Governance Challenges of Listed State- Owned Enterprises Around the World: National Experiences and a Framework for Reform

Cornell International Law Journal

Volume 50 Number 3Fall 2017

Article 3

Governance Challenges of Listed State- Owned Enterprises Around the World: National Experiences and a Framework for Reform

Curtis J. Milhaupt Columbia Law School

Mariana Pargendler New York University School of Law

Part of theBusiness Organizations Law Commons, and theComparative and Foreign Law

Commons

Governance Challenges of Listed StateOwned Enterprises Around the World: National Experiences and a Framework for Reform

Curtis J. Milhaupt & Mariana Pargendler†

Despite predictions of their demise in the aftermath of the collapse of socialist economies in Eastern Europe, state-owned enterprises (SOEs) are very much alive in the global economy. The relevance of listed SOEs— firms subject to government ownership, but with a portion of their shares traded on public stock markets—has persisted and even increased around the world, as policymakers have encouraged the partial floating of SOE shares either as a first step toward, or as an alternative to, privatization. In this Article, we evaluate the governance challenges associated with mixed ownership of enterprise, and examine a variety of national approaches to the governance of listed SOEs, with a view to framing a robust policy discussion in many countries where SOE reform is a topic of major significance. We describe the evolution and current status of the institutional framework applicable to listed SOEs in eight different jurisdictions which reflect a variety of economic, legal, and political environments: France, the United States, Norway, Colombia, Brazil, Japan, Singapore, and China. We leverage the lessons from this comparative analysis by critiquing the policy prescriptions of international agencies such as the OECD and framing our own policy suggestions. Introduction.....................................................474 I.Governance Challenges...................................477 II.National Experiences.....................................480 A.France................................................480 B.United States.........................................487

C.Norway...............................................494 D.Colombia.............................................500 E.Brazil.................................................505 F. Japan.................................................512 G.Singapore.............................................518 H.China................................................524 III.Patterns of SOE Regulation?..............................529 IV.Critique of Best Practice Guidelines......................533 V.Policy Responses.........................................535 A.Ownership Structure..................................536 B.Board Composition and Independence.................537 C.The Role of the Board.................................538 D.Remuneration.........................................538 E.Structural Incentives..................................539 F.Disclosure and Treatment of Non-Commercial Decisions

......................................................540 G.Enforcement..........................................541 Conclusion......................................................542

Introduction

The state is still in business. Despite predictions of their demise in the aftermath of the collapse of socialist economies in Eastern Europe, stateowned enterprises (SOEs) are very much alive in the global economy. As of 2010, state-owned firms accounted for about one-fifth of world market capitalization.1 Since then, listed SOEs have generally suffered from the recent decline in commodity prices, but otherwise continue to play a major role in their respective economies and capital markets. The relevance of listed SOEs—firms subject to government ownership, but with a portion of their shares traded on public stock markets—has persisted or even increased around the world, as policymakers have encouraged the partial floating of SOE shares either as a first step toward, or an as alternative to, privatization.

The “mixed ownership” model presented by listed SOEs is spreading. For example, Saudi Arabia, which previously relied on a system of whole ownership by the state, has recently announced a plan to publicly list shares of its oil giant Aramco.2 In China, the Xi Jinping administration has championed mixed ownership as a means of reforming the massive state sector of the economy.3 The motivation behind this approach is the notion that injecting private capital into SOEs will improve their management and expose them to much-needed market discipline.

Yet mixed-ownership structures add a distinctive layer of governance challenges atop the standard corporate governance problems faced by any listed firm. This is because “the state” is a distinctive type of owner. State ownership creates its own agency problems, which are caused by the separation of the politicians and bureaucrats who oversee SOEs from “the citizens” on whose behalf the enterprises are ostensibly owned. From time to time, problems caused by these distinctive governance challenges explode into scandals, as exemplified by recent events in Brazil. Brazil’s oil giant Petrobras, a state-controlled firm whose shares are listed on both Brazil’s stock exchange (B3) and the New York Stock Exchange, was previously held up as an international model of solid corporate governance and performance.4 But it has recently been embroiled in crisis on two fronts.

First, the government used Petrobras as a tool of macroeconomic policy in the last decade by keeping oil prices significantly below the international market price to fight inflation. This resulted in major losses to the company and a reduction in its investment capacity.5 Second, Petrobras is at the center of the Lava Jato (Carwash) investigation, which uncovered the largest corruption scandal in Brazil’s history. The then ruling Workers Party (PT) and its coalition partners appointed many of Petrobras’s most important executives.6 A massive bid rigging scheme orchestrated by some of these appointees generated a campaign slush fund for PT politicians while providing lavish payouts to the corrupted insiders.7 Petrobras has so far written off more than two billion dollars in direct corrupt payments associated with Lava Jato,8 but analysts estimate that the total losses will be far greater.9

In this Article, we evaluate the governance challenges associated with mixed ownership of enterprise, and examine a variety of different international approaches to the governance of listed SOEs, with a view to framing a robust policy discussion in the many countries where SOE reform is a topic of major significance. Brazil is one such country, where the crisis just described has motivated both private sector and government initiatives

to restore investor and popular confidence in these firms.10 In September 2015, B3, the Brazilian stock exchange (which was then called BM&FBovespa), launched an innovative SOE Governance Program (Programa de Governan¸ca de Estatais), which offers special certifications to listed SOEs that agree to comply with various sets of corporate governance best practices.11 In June 2016, Brazil enacted a new statute on SOEs, which includes a number of special governance rules for these firms.12

Although the governance challenges associated with listed SOEs are common to all mixed-ownership firms, countries have responded to these challenges in myriad ways. In this Article, we describe the evolution and current status of the institutional framework applicable to listed SOEs in eight different jurisdictions, which reflect a variety of economic, legal, and political environments: France, the United States, Norway, Colombia, Brazil, Japan, Singapore, and China. We leverage the lessons from this comparative analysis by critiquing the policy prescriptions of international agencies and framing our own policy suggestions.

The selection of countries for our study is based on several considerations. We chose to include a diverse sample of countries across a range of dimensions: geography, stage of economic development, governmental structure and philosophy, and the relevance of SOEs to the domestic political economy. We included Singapore and Norway, countries that are highly regarded for their SOE governance. We selected China because it is a large, developing country in which SOEs play a particularly important economic role and in which a mixed-ownership strategy has long been its preferred path toward enterprise reform. China’s experience is also interesting because at times it has looked to Singapore as a model for its reform of mixed enterprises. In Norway, Brazil, and Colombia, SOEs operating in the oil industry are particularly important to the national economy, despite vast differences among these countries along a host of other dimensions. France was chosen as an example of an advanced democracy in which SOEs play an important role in the political economy. We included the United States and Japan in part due to the size and importance of their economies. In addition, the U.S. is often viewed (justifiably or not) as a leader in corporate governance reform, while Japan has adopted a distinctively gradual plan of partial privatization for important firms in its economy, a process that is ongoing to this day.

The national experiences we discuss are rich and diverse. We use them to provide a perspective from which to critically evaluate the guidelines on SOE governance prepared by the Organisation for Economic Cooperation and Development (OECD) and the World Bank. They also help frame a list of subjects ranging from ownership structures and board composition to executive compensation and the noncommercial, “public pol-

icy” role of mixed-ownership enterprises that policy makers may find useful as they seek to address the governance challenges of listed SOEs.

The Article is organized as follows: Part I outlines the rationale for mixed ownership of enterprise and the distinctive governance challenges associated with this organizational form. Part II provides analytical narratives on the evolution and key attributes of governance arrangements for listed SOEs in eight countries in different parts of the world and at different stages of development. Part III analyzes the governance patterns (or lack thereof) that emerge from the various national experiences we examine. Part IV critiques the prescriptions of best practice guidelines of international organizations. Part V offers a conceptual framework for improving the governance of listed SOEs.

I.Governance Challenges

Why are mixed-ownership enterprises so prevalent around the world, and how do their governance challenges differ from those of private firms?13 The prevalence of listed SOEs—in which the government retains a controlling or blocking stake in a firm with private investors—may at first blush appear to present a puzzle: If a government wants to be directly engaged in commercial activity, why cede partial ownership of the firm to private investors? The answer is that listing of SOE shares on stock exchanges has a number of potential advantages over whole ownership by the state. SOE performance may benefit from both the monitoring efforts of outside investors and the tighter governance and regulatory constraints applicable to publicly traded firms. Moreover, capital obtained through partial listings of shares may permit states to reallocate funds to other uses, such as paying off public debt or performing critical government functions, while not relinquishing control over corporate operations or the other perceived benefits of government ownership of enterprise.14

Nevertheless, as the recent Brazilian experience makes clear, the listing of SOE shares does not eliminate, and often exacerbates, corporate governance challenges. Listed SOEs generally take the form of a joint-stock business corporation. The corporate form, in turn, gives rise to two primary agency problems: (1) between managers and shareholders (which is more severe if corporate ownership is dispersed) and (2) between controlling shareholders and non-controlling shareholders (which is more severe if corporate ownership is concentrated). These problems do not disappear, and in fact intensify, when the state is a large shareholder, though the relative strength and contours of these problems will depend on how the state acts as an owner.

The state is a distinctive type of owner. It is an economic and political organization in its own right, giving rise to another layer of agency costs— which might be called the “agency costs of state capitalism.”15 The main agency cost within the state is that between government officials (elected or not) and citizens, who should in theory be the ultimate beneficiaries of state action. As is well known, the agency costs within the state are particularly severe, for various reasons. First, the exit options enjoyed by citizens are far weaker than those available to shareholders (and, in non-democratic societies, the voice option is also virtually non-existent). Second, the collective action problem faced by citizens in monitoring politicians is more intense than the one facing shareholders in private firms. Third, the lack of a clear consensus on which objectives the government should pursue— as well as on the means to accomplish such objectives—hinders the development of effective mechanisms of accountability.Consequently, there is great risk that the actions by government officials will serve their own interests in enhanced power and wealth, rather than the interests of citizens.

Listed SOEs may face different problems depending on how the state behaves as a shareholder. If it acts as a passive or absentee owner, managerial agency problems will prevail, so SOEs may suffer from managerial slack and managerial tunneling (i.e., theft of corporate assets). Although these problems are common to private firms as well, they tend to be more severe in SOEs, given the absence of a market check on managerial agency costs. The presence of the state as a large shareholder prevents the operation of a market for corporate control through hostile takeovers. In addition, implicit state guarantees undermine the threat of bankruptcy as a source of managerial discipline.

If the state is actively engaged as a shareholder, this can, in theory, reduce managerial agency problems but at the cost of increasing the potential for abuse by the controlling shareholder. As in private firms, there is a risk that the state will appropriate to itself a disproportionate share of SOE returns—what the corporate governance literature calls pecuniary private benefits of control16—to the detriment of minority shareholders. State ownership exacerbates other risks, however, such as the possibility that government officials will appropriate financial value to themselves to the detriment of both citizens and minority shareholders—i.e., the risk of corruption. Moreover, compared to private controlling shareholders, the state has stronger incentives to pursue non-pecuniary private benefits of control. These can be benign, as in the pursuit of corporate policies that enhance social welfare, even if they fail to maximize shareholder value. However,

they can also be malign, as when SOEs favor political allies in awarding contracts to the detriment of both citizens and minority shareholders. Consequently, while the strong role of the state as a shareholder may mitigate managerial agency problems, it opens the door to private benefits of control, political favoritism, and corruption.

The combination of these two layers of agency costs creates areas of both alignment and misalignment between the interests of citizens and outside shareholders in listed SOEs. Both shareholders and citizens have a shared interest in (1) increasing managerial effort, (2) reducing managerial tunneling, and (3) curbing rent-seeking behavior by politicians (from favoritism to outright corruption). Mixed ownership, however, also creates conflicts of interest between shareholders and citizens with respect to other dimensions, such as (1) “policy channeling”—a government’s pursuit of social welfare or other non-financial policy objectives through ownership of SOEs,17 which favors citizens but not shareholders; (2) the appropriation of disproportionate financial benefits by the state (which favors citizens, at least in the short term, but not shareholders); and (3) the award of subsidies to SOEs, which favors shareholders, but not necessarily citizens (who pay for these subsidies), which may disrupt the level playing field between SOEs and private firms and hamper competition.

The main mechanisms to address these two layers of agency costs are general corporate laws, on the one hand, and general political and legal institutions, on the other. Strong corporate laws, backed up by effective enforcement mechanisms, tend to benefit minority shareholders of SOEs as well. Moreover, the governance of SOEs (including listed SOEs) is also inherently dependent on how the existing legal and political institutions reduce or magnify the agency problems between politicians and citizens. For instance, the challenges associated with Brazil’s political system and mechanisms of campaign finance likely played a significant role in the recent SOE corruption scandals. There is little question that SOEs will tend to work much better both when corporate laws mitigate private benefits of control and when political institutions effectively constrain corruption and mismanagement by government officials.

Nevertheless, despite their obvious importance, neither general corporate laws nor general political institutions are the primary object of our

attention here. Instead, our focus is narrower, centering on institutions and strategies that target more directly the peculiar challenges facing listed SOEs. In contrast to other areas of corporate law and governance, the scholarly literature on the institutional and governance frameworks applicable to listed SOEs is relatively scarce. We seek to redress the imbalance in the sections that follow.

II.National Experiences

National experiences with the governance challenges of mixed-ownership enterprises have rarely been subjected to side-by-side comparison. Even beyond the purely informational benefits of the exercise, understanding how different countries have approached mixed-ownership enterprises promises a policy payoff, due to the common nature of the governance challenges faced by all countries with this form of economic organization. A side-by-side analysis also provides a useful perspective from which to evaluate the best practice guidelines for SOEs adopted by international organizations such as the OECD, a subject we address later in the article.

A.France

Famous for its state-centered society and dirigiste economic policy, France has a long history of government involvement in business, dating back to the trading companies from the era of Louis XIV. Yet corporatized SOEs were virtually non-existent in France until the twentieth century.18 In the few instances in which the state engaged in economic activity in the nineteenth century, it did so directly—as in the case of its tobacco monopoly (1810) and the postal service (1851)—without resorting to a separate business entity.19 Mixed enterprises first appeared in France in the interwar period as a transplant and heritage from German experience, following the return of Alsace-Lorraine to France.20 The French state initially participated as a minority shareholder in the first mixed enterprises of the 1920s.21 However, following the Great Depression and World War II, majority ownership and control became more common, as the government sought to rebuild the economy and bridge France’s historical gap in certain economic sectors.22

Under the socialist government of Fran¸cois Mitterrand in the early 1980s, France witnessed its most recent wave of nationalizations, when the state came to own thirteen of the country’s twenty largest corporations and almost the entire credit sector.23 France then underwent two waves of privatizations in 1986 and 1993 and, since then, has alternated between further privatizations, new equity investments by the state in private firms, and periods of the so-called “ni-ni” policy of “ni privatisation, ni nationalisation” (neither privatization nor nationalization). The weight of the state’s participation in French listed companies has oscillated between thirteen percent in the early 1980s to three percent in 1996 and ten to twelve percent following the financial crisis.24 Minority holdings by the French state are increasingly prevalent. As of 2014, France ranked sixth among OECD countries in terms of the value of enterprises under majority state control, but first in view of the market value of firms subject to minority ownership by the state,25 amounting to government shareholdings of over eighty-three billion euros in listed firms alone.26 Recent laws have encouraged the state to sell partial stakes in SOEs to raise revenue without necessarily relinquishing control and the pursuit of public policy objectives.27

Given the traditional centrality of state ownership to France’s style of capitalism, dedicated systems of SOE governance have emerged and evolved over time. We will address them in two parts, first focusing on the special governance arrangements applicable to SOEs, and then turning to the recent experience with the centralization of the state’s shareholding function in a specialized government agency since 2004.

