6) of Table 3. The first-stage results in columns (3) and (5) reveal a significant association between IVs and institutional cross-ownership. The F-statistic of the weak identification test is 31.223 and 33.150, exceeding the empirical value of 10. Therefore, there is no weak instrumental problem. In the second-stage regression results presented in columns (4) and (6), the coefficients on $\widehat{Number}$ and $\widehat{Percent}$ are both significantly negative, indicating that the baseline results remain consistent after considering the endogeneity issue arising from omitted variables.

4.2.3. Heckman Two-Stage Approach

We apply the Heckman two-step method to address the potential sample selection bias issue. In the first-stage regression, a probit model is constructed to estimate the Inverse Mills Ratio (IMR), examining whether Robeco’s rating decisions for listed companies are influenced by firm characteristics. Given that institutional investors can only select investment firms based on previously disclosed information, we utilize lagged firm characteristic variables as independent variables. As shown in columns (7) and (8) in Table 3, the regression results for Number and Percent are consistent with the baseline findings. However, the statistically insignificant coefficients of the IMR across all specifications suggest the absence of sample selection bias in our analysis.

4.2.4. Propensity Score Matching (PSM) Method

Finally, to mitigate the impact of inherent differences in firm characteristics on regression results, we employ propensity score matching (PSM) to match firms. Specifically, we construct a cross-ownership dummy variable that equals 1 if such holdings exist and 0 otherwise, which serves as the dependent variable. Subsequently, using previous-period firm characteristic variables as matching covariates, we compute propensity scores through logistic regression and implement 1:1 nearest-neighbor matching. Regression analyses are then performed using the matched sample. The regression results are reported in columns (9) and (10) of Table 3. The results demonstrate that the coefficients of both Number and Percent are significantly negative, which aligns with the baseline results.

4.3. Robustness Tests

4.3.1. Replace Independent Variables and Dependent Variables

To strengthen the reliability of our findings, we re-examined the main findings by adopting ESG indicators commonly used in the literature to evaluate corporate sustainability performance. Specifically, we substituted the dependent variables in the baseline model with ESG ratings from three rating agencies: Huazheng (Hzesg), SynTao Green Finance (Sdesg), and Wind (Windesg). As shown in

Table 4. Robustness results of replacement of variables and time window shortening.

	Replacing of Dependent Variables						Replacing of Independent Variables		Shortening Time Window
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)
	Hzesg		Sdesg		Windesg		CSP		CSP
Number	0.115 ***		−0.317 ***		−0.364 ***		−0.366 ***		−0.209 **
	(2.953)		(−4.265)		(−8.362)		(−5.650)		(−2.073)
Percent		0.626 **		−0.922 **		−0.712 **		−1.020 ***		−0.684 ***
		(2.376)		(−2.361)		(−2.231)		(−4.304)		(−2.752)
Controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Firm/Year/Industry FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	9777	9777	3739	3739	6291	6291	2256	2256	3332	3332
R2	0.109	0.125	0.335	0.316	0.085	0.069	0.118	0.109	0.081	0.083

Note: Standard errors are clustered at the firm level. The t-statistics are reported in parentheses. *** and ** indicate two-tailed tests of significance at the 1% and 5% levels, respectively.

, Number and Percent exert significantly negative effects on Sdesg and Windesg, corroborating our baseline results. Notably, the impacts on Hzesg are significantly positive. The potential reason for this result is that Rebecco’s SDG scoring adopts a “Do No Harm” principle with a one-vote veto mechanism: severe negative impacts on any single SDG override positive contributions via min-max aggregation, emphasizing absolute sustainability thresholds. In contrast, the Huazheng ESG Rating allocates substantial weight (about 30%) to disclosure quality, rewarding transparency in reporting practices irrespective of actual environmental impacts. Moreover, in contrast to strict “one-vote veto” mechanisms, it employs gradual penalty measures for controversies, allowing firms to offset negative outcomes through strengths in other dimensions. These findings align with the ESG rating divergence literature, highlighting the limitations of the existing ESG rating framework in assessing corporate sustainability performance.

We then raise the shareholding ratio of institutional cross-shareholders from 5% to 10% and reconstruct the Number and Percent indicators to re-estimate the baseline model. The results are presented in columns (7) and (8) of Table 4, where Number and Percent still exert significant negative effects on SDG. This finding further confirms the robustness of our conclusions.

