One important dimension of China’s economic development has been the privatization of state-owned enterprises and the opening of markets to private firms. Two bodies of theory have dominated the debate. Economic theories emphasize the inefficiency of state ownership, while political theories stress ideology and regulatory capture. Yet neither body of theory effectively explains the complex distribution of state ownership: why state ownership recedes in some sectors but persists, or even expands, in others. Professor James Si Zeng’s new book offers a powerful answer. Its Legal Theory of State-Owned Enterprises (LTS) provides an analytical framework unusual in both theoretical ambition and empirical breadth.

At the centre of the book is a trade-off between ownership costs and regulatory costs. State-owned enterprises generally incur higher ownership costs because of soft budget constraints and agency costs. At the same time, state ownership gives the government better access to internal information and greater decision-making flexibility through continuing relationships and residual control. Zeng characterizes these advantages as lower regulatory costs. If the state declines to bear ownership costs, it must rely more heavily on the enactment and enforcement of law to regulate private firms, thereby bearing the resulting regulatory costs. Zeng argues that regulatory costs rise where the law is more incomplete, increasing the relative advantage of state ownership. Across different industries and organizational forms, the book shows how this trade-off generates an uneven distribution of state ownership. This account also places LTS within the broader tradition of contract theory: as formal contracts become more incomplete, relational contracting and organizational authority may become relatively more attractive.

Because this trade-off is the core of LTS, it is worth asking who makes it, and why. If LTS were a normative account, the identity of the decision-maker would matter less. The decision-maker might simply be modelled as a hypothetical benevolent social planner. Yet LTS is expressly offered as an explanatory theory. It does not fully develop its account of who makes the trade-off. The book explicitly analyses private investors, whose decisions may indirectly shape the distribution of state ownership. Facing lower ownership costs but potentially higher regulatory burdens than state-owned incumbents, they may decide whether to enter or exit an industry, or to bring in a partial state shareholding. Private investors, however, do not directly make the institutional choice between state ownership and legal regulation. The LTS trade-off itself therefore belongs to a public authority with the relevant powers and incentives. This raises a further question: which public authority has an incentive to make the trade-off? The book’s discussion of public authority ranges across state asset regulators, sectoral regulators, and central and local governments. In the case of local state ownership, however, I would identify the relevant decision-makers more precisely: Chinese local officials subject to a performance-measurement system.

Performance measurement is a central feature of China’s political system. Superiors assess subordinate officials, especially principal local officials, by reference to their jurisdictions’ performance in areas including economic growth, environmental protection, social stability, and workplace safety. The results can affect officials’ promotion prospects and exposure to political accountability (Xu, Reference Xu2011).

Performance measurement explains how ownership costs and regulatory costs enter local officials’ objective functions. Under performance measurement, local officials are multitasking agents who must attend simultaneously to economic growth, social stability, and other tasks. They will therefore engage in an LTS-style trade-off. If, for example, an official captured by a state-owned enterprise insists on state ownership in a field where the law is relatively complete and regulatory costs are correspondingly low, the higher ownership costs may eventually appear as slower GDP growth and lower fiscal revenue. This may lead to a weaker performance assessment in the economic domain. Conversely, if an official continues to rely primarily on legal regulation of private firms in a sector with high regulatory costs, that choice may lead to underinvestment and increase the risk of accidents and disorder. Those outcomes may in turn count against the official under welfare and stability measures. Of course, the translation of social costs into political consequences through performance evaluation is inevitably imperfect, in part because some information is unverifiable. Yet, as multitasking agents, local officials may still make the trade-off in an effort to secure favourable performance assessments and advance their political careers. Performance measurement thus allows LTS’s social-cost calculus to enter actual political decision-making. The book discusses the performance-measurement literature, but it could more explicitly identify this mechanism as the microfoundation of the trade-off between ownership costs and regulatory costs, especially in its account of local state ownership.