Special Legal Regime: As in other countries, mixed enterprises in France have traditionally been subject to a hybrid legal regime. As a rule, these companies take the form of a business corporation (soci´et´e anonyme) and are subject to the same legal regime governing private firms. However, numerous exceptions apply. First, industry regulations and statutory charters can create firm-specific carve-outs from general corporate laws, which led French jurist George Ripert to decry the system of “une loi par soci´et´e” as early as 1946.28 For instance, the Civil Aviation Code allows the minister in charge of civil aviation to appoint a government commissioner to attend board meetings of listed firm A´eroports de Paris in an advisory capacity.29 Second, as will be further explained below, there are a number of general laws subjecting SOEs to distinct rules that set them apart from privately owned enterprises (POEs).30 While we focus here on the rules of corporate law and governance, it is important to recognize that, unlike POEs, SOEs (notably those under majority state control) are also subject to various public law constraints, including the broad supervisory authority of the Court of Auditors (Court des comptes),31 the exercise of economic and financial control by the state,32 and requirements of prior state approval of certain decisions.33

Starting in the mid-twentieth century, the state enjoyed the right to appoint board representatives in proportion to its equity ownership whenever it held more than ten percent of the shares, subject to a minimum of two members and a maximum of two-thirds of the board seats.34 In 1983, however, the prominent “law for the democratization of the public sector” (known as “loi DSP”) established a number of special rules for majority-

owned SOEs, including listed firms.35 In companies where the state held a majority but less than ninety percent of the shares, workers had the right to appoint one-third of the board members (or three out of eighteen board members in companies employing up to 1,000 employees).36 The remaining two-thirds of members were appointed by the shareholders meeting subject to the state representatives.37 However, the number of state representatives was not specified by the loi DSP and, in practice, it so happened that the state was under-represented in SOE boards.38

to

The state representatives

the board of directors (conseil d’administration) were appointed by Presidential decree, as was the officer combining the roles of chairman of the board and president (pr´esidentdirecteur general (PDG —a position that has traditionally concentrated significant power under the French system of corporate governance.39 This, in turn, has enabled politically motivated changes in management following electoral cycles.40 The loi DSP formally imposed a five-year term for board members (which is lower than the six years permitted under the commercial code but higher than the four years recommended by France’s Corporate Governance Code Afep-Medef), though in practice there has been a high turnover of state appointees.41

This system underwent a significant transformation in 2014 with the enactment of new laws ostensibly aimed at simplifying SOE governance and approximating it to the regime applicable to POEs.42 The new rules withdraw all entities taking the form of business corporations (including majority-owned SOEs) from the scope of the loi DSP.43 They eliminate prior requirements regarding board size and director terms, and permit the state to appoint certain board members who are not public servants.44 Yet the key goal of this reform was to strengthen the government’s influence as a shareholder. First, pursuant to the new regime, the state may appoint directly by ministerial order one state representative (who must be a public

servant) in all majority-owned SOEs.45 The state may also appoint one representative in all firms in which it directly holds more than ten percent of the shares.46 This is in contrast to the former rules, according to which the shareholders meeting appointed the state representatives upon the proposal of the state.47 Second, in firms in which the state holds a direct participation between ten percent and fifty percent of the shares, the government also has the right to propose at the shareholders meeting a number of board members in proportion to its equity interest in the firm.48 These rules help put “an end to the paradox that led the state to have less influence as a shareholder in state-owned enterprises than a private shareholder.”49 Nevertheless, convergence to the private legal regime remains elusive, in view of the generous appointment rights enjoyed by the state as well as by employees, who retain one-third of board seats in majorityowned SOEs.50

Another area of both traditional and recent concern under French law is executive compensation in SOEs. Under the loi DSP, state-appointees to SOE boards did not receive any remuneration for their services (though expenses were reimbursed).51 Since 2001, companies must pay for the services of state representatives on the board, but the amounts revert to the government budget.52 Moreover, since 2012 the remuneration of managers in majority-owned SOEs is subject to a cap, currently set at C=450,000.53 The new state appointees who are not public servants, as permitted by the 2014 reform, may receive compensation up to a certain cap to be determined by the Minister of the Economy, with the excess amounts also reverting to the government.54

Beyond the special legal regime governing SOEs, the state’s interests as a shareholder have also shaped France’s general corporate laws, a pattern

that is particularly clear in the area of shareholder voting rights.55 While French law prohibits the issuance of multi-voting stock, it has long permitted companies to grant double voting rights to investors holding shares for at least two years.56 More recently, the openly protectionist loi Florange of 2014 has provided for the automatic application of double voting rights to shares held for at least two years (unless shareholders opt out by securing a two-thirds majority vote)—a reform that was explicitly driven by the state’s interests as a shareholder.57 The broader attribution of double voting rights would permit the French state to divest some of its stockholdings to reduce the national debt while at the same time preserving or magnifying its influence as a shareholder in companies of strategic importance. In a controversial and widely followed episode, the French state acquired additional shares in Renault and Air France-KLM in order to secure the automatic application of the double voting rights under loi Florange.58

Government Shareholding Agency: The second pillar of the SOE infrastructure relates not to the rights granted to the state as shareholder under corporate laws and industry regulations, but to the institutional framework for the state’s exercise of the ownership function. In 2003, the influential Report by Ren´e Barbier de La Serre and others identified a number of shortcomings of the existing system of SOE governance, including considerable confusion in the various roles played by the state with respect to its enterprises, the poor functioning of the board of directors, and the excessive presence of the state in day-to-day management.59 As a result of the Report’s recommendation, France established its Government Shareholding Agency (Agence de participation de l’´Etat (APE with the goal of dissociating the state’s shareholding function from its role as a regulator.60

Associated with the Ministry of the Economy, the APE centralizes the shareholding responsibilities with respect to numerous companies subject to whole or partial ownership by the state.61 Not all government equity

stakes are held through APE, however. For example, while the French state accounts, directly or indirectly, for over eighty-five percent of the capital of AREVA, the APE holds only 28.83% of its shares; the main shareholder is the CEA, France’s Alternative Energies and Atomic Energy Commission.62

The APE should consider the financial interests of the state in exercising its role as a shareholder, but it exercises this mission in conjunction with other ministries in charge of the regulatory responsibilities of the state.63 The APE may be consulted for the appointment and removal of board members nominated by decree or ministerial orders.64 The APE also represents the state in shareholder meetings and ensures, together with the government commissioner (if any), the coherence of the positions of government representatives.65 Moreover, the APE participates, as and when required, in drafting the relevant contracts between its portfolio companies and the government.

In contrast to Singapore’s holding company Temasek (discussed below), the APE does not limit its role to managing its portfolio companies under strictly commercial terms.66 Instead, it also focuses, especially in recent years, on “managing its investments from an industrial perspective, and on establishing a clear, long-term industrial and economic development strategy for the companies concerned.”67 Such a strategy must “simultaneously optimize the value of its assets and the specific business and social aims of each of the companies concerned,” especially in strategic sectors such as national defense, energy, and the automobile industry.68 The most recent APE report specifically mentions that the 2014 legal reforms permitted the state to embrace a stronger and more active role as a shareholder in consideration of the “general interest.”69 The Agency also takes pride in its responsible and stable dividends policy, calibrated to the risks involved and the practices of comparable companies in regulated sectors.70 Nevertheless, its above-market rate of dividend payments has attracted criticism by the Court des comptes, which has raised concerns over a short-term orientation by the state as a shareholder to the

detriment of long-term investments.71

Finally, in 2014 France disclosed its new policy for intervening as a shareholder, which is comprised of four key objectives reflecting a mixture of industrial policy, sovereignty, and macroeconomic considerations: (1) ensuring sufficient government control over companies of strategic public interest in areas key to France’s sovereignty; (2) guaranteeing the companies’ ability to fulfill the country’s basic requirements; (3) supporting corporate growth and consolidation, particularly in sectors and industries that are essential to French and European economic concerns; and (4) subject to EU regulations, undertaking corporate bail-outs on an ad hoc basis in systemic risk cases.72

B.United States

The United States boasts the world’s largest capital market and is the principal originator of “best practices” in corporate governance. The United States is also exceptional in its traditional hostility to government ownership of enterprise.73 At present, there is not a single domestic SOE (conventionally defined) among the thousands of companies listed on U.S. exchanges, although, as discussed throughout this study, numerous foreign SOEs cross-list their shares in the United States and are subject to U.S. regulatory authority.

This does not mean, however, that the United States is devoid of experience with mixed enterprises or other forms of government involvement in corporations. In fact, government shareholdings in private firms were particularly common in early periods of U.S. history, even if intentionally eliminated thereafter.74 The twentieth century saw the emergence of important government-sponsored enterprises (GSEs), which, like SOEs, also entail the coexistence of public support and mission on one hand, and outside shareholders on the other.75 More recently, the U.S. government temporarily acquired equity stakes in various private firms in the bail-outs

following the financial crisis of 2008.76

While mixed enterprises have been rare to non-existent in recent times, U.S. states were frequently shareholders in local banks, railroads, and canals in the late eighteenth and early nineteenth centuries.77 Irrespective of any potential benefits of government support in the early days of American capitalism,78 these investments posed a clear conflict-of-interest problem from the perspective of the states’ role as regulators. Various studies have documented the reluctance of different states to charter new corporations when they had shareholdings in the same industry, due to the fear that increased competition could jeopardize the government’s financial return from existing investments.79

Although systematic studies on the governance practices of these early mixed enterprises are lacking, existing evidence suggests that government interference in corporate decision-making was fairly limited from the outset.80 States routinely appointed directors to corporate boards,81 but generally refrained from exercising their voting rights to interfere with management by private managers.82 Even so, these public stakes in business firms eventually became controversial as the 1837 depression left states with bad investments.83 The distinctive U.S. solution at the time was to eliminate these hybrid entities.84 As part of a broader set of reforms that sought to encourage fiscal responsibility, most state constitutions in the mid-nineteenth century came to prohibit state shareholdings in private companies.85

Early instances of stock ownership by the federal government were fewer, but the companies involved played a major role in the U.S. economy. One example is the U.S. government’s participation in the First Bank of the United States of 1791.86 Conceived by Treasury Secretary Alexander Ham-

ilton, the Bank displayed the combination of public mission with essentially private management—a recipe that would become the hallmark of the U.S. model of hybrid enterprises.87 The bank Hamilton envisioned would not be “a mere matter of private property, but a political machine of the greatest importance to the State.”88 At the same time, he vigorously argued that the Bank should be “under a private not a public [d]irection, under the guidance of individual interest, not of public policy,” given the well-recognized risks of governmental abuse.89

This suspicion of government involvement, in turn, also translated into a more limited degree of state ownership. Hamilton advocated against both government involvement in management and whole or majority stock ownership by the state.90 The main right of the government was to inspect the Bank’s operations.91 While the federal government held twenty percent of the Bank’s stock at its founding, it soon sold its shares in order to repay its debt to the Bank.92

The federal government was also a twenty percent shareholder at the founding of the Second Bank of the United States in 1816.93 Unlike its predecessor, this new Bank’s charter followed then-prevailing practice in the United States by providing for the government to appoint five (out of twenty-five) directors.94 Although commensurate with the federal government’s initial financial investment in the Bank, the presence of government appointees proved controversial and of doubtful practical importance. The Bank reportedly concealed information from the government-appointed directors, fearing a conflict of interest.95

The Union Pacific Railroad, chartered by Congress in 1862, was the last grand experiment with federal support of corporations in the nineteenth century.96 This time, however, government ownership of stock was replaced entirely in favor of a model based on generous public subsidies through land grants and loans, coupled with the appointment of a minority of government directors (initially two and later five out of twenty board members).97 As with the Second Bank of the United States, government appointees to the board reported that they were kept out of meetings and generally antagonized by the other directors.98

Government-Sponsored Enterprises: Fast-forwarding to the twentieth century, the U.S. largely resisted the growing international trend toward

state ownership of industry. In response to the proliferation of government corporations following the Great Depression, Congress enacted the Government Corporation Control Act (GCCA) of 1945, which sought to restrain the formation of government corporations and enhance their accountability, by, among other things, requiring Congressional authorization for the establishment of new government corporations.99 Despite these constraints, new hybrid entities, which came to be known as government-sponsored enterprises (GSEs), gradually emerged. GSEs stayed true to the earlier model of state support without ownership: they were chartered by the federal government, imbued with a public mission, but owned and controlled by private shareholders. The government’s role in their governance was formally limited to the appointment of a minority of directors by the President of the United States.100

The most prominent GSEs are Federal National Mortgage Association (Fannie Mae) and Federal Home Loan Mortgage Corporation (Freddie Mac), which became publicly traded on the New York Stock Exchange in 1968 and 1989, respectively.101 Their mission is to promote access to housing by supporting the secondary market for residential mortgages generally, and, since the 1990s, also by facilitating financing to low- and middle-income borrowers.102 The federal government chartered Fannie Mae and Freddie Mac, and they enjoyed different legal privileges, such as exemptions from state and local income taxation as well as from federal securities laws.103 However, the main form of public support to these entities came in the form of an implicit government guarantee. Even though their federal charters explicitly disclaimed any liability or guarantee from the U.S. government for their debts, market participants widely recognized that the government would stand behind their obligations in the event of default—a perception that the financial crisis of 2008 proved to be accurate. As a result, these entities were able to borrow at significantly lower cost their expansion.104

firms, which contributed

than comparable private

to

Even prior to the financial crisis, Fannie Mae and Freddie Mac were politically controversial on both sides of the ideological spectrum. Fannie Mae, in particular, was embroiled in accusations of accounting fraud in the

early 2000s.105 Partly in response to this scandal, there was even greater governance forbearance on the part of the federal government. In 2004, President George W. Bush decided to stop using his prerogative to appoint government directors to Fannie Mae and Freddie Mac, a policy that President Obama continued, to the effect that private shareholders could then elect all directors.106

Post-Financial Crisis Experience: It was not until the financial crisis of 2008 that the U.S. government would make unprecedented acquisitions of shares in private corporations, even if with significant hesitation.107 The idea of having the federal government acquire equity stakes in distressed financial institutions emerged as a policy transplant from England, a country with far greater experience with mixed enterprises.108 Nevertheless, the U.S. government consistently described itself as a “reluctant shareholder” whose role was meant to be temporary.109 This reluctance manifested itself in the structure of the government’s investments, which reflected lower equity stakes and control rights than standard commercial practice would have warranted.110

The government bailouts during the financial crisis did not follow a single model. On the contrary, deal structures were crafted ad hoc, leading to significant heterogeneity in governance arrangements.111 In the case of General Motors, the federal government acquired a controlling stake by exchanging its existing loan for an equity interest in the firm during its bankruptcy proceeding.112 The government held its stock in the company directly and appointed ten out of twelve board members, but generally refrained from intervening in the daily management of the firm.113 By mid-2009, all directors appointed by the government were required to be independent, and would take responsibility for further director nominations, but the government retained the right to vote against or remove them.114

In the case of AIG, the government acquired a large equity stake of nearly eighty percent of voting stock, but surrendered direct control by transferring its shares to a trust—a move largely motivated by doubts about

the federal government’s legal authority to hold shares in corporations without prior Congressional authorization.115 The Trust Agreement also specifically mentioned the Federal Reserve Bank of New York’s intention of not exercising its voting rights for AIG stock in order to “avoid any possible conflict with its supervisory and monetary policy function.”116 Although the trustees retained sole discretion with respect to the management of the trust, the Trust Agreement explicitly mentioned the government’s expectations that the trustees would not intervene in the daily management of the company.117 The Agreement also referred to the government’s non-binding view that maximizing AIG’s ability to repay its debt to the government and not disrupting financial markets were consistent with maximizing the value of the trust stock.118

By contrast, in the case of Citigroup, the federal government came to hold its large equity interest of approximately thirty-four percent of the shares directly.119 The government still limited its control power by agreeing to vote in the same proportion as the other shareholders in routine matters, but preserved its voting rights with respect to the election of directors and major corporate transactions.120

Finally, the GSEs Fannie Mae and Freddie Mac were subject to the greatest degree of government intervention. The federal government took them into conservatorship in 2008, thus acquiring control and suspending the voting rights of shareholders.121 In 2012 the Treasury entered into a controversial “net worth sweep” with the companies, which essentially transferred their future profits to the government at the expense of private shareholders.122 This development attracted significant criticism as an unlawful expropriation of private investors,123 but it has thus far withstood

judicial scrutiny.124

With the exception of Fannie Mae and Freddie Mac, the government has since sold the shareholdings acquired during the financial crisis, and the various firms have returned to wholly private ownership.125 In fact, the decision to divest as promptly as possible was a governance decision aimed at curbing the risks of political interference. From the perspective of maximizing the government’s financial return on its investment, the timing of its exit may have been premature.126 In any case, the short duration of this experiment, combined with the great variety of institutional arrangements employed, makes it difficult to draw definitive conclusions about the effectiveness of the different governance arrangements employed in the wake of the crisis.