4.3.2. Time Window Shortening

In 2015, the UN Millennium Development Goals expired, and the 17 Sustainable Development Goals (SDGs) were launched, aiming to holistically tackle social, environmental, and economic challenges by 2030. The UN’s unanimous adoption of the SDGs marked a critical phase in global sustainability efforts, with potential policy improvements influencing corporate sustainability performance and our conclusions. Consequently, we restricted our sample to post-2015 and re-estimate the baseline model. As shown in columns (9) and (10) of Table 4, the absolute coefficient on Number (Percent) drops from 0.439 (1.107) to 0.209 (0.684). This corresponds to a reduction in explanatory power attributable to cross-ownership by 52.39% and 38.21% ((0.439−0.209)/0.439; (1.107–0.684)/1.107). Such declines suggest that the SDGs framework has meaningfully constrained institutional investors’ ability to engage in short-termism through cross-ownership strategies, likely due to stricter disclosure norms and accountability pressures under the new regime.

4.4. Mechanism Analysis

The baseline results indicate that institutional cross-ownership significantly reduces corporate sustainability performance. The possible reason for this phenomenon is that institutional cross-shareholders may reduce market competition through collusion facilitated by information sharing, which prevents firms from prioritizing sustainability performance as a key competitive advantage, ultimately resulting in decreased resource allocation for promoting corporate sustainability performance. This explanation aligns with Azar and Vives [49], who conducted a theoretical analysis of the factors influencing collusion through a production decision model, identifying investor preference and market power as two key factors.

To investigate the impact of institutional investors’ investment style preferences on corporate sustainability performance, we classify institutional investors into six categories, including funds, social security funds, insurance companies, securities firms, trust institutions, and Qualified Foreign Institutional Investors (QFII) [51]. We subsequently examine the relationship between the shareholding proportions of various types of institutional investors (Ini_percent) and corporate sustainability performance, with regression results presented in columns (1)–(6) of

Table 5. Channel analysis based on institutional investors’ short-termism.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)
	Funds	QFII	Brokerages	Insurance	Social Security Funds	Trusts	Envest	Envest	Lmvest	Lmvest
Ini_percent	−1.113 **	1.238	0.296	−0.907	−0.079	−0.780
	(−2.231)	(1.161)	(0.374)	(−0.808)	(−0.054)	(−0.749)
Number							−0.052 **		−0.112 **
							(−2.015)		(−2.315)
Percent								−0.083 **		−0.131 **
								(−2.213)		(−2.412)
Controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Firm/Year/Industry FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	3828	3828	3828	3828	3828	3828	3375	3375	3375	3375
R2	0.079	0.078	0.079	0.077	0.079	0.078	0.068	0.070	0.071	0.072

Note: Standard errors are clustered at the firm level. The t-statistics are reported in parentheses. ** indicates two-tailed tests of significance at the 5% level.

. As shown in Table 5, only fund-type institutional investors have a significant negative impact on sustainability performance, with a regression coefficient of −1.113 and significant at the 5% level, whereas the other types of institutional investors show insignificant effects on sustainability performance. The likely explanation for this result is the short-term performance pressures and liquidity constraints faced by fund managers [52]. And such pressures may be transferred to portfolio firms, forcing them to prioritize short-term profit maximization over long-term sustainability investments, thereby undermining corporate capacity for systematic sustainable development implementation.

To further verify this finding, we follow the methodologies of Zhang et al. (2019) and Jin et al. (2024) to measure firms’ environmental investment (Envest) and long-term investment intensity (Lmvest) [53,54]. These two metrics, respectively, reflect a firm’s expenditure on environmental protection and its commitment to future growth. We then conduct regression analyses using institutional cross-ownership as the explanatory variable and these two indicators as dependent variables, the results of which are presented in columns (7)–(10) of Table 5. The coefficients of institutional cross-ownership are consistently negative and statistically significant, indicating that such cross-ownership undermines corporate environmental and long-term investments. This empirical evidence further supports the hypothesis of institutional investors’ short-termism.

