Venture capital plays a vital role in fostering entrepreneurship and innovation, yet concerns exist regarding collusive behavior between venture capitalists and entrepreneurs. This study focuses on China, where "guanxi," a system of social relationships based on trust and reciprocity, significantly influences business practices. Existing research often highlights the positive aspects of "guanxi" in facilitating investment and financing, but it overlooks the potential for negative consequences arising from collusion. This research addresses this gap by investigating how "guanxi" impacts the relationship between venture capitalists and entrepreneurs, particularly during the exit stage of venture capital investments. The primary research questions are: (1) Does "guanxi" contribute to collusive behavior between venture capitalists and entrepreneurs during the exit stage? and (2) What factors can mitigate this collusive behavior? The choice of China as the empirical setting is driven by its unique characteristics: the crucial role of venture capital in economic transformation, a speculative secondary securities market incentivizing stock price manipulation, and the pervasive influence of "guanxi" in social and business interactions. This study offers a novel perspective by integrating principal-agent theory, social exchange theory, social identity theory, and the theory of unethical pro-relational behavior to analyze the complex interplay between "guanxi" and collusive behavior in venture capital transactions. The research is expected to enrich existing theoretical frameworks, provide insights into the ethical implications of "guanxi," and offer practical recommendations for mitigating the adverse effects of collusion.

The literature review examines three key areas: the collaboration between venture capitalists and entrepreneurs, the role of "guanxi" in organizational behavior, and collusive behavior in transactions. Regarding venture capitalist-entrepreneur collaboration, the review explores the limitations of principal-agent theory and contractual cooperation in fully addressing the inherent conflicts. Organizational relationship theory is introduced as a complementary framework, highlighting the importance of governance and trust in fostering cooperation and restraining opportunistic behavior. The role of trust in reducing agency problems and conflicts of interest is emphasized. Regarding "guanxi," the review explores both its positive and negative impacts. While "guanxi" can facilitate information exchange and resource acquisition, it can also weaken supervision and increase business risks. The literature on collusive behavior highlights incentives for self-interest, low collusion costs, and poor corporate governance environments as contributing factors. Prevention efforts mainly focus on regulatory agencies and improved corporate governance.

This study employs an empirical approach using data from Chinese Growth Enterprise Market (GEM)-listed companies from 2009 to 2021. The sample includes 462 firms and 1600 observations after rigorous selection criteria, excluding non-financial firms, ST and *ST firms, and samples with incomplete data. The dependent variables representing collusive behavior include: entrepreneurial self-interest behavior (measured by executive perks consumption and executive overcompensation), venture capitalists' returns (measured by the annual average investment return rate), minority shareholders' interests (measured by the presence of online voting and cumulative voting systems), and corporate operational risks (measured by earnings volatility). The independent variable is "guanxi" (measured by the sum of different types of relationships: alumni, colleague, hometown, political, and association). Equity incentives (measured by the proportion of shares held by the chairman) serve as the moderating variable. Control variables encompass characteristics of entrepreneurs, firms, venture capitalists, and venture capital institutions. Different econometric models are employed based on the nature of the dependent variables: fixed-effects models for continuous variables (entrepreneurial self-interest behavior and corporate operational risks), OLS for continuous cross-sectional data (venture capitalist returns), and xtlogit for binary panel data (minority shareholders' interests). To examine the moderating effect of equity incentives, a hierarchical regression model is used. To address potential endogeneity issues, propensity score matching and system GMM are used for robustness checks. The study also performs a robustness check by changing the proxy for the independent variable, using a dummy variable indicating the presence or absence of "guanxi".

The empirical analysis supports the hypotheses. The presence of "guanxi" between venture capitalists and entrepreneurs is positively associated with entrepreneurial self-interest behavior (both executive overcompensation and perks consumption), higher venture capitalist returns, harm to minority shareholders' interests (a negative association with minority shareholder protection mechanisms), and increased corporate operational risk. The moderating effect of equity incentives is significant. Equity incentives weaken the positive relationship between "guanxi" and entrepreneurial self-interest, venture capitalist returns, and corporate operational risk. They also weaken the negative relationship between "guanxi" and minority shareholders' interests. Robustness checks using propensity score matching, system GMM, and an alternative proxy for "guanxi" confirm the findings.

The findings contribute to the understanding of the complex interplay between "guanxi," collusive behavior, and corporate governance in the context of venture capital. The positive association between "guanxi" and collusive outcomes underscores the potential downsides of informal social networks in business transactions. The moderating role of equity incentives suggests that strong corporate governance mechanisms can help mitigate the negative effects of "guanxi". These results challenge the simplistic view that "guanxi" is always beneficial in business dealings and highlight the importance of considering potential ethical and agency problems arising from close relationships. The findings are particularly relevant for understanding the specific challenges in managing venture capital transactions in relational societies like China.

This study contributes significantly to the literature by examining the impact of "guanxi" on collusive behavior in venture capital transactions. It demonstrates a clear link between "guanxi" and several negative outcomes, including harm to minority shareholders and increased risk. The moderating effect of equity incentives provides crucial insights for corporate governance. Future research could focus on expanding the sample to include non-listed firms, refining the measurement of "guanxi," exploring the role of external governance mechanisms, and analyzing the venture capital ecosystem holistically.

The study's limitations include the sample being limited to GEM-listed companies in China, potentially affecting the generalizability of the findings. The measurement of "guanxi" might not capture the full complexity of social relationships. The study does not explicitly examine the role of other external governance mechanisms, which could also impact collusion behavior. Finally, while the study uses advanced econometric techniques to address potential endogeneity issues, the possibility of residual confounding effects remains.