However, it is still possible to discern certain lessons. First, the U.S. has adopted a distinctive approach to mixed enterprises: it sought to avoid its predictable challenges by shunning government ownership through clear constraints in state constitutions and federal legislation. To be sure, recent works have argued that state collaboration with, and support of, businesses in the United States is far greater than usually acknowledged.127 Nevertheless, such support generally took the form of government loans and grants, rather than equity investments.128

Second, even if helpful, the various mechanisms to curb government influence, such as the transfer of ownership interests to a trust, and strong reliance on independent directors, were insufficient to eliminate the specter of political intervention in the companies’ governance and management.129 Admittedly, there is no consensus on the optimal degree of political influence over business corporations owned by the state, and certain commentators have criticized the bailout deal structures for bestowing too little control on the government.130 Nevertheless, to the extent that the objective was to avoid any form of politically motivated intervention, it is not obvious that the governance mechanisms employed were effective.

Third, the existing legal infrastructure has largely failed to constrain

the government’s role in the bailouts.131 Scholars have noted that U.S. corporate law is ill-equipped to address the particular political risks created by government ownership—a problem that is only aggravated by the relatively broad scope of the doctrine of sovereign immunity in the United States.132 Moreover, the conflict of interests between the government’s role as regulator and shareholder remained apparent. The recent transaction permitting the government to appropriate all the profits of Fannie Mae and Freddie Mac is illustrative in this regard.133 Allegedly authorized by recent federal legislation, this move clearly belies the widespread notion that the government would not attempt to enrich itself at the expense of private investors.134

C.Norway

The level of state ownership in Norway is higher than in any other OECD country and comparable to that of large emerging economies.135 The motivation for state involvement in business has varied over time.136 Significant government ownership of enterprise in Norway dates back to the post-World War II period, when the weakness of local capital markets prevented private firms from financing industrial development.137 Following the discovery of vast oilfields off its shore, Norway established stateowned enterprise Statoil in 1972 as part of its institutional infrastructure to retain control over natural resources while shielding its economy from the black gold curse. Government ownership in the banking industry increased dramatically in response to a financial crisis in the late 1980s.138 In the last decade, state shareholdings have represented roughly thirty-five to forty percent of the Oslo Stock Exchange (OSE) market capitalization, up from approximately fifteen percent in the 1990s.139 The increase is largely due to IPOs of major SOEs—including Statoil and telecom company

Telenor—during the early 2000s.140

The Norwegian Model: In addition to the sheer significance of state ownership in its advanced economy, Norway also stands out in view of its acclaimed institutional arrangements. The “Norwegian model” for the oil sector—premised on the separation of policy, regulatory, and commercial functions—has become a blueprint for resource-rich countries (though questions persist about its adequacy in different contexts).141 Here we will focus on two dimensions of the institutional framework: (1) the exercise of the shareholder function by the state and (2) the corporate law and governance framework applicable to listed SOEs.

Shareholder Function: Under Norway’s constitution, state-owned enterprises fall under the administration of government ministries, but the Norwegian Parliament (Storting) has express authority to instruct the government with respect to SOEs.142 This framework requires the prior consent of the Storting for changes in the state’s shareholdings (acquisitions and divestitures), as well as for capital increases entailing disbursements by the state.143 However, SOEs are generally able to buy and sell shares in other companies without Storting approval when the transactions are part of their regular business activities.144 The Office of the Auditor General of Norway oversees the administration of SOEs by the relevant ministry and provides annual reports to the Storting.145

The pursuit of formal differentiation between the state’s role as shareholder and regulator is a hallmark of the Norwegian model. Norway has followed the trend toward centralization of shareholdings by allocating interests in most commercial SOEs to the “ownership department” of the Ministry of Trade, Industry and Fisheries, especially since 2001.146 For instance, the ownership department exercises the shareholding function of Telenor, while the Ministry of Transportation and Communication serves as the company’s regulator.147 Nevertheless, important exceptions to centralization persist, as in Statoil, whose shareholdings are administered by the Ministry of Petroleum and Energy.148 Still, even here there is a formal

separation of functions: Statoil, like private oil companies, is subject to regulatory oversight by the Norwegian Petroleum Directorate, a technical advisory agency.149

In the mid-2000s, Norway carefully considered but ultimately rejected the possibility of instituting a holding company model.150 A 2004 report by the preparatory committee in charge of reviewing the organization and administration of state ownership offered only timid support for the establishment of a holding company to manage state shareholdings with purely value-maximization objectives, and pointed to the need for further assessments.151 The same committee counseled against the use of a holding company structure for SOEs serving the goal of keeping head office functions in Norway, in view of the perceived need for continued political governance and supervision.152 The government eventually declined to create a holding company, which it regarded as creating unnecessary duplication of functions and confusion of responsibilities.153 It argued that “ownership matters are of such a character that they need to be handled through a political body,” and that “[t]he current ministerial affiliations ensure transparency concerning ownership and ensure considerations of democratic control.”154

Governance Regime: Norway’s listed SOEs are subject to the same corporate and securities laws governing private firms, including the Public Limited Liability Companies Act and stock exchange regulations.155 Norway’s corporate law provides for a strong principle of equal treatment.156 Majority shareholders have significant decision-making powers under Norwegian law, but may not act in abuse of power to the detriment of the company and other shareholders.157 There is special concern that the state, in its role as a shareholder, not receive preferential access to information vis-`a-vis private shareholders.158 The Oslo Stock Exchange has also contributed to enforcement of the equal treatment norm by issuing letters questioning whether statements by Statoil management reflected private information not available to public investors or resulted from non-financial

considerations.159 The Oslo Stock Exchange and the Norwegian Annual Accounts Act also require all listed companies to report on their adoption of the Code of Practice for Corporate Governance on a “comply or explain” basis.160

The state’s involvement as a shareholder must take place through the shareholder meeting, in accordance with the Norwegian Public Limited Liability Companies Act, though the Ministry of Oil and Energy has admitted to exercising influence through informal meetings as well.161 At the shareholder meeting, shareholders of large companies elect two-thirds of the members of the corporate assembly, with workers electing the remaining one-third.162 The corporate assembly elects two-thirds of shareholder representatives and one-third of worker representatives to the board.163 Companies may, however, opt out of the requirement of a corporate assembly by obtaining workers’ consent, in which case shareholders and workers directly elect approximately two-thirds and one-third of board members, respectively.164

The board of directors appoints the CEO, who may not be a board member, and sets his or her salary.165 There are no state representatives on the board of listed SOEs, but ministry representatives engage, directly or indirectly, with the nomination committee.166 The nomination committee, which is comprised of shareholders or shareholder representatives, submits recommendations to the shareholder meeting and the corporate assembly (if there is one) for the appointment of the respective shareholder-elected members and the setting of their compensation.167 In practice, there seems to be moderation in the state’s involvement: even though the state holds sixty-seven percent of the equity in Statoil, it has recently appointed only one member (from the Ministry of Petroleum and Energy) out of four members of the nomination committee.168

A distinctive trait of the Norwegian system compared to other countries is that currently serving politicians and public servants from the central government may not serve on SOE boards.169 This prohibition traces back to a 1962 fatal accident involving a state-owned mining company, which had the Minister of Industry serving on its board.170 The incident, which came to be known as the “King’s Bay affair,” resulted in allegations of negligence and ultimately brought down the ruling labor government.171 The original rationale for the ban was not primarily to prevent political interference in management, but rather to mitigate conflicts of interest in the government’s oversight of SOEs172 and to discourage Storting from holding the government accountable for the business decisions of stateowned companies.173 The existing restriction on board membership, however, does not encompass former politicians, whose participation in SOE boards remains relatively common.174

Norway has continuously strived to strengthen the corporate governance of SOEs, which is touted as “of vital importance for the market’s confidence in the companies and hence also for the companies’ capital costs.”175 Moreover, the state has repeatedly acknowledged that, given its large participation in listed companies, “[t]he manner in which the State acts as an owner therefore has great influence on public and investor confidence in the Norwegian capital market.”176 The state has chosen to follow the principle of proportionality between capital invested and voting rights in SOEs, avoiding the introduction of special rights to the state as shareholder.177 In the International Monetary Fund’s assessment of SOEs in Norway, “[m]ost large enterprises operate on a commercial basis and are profitable.”178

In 2002, the Norwegian government formulated ten principles of corporate governance for SOEs, with the goal of increasing predictability in the exercise of ownership by the state.179 These principles were subject to modest amendments in 2014 to further underscore the role of the board of directors in SOE governance and administration and the commitment to corporate social responsibility.180 As emphasized by the Norwegian gov-

ernment, its principles essentially correspond to the OECD Guidelines on the Corporate Governance of State-Owned Enterprises.181

Since 2006, the government has also issued guidelines on the remuneration of senior executives of SOEs, partly out of concern over a widening remuneration gap between the senior employees and the rest of the workforce.182 The 2015 version of the guidelines, which are applied on a “comply or explain” basis, instructs that (1) executive salaries should be competitive, though not wage-leading; (2) the main element of compensation should be the fixed salary; and (3) the use of stock options and similar arrangements is prohibited.183 Norway’s Code of Practice for Corporate Governance encourages stock ownership by board members, and the state takes a positive view of this strategy.184

Also starting in 2006, Norway implemented the Preparatory Committee on State Ownership’s prior recommendations to clarify the fundamental objectives served by state ownership in each case, with the objective of reducing uncertainty in capital markets and consequently lowering financing costs.185 The Committee proposed to classify SOEs into four categories: (1) companies with commercial value maximization objectives; (2) companies with commercial value maximization objectives and ensuring head-office functions in Norway; (3) companies with commercial value maximization objectives along with other specifically defined objectives; and (4) companies with sectoral policy objectives.186 From the eight listed

SOEs in Norway, six fall within the second category (including Statoil, Telenor, and financial firm DNB), and two fall within the first category.187 The third and fourth categories are composed exclusively of non-listed SOEs.188

The 2010–2011 Report to Storting highlighted the contribution of extensive state ownership to the success of the Norwegian economy and the plan to strengthen the state’s ownership administration.189 More recently, however, the current conservative administration has vowed to reduce the level of state ownership in the economy, despite the fact that “[i]n the government’s assessment, the governance of direct state ownership is handled in a professional and responsible way.”190 The administration’s objective is to partially or fully divest companies in the first category, while maintaining at least thirty-four percent of stockholdings (enabling “negative control” through veto rights) in companies within the second category.191

The Norwegian government argues that three particular challenges associated with state ownership persist: (1) conflicts between ownership of companies and the state’s other roles; (2) the risk of a concentration of powers weakening the private sector; and (3) limitations in industry expertise.192 Specifically, the government has suggested that, notwithstanding effective governance arrangements, “[a]s long as the state has ownership interests, it is . . . effectively impossible for the state to be organised and to act in such a way as to prevent or discourage doubt being raised about its neutrality in exercising authority.”193 This initiative resulted in the 2013 IPO of Entra, a state-owned commercial real estate company.194

D.Colombia

State-owned firms are pervasive in Latin America, but the significance of listed SOEs to the respective local economies varies widely. Mexico, the only OECD country in the region195 and home to the national oil giant Pemex, has no listed SOEs.196 Chile and Peru, regarded by the World Bank

as the “regional frontrunners” in SOE governance reform,197 boast a few listed SOEs, but these SOEs represent a comparatively small share of the trading and capitalization of their respective markets.198 Thus, the next two sections focus on Colombia and Brazil, countries in which state-controlled firms play important roles in the stock market and the general economy, where significant strides have been made in SOE governance reform. Until recently, Colombia had four listed SOEs—Ecopetrol, ETB, ISA and ISAGEN—which together accounted for over fifteen percent of market capitalization and fifty percent of the total value of SOEs.199 These firms went public in the 2000s but the government kept a stake of nearly eighty percent on average.200 ISAGEN, however, was fully privatized in January 2016.201

Of these four SOEs, the main player is national oil company Ecopetrol, which has been the largest listed company in Colombia, with a budget representing almost seven percent of GDP and twenty percent of the government budget as of 2014.202 At one point, the company made headlines by overtaking Petrobras as Latin America’s largest oil company by market capitalization—an outcome partially motivated by perceptions about the quality of the institutional environment in Colombia.203 Recently, however, Ecopetrol’s economic performance has faltered due to the decline in inter-

national oil prices. It has also been the subject of a corruption probe relating to the payment of bribes by a foreign company to secure a lucrative contract.204

Under Colombia’s constitutional framework, the creation of SOEs, including mixed enterprises, must be authorized by statute.205 The Constitution also provides that divestitures of state shareholdings should seek to democratize ownership, especially among workers.206 Moreover, the Comptroller General’s Office (CGR) exercises fiscal control over SOEs under the Constitution and laws.207 Nevertheless, the OECD identified overly zealous oversight by the CGR as a potential problem that arguably has deterred boards of directors from making professional and independent decisions and assuming appropriate levels of risk.208