We then investigate the association between market power and corporate sustainability performance. In accordance with Azar and Vives [49], we measure market power (Con_number) by using the number of firms connected to a focal firm through institutional cross-shareholders. The regression results are presented in Column (1) of

Table 6. Channel analysis based on corporate market power and green innovation.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)
	CSP	Lerner index	Lerner index	Markup rate	Markup rate	EnvrPa	EInvPa	EUtyPa	EnvrPa	EInvPa	EUtyPa
Con_number	−0.740 ***
	(−7.453)
Number		0.165 ***		0.324 ***		−0.346 ***	−0.248 **	−0.189 **
		(3.162)		(4.320)		(−3.018)	(−2.337)	(−2.148)
Percent			0.226 ***		0.415 ***				−0.178	−0.178	−0.214
			(3.453)		(5.125)				(−0.641)	(−0.690)	(−1.003)
Controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Firm FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	3375	3375	3375	3375	3375	2550	2550	2550	2550	2550	2550
R2	0.081	0.094	0.098	0.102	0.106	0.050	0.038	0.041	0.045	0.035	0.039

Note: Standard errors are clustered at the firm level. The t-statistics are reported in parentheses. *** and ** indicate two-tailed tests of significance at the 1% and 5% levels, respectively.

, showing a statistically significant negative coefficient of −0.740 for Con_number at the 1% significance level. This finding indicates that enhanced market power is associated with poorer sustainability performance, thereby supporting hypothesis H3. The results suggest that as firms are connected through institutional cross-shareholders with a larger number of firms, the enhanced market power and higher collusive benefits reduce firms’ incentives to improve sustainability performance for competitive advantages, ultimately leading to poorer sustainability performance.

Beyond this network-based mechanism, following Hu et al. (2024) and Tang et al. (2022) [55,56], we construct two additional proxies for firm-level market power: the Lerner index and the markup rate. The Lerner index captures a firm’s market share and pricing power, with higher values indicating stronger monopolistic positions. The markup rate reflects a firm’s ability to sustain prices above marginal costs, serving as a critical indicator of market competitiveness. Subsequent regression analyses using institutional cross-ownership as the independent variable and these two market power indicators as dependent variables reveal that institutional cross-ownership significantly strengthens firms’ market power. The results are presented in columns (2)–(5) in Table 6. This further validates the mechanism through which cross-ownership amplifies market influence.

Furthermore, green innovation is an essential way for firms to exercise the concept of sustainable development, and the progress of green innovation ability can effectively enhance corporate sustainability performance [57]. Therefore, the inhibitory effect of institutional cross-ownership on sustainability performance may manifest as a reduction in corporate green innovation. To confirm this assumption, we further examine the effect of institutional cross-ownership on corporate green innovation. Referring to Wang et al. (2021) [58], the indicators of corporate green innovation are defined as follows: (1) EnvrPa, the number of corporate green innovation patents; (2) EInvPa, the number of corporate green invention patents; and (3) EUtyPa, the number of corporate green utility patents. All the above indicators are adjusted by adding 1 and applying the natural logarithm.

The regression results presented in columns (6)–(11) Table 6 indicates that the coefficients of Number on the three green innovation indicators are −0.346, −0.248, and −0.189, respectively, and are all statistically significant; the effect of Percent is not significant, while still negative. It can be seen that institutional cross-ownership diminishes corporate initiatives to promote green innovation, thereby reducing sustainable development investment and ultimately undermining corporate sustainability performance.

4.5. Heterogeneity Analysis

4.5.1. Heterogeneity Analysis on New Quality Productivity (NQP)

New Quality Productivity (NQP) refers to an advanced productivity paradigm driven by transformative technological innovation, efficiency-optimized resource allocation, and structural industrial upgrading. Characterized by knowledge-intensive and technology-driven industries, NQP represents the direction of green, innovative, and modernized development. Enterprises operating in regions with different NQP levels exhibit marked differences in development concepts and governance structures, which may lead to diverse outcomes in corporate sustainability performance.

Following Liu and He (2024) [59], we calculate annual NQP indices for 31 Chinese provinces and municipalities and categorize regions into high- and low-NQP categories based on the mean value over the sample period. We then conduct subgroup regressions for firms in these two categories. The results are presented in columns (1) and (2) of Panels A and B in

Table 7. Results of heterogeneity tests.