Clarifying which agent makes this trade-off has several advantages. First, it distinguishes LTS more sharply from competing theories of state ownership. Theories of regulatory capture and ideology do not, by themselves, give officials an incentive to make the LTS-style trade-off. Moreover, not all Chinese officials with the authority to influence the establishment of state-owned enterprises face this incentive. Officials who fall outside the jurisdiction-based allocation of responsibility under this performance regime will not make the LTS-style trade-off. This specification also generates testable implications. Changes in evaluation criteria, as well as variation in officials’ career incentives across age and tenure, should affect how the LTS trade-off is made. Evaluation criteria, in particular, change often enough and are documented well enough to support empirical work.

Identifying the agent who makes the trade-off also addresses a classic question in Chinese political economy: are private firms discriminated against? One recent strand of the performance-measurement literature suggests that local officials need not discriminate against private firms simply because of private ownership. The principal economic measures, including GDP growth, tax revenue, and investment inflows, are largely ownership-neutral. Contributions from private and state-owned enterprises alike can be credited to an official’s performance. The more immediate distinction in this literature is therefore often between local and non-local firms, because the performance of firms within the jurisdiction is more readily translated into the official’s own record (Zhou, Reference Zhou2018). Yet state-owned enterprises continue to enjoy substantial preferential treatment. In practice, state-owned and private firms are plainly not treated alike. LTS offers a new explanation for this apparent puzzle. The preferential treatment that local officials grant to state-owned enterprises may be understood as the price of securing the lower regulatory costs associated with state ownership.

Finally, identifying who makes the LTS trade-off reveals the theory’s broader potential. Who, besides Chinese local officials, might make a similar trade-off between ownership costs and regulatory costs? One possible answer is shareholders with environmental, social, and governance (ESG) preferences. Shareholder-led ESG has become one of the most prominent issues in contemporary corporate governance.Footnote ¹

The relevance of LTS becomes clearer once the institutional choice facing these shareholders is specified. Pro-social shareholders may leave the pursuit of environmental and labour goals to external law, or they may use ownership and corporate governance rights to pursue those goals from within the firm. The logic of LTS can, therefore, be extended from the choice between state ownership and law to the choice between ownership and law more generally.

A common objection to corporate ESG is that environmental and labour issues are better addressed through environmental or labour law, rather than through corporate governance. On this view, asking companies to pursue ESG goals is either symbolic or inefficient. LTS suggests a different possibility by directing attention to the comparative costs of external law and internal governance. Pro-social shareholders may have preferences that are not reducible to their financial interests (Hart and Zingales, Reference Hart and Zingales2017). When they use the control rights associated with ownership to pursue environmental or labour goals, they may incur additional ownership costs. Agency costs may rise when managers serve shareholders with multiple and difficult-to-verify objectives, while heterogeneous shareholder preferences may increase collective decision-making costs. At the same time, ownership gives shareholders access to internal information and residual control, potentially reducing the regulatory costs of pursuing these goals relative to external law. Zeng’s account of the informational and control advantages of state ownership can thus explain why internal governance by pro-social shareholders may, in some settings, be less costly than external regulation.

Pro-social shareholders may, therefore, assess whether the regulatory costs saved through internal control exceed the additional ownership costs. This entails a trade-off between ownership and law analogous to the state’s choice in LTS. Viewed in this way, extending LTS from state ownership to pro-social shareholding offers a new efficiency-based explanation of shareholder-led ESG. Shareholder-led ESG may be effective and efficient rather than merely symbolic.

This extension to ESG illustrates the potential for Zeng’s framework to reach beyond its original setting of Chinese state-owned enterprises. In our conversations, Zeng told me that his aim was to explain the Chinese experience, which remains far from fully understood. Yet in explaining that experience, Zeng identifies a general trade-off between ownership and law with much broader application. An LTS may ultimately prove too narrow a name for the theory he has developed.

That broader reach underscores the significance of the book. In his first monograph, Professor Zeng places ownership and regulation within a common analytical framework and offers a compelling theory with substantial explanatory power. Like his Yale adviser Henry Hansmann, Zeng seeks to explain why law has the institutional content it does. LTS succeeds as a theory of the law itself, not merely of its consequences (Cooter, Reference Cooter2018). It sheds new light on the complex relationship between ownership, whether public or private, and law.