Mixed enterprises in which the state holds less than ninety percent of the shares are generally subject to a private law regime.209 Although mixed enterprises may adopt any of the organizational forms available under the Commercial Code, the corporation is usually the organizational form of choice.210 Listed SOEs are also subject to Colombia’s Securities Market Law and its corporate governance code (C´odigo Pa´ıs) on a “comply or explain” basis. According to a survey in 2011, listed SOEs ranked among the top companies for implementation of C´odigo Pa´ıs.211

Special Protections: The Colombian state, as a controlling shareholder, has opted to grant additional protections to minority investors. In a practice that started with the IPO of energy company ISA in 2000, the government has signed declarations committing it to adopt certain governance safeguards that go beyond those required by law.212 For instance, under Ecopetrol’s declaration, these protections include the right of minority shareholders to elect one independent director (out of nine board members), as well as to seek a valuation by an investment bank appointed by the Bogot´a Chamber of Commerce to determine the value of the shares in appraisal proceedings.213 Although the declaration is valid for an initial term of ten years expiring in 2017, the prior practice in other SOEs was to make these protections permanent by incorporating similar content into the companies’ charters.214

Colombia’s listed SOEs grant one vote per share. The board of directors appoints and removes the CEO in accordance with general corporate law.215 The boards of SOEs are comprised of a majority of independent directors,216 a practice that exceeds the twenty-five percent threshold imposed by securities laws.217 Ecopetrol has traded its ADRs on the New York Stock Exchange since 2008, and is therefore also subject to regulatory oversight by the U.S. Securities and Exchange Commission.218Ecopetrol traded ADRs on the Toronto Stock Exchange as well, but it voluntarily delisted from that exchange in 2016.219

The state has long employed Ecopetrol to subsidize the sale of fuel to consumers.220 In anticipation of the company’s partial privatization in 2007, the government assumed responsibility for the price subsidies.221 Under the current regulatory framework, Ecopetrol sells gasoline and diesel at a regulated price, but the government then reimburses the company for the subsidy.222 Although there were initial delays in the early reimbursements,223 this mechanism has apparently worked smoothly since, as Ecopetrol stopped mentioning the risk of such delays in its securities filings.224

Until its financial condition deteriorated in recent times, Ecopetrol has practiced generous dividend payments.225 As a majority shareholder holding 88.49% of the shares, the state stood to gain from high dividends as a

source of revenue, giving rise to suspicions of a short-term orientation on the part of the government.226 In fact, Ecopetrol’s securities filings warned that the state as a controlling shareholder “may approve dividends at the ordinary general shareholders’ meeting, notwithstanding the interest of minority shareholders, in an amount that results in us having to reduce our capital expenditures, thereby negatively affecting our prospects, results of operations and financial condition.”227

International Recognition: The general assessment of the firm-level practices of Colombia’s listed SOEs has been very positive.228 Observers have hailed Ecopetrol, together with the other listed SOEs in Colombia, as models of strong corporate governance.229 An OECD review of governance practices of SOEs in Colombia underscores the prevailing opinion by Colombian experts that the three listed firms (including ISAGEN, not yet privatized at the time of the review’s publication) “are good examples of professional management and excellent corporate governance,” a view that “is also borne out by the different awards they have received for good corporate governance” and “shared by public opinion in general.”230 Beyond the particular institutions, however, the OECD review also noted the perception that individual leadership played an important role in such success, singling out the contribution of Javier Guti´errez, longtime CEO of ISA and later of Ecopetrol.231

While the OECD concluded that Colombian practices are “good examples” and “internationally recognized” when it comes to the equal treatment of shareholders or stakeholders’ engagement, it identified a number of weaknesses in the exercise of the ownership function by the state.232 By law, the exercise of the ownership function lies with the minister or head of department to which the company is linked.233 Colombia has traditionally followed a decentralized model of SOE administration, with companies falling under the supervision of different ministries according to industry and regulatory affinity.234 It is however common for an SOE to be owned by a

certain ministry but regulated by another, which creates the potential for administrative conflicts.235 Representation of ministers and government officials on SOE boards, including those of ISA and Ecopetrol, is also common.236 According to the World Bank’s report, some director and managerial appointments are made on the basis of political allegiance rather than technical competence.237

In anticipation of joining the OECD in 2017, Colombia announced a plan to overhaul its system of SOE governance.238 The goal is to strengthen the role of the state as shareholder by centralizing the ownership function, initially under the Finance Ministry as a pilot project, and starting in 2019 under a new dedicated national entity that will act as the shareholder of all SOEs.239 Colombia has also committed to developing a general policy that clarifies the objectives of state ownership, communicating a clear mandate to firms, and institutionalizing the mechanisms for selection and evaluation of board members.240

In view of the goal of separating the roles of state as both shareholder and regulator, Colombia will phase out the participation of ministers and other government officials on SOE boards, starting with ISA and other nonlisted SOEs.241 In light of Ecopetrol’s importance to the Colombian economy, changes to its board practices will come last, in order to take advantage of the lessons learned with other SOEs.242 Finally, the government intends to publish a corporate governance code for SOEs.243

E.Brazil

While Brazil had early experiments with state-owned corporations in the nineteenth century, it was in the second half of the twentieth century that SOEs witnessed a significant expansion.244 Since then, listed SOEs have played a key role in the Brazilian economy and capital markets. Mixed enterprises in which the state held a majority of the voting rights accounted for a staggering seventy percent of stock market capitalization in the 1970s, the decade in which most of the current legal framework went

into effect.245 Even after the wave of privatizations in the 1990s and the IPO boom for private firms of the 2000s, SOEs still accounted for roughly one-third of Brazil’s stock market value in 2008.246 As of 2015, this proportion was down to approximately fourteen percent.247 Yet the reduction was not due to a major retreat of state ownership during this period, but rather to the decline in the stock prices of SOEs in recent years—in no small measure due to governance problems, as described below.

Legal Framework: Like other jurisdictions, mixed enterprises in Brazil are subject to a combination of both public and private law constraints. Brazil’s Constitution of 1988 conditions the direct undertaking of economic activity (other than general public services) by the state through SOEs on the existence of “national security imperatives” or “relevant national interest.”248 Mixed enterprises must be created by statute,249 and are subject to oversight by Tribunais de Contas, an external body that has a constitutional mandate to control the government’s activities and expenditures.250 Brazil’s Constitution requires mixed enterprises engaging in economic activity to be governed by the legal regime applicable to private firms, including as to civil, commercial, labor, and tax matters.251 However, statutory law has from time to time exempted SOEs from bankruptcy laws, though the constitutionality of this special regime remains the subject of debate.252 There are also constitutional exceptions to the application of the general corporate regime, such as the requirement that the annual federal budget law include the investment budget of SOEs controlled by the federal government.253

Mixed enterprises in Brazil (sociedades de economia mista) must be organized as a sociedade anˆonima (business corporation) and have been largely governed by general corporate and securities laws, except to the extent to which their statutory corporate charters abrogate the standard private law regime.254 Until the recent enactment of a special SOE statute

in 2016 (as discussed below), Brazil’s Corporations Law contained only a few specific provisions tailored to SOEs.255 Brazilian law had long afforded minority board representation in SOEs—a mechanism that is currently prescribed by Brazil’s constitution.256 Shareholders in private firms are entitled to appraisal rights in case of a subsequent government taking of control.257 Yet the most prominent and distinctive feature of the legal regime applicable to mixed enterprises in Brazil, among the jurisdictions examined in this Article, is a rule contained in Article 238 of the Corporations Law (now reproduced in modified form in the new SOE statute258). This rule specifically provides that the legal entity controlling a mixed enterprise has the same rights and duties as those of a controlling shareholder of a privately owned corporation, but it “may steer the company’s activity in order to satisfy the public interest that justified its creation.”259

Beyond these special rules applicable to SOEs under majority state ownership, the interests of the state as a shareholder in mixed enterprises have also shaped the content of the general corporate law regime in Brazil. The strong stakeholder orientation of fiduciary duties under Brazilian law—which is especially accommodating to the interests of the state as a controlling shareholder—was conceived against the background of equity markets populated by SOEs.260 However, the most conspicuous example of the influence of the state as shareholder in general corporate laws took place during the wave of privatizations in the 1990s, when a legal reform to the Corporations Law eliminated various minority shareholder rights in control transfers and spin-offs in order to facilitate privatizations and permit the appropriation of the entire control premium by the government as selling shareholder.261 Although listed SOEs have populated Brazilian

markets since the mid-twentieth century,262 the federal and state governments have made various new issuances of SOEs’ stock to the public in the late 1990s and 2000s. These included for the first time the issuance of ADRs traded on the New York Stock Exchange, subjecting these firms to U.S. securities laws and enforcement mechanisms.263

Recent Governance Challenges. While Brazil’s listed SOEs—and especially Petrobras—were hailed as models of good corporate governance and performance not long ago,264 they have since entered a period of crisis, experiencing the full array of governance challenges plaguing mixed enterprises. First, listed SOEs have been at center stage of the main public corruption scandals of the last decade, such as the Congressional vote-buying scandal orchestrated by government officials (mensal˜ao), which implicated Banco do Brasil, and the more recent and large-scale corruption charges involving Petrobras.265

Second, these firms have experienced a clear conflict between the social and political objectives of the government as a controlling shareholder, on the one hand, and the interests of outside investors in the firm, on the other. In recent years, Brazilian SOEs have engaged in actions that were widely perceived as detrimental to the interests of the company and its private shareholders. The use of price controls by oil giant Petrobras and the renegotiation of concession contracts by power company Eletrobras illustrate the tendency to pursue objectives that are not in the financial interest of the firm.

Third, Brazilian SOEs also face problems that are not unique to state control, but common in private firms as well (and especially in companies with a controlling shareholder), such as engaging in related-party transactions that may harm minority investors.266

Finally, the very definition (and legal treatment) of government control raises difficult questions in the Brazilian context. Like France, Brazil has increasingly relied on minority, rather than majority, shareholdings by the state.267 The prevalence of state-controlled institutional investors (such as pension funds of SOEs and BNDESPAR, the equity arm of Brazil’s development bank) in the Brazilian stock market, as well as the widespread use of shareholder agreements by state actors, raise the specter of government intervention even in firms where the government does not directly hold a majority of the voting capital.268 However, the special legal regime applica-

ble to SOEs under the Brazilian constitution, the Corporations Law and the new SOE statute apply exclusively to mixed enterprises in which the state holds a majority of the voting stock.269 Despite the significant governance clout of the state as a minority shareholder, these firms continue to be governed by the private legal regime alone, without commensurate mechanisms of public oversight.270

Reform Initiative: Recently, both private sector and legislative initiatives have emerged in an attempt to restore the confidence of investors and society in listed SOEs. In 2015, B3 launched its SOE Governance Program,271 which to our knowledge is the first initiative in which a stock exchange has provided a governance platform specifically tailored to listed SOEs. The Program follows the exchange’s successful experience with the Novo Mercado, a premium corporate governance listing segment, in fostering governance reform and attracting investor confidence without legislative or regulatory change.272

Like the Novo Mercado and the other premium corporate governance segments, participation in the SOE Governance Program, requiring adherence to stricter corporate governance rules, is voluntary. The program, which contains twenty-five corporate governance requirements, reflects four lines of action: (1) disclosure and transparency; (2) internal control structures and practices; (3) composition of boards and management; and (4) commitment of the government shareholder to include in the state’s code of conduct rules concerning the protection of inside information and the disclosure of nonpublic information. Under the Program, B3 will grant certification in Category 1 to SOEs complying with all of the twenty-five corporate governance requirements of the Program.273 Category 2 is reserved to companies that comply with six mandatory requirements while earning a total of at least twenty-seven points (out of thirty-seven) by adopting some of the remaining corporate governance rules, weighted based on their importance.274 However, in contrast to the Novo Mercado

and the other premium listing segments, the SOE Governance Program is not a listing regime, and therefore does not impose any sanctions for noncompliance beyond the loss of certification.275 Since SOEs may withdraw from the Program at any time, the only potential sanctions for noncompliance or withdrawal are reputational in nature, which significantly weakens the promise of the program as a credible commitment device. The Program attracted its first three SOEs in 2017: Petrobras, Banco do Brasil and BB Seguridade.

More broadly, in 2016 Brazil enacted a statute providing for a special legal regime for public and mixed-enterprises,276 as required by a 1998 constitutional amendment.277 The statute imposes a number of new governance rules on SOEs, some of which partially overlap with the SOE Governance Program. First, the statute enhances disclosure requirements for SOEs, including the obligation to spell out the SOE’s public policy objectives and to quantify the financial consequences of pursuing such objectives in an annual letter.278 SOEs must also produce and disclose a relatedparty transactions policy, as well as formulate a dividend policy in view of the public interest that justified its creation.279 Second, the statute requires the SOE to adopt internal control systems that include an internal audit, a permanent audit committee, and a risk and compliance unit, as well as practices that preserve the independence of the board of directors in the exercise of its functions.280 Third, the statute contains numerous rules on board composition.281 It sets forth minimum qualifications for and restrictions on the appointment of directors and officers.282 At least twenty-five percent of the board must be comprised of independent directors (including minority shareholder representatives), or at least one independent director if shareholders opt for cumulative voting.283 It also limits the remunerated participation of government officials on SOE boards to a maximum of two.284 Finally, the statute attempts to define the social function of SOEs.285

Exercise of the Ownership Function: Beyond the reforms of the governance framework applicable at the firm level, Brazil has undergone numerous changes to the government bodies in charge of overseeing SOEs and exercising shareholder rights. While SOEs in Brazil are formally linked to the ministry with jurisdiction over their market activity, they are also subject to different forms of centralized oversight and control. In 1979, the government created the Secretariat of Control of State-Owned Enterprises (Secretaria de Controle de Empresas Estatais (SEST) under the Ministry of Planning, with the aim of gathering information on SOEs and enabling greater control over their budget and management in a time of international crisis.286 In 1999, in the aftermath of the wave of privatizations, SEST lost its status as Secretariat and became the Department of Coordination and Control of State-Owned Enterprises (Departamento de Coordena¸c˜ao e Controle das Empresas Estatais (DEST .287

Since then, there have been two changes to this structure. In 2007, the federal government instituted the Interministerial Commission of Corporate Governance and Administration of the Federal Government’s Shareholdings (Comiss˜ao Comiss˜ao Interministerial de Governan¸ca Corporativa e de Administra¸c˜ao de Participa¸c˜o>es Societ´arias da Uni˜ao (CGPAR , with the goal of tackling strategic and corporate governance issues relating to federal SOEs.288 In July 2016, the government transformed DEST into the Secretariat of Coordination and Governance of State-Owned Enterprises (Secretaria de Coordena¸c˜ao e Governan¸ca das Empresas Estatais (SEST , with a view to strengthening their governance and monitoring, and possibly paving the way for future privatizations.289 While it is too early to

assess the end results of these changes, Brazil continues to lack full centralization of the state’s shareholding function in the form observed in some other countries, relying on relevant line ministries to play an important role in SOE governance and management.

F.