Panel A: The Independent Variable is Number
		Number	Controls	Firm/Year/Industry FE	N	R2
Levels of New Quality Productivity heterogeneity	(1) High	−0.444 *** (−4.255)	Yes	Yes	1535	0.206
	(2) Low	0.223 (1.479)	Yes	Yes	1031	0.183
Region heterogeneity	(3) Central-Eastern	−0.230 *** (−2.973)	Yes	Yes	2162	0.221
	(4) Northwestern	−0.112 (−0.437)	Yes	Yes	404	0.106
Ownership Structure heterogeneity	(5) State owned	−0.470 *** (−3.679)	Yes	Yes	1200	0.208
	(6) Nonstate owned	0.008 (0.091)	Yes	Yes	1366	0.114
Intensity of environmental regulation heterogeneity	(7) High	−0.060 (−0.421)	Yes	Yes	769	0.253
	(8) Low	−0.265 *** (−3.000)	Yes	Yes	1797	0.139
Panel B: The Independent variable is Percent
		Percent	Controls	Firm/Year/Industry Fixed Effects	N	R2
Levels of New Quality Productivity heterogeneity	(1) High	−0.979 *** (−4.015)	Yes	Yes	1535	0.202
	(2) Low	0.019 (0.049)	Yes	Yes	1031	0.186
Region heterogeneity	(3) Central-Eastern	−1.029 *** (−4.764)	Yes	Yes	2162	0.228
	(4) Northwestern	0.254 (0.446)	Yes	Yes	404	0.104
Ownership Structure heterogeneity	(5) State owned	−1.079 *** (−3.752)	Yes	Yes	1200	0.222
	(6) Nonstate owned	−0.224 (−0.742)	Yes	Yes	1366	0.111
Intensity of environmental regulation heterogeneity	(7) High	−0.178 (−0.582)	Yes	Yes	769	0.259
	(8) Low	−1.041 *** (−4.051)	Yes	Yes	1797	0.146

Note: Standard errors are clustered at the firm level. The t-statistics are reported in parentheses. *** indicates two-tailed tests of significance at the 1% level.

. The results show that institutional cross-ownership significantly reduces corporate sustainability performance at the 1% significance level in high-NQP regions (e.g., Beijing, Shanghai, Guangdong), whereas this effect is insignificant in low-NQP regions. A plausible explanation is that firms in high-NQP regions possess mature governance frameworks that facilitate information exchange through institutional cross-ownership. However, this connectivity may foster collusive behaviors, thereby decreasing firms’ incentives to enhance sustainability performance as a competitive advantage. Ultimately, this consequence leads to reduced corporate sustainability performance in technologically advanced regions.

4.5.2. Heterogeneity Analysis on Geographical Regions

Given China’s economic development pattern, the influences of geographical differences on the results are similar to the varying impacts of NQP levels. Moreover, inter-regional enterprise differences are influenced by factors such as local customs, policies, and institutional environments. To ensure the rigor of the analysis, this study classifies the sample into two subgroups based on corporate registration locations: firms in the Central-Eastern Region and those in the Northwestern Region. We then conduct subgroup regression analyses for these two groups. Columns (3) and (4) of Panels A and B in Table 7 show that institutional cross-ownership negatively affects corporate sustainability performance exclusively in the Central-Eastern Region, aligning with the earlier findings.

4.5.3. Heterogeneity Analysis on Corporate Ownership Structure

In contrast to non-state-owned firms, state-owned enterprises (SOEs) bear significant responsibilities in promoting China’s economic structural transformation and fulfilling national strategic objectives [60]. Meanwhile, firms with different ownership structures exhibit pronounced differences in corporate governance structures, resulting in heterogeneous firm characteristics. Therefore, we classify the sample firms into state-owned and non-state-owned categories and examine the impact of ownership structure on the results.

Columns (5) and (6) of Table 7 reveal that institutional cross-shareholders significantly negatively impact the sustainability performance of state-owned enterprises (SOEs), but not that of non-state-owned enterprises (non-SOEs). A credible explanation for this is that SOEs possess strong market competitiveness due to their irreplaceable role in policy-led sectors, hence reducing their reliance on external governance mechanisms. Therefore, the influence of institutional cross-shareholders on SOEs is more likely to result in collusion rather than collaborative governance, thereby reducing their incentives to improve sustainability performance.