Japan

In Japan, “special public corporations” (tokushu hˆojin) are used to deliver government services or operate a monopoly. These enterprises, whose shares are wholly owned by the Ministry of Finance, are not subject to the corporate laws that govern private corporations; rather, their establishment, governance structure and operations are regulated by special laws subjecting them to oversight by specified government ministries.290 “Privatization” in the Japanese context first entails converting special public corporations into regular joint stock corporations governed by the Companies Act so that their shares can be offered to the public.291

Privatization of special public corporations in Japan has come in two waves. The first, influenced by the Thatcher Revolution in Britain, was launched by Prime Minister Yasuhiro Nakasone in the 1980s in response to a national debt crisis. Large-scale privatizations of three special public corporations were initiated in this era: Nippon Telegraph and Telephone Public Corporation (NTT), Japan Tobacco & Salt Public Corporation, and Japan National Railways (JNR). The second major privatization push, involving financial institutions affiliated with Japan Post, was initiated by reformist Prime Minister Junichiro Koizumi in 2005, but for political reasons shares were only offered to the public for the first time in 2015. An exploration of these privatizations provides a window onto some distinctive features of Japan’s approach to privatization and post-privatization governance of enterprises still partially owned by the government.

JNR: The railway industry in Japan was nationalized in 1906.292 In 1949, Japan National Railway (JNR) was created under the U.S. occupation as a special public corporation to manage all railway operations in the country.293 Under the Japan National Railway Law, JNR received appropriations from the legislature, Japan’s National Diet. Discretion in the use of funds was left to JNR’s managers, but the governance structure mandated by the law provided an avenue for government control over managerial

decisions.294 In response to a massive build-up of debt in JNR and a government-wide debt crisis, a Provisional Committee on Administrative Reform appointed by the Prime Minister recommended the privatization of JNR, together with that of NTT and Japan Tobacco in 1981. With respect to JNR, the committee recommended not only privatization but also division of the company into several smaller, regional companies (JR companies) that would be more manageable and tailored to local conditions, as well as one freight railway company operating throughout the country. After five years of preparation and planning, the JNR Restructuring Act was enacted on the premise that the JR companies would be completely privatized. The law changed the mission of the companies from “[i]mproving the welfare of the general public” to “responding to market needs and effective management.”295

In the first step of the corporatization process, JR companies were formed as wholly owned subsidiaries of a newly established and government-owned JNR Settlement Corporation (JNRSC), which assumed the debt of JNR. JNRSC then began to sell shares in the JR companies in the early 1990s. In 1998, JNRSC was dissolved and the Japan Railway Construction Public Corporation was formed to settle the remaining obligations of the JNRSC.296 The sale of shares in three JR companies was completed in the early 2000s. The shares in four JR companies with less attractive assets and railway routes remained wholly owned by the government, but the government plans to fully privatize two of these companies in the near future. When the JR companies were initially corporatized, important matters required approvals from the Minister of Land, Infrastructure, Transport and Tourism, including the selection of the CEO and corporate auditors, issuance of stock and bonds, long-term borrowings, business plans, sales of important assets, and revision of articles of incorporation.297 These restrictions are effective for the four government-owned JR companies, but they are no longer applicable to JR companies that have been fully privatized. The privatized JR companies have a standard Japanese corporate governance structure provided by the Companies Act, consisting of a board of directors (with a number of independent directors298

that exceeds Japanese legal requirements) and a separate board of corporate auditors (kansayaku) (comprised of a majority of independent auditors—also in excess of the legal requirement).299

There is widespread agreement that the privatization of JNR and regional division of the resulting firms were highly successful.300 Dividing JNR into regional companies promoted business policies tailored to specific local markets. The new corporate culture promoted profitability, and deregulation allowed for more cost-effective policies. Moreover, perhaps most significantly, the new companies were relieved of the massive debt burden that had weighed down JNR.

NTT and Japan Tobacco: In contrast to the strategy of complete (if very gradual) privatization of the railway system, the privatization of the other two special public corporations launched in the 1980s, NTT and Japan Tobacco, was by design only partial. Until 1986, the Japanese government owned 100% of the shares of NTT. Through a series of share offerings taking place over a long period of time, the government’s ownership of shares gradually fell to its current level of about thirty-five percent. Under the NTT Act of 1984, enacted as part of the privatization process, the government is required to retain at least one-third of the shares of NTT Corporation, the holding company for several regional NTTs.301 In 1997, the government stated before the Diet that it did not intend to actively use its position as a shareholder to direct the management of NTT, and in fact, the government has not historically used its power to do so.302 Nonetheless, the government retains significant control rights in the firm wholly apart from its status as a shareholder, as the NTT Act requires approval of the Minister of Posts and Telecommunications (subsequently restructured and renamed the Ministry of Internal Affairs and Communications) with respect to the appointment or dismissal of directors or corporate auditors, and submission of the business plans of the regional subsidiaries of NTT to the Minister.303

The privatization plan for Japan Tobacco was similarly designed to be only partial. The Japan Tobacco Act established a new corporation, JT, as a joint stock corporation subject to the Companies Act.304 The JT Act pro-

vides that the Japanese government must continue to hold one-third of JT’s issued shares, except for shares that have no voting rights.305 The Act also provides that issuance of new shares and a variety of other important matters, including the appointment or dismissal of directors and corporate auditors, requires the approval of the Minister of Finance.306 The current chairman of the board of JT is a former senior official of the Ministry of Finance, and there is a long history of personnel connections between the ministry and JT.307

As noted, the Japanese government retains significant share ownership and governance rights in both NTT and JT as required by law.308 It is not entirely clear why retention of some level of government ownership of telecommunications and tobacco was legally mandated while railways were slated for complete privatization.309 Regardless of the reason, while the government’s retention of the potential to control post-privatization NTT and JT is of course not without significance, there is no evidence that either of these companies has pursued non-commercial objectives at the behest of political actors or subjected public shareholders to transactions that extract corporate value to the benefit of the government or other governmentfavored interests. However, at least with respect to tobacco, it has been argued that the reverse phenomenon has occurred: namely, that the significant level of ongoing government ownership in the tobacco industry has reduced the government’s incentives to regulate smoking more aggressively in the interests of public health.310

Japan Post: The second privatization wave, initiated in the mid-2000s, centered on financial services provided by Japan’s extensive network of post offices. The government established postal operations and a postal savings system in the 1870s and a postal life insurance system in 1916.311 The Ministry of Communications, formed in 1885 to run these enterprises, was superseded in 1949 by the Ministry of Posts and Telecommunications.312 Through the postal savings system, the Japanese government is

one of the largest holders of private assets in the world. In 2000, shortly before plans for privatization were first developed, the postal savings system held 260 trillion yen, which was about forty percent of all Japanese household savings.313 Profits from the postal savings and insurance system accrue to the Fiscal Investment and Loan Program (FILP), which is managed by the Financial Bureau within the Ministry of Finance. FILP (often referred to as the “second budget”) provides funding for public works projects and local governments.314 This renders postal finance privatization a highly fraught political topic in Japan.315 As with the privatizations of JNR, NTT and JT, the process began with the formation by the Prime Minister of a council to evaluate the benefits of privatization and the plan for carrying it out.316 Koizumi’s effort to privatize the postal savings and insurance system was motivated by criticism of FILP as a font of mismanagement and pork barrel politics.317 Privatization proponents also argued that moving the enormous assets of the postal finance system from the government’s balance sheet into the private sector would boost Japan’s struggling financial industry.318

In 2003, the government established Japan Post Corporation to manage the three services of postal operations, banking, and insurance.319 In 2005, the Diet passed the Postal Service Privatization Act, which abolished the Japan Post Corporation and established Japan Post Holdings (JPH), a holding company owned by the government, with four subsidiaries: Japan Post Bank, Japan Post Insurance, Japan Post Service, and Japan Post Network.320 Assets and operations of Japan Post Corporation were divided among the four subsidiaries. The original law required the government to reduce its ownership interest in JPH to one-third “as early as possible.”321 It also required JPH to sell its entire interest in Japan Post Bank and Japan Post Insurance by 2017.322 The two subsidiaries that operate the post offices and postal delivery service were to remain wholly owned by JPH.323 Due to a change of government in Japan, Koizumi’s privatization plan

stalled after he left office.324 However, following the earthquake and nuclear disaster affecting the Fukushima area in 2011, plans for Japan Post’s privatization were resuscitated to raise funds for the rebuilding effort, and in fact four billion dollars of proceeds from the offering were earmarked for reconstruction.325 The new framework called for disposing of all shares in Japan Post Bank and Japan Post Insurance “as soon as possible” rather than the original deadline of 2017.326 The two postal service subsidiaries were combined into a single entity, Japan Post Co., which will remain wholly owned by JPH.327 In an initial public offering in the fall of 2015, the government sold eleven percent of its shares in JPH, and eleven percent of JPH’s shares in both Japan Post Bank and Japan Post Insurance.328 All three offerings were popular among Japanese retail investors.329 Under the Privatization Law, the government is required to gradually sell shares in JPH until its ownership interest falls to one-third, and it must gradually dispose of all of its shares in Japan Post Bank and Japan Post Insurance, although the legislation does not set out a timeframe for any of these sales.330

In contrast to the traditional Japanese corporate governance structure adopted by NTT, JT and JR companies, the Japan Post privatization utilized a board structure option that was not available during the earlier privatizations.331 Under the Companies Act, firms selecting this alternative board structure eliminate the board of auditors and establish three mandatory committees of the board of directors (audit, compensation, and nomination), with each committee comprised of a majority of independent directors.332 JPH has a board of fifteen directors, nine of which are independent—well above the two independent directors recommended by the Japanese Corporate Governance Code.333 Japan Post Bank and Japan Post Insurance have the same corporate governance structure.334

As noted, although the initial public offering was considered highly successful, Japan Post’s privatization has drawn some criticism for the vagueness of its timetable, and the success of the resulting companies is not without some doubt.335 Statutory constraints on the businesses of JPH’s subsidiaries limit their ability to compete with other financial institutions, a legacy of their government ownership.336 For example, Japan Post Bank is not allowed to offer many standard types of loans, such as home mortgages, while Japan Post, which will remain a wholly owned subsidiary of the holding company, is required to maintain post offices in every Japanese locality without regard to profitability.337 Another potential problem is a divergence of interests between the public shareholders and the government, and between shareholders of the holding company and those of its subsidiaries. Due to the recent and small-scale nature of the initial public offering, it may be many years before the success of the Japan Post privatization plan can be fully assessed.

As these episodes reveal, privatization in Japan is characterized by a highly deliberate, gradual process. In each case, a committee outside the formal government bureaucracy was established to formulate a privatization plan. This had the benefit of insulating the process from the vicissitudes of politics to the extent possible, so that the results were effective and the burdens of privatization were fairly distributed.338 Although privatization of special public corporations in Japan has been partial rather than complete, post-privatization governance has not been plagued by political interference or extraction of wealth from private shareholders.339 It should be noted, however, that this benign outcome appears to be the result of healthy forbearance on the part of the Japanese government rather than the inevitable consequence of robust institutional design: the continued application of special laws governing mixed-ownership corporations provides an avenue for government interference, if it were so inclined. At least under conditions of relatively clean and responsible government, privatization in Japan appears to demonstrate the viability of prolonged partial government ownership as an alternative to the polar extremes of state and private ownership. This approach appears to be effective in improving management and profitability, and retaining some level of potential government control over the provision of services important to public welfare.

G.Singapore

Since its independence from Malaysia and the withdrawal of the British military, Singapore’s economic development strategy has relied heavily on what are known in that country as government-linked companies

(GLCs).340 The GLC-centered strategy, developed in the 1960s, grew out of “the ruling PAP [People’s Action Party] government’s perceived need to support the transformation of the Singapore economy,” based on the conclusion that “control over key domestic markets and institutions [was] the most effective way to . . . meet the main planning objectives of absorbing surplus labour and promoting economic growth.”341 As has been the case in China for the past several decades, the political legitimacy of the PAP ruling in Singapore is strongly tied to successful economic development.342 Shares of Singapore’s GLCs are held by Temasek Holdings Pte. Ltd. (Temasek), which was formed in 1974 as a wholly owned subsidiary of the Ministry of Finance (MOF).343 Temasek is an exempt private investment holding company governed by the provisions of the Singapore Company Act.344 The government interposed Temasek between itself and GLCs to insulate the latter from political influence and reinforce the GLCs’ commercial orientation.345 Upon its establishment, Temasek took control of a number of companies that had been held by other government bodies.346 These companies had already been formed into groups by the government in order to foster national champions, an approach that China later replicated with its national SOEs.347 Thus, each of the companies in Temasek’s portfolio is the head of its own corporate group with numerous affiliated companies.348 Today, Temasek is the controlling shareholder of twentythree of Singapore’s largest companies, which collectively account for almost forty percent of Singapore’s total market capitalization.349

Temasek has two closely-related defining features that distinguish it from China’s State-Owned Assets Supervision and Administration Commission (discussed in the following section): (1) an unambiguously commercial orientation articulated in public documents and verified by its

performance; and (2) a high degree of independence from direct political influence over the companies in its portfolio.

Commercial orientation: Temasek refers to itself as an active investor and steward of state assets.350 This claim is backed up by its performance: it has achieved a total shareholder return of sixteen percent compounded annually since its inception in 1974.351 Temasek uses various devices to increase its financial discipline, such as issuing bonds (rated AAA by Standard & Poor’s) and using market benchmarks to structure incentive compensation for its managers.352 These performance incentives are deferred over a number of years and subject to clawbacks.353

Independence: Temasek’s board of directors is highly professional and non-political in its orientation. The thirteen-member board (expanded from ten in January 2015) is presently comprised of a majority of independent, private-sector directors, three of whom are non-Singapore nationals, including Robert Zoellick, former president of the World Bank, and Peter Voser, former CEO of Royal Dutch Shell.354 There is no ministerial representative on the board.355 Additionally, approximately forty percent of the senior management team of Temasek is comprised of non-Singapore nationals,356 whose political allegiances likely have little bearing on their motivations and performance.

To mitigate the risk that the GLCs will be politically rather than commercially driven, the Singapore government has “constructed a highly visible and well-tailored regulatory regime, which aims to prevent the government from abusing its position as the ultimate controlling shareholder” of the GLCs.357 This regulatory regime has multiple, complementary components.

First, a variety of legal constraints are imposed on MOF’s rights as a shareholder.358 For example, the Singapore Constitution provides that MOF’s appointment, reappointment, or removal of Temasek directors must be approved by the President of Singapore.359 Moreover, Temasek’s Articles of Incorporation provide that its board of directors, not MOF as the controlling shareholder, has authority to determine the amount of dividends to be paid to the government.360

Second, restrictions are placed on Temasek’s board to minimize the

potential for politically motivated intervention.361 Temasek, as a Fifth Schedule entity—a constitutional designation signifying that it is a key government company—is accountable to the President of Singapore to demonstrate that the disposition of an investment is done at fair market value, and the President must approve of a drawdown of accumulated reserves.362 In addition, the Temasek Charter, a public corporate governance policy statement, places voluntary restrictions on Temasek’s involvement in portfolio companies. It is worth quoting the investment section of the Charter in full:

Temasek’s public statements also emphasize the consultative nature of its interactions with its portfolio firms: “[W]e identify value creation opportunities across our portfolio and share our perspectives, at appropriate junctures, with the boards and management of our portfolio companies, for their consideration.”364 In view of its investment orientation and approach to its portfolio firms, commentators have suggested that Temasek is akin to an engaged pension fund—promoting good corporate governance and actively voting its shares, but not becoming directly involved in management.365

Finally, although Temasek is legally exempt from public reporting requirements, it has chosen to publish detailed disclosures of its portfolio and performance, and subjects its financial statements to annual audits by an international audit firm.366 As a result, Temasek is often viewed as setting the “gold standard” for transparency by a government-owned investment fund, receiving the highest possible rating in the Linaburg-Maduell

Transparency Index.367

An additional form of protection for the commercial orientation and political independence of Singapore’s GLC infrastructure is extensive reliance upon independent directors.368 The listed GLCs in Temasek’s portfolio are subject to Singapore’s Code of Corporate Governance, which sets out best practices for public companies on a “comply or explain” basis.369 The Code provides that at least one-third of the board of a listed company should be independent, increasing to one-half in companies where the chairman is not independent of management.370 As previously noted, a majority of Temasek’s board is comprised of independent directors. In addition, Temasek splits the positions of CEO and Chairman, with the Chairman being a non-executive director independent of Temasek’s management.371 Temasek also promotes the independence of the boards of its GLC portfolio firms. Recent research indicates that nearly sixty-five percent of the directors at the GLCs in Temasek’s portfolio are identified as “independent.”372 The authors of this study, who examined the publicly available biographical information of every director identified as independent, concluded that they are “generally highly skilled, prominent figures in the Singapore business community who appear to be independent from the management of their respective Government-Linked Companies.”373

Notwithstanding these safeguards, it may be inaccurate to conclude that Temasek and the GLCs are entirely free from political influence. Temasek’s current CEO is the daughter-in-law of modern Singapore’s founding father Lee Kwan Yew.374 Members of the boards of both Temasek and its portfolio companies have historically been drawn from Singaporean civil service and the military.375 Even today, the managers of Temasek’s portfolio firms are chosen from the “ruling strata of Singapore.”376 “As a result, there [is] widespread agreement about the developmental objectives of the government, which has remained in the hands of the Peoples Action Party since independence.”377 Consistent with this view, fifty percent of the 148 directors in the portfolio GLCs identified as independent currently hold or previously held positions in the Singapore government and/or gov-

ernment bodies.378 Given the shared backgrounds, world view and objectives of the controlling shareholder on the one hand, and directors and managers of the GLCs on the other, direct intervention by the government would not be necessary to ensure that Temasek’s GLC investment strategy fulfills the government’s policy goals and strategic objectives. Yet at the same time, the unalloyed commercial objective of the government in holding shares of the GLCs provides great clarity to managers operating within the system, and stands in considerable contrast to the mixed commercial and social motives of many governments in holding shares in business enterprises.379

It should also be noted that Temasek’s excellent long-term investment returns have not completely insulated the Singapore government’s GLC strategy from criticism. Some view an economic development strategy centered on GLCs as misguided, since these firms compete with the private sector and potentially crowd out private businesses in new markets, products and technologies.380 Temasek has also been embroiled in controversy over an investment in a politically connected Thai conglomerate.381

These side notes and criticisms, however, do not outweigh the remarkable overall success of the Singapore approach on almost any measure— financial performance, transparency, lack of corruption, and protection of minority shareholder interests. Singapore appears to have achieved the “best of both worlds” with its GLC strategy. The government, acting through Temasek, achieves the monitoring benefits of being the controlling shareholder of its portfolio companies. At the same time, due to a host of institutional constraints and what might be called an “ethos of cleanliness” pervading the state sector, it has avoided the problem of minority shareholder exploitation found in most controlling shareholder regimes.

Although their impact may be impossible to quantify, Singapore’s unique qualities—its small size, extensively globalized domestic economy, and shared work ethic among managerial elites—may have contributed to the success of the government’s GLC strategy. If so, it will be difficult to replicate the Singapore model elsewhere. The issues of replicability loom largest for China, which has consciously modeled its SOE holding com-

pany structures on Temasek.382 But the question whether the Temasek model can be scaled up and transplanted into a very different institutional and cultural environment is relevant for any country that is seeking to emulate Singapore’s approach.

H.China

Structure of State Ownership: Although, as discussed below, a mixedownership strategy is currently the focus of China’s approach to SOE reform, this is not a new approach. Since the inception of economic reforms in the late 1970s, the boundary between state-owned and private firms in China has often been blurred.383 One of the main drivers of China’s economic miracle during the 1980s and the early 1990s was the emergence of so-called “non-state” firms, whose share of national industrial output increased from twenty-two percent in 1978 to forty-two percent in 1993.384 One major category of non-state firms was “collectively owned” firms, ostensibly owned by “all residents” in a community.385 Many of these collectively owned firms were in fact privately owned and operated and were registered as collectively owned only because, at the time, there was no legal framework for the registration of private firms.386 With the adoption of the Company Law in 1994, the government started converting SOEs to corporate forms.387 This corporatization campaign created not only SOEs whose corporate shares were wholly owned by the state, but also mixed-ownership firms where the ownership and management of the firms were shared among state and private shareholders. In 1997, China announced a massive program to privatize all but the largest SOEs under the slogan “grasping the large, letting go of the small.”388 In practice, however, the newly “privatized” SOEs under this program did not become private firms as that term is commonly understood; instead, they became firms with mixed ownership. It was estimated that as of 2003, mixed-ownership firms accounted for forty percent of China’s GDP.389 Some of the best-known Chinese firms, such as Haier, TCL, and Lenovo, are of the mixed-ownership type. Mixed ownership is currently an important ownership form among China’s SOE groups at the subsidiary level. For example, almost all of the thirty-four subsidiaries of China National Offshore Oil Corporation (CNOOC) are mixed-ownership firms, with an

average state-share percentage ranging from forty to sixty-five percent.390 Chinese SOEs at the national level are organized into business groups comprised of numerous separate corporations arranged in hierarchical order.391 The business group concept has been enshrined in regulations that permit registration as a business group if it has certain required components and layers of entities.392 Registration as a business group affords eligibility to establish a finance company to handle lending, underwriting, cash management and other financial functions that are otherwise prohibited on an inter-company level.393

The parent (holding) company of a Chinese SOE business group is legally organized as a “wholly state-owned limited liability company” (WSOLLC) under the Chinese Company Law.394 A WSOLLC is not required to have an annual shareholders meeting and has only one shareholder—the State-Owned Assets Supervision and Administration Commission (SASAC).395 SASAC was established directly under the Chinese State Council (cabinet) in 2003 in an attempt to consolidate control over all central SOEs.396 SASAC’s formal role, set out in legislation,397 is to serve as the investor on behalf of the State Council in the approximately 110 large corporate groups under its supervision.398 SASAC has a broad mandate: its formal functions include preserving and enhancing the value of stateowned assets, appointing, removing and setting remuneration of top SOE executives, dispatching supervisory panels to SOEs, and drafting regulations on the management of state-owned assets.399 SASAC shares decision rights on senior management appointments with the Chinese Communist Party (CCP) in a highly institutionalized arrangement whereby the top positions in the most important SOEs are evaluated by the Organization

Department of the CCP—the party’s premier personnel department.400 Deputy positions at these firms and positions in the remaining SOEs are handled by other party committees within SASAC.401 Given its broad remit, SASAC is part investor, part regulator and compliance department, and part conduit for CCP influence and industrial policy dissemination. Arguably the closest model for Chinese SOE ownership structure can be found in Singapore. The basic structural similarities between Temasek and SASAC reflect similarities in the two government’s motivations for adopting a state capitalist approach. As previously noted, Singapore’s GLC-centered strategy grew out of “the ruling PAP government’s perceived need to support the transformation of the Singapore economy,” based on the conclusion that “control over key domestic markets and institutions [was] the most effective way” to promote economic growth.402 Moreover, in both countries, there is a strong link between economic success and political legitimacy. Particularly given the strength of these parallels and the success of Singapore’s approach, it is not surprising that the CCP, along with many other analysts concerned with Chinese SOE reform, have continued to look toward Temasek as a model for SOE share ownership and supervision.403

But China has only selectively adopted the Singapore holding company structure: SASAC is far from a copy of Temasek, which as discussed previously, has two closely related defining features that signal its role as a true holding company: (1) an unambiguously commercial orientation articulated in public documents and verified by its performance; and (2) a high degree of independence from direct political influence vis-`a-vis the companies in its portfolio.404 Chinese SOEs under SASAC supervision are in the main commercially oriented, but some distinctive features of the Chinese system raise questions about the purely commercial orientation of the state sector. For example, SASAC, in consultation with CCP organs, rotates senior corporate leaders among SOE business groups.405 On occasion, it has simultaneously rotated the CEOs of several SOEs in a given industry in a “musical-chairs” fashion.406 In addition, the SOEs are sometimes called upon to perform social functions on behalf of the state, such as maintaining employment.407 These practices suggest that at least for some purposes, the interests of the national SOE business groups are viewed collectively—that is, the important consideration for SASAC and the CCP is to maximize the interests of the state sector as a whole, rather than at the individual firm or group level. Perhaps an even more stark contrast with the Singaporean approach relates to the lack of a clear separation of

politics from business in the Chinese SOE sector. Within every firm throughout a Chinese SOE business group, there is a CPC committee responsible for managerial appointments, promotions, and party discipline. Senior executives of Chinese SOEs are uniformly members of the CCP, and simultaneously hold positions of equivalent rank within the corporation and the party. “Party centrality” is thus a defining characteristic of the Chinese state sector.408

Mixed Ownership Structure: Shareholding in Chinese SOE business groups is hierarchical: firms higher in the structure, beginning with the WSOLLC parent company in which strategic and managerial decisionmaking are concentrated, own downstream subsidiaries, but there is very little upstream or cross ownership within the group.409 Typically, one or more of the subsidiaries below the parent company are publicly listed on a Chinese stock exchange, and often cross-listed on the Hong Kong Stock Exchange and/or a major foreign stock exchange as well.410 The percentage of shares retained directly or indirectly by the parent company varies, but it is always sufficient to retain ultimate control over the listed firm(s) in the group, particularly given the dispersed nature of the private shareholdings.411 These public listings are intended to provide not only capital, which often is available at low cost from state-owned banks, but also managerial discipline and global visibility.412

As urgency to overhaul China’s economy has built up due to problems of overcapacity, high levels of corporate indebtedness, and sluggish domestic demand, SOE reforms have become a priority of China’s political leadership. A central component of the current SOE reform efforts, announced with considerable fanfare at the Third Plenum in 2013, is to convert more SOEs to so-called “mixed-ownership” firms—firms in which the state and private shareholders hold joint equity stakes.413 In September 2015, the State Council promulgated detailed “Opinions on the Development of

Mixed Ownership Economy by State-Owned Enterprises.”414 The “starting point” of the Opinions is that “cross-shareholding and mutual integration of state-owned capital, collectively-owned capital, [and] private capital are an important manifestation of China’s basic economic regime.”415 They seek to “treat enterprises as market players,” and combine efforts to attract capital with efforts to diversify property ownership and improve corporate governance structures.416 In order to implement these goals, the Opinions call for a host of initiatives, including (1) encouraging private capital and foreign investors to participate in the mixed-ownership reforms, while encouraging state-owned capital to invest in non-SOEs; (2) promoting the public-private partnership (PPP) model; (3) exploring use of preferred shares and golden shares, which to date have not been used in China; (4) exploring employee stock ownership plans; and (5) improving corporate governance in mixed-ownership enterprises and promoting professional management thereof.417 The National Development and Reform Commission, China’s economic planning agency, is reportedly developing a plan to “more or less complete” mixed-ownership reforms for all SOEs by the year 2020.418

The reform agenda of the Third Plenum aims to expand mixed ownership to all levels of the SOE structures, including the central SOE groups themselves.419 While the goal is bold and the political rhetoric surrounding the plan is emphatic, it is important to recognize that mixed ownership is the path that China has been pursuing over the past two decades.420 Thus, it is fair to question whether the current ownership-based reform strategy holds new potential for improving governance and performance in China’s state sector.

In fact, there would appear to be serious limitations to this strategy. Injecting more private capital into SOEs may do relatively little to improve the performance of individual SOEs or to increase the competitiveness of the Chinese economy.421 The profitability gap between the state sector and the private sector is widening.422 A mixed ownership strategy, bringing more private capital into the state sector, will not transform the role of the state from major participant in the market to an impartial arbiter of

market competition. On the contrary, the reforms to date seem primarily intended to create larger SOEs on the theory that global competition requires scale.423 While the number of SOEs under SASAC supervision is declining, total assets under its control have increased.424 Equally important, the current reforms will do little to eliminate political intervention in the state sector. A Temasek-style structure creating a firewall between the state in its role as investor and the management of its portfolio companies would require the CCP to withdraw from its role in personnel appointments and elimination of the firm-level party committees in favor of depoliticized internal control and reporting structures used in major western firms. Thus far, there is no sign that the party is withdrawing from state-invested firms; to the contrary, the grip of the party seems to be tightening.425

III.Patterns of SOE Regulation?

Having surveyed national experiences with SOE governance, we pause to consider whether any patterns can be discerned in the way that the countries in our study have regulated mixed enterprises. To facilitate this effort, we have created Chart 1, which indicates whether a given country’s regulatory regime for listed SOEs has the characteristics indicated in the top row.

Several commonalities in regulatory approach are apparent. As Chart 1 shows, consistent with the recommendations of the international guidelines (but also subject to our critique), every country generally subjects SOEs to the same corporate law and securities law regime that governs private corporations. While various jurisdictions subject listed SOEs to the oversight of different state actors (such as Parliament or a Court of Auditors), judicial enforcement of corporate and securities laws does not appear to play a central role in SOE governance.

Subject to various exceptions, our study finds that the centralization of state ownership of SOE shares—another key recommendation of the international guidelines—exists in all of the countries other than the United States and Brazil. It is noteworthy, however, that only Singapore and China have centralized SOE share ownership in the form of a holding company, arguably the most robust formal means of separating the state’s distinct roles as regulator and shareholder. Chart 1 also indicates that cross-listing of SOEs on foreign stock exchanges is prevalent. Cross-listing is perceived as a means by which managers bond themselves to higher standards of disclosure and corporate governance than those required in the company’s home country, and credibly signal conformance with high governance standards to foreign investors.426 Yet we know from global experience that cross-listing turns out to be far from a failsafe method of securing good corporate governance.427 It is not unusual for SOEs cross-listed on foreign stock exchanges to be the subject of serious corporate governance scandals.428

Chart 1 also reveals an interesting negative commonality: other than Brazil, where it is a very recent and still untested development, none of the countries in our study uses stock exchange rules to create a special governance regime for mixed enterprises. This observation highlights a potential policy area for further inquiry and discussion, considered in the next section.

We can see from Chart 1 that few states possess formal control rights in excess of their equity interest in the firm. In its short-term experiment with state ownership following the financial crisis, the U.S. government intentionally chose to exert less influence than a similarly situated private investor would legally possess.429 Among the jurisdictions examined, France and Brazil are outliers in this regard. France not only provides special governance rights for the state as shareholder but also applies to SOEs the disparate voting rights available to long-term shareholders, which effectively amplifies the power of the state as shareholder.430 The Brazilian state, in turn, exercises voting control over SOEs that well exceed its cash-flow rights by relying exten-

sively on non-voting shares and pyramidal structures.431 China also permits disproportionate state influence, due to the formal powers provided to SASAC and to the role of the Chinese Communist Party in SOE governance.432

Perhaps the principal takeaway from the effort to discern patterns in SOE governance in the countries we have surveyed is the lack of pattern. Diversity appears to be the hallmark of governance of listed SOEs around the world. This finding is probably not surprising, given that SOEs are products of states and often enjoy privileged access to lawmakers and regulators by virtue of their provenance. As such, it is natural for SOE governance to reflect the characteristics of national governance—that is, the characteristics and quality of a national regulatory regime for SOEs is deeply influenced by prevailing national philosophy about the proper scope of state ownership of enterprise, separation of powers, the level of corruption in society, and related factors. There is little reason to believe that resorting to mixed ownership—by offering a portion of an SOE’s shares to private investors—is sufficient to transform a product of the state into a pure product of private market transactions.