4.5.4. Heterogeneity Analysis on Environmental Regulation Intensity

The rigor of environmental regulation in an area substantially influences corporate sustainability performance [61]. Firms in regions with higher environmental regulation intensity face stricter external monitoring and greater pressure from public opinion [62], compelling them to adopt environmentally friendly and socially responsible practices. Following Berman and Bui [63], we measure the intensity of regional environmental regulation by calculating industrial pollution treatment investment per CNY 10,000 of gross industrial production value (calculated as 10,000 × pollution control investment/regional gross output value). We classify samples into high- and low-intensity groups based on the mean value during the sample period. As shown in columns (7) and (8) of Table 7, institutional cross-ownership has a markedly negative impact on sustainability performance solely for firms located in areas with low environmental regulation intensity. These firms bear substantial upfront costs to improve sustainability performance with minimal short-term returns. Meanwhile, diminished external monitoring and lax environmental regulations make investments in sustainable development financially impractical. Moreover, institutional cross-shareholders weaken market competition, thereby reducing firms’ incentives to improve sustainable development initiatives and ultimately leading to poorer sustainability performance.

4.6. Differential Impacts on Sustainability Dimensions

According to Banerjee et al. (2020) [64], corporate sustainability performance can be classified into three dimensions: economic (ECO), social (SCO), and ecological (BIO). The score for each dimension is determined by the average of a firm’s performance on the corresponding subgoals. The ECO includes SDGs 8 (Decent Work and Economic Growth), SDGs 9 (Industry, Innovation, and Infrastructure), SDGs 10 (Reducing Disparities), SDGs 12 (Responsible Consumption and Production), and SDGs 17 (Partnerships for Goal Achievement). The SCO includes SDGs 1 (Eradicate Poverty), SDGs 2 (Eradicate Hunger), SDGs 3 (Good Health and Well-being), SDGs 4 (Quality Education), SDGs 5 (Gender Equality), SDGs 7 (Affordable and Clean Energy), SDGs 11 (Sustainable Cities and Communities), and SDGs 16 (Peace, Justice, and Strong Institutions). The BIO includes SDGs 6 (Clean Water and Sanitation), SDGs 13 (Climate Action), SDGs 14 (Underwater Organisms), and SDGs 15 (Terrestrial Organisms).

The regression results are presented in

Table 8. Results of the impacts of institutional cross-ownership on different dimensions of sustainability performance.

	(1)	(2)	(3)	(4)	(5)	(6)
	ECO		SCO		BIO
Number	−0.496 ***		−0.314 ***		0.169
	(−6.020)		(−2.602)		(0.906)
Percent		−0.501 ***		−0.691 ***		−0.756 **
		(−3.580)		(−2.957)		(−2.061)
Controls	Yes	Yes	Yes	Yes	Yes	Yes
Firm/Year/Industry FE	Yes	Yes	Yes	Yes	Yes	Yes
N	1449	1449	1784	1373	658	658
R2	0.175	0.147	0.233	0.037	0.291	0.297

Note: Standard errors are clustered at the firm level. The t-statistics are reported in parentheses. *** and ** indicate two-tailed tests of significance at the 1% and 5% levels, respectively.

. Columns (1), (3), and (5) report the effects of Number on ECO, SCO, and BIO, respectively. The most pronounced negative impact is evident on ECO, with a statistically significant coefficient of −0.496 at the 1% level, followed by SCO showing decreased coefficient of −0.314 at the 1% level. However, the effect of Number on BIO is not significant. Columns (2), (4), and (6) show the effects of Percent on ECO, SCO, and BIO, respectively. The greatest significant negative effect is observed on SCO, followed by ECO and BIO. These results suggest that institutional cross-ownership diminishes corporate sustainability performance across economic and social dimensions. This finding may stem from two mechanisms: First, institutional cross-shareholders facilitate inter-firm collusion, enabling coordinated efforts to enhance product pricing power and market dominance. Second, the substantial transition costs associated with sustainable development disincentivize firms from improving economic sustainability, thereby diminishing overall sustainable performance.