This point leads to a normative conclusion that may disappoint policy makers: there appears to be no single formula for achieving a high quality regulatory structure for SOE governance. Compare Norway and Singapore, for example, the two countries in our study that are frequently cited as setting the global standard for SOE governance. As Chart 1 indicates, although they share some regulatory traits, these countries have not followed a single template for SOE governance. Most prominently, while Singapore’s approach revolves around Temasek, the state holding company, Norway has eschewed a holding company approach, and has not even completely centralized the ownership of SOE shares. What these countries have in common, however, is a reputation for clean government and the rule of law.

Another comparison is revealing from a different angle: Singapore and China, countries at some distance from one another in terms of the quality of SOE governance. Yet as Chart 1 shows, they have many regulatory traits in common. This is not accidental: as noted, Chinese economic reformers have looked to Singapore as a model in the governance of its SOEs. But the two countries have vast differences in “Governance with a Capital G,” including among other things the levels of corruption and political intervention in the economy and the quality of legal institutions. These differences probably influence the quality of SOE governance far more heavily than the specific factors identified in Chart 1.

Yet from a policy perspective, there is a bright side to the lack of a fixed template for SOE governance: the diversity of successful approaches highlighted by our study suggests that each country is free to develop governance structures most suitable to local conditions. Path dependence does not doom any country’s system of SOE governance to failure; rather, policy makers have

room for experimentation and creativity in addressing the distinctive governance challenges posed by state ownership of enterprise.

IV.Critique of Best Practice Guidelines

International organizations such as the OECD and the World Bank have produced a plethora of policy prescriptions for best practices in SOE corporate governance. Typical examples are the OECD Guidelines for State-Owned Enterprises, originally adopted in 2005 and revised in 2015, and the World Bank Toolkit for State-Owned Enterprises, published in 2014.433 Although the policy recommendations of the various organizations are numerous, they exhibit a high degree of uniformity around the following basic principles:

These principles reflect laudable aspirations at the high level of generality at which they are framed. The basic impulse behind the recommendations is to ensure that SOEs are operated as if they were private enterprises.434 Beyond improving operational performance and the protection of minority investors, a primary concern of these initiatives is to level the playing field vis`a-vis private competitors.435 Perhaps the most important addition to the OECD Guidelines in the 2015 version is the principle, set out at the very beginning of the Guidelines, that the state “should carefully evaluate and disclose the objectives that justify state ownership and subject these [objectives] to a recurrent review.”436 Although there may be no direct legal consequences to a state for failing to live up to its disclosed objectives justifying state ownership, disclosure requires the formulation and articulation of concrete justifications for state ownership, and may help tie the hands of state actors with respect to the operation of SOEs. Disclosure promotes public scrutiny and may thereby act as an informal constraint on the use of SOEs to fulfill ill-defined or politically motivated objectives.437

While we applaud the goal of promoting good corporate governance of SOEs, closer inspection of the principles espoused by international organizations reveals a number of potential problems with this type of initiative, particularly in view of the possibility of government failure. First, whereas the guidelines of the international organizations enunciate numerous objectives, they contain comparatively little guidance on the institutional practices that are necessary to achieve them. In other words, although the destination is clearly marked, there is no road map provided to assist in reaching the destination.

Second, as noted above, the guidelines reflect a strong view that SOEs should be subject to the same legal regime as private firms,438 a policy linked to the goal of leveling the playing field for private competitors. Yet the state is a very distinctive kind of controlling shareholder as compared to an individual or a POE. The guidelines attempt to deal with this reality by encouraging the state to behave like a “standard” controlling shareholder.439 This is to be accomplished by (1) centralizing the exercise of ownership rights in a single state shareholding entity or holding company; and (2) subjecting SOEs to the same corporate law regime as POEs.440 This policy approach, however, has potentially serious weaknesses. It is not obvious why centralization is an effective antidote to the unique problems posed by a state controlling shareholder, such as political interference in management and the pursuit of noncommercial objectives. Centralized ownership has been very effective in Singapore, but it has arguably not been relevant to good governance in the other “gold standard” country, Norway.441 The second approach, a unitary legal regime for SOEs and POEs, may have serious political economy consequences if the state’s interests as a controlling shareholder cloud the content and distort the development of the corporate law and other market regulatory regimes such as competition law. Real-world examples of this phenomenon can be found in Brazil, France and China.442

Third, the guidelines simultaneously stress the importance of “active ownership” on the part of the state and the need to avoid politically motivated interference in management.443 But there is a tension inherent in pursuing these two goals, since active ownership by the state creates the mechanism for political interference in management and the pursuit of noncommercial objectives by SOEs. A closely related tension exists between the principle of active state ownership on the one hand, and board independence on the other. Given that “the state” is by definition a political actor—moreover, a political actor that is not a monolith but a diverse agglomeration of interests—favorably resolving these tensions is a highly complex undertaking.

We highlight these issues not to denigrate attempts to formulate best practices in the governance of SOEs, but to highlight the complexity of the task, and the inherent limitations of a general-principles-based-approach to SOE governance in countries with widely divergent economic and political institutions.

V.Policy Responses

In reviewing the national experiences for policy suggestions, we avoid attempting to crystallize conclusions as to “what worked” and “what did not work” in these countries. A number of reasons justify this approach. First, defining success in the SOE context is difficult, given the variety of normative objectives that these firms may pursue. Is “success” to be measured by financial results and shareholder returns, or by the effectiveness of what we have called “policy channeling”—government ownership of an SOE as a means of implementing social or industrial policy? Second, it is problematic to make strong causal claims about the effects of different governance mechanisms, regardless of how success is defined. As suggested above, it is possible, indeed likely, that other features of the institutional and economic environment in a given country affect the performance of SOEs. Third, even if it were possible to unambiguously identify a “successful” SOE governance mechanism in a foreign jurisdiction, adopting that mechanism as a policy roadmap for reform in any given country may be problematic, given the well-known role of local context in determining the success of legal transplants. We therefore approach the foreign narratives as an instrument for institutional critique and imagination in the hope of providing policymakers with ideas about a range of governance arrangements that go beyond those usually considered in domestic circles and international organizations.

Similarly, the diversity of approaches we have encountered in this study makes us reluctant to recommend a specific formula or checklist of best practices in the governance of mixed-ownership enterprises. Indeed, the thrust of our analysis would counsel against such an approach. Short of complete privatization, there is no fail-safe method of resolving the distinctive governance problems of listed SOEs. And, of course, private firms also face significant governance challenges, which are similarly resistant to a check-the-box approach.

Recognizing this reality, however, does not mean that SOE governance is in a hopeless state around the world. Quite to the contrary, several of the countries in our study—countries with a diverse array of political systems and institutional structures for SOEs—are providing a healthy governance environment for their mixed enterprises.

The principal objective of any governance framework for SOEs is to insulate the management of the enterprise from political interference that distorts its public mission and commercial orientation and makes public (including both investor and citizen) understanding and oversight of the firm more difficult. The specific means a given country chooses to accomplish this objective are not particularly important, provided that they fit the institutional setting

in that country. In Singapore, insulation from political intervention has been accomplished structurally, through the design of the Temasek holding company and the surrounding laws that support a purely commercial orientation of the GLCs. In Japan and, to a lesser extent, in Norway, insulation of mixed enterprises from political interference appears to be less the result of robust structural design than of a “hands off” approach of political and bureaucratic actors. Brazil is experimenting with innovative private sector and legislative initiatives on SOE governance that find no direct parallel elsewhere.

To be sure, designing and implementing the optimal institutional infrastructure for SOEs is no easy task. In fact, as noted above, the quality of public governance itself is arguably as important, if not more important, than the quality of specific corporate governance arrangements for SOEs. Moreover, the challenges associated with the state’s involvement in capital markets go beyond its role as a controlling shareholder of SOEs. In Brazil, state-controlled institutional investors such as SOE pension funds and BNDESPAR play a major role as prominent blockholders in public companies.444 In Japan, the giant Government Investment Pension Fund owns shares in hundreds of Japan’s leading companies, while the central bank owns sixty percent of Japan’s exchange traded fund (ETF) market.445 The influence of state-controlled institutional investors on corporate governance raises important questions, whose treatment we leave for future research and policy discussion.

Based on the foreign experience we have surveyed and our own analysis of the governance challenges of listed SOEs, we suggest that the following subjects are worth careful consideration by policy makers seeking to improve the regime for mixed-ownership enterprises. Of course, not every subject will be relevant for every country, and not every proposal will be optimal for the particular circumstances of an existing regime. The list that follows is offered as a way of extending and reframing the policy discussion outside the parameters of the Guidelines on SOE governance set by the international agencies.

A.Ownership Structure

Ownership structure is a key determinant of behavior in corporate governance. Scholars have suggested that the separation between voting and economic rights in the firm can increase the incentives of controlling shareholders to extract private benefits of control.446 The use of “controlminority structures” that permit shareholders to exercise uncontested control

over the company while holding only a minority of the firm’s equity interest exacerbates the agency problem between controlling and minority shareholders.447 Such a problem tends to be even more severe in SOEs, given the greater misalignment of incentives between private and public shareholders, with the state being even more tempted to promote political objectives through the firm when it does not have a commensurate financial interest in the corporation.

With the notable exceptions of France and Brazil, all other jurisdictions have either granted one vote per share or, in the case of the United States, affirmatively restricted the voting rights of the state. In Colombia, the state has not only adopted a single class of shares but has also maintained a high equity stake in its controlled firms, which has likely reinforced the state’s interest in their financial performance.448 France has historically employed formal mechanisms that magnify the power of the state as a shareholder, such as special appointment rights and a tenured voting regime giving double voting rights to long-term shareholders.449 Brazil, in turn, is exceptional among the countries examined in making broad use of dual-class shares and pyramidal structures in SOEs that greatly enhance the state’s power.450

In view of the theoretical findings and the predominant national experience, SOEs should avoid the use of non-voting preferred shares, pyramidal structures, and other means of magnifying state power given their effect on the controlling shareholder’s (i.e., the government’s) incentives to pursue private benefits to the detriment of the SOE and its minority shareholders.

B.Board Composition and Independence

Should politicians and regulators be allowed to serve on the boards of listed SOEs? Countries have answered this question in different ways. Norway has a broad ban that encompasses not only politicians but also civil servants of the federal government.451 Brazil has newly banned the appointment of regulators, politicians and union leaders to SOE boards.452 An alternative approach is to increase the participation of appointees by minority shareholders.453 Unlike directors whom the government as controlling shareholder can appoint and dismiss at any time, minority-shareholder appointees enjoy structural independence, and thus may be well suited to monitor firm insiders.

These approaches seek to increase the independence of SOE boards, a major principle of the international guidelines on SOE governance. But there is a tension between a board’s independence from the government as controlling shareholder (thought to be healthy from a corporate governance perspective) and a board’s independence from the government as public policy maker (problematic if one assumes that the reason for continued government ownership is the accomplishment of a public policy objective). Recognizing this tension suggests that board composition and independence should not be considered solely within the standard corporate governance rubric.

C.The Role of the Board

It is important to recognize that SOE reforms focusing on board composition must take account of the legal balance of power in the corporation in a given jurisdiction. The recent SOE Statute in Brazil illustrates the issue.454 As noted, board composition is a central feature of the statute. Yet, from a comparative perspective, Brazil’s system of corporate governance is unusually shareholder-centric—with shareholders possessing far more, and the board of directors proportionately less, decision-making power than in other jurisdictions.455 This suggests that board reform will be of limited effectiveness in the Brazilian context, unless the reform is coupled with a strengthening of the board’s role in SOE governance.

D.Remuneration

Managers of listed SOEs serve two masters—minority private shareholders and “the citizens.” As we noted in Part I, the interests of these two masters sometimes converge but often conflict. To date, compensation issues in SOEs have focused on the question of whether the compensation of government officials serving on SOE boards should be subject to special regulation. Again, countries have answered this question in different ways. Despite its statist tradition, France does not permit government officials serving on SOE boards to receive additional compensation for their board service.456 By contrast, board appointments in Brazilian SOEs have historically served as a means of supplementing the earnings of government officials, creating incentives to propagate insider-dominated boards and maintain SOEs in non-core sectors. Brazil’s new SOE statute now limits the government officials’ remu-

nerated participation to two boards of directors or boards of supervisors of SOEs.457 In China, senior SOE executives’ compensation is regulated by SASAC—the state holding company—in cooperation with a senior Party organ. Compensation amounts are low by international standards. Nevertheless, beneath the surface of state control over SOE compensation lies a vast domain of managerial autonomy. In fact, informal forms of compensation often exceed the amounts formally provided for by regulation.458

Perhaps a more important, rarely asked question is how SOE managers’ compensation should be structured. In private firms, compensation arrangements should be designed to incentivize managers to maximize shareholder value. But what is to be maximized in an SOE? The continued presence of state ownership in a listed SOE suggests that something other than or in addition to maximizing shareholder value should be the goal of SOE managers. The design of optimal contracts for SOE compensation is well beyond our ambition here, and in any event, optimality will depend on the specific goals of a given SOE. Our point is that, as with board composition and independence, compensation issues should not be cabined exclusively within a binary framework. The important question is not whether government officials should be compensated for SOE service. The question is how to structure compensation arrangements that advance the commercial (and, where appropriate and adequately disclosed) non-commercial objectives of a listed SOE.

E.Structural Incentives

SOEs are not only subject to relevant economic and industry-specific regulations along with private firms, they are also frequently impacted by their home government’s fiscal and industrial policies. In Japan, the postal savings system grew to enormous size, and its gradual privatization took decades to initiate, because it provided funding for the government’s “second budget,” a font of pork barrel politics. Recent reforms in Brazil’s budgetary laws indirectly encouraged the use of SOEs to promote non-economic objectives.459 Since 2009, the legislature has excluded SOEs Petrobras and Eletrobras from the calculation of Brazil’s primary surplus or deficit—a key concept under existing fiscal responsibility laws.460 With this exclusion, investments and expenditures by these SOEs were taken “off-the-books” from the perspective of the government.461 This created an incentive for the government to pursue costly public policies through SOEs rather than directly.462 Even in the U.S., where staunch resistance to government ownership prevails, federal policies promoting home ownership greatly influenced the risk-taking behavior of Fannie and Freddie, leading to government intervention when they ran into

serious trouble during the global financial crisis.463

As these examples indicate, policy makers and investors must be conscious of the incentive effects of the government’s fiscal and industrial policies on SOE behavior. While there is no way to completely insulate SOEs from these structural incentives, as discussed below, mandating disclosure of SOE conduct that serves non-commercial purposes, and subjecting such conduct to special approval regimes, may mitigate the potential negative consequences of these incentives.