Notably, the insignificant effect of Number on BIO suggests that the influence of institutional cross-ownership on environmental sustainability is not robust. The muted environmental effects of institutional cross-ownership reflect China’s unique regulatory regime. Since the 2015 Environmental Protection Law established the world’s most strict enforcement infrastructure—including Central Environmental Inspection tours, industry-specific emission deadlines, and corporate environmental credit systems—firms face near-zero tolerance for non-compliance. Even collusion-prone resulting from institutional cross-ownership cannot circumvent real-time pollution monitoring and immediate penalties. Conversely, social and economic governance exhibit elasticity. Economic sustainability indicators (e.g., R&D intensity) remain vulnerable to strategic accounting policy selection. Social responsibility requirements, such as labor welfare and supplier ESG audits, lack standardized implementation across provinces. This regulatory asymmetry enables cross-ownership to amplify short-term profit extraction. Reduced market discipline allows firms to redirect resources from long-term capacity building to quick-win strategies like supply chain cost compression, thereby decreasing social and economic performance.

5. Conclusions

This study systematically investigates the impact of institutional cross-ownership on corporate sustainability performance using micro-level data from China’s A-share listed companies (2012–2023). The findings reveal that institutional cross-ownership significantly undermines corporate sustainability performance through multiple channels, with its negative effects remaining robust across endogeneity tests such as instrumental variable approaches, Heckman two-stage models, and propensity score matching. Specifically, the short-termism of institutional investors pressures firms to reduce green innovation investments in pursuit of immediate financial returns, while market power consolidation fostered by cross-ownership networks weakens corporate incentives to gain competitive advantages through sustainability practices. Further analysis indicates that these adverse effects are particularly pronounced in high-productivity regions, central-eastern enterprises, areas with lax environmental regulations, and state-owned enterprises (SOEs), reflecting the amplified collusion dynamics under China’s institutional context, characterized by a “weak supervision-high concentration” paradox. Notably, the influence caused by cross-ownership is most evident in economic dimensions and social dimensions, while its impact on ecological dimensions remains limited. This suggests a strategic selectivity in the negative effects—institutional investors tend to prioritize short-term gains at the expense of economic and social sustainability goals.

Based on the above findings, this study proposes a series of policy recommendations. First, the anti-monopoly regulatory framework must be refined. At the ex-ante prevention level, the Anti-Monopoly Law should be revised to explicitly regulate market power expansion induced by institutional cross-ownership, with industry concentration warning thresholds established. For real-time monitoring, a dynamic surveillance system should be implemented to mandate institutional investors to disclose cross-firm equity linkages, thereby enhancing market transparency. In terms of post hoc accountability, entities exploiting cross-ownership for price manipulation or exclusionary agreements should face heavy fines and market access restrictions. Second, green financial incentive mechanisms should be constructed. For instance, tax incentives such as exemptions or reductions in capital gains taxes could be granted for long-term holdings of green assets (e.g., equity in low-carbon projects). Corporate SDG performance should be integrated into credit rating systems to guide capital allocation, while innovative financial instruments like Sustainability-Linked Bonds (SLBs) could attract institutional investors to participate in sustainability transitions. Furthermore, strengthening information disclosure and evaluation systems is critical. Companies should be required to compile sustainability reports under the UN SDG framework, detailing environmental and social metrics. International third-party certification bodies should be introduced to curb greenwashing practices. Concurrently, investor education programs should be promoted to shift institutional investors from acting as mere financial agents to becoming stewards of sustainable governance, actively influencing corporate strategies aligned with long-term sustainability goals.

While this study provides emerging-market evidence for understanding the sustainability paradox of institutional cross-ownership, several limitations warrant acknowledgment. First, reliance on publicly listed firms’ data may underestimate the collusion effects among unlisted enterprises, which often operate with less transparency. Second, the dynamic mechanisms underlying equity network evolution—such as institutional exits or ownership restructuring—and their long-term implications for sustainability remain underexplored. Third, the absence of international comparisons limits the generalizability of our conclusions; future research could extend to emerging markets like India and Brazil to test the universality of capital concentration-SDG conflicts. Fourth, theoretical frameworks could be enriched by integrating game-theoretic models to simulate strategic interactions between institutional investors and corporate managers, offering micro-foundations for policy design. These limitations not only chart pathways for subsequent studies but also underscore the need for deeper exploration into balancing capital efficiency with social welfare in practice.