F.Disclosure and Treatment of Non-Commercial Decisions

The distinctive governance challenges of listed SOEs result from the Janus-faced nature of a mixed-ownership enterprise: part commercial actor; part public policy actor. Improving the governance of SOEs requires attention to both the commercial and non-commercial dimensions of these enterprises. The OECD Guidelines for State-Owned Enterprises address the second dimension in providing that “costs related to public policy objectives should be funded by the state and disclosed.”464 Colombia follows this approach.465 Its state owned oil giant Ecopetrol sells subsidized fuel to the public.466 But since the company’s public listing, the government has consistently reimbursed the company for the cost of the subsidy.467 In Brazil, B3’s SOE Governance Program and the new SOE Statute require the specification of the public policy objectives pursued by each company, as well as disclosure of the costs of such policy interventions.468 Neither the Program nor the Statute, however, imposes any substantive constraints on the pursuit of public policies that lack a commercial justification, nor any requirement that the state compensate the SOE for the costs of such policies.469 Thus, the government could continue to direct Petrobras to sell fuel at below market prices as long as this practice is publicly disclosed. Colombia’s approach in reimbursing the SOE for the cost of pursuing a public policy goal is both fairer to minority shareholders and a more effective constraint on expansive use of SOEs to pursue policy goals than the Brazilian approach of disclosure alone.

It may be useful to consider responses to the public policy dimension of SOEs that go beyond disclosure and reimbursement. For example, directors of SOEs may be required to consider, justify, formally approve, and disclose to shareholders the pursuit of specific public policy objectives in a manner similar to the process used by boards of directors for conflict-of-interest transactions.470

G.Enforcement

One striking finding from our review of the national experiences with SOE governance is the relative rarity of public enforcement efforts against SOEs.471 Perhaps it is unsurprising that public agencies are reluctant to bring enforcement actions against an SOE—where one instrumentality of the government challenges the actions of another. But this suggests the importance of private enforcement. Yet shareholder-plaintiffs often face significant obstacles in bringing claims against an SOE. These range from extreme reluctance on the part of the judiciary to hear certain categories of claims against politically powerful SOEs472 to procedural rules that increase the cost and risk of minority shareholder claims against the state in its role as controlling shareholder.473 In Brazil, both the premium stock exchange listing segments and the charters of various SOEs impose arbitration as the mandatory method of dispute resolution between the company, its shareholders and managers.474 This is problematic insofar as arbitration proceedings in Brazil can be more expensive than judicial lawsuits.475 Moreover, arbitration proceedings are confidential, reducing the amount of information available to outside investors as well as the reputational cost to the government associated with their filing.

At least at a conceptual level, strengthening public and private enforcement capacity against listed SOEs is the low-hanging fruit of SOE reform and an obvious means of supplementing the diminished role of market forces in disciplining mixed-ownership firms.

Conclusion

In this Article, we have provided insights from national experiences and theoretical analysis relevant to policymakers seeking to improve the governance of listed SOEs. Several sobering facts emerge from our study. In contrast to the implicit message of the international best practice guidelines, there is no set “roadmap” for successful SOE governance, and indeed “success” in this field may be a contestable term. Moreover, at least a level of informal observation, the quality of SOE governance appears to be quite closely correlated with the quality of political governance in a given country. But clearly there are examples of sound SOE governance around the world and innovative approaches to the challenge of SOE governance that may offer guidance to policymakers elsewhere. The diversity of approaches to SOE governance revealed by our study—even among countries that have managed the challenges relatively well—may be cause for optimism, by suggesting that effective governance strategies can be forged with the tools at hand in a given institutional environment, when coupled with appropriate doses of imagination and political will.

--- FOOTNOTES ---

Seq: 1

†Milhaupt is the Parker Professor of Comparative Corporate Law and the Fuyo Professor of Japanese Law at Columbia Law School. Pargendler is Professor of Law at Funda¸c˜ao Getulio Vargas School of Law in S˜ao Paulo (FGV Direito SP) and Global Professor of Law at New York University School of Law. This Article is based on a report prepared by the authors for Associa¸c˜ao BM&F—a non-profit organization linked to B3 S.A., which operates Brazil’s primary stock and futures exchange. The authors are grateful to Sofie Cools, Aur`ele Delors, Flavia Fernandes, Gen Goto, Gunnar Norden, Caio Figueiredo Cibella de Oliveira, Mario Engler Pinto Jr., Daniel Puchniak, Francisco Reyes, Mario Schapiro, Beate Sj˚afjell, and Thomas Trebat for their invaluable comments on a draft of the report. We also benefitted from comments at workshops at FGV Direito SP, B3, and the Securities and Exchange Commission of Brazil (Comiss˜ao de Valores Mobili´arios—CVM). Bruno Becker, Janet Kanzawa, and Yasmin Saba provided excellent research assistance. All errors are our own.

50 CORNELL INT’L L.J. 473 (2017)

Seq: 2

1.China Buys up

the World, THE ECONOMIST (Nov. 11, 2010), http://

www.economist.com/node/17463473 [https://perma.cc/422Z-SAXY].

2.Anjli Raval, Saudi Arabia Considers Aramco Share Sale, FIN. TIMES (Jan. 7, 2016), [https://

https://www.ft.com/content/5709ad0a-b555-11e5-8358-9a82b43f6b2f perma.cc/VMK7-M42S].

3. See infra Part II. H.

Seq: 3

4.See, e.g., World Bank Group [WGB], Corporate Governance of State-Owned Enterprises: A Tool Kit, at 44, WB456286/91347 (2014), http://documents.worldbank.org/ curated/en/228331468169750340/pdf/913470PUB097810B00PUBLIC00100602014. pdf [https://perma.cc/2QRE-24P9] (citing Petrobras among “several successful listed SOEs [that] are recognized as world leaders”) [hereinafter World Bank Group, Corporate Governance of SOE]; Francisco Flores-Macias & Aldo Musacchio, The Return of StateOwned Enterprises: Should We Be Afraid?, HARV. INT’L REV. (Apr. 4, 2009), http://hir.har vard.edu/article/?a=1854 [https://perma.cc/FSD6-TAW2].

5.Edmar Luis Fagundes de Almeida, Patricia Vargas de Oliveira & Luciano Losekann, Impactos da conten¸c˜ao dos pre¸cos de combust´ıveis no Brasil e op¸c˜oes de mecanismos de precifica¸c˜ao [Impacts of Fuel Price Containment in Brazil and Options for Pricing Mechanisms], 35 REVISTADE ECONOMIA POLITICA [REV. ECON. POL.] 531 (2015) (Braz.). 6.See Joe Leahy, What Is the Petrobras Scandal that Is Engulfing Brazil?, FIN. TIMES (Mar. 31, 2016), https://www.ft.com/content/6e8b0e28-f728-11e5-803c-d27c7117d132 [https://perma.cc/8MBU-V5TD].

7. 8.Petrobras, Annual Report (Form 20-F) 30 (Apr. 27, 2016). 9.See Leahy, supra note 6.

Seq: 4

10.See infra Part II. E. 11.B3, Regimento do Programa Destaque em Governanca de Estatais [State-Owned Enterprise Governance Program] (Braz.), http://www.bmfbovespa.com.br/lumis/portal /file/fileDownload.jsp?fileId=8AA8D0975BE9A0FB015BF91674D55660.

12.

Seq: 5

13.For a detailed discussion of the tradeoffs associated with mixed ownership, see Mariana Pargendler, Aldo Musacchio & S´ergio G. Lazzarini, In Strange Company: The Puzzle of Private Investment in State-Controlled Firms, 46 CORNELL INT’L L.J. 569, 574 (2013).

14.Governments have a wide range of motivations for retaining control over commercial enterprises, ranging from opportunities for employment or theft by political insiders to national security concerns.

Seq: 6

15.See Ronald J. Gilson & Jeffrey N. Gordon, The Agency Costs of Agency Capitalism Activist Investors and the Revaluation of Governance Rights, 113 COLUM. L. REV. 863 (2013) (using the expression “the agency costs of agency capitalism” to describe the second layer of agency costs created by the rise of institutional ownership of corporate shares).

16.See Ronald J. Gilson, Controlling Shareholders and Corporate Governance: Compli-

cating the Comparative Taxonomy, 119 HARV. L. REV. 1641, 1663 (2006).

Seq: 7

17.

In a separate paper, we develop the concept of policy channeling to describe the state’s control of an SOE (as opposed to resort to regulatory processes or other mechanisms of government) to achieve a policy objective. Although policy channeling may not be in the financial interests of the private investors in the SOE, state actors may engage in this conduct because it generates a distinctive type of private benefits of control— what we call political private benefits of control.See Curtis J. Milhaupt & Mariana Pargendler, RPTs in SOEs: Tunneling, Propping and Policy Channeling, in THE LAWAND FINANCEOF RELATED PARTY TRANSACTIONS (Luca Enriques & Tobias Tr¨oger eds., forthcoming). In theory, it is clear that SOEs should be permitted to pursue welfare-enhancing public policies, such as price moderation in non-competitive industries or the pursuit of countercyclical strategies. After all, this is one of the most common theoretical justifications for state ownership of enterprise. In practice, however, there is often disagreement about whether any given strategy constitutes a welfare-enhancing public policy or rent-seeking by politicians and interest groups.

Seq: 8

18.ANDR´E DELION, DROITDESENTREPRISESETPARTICIPATIONSPUBLIQUES [CORPORATE LAWAND PUBLIC PARTICIPATION] 13 (2003) (Fr.) (citing the Compagnie du chemin de fer de l’Etat of 1887 as the only example of public enterprise in its modern meaning in nineteenth century France).

19. 20.See JEAN-DENIS BREDIN, L’ENTREPRISESEMIPUBLIQUEETPUBLIQUEETLEDROITPRIV´E [THE SEMI-PUBLICAND PRIVATE ENTERPRISEAND PRIVATE LAW] 19 (1957) (Fr.); see also GEORGES RIPERT, ASPECTS JURIDIQUEDU CAPITALISME MODERNE [LEGAL ASPECTSOF MODERN CAPITALISM] 315 (1946).

21.Mariana Pargendler, State Ownership and Corporate Governance, 80 FORD. L. REV.

2917, 2953 (2012) [hereinafter Pargendler, State Ownership].

22.COMMISSIONDES PARTICIPATIONSETDES TRANSFERTS [CPT] [COMMISSIONON INVESTMENTAND TRANSFERS], Les Privatisations en France [Privatization in France] 31–32 (March

Seq: 9

2016) (Fr.), http://www.economie.gouv.fr/files/cptexpo2016.pdf [https://perma.cc/ PW5T-D9ZS].

23.Pepper D. Culpepper, Capitalism, Coordination, and Economic Change: The French Political Economy since 1985, in CHANGING FRANCE 29, 34 (Pepper D. Culpepper et al. eds., 2008).

24.Observatoire de l’Epargne Europeene [OEE] [The European Saving Institute] & INSEAD OEE Data Services, Who Owns the European Economy?, 42 (Aug. 2013), https:// ec.europa.eu/info/sites/info/files/file_import/1308-report-who-owns-european-econo my_en_0.pdf [https://perma.cc/C8Z6-BXQA].

25.

Jon Sindreu, GE and Alstom: In a World of State Ownership, France Prefers Minority Stakes, WALL ST. J. BLOG (June 20, 2014, 2:13 PM), https://blogs.wsj.com/corporateintelligence/2014/06/20/in-a-world-of-state-owned-enterprises-france-prefers-minoritystakes [https://perma.cc/C76F-JLQC].

26.Agence des participations de l’´Etat [APE] [State Shareholding Agency], L’´Etat Actionnaire [State Shareholder] 2014–2015, 8 (Fr.), https://www.economie.gouv.fr/ files/files/directions_services/agence-participations-etat/Rapport_APE.pdf [https:// perma.cc/G7UA-8SRD] [hereinafter APE Report of 2014–2015].

27.Ord. 2014-948 du 20 aoˆut 2014 relative `a la gouvernance et aux op´erations sur le capital des soci´et´es `a participation publique [Order 2014-948 of Aug. 20, 2014 on Governance and Capital Transactions of Publicly-Owned Companies], JOURNALOFFICIEL DELAR´EPUBLIQUE FRAN¸CAISE [J.O.] [OFFICIAL GAZETTEOF FRANCE], 14012 (Aug. 23, 2014); Loi 2015-990 du 6 aoˆut 2015 pour la croissance, l’activit´e et l’´egalit´e des chances ´economiques [Law 2015-990 of Aug. 6, 2015 for the Growth, Activity, and Equal Economic Opportunity], JOURNALOFFICIELDELAR´EPUBLIQUE FRAN¸CAISE [J.O.] [OFFICIAL GAZETTEOF FRANCE], 13537 (Aug. 7, 2015). For a discussion of these recent developments, see Gr´egory Kalfleche, Les soci´et´es `a participation publique et l’Etat actionnaire apr`es la loi Macron, REVUEDEJURISPRUDENCECOMMERCIALE 1, 6 (Oct. 8, 2015) (Fr.).

Seq: 10

28.Ripert, supra note 20, at 317. 29.CODEDEL’AVIATIONCIVILE [C. AVIAT.] [CIVIL AVIATION CODE] art. R251-1 (Fr.). Moreover, specific decrees also provide for the appointment of government commissioners and other officials to attend in advisory capacity the board meetings of major SOEs (such as energy firms EDF and AREVA), for purposes of conveying the government’s policies. 30.

In France, this special legal regime is typically triggered when the state acquires ten percent of the shares and is strengthened when it holds a majority of the capital.See Order 2014-948 of Aug. 20, 2014, supra note 27.

31.CODEDESJURISDICTIONSFINANCIER`ES [C. JUR. FIN.] [CODEOF FINANCIALJURISDIC-

TION] art. L133-1 (Fr.).

32.D´ecret 55-733 du 26 mai 1955 relatif au contrˆole ´economique et financier de l’Etat [Decree 55-733 of May 26, 1955 on the economic and financial control of the State], JOURNALOFFICIELDELAR´EPUBLIQUE FRAN¸CAISE [J.O.] [OFFICIAL GAZETTEOF FRANCE], 5547 (June 6, 1955).

33.D´ecret 53-707 du 9 aoˆut 1953 relatif au contrˆole de l’Etat sur les entreprises publiques nationales et certains organismes ayant un objet d’ordre ´economique ou social [Decree No. 53-707 of Aug. 9, 1953 on the Control of the State over National Public Enterprises and Certain Bodies Having an Economic or Social Purpose], JOURNAL OFFICIELDELAR´EPUBLIQUE FRAN¸CAISE [J.O.] [OFFICIAL GAZETTEOF FRANCE], 7051 (Aug. 10, 2014) (discussing companies taking the form of a business corporation, budgets, income statements, balance sheets, the allocation of profits, and the remuneration of managers require the prior approval of the Minister of the Economy).

34.D´ecret-loi du 30 octobre 1935 organisant le contrˆole de l’Etat sur les soci´et´es, syndicats et associations ou entreprises de toute nature ayant fait appel au concours financier de l’Etat, Article 2, as amended by Loi 49-985 1949-07-25 Art. 12 and Loi 2001-420 2001-05-15 Art. 139.

Seq: 11

35.Loi 83-675 du 26 juillet 1983 relative `a la d´emocratisation du secteur public [Law 83-675 of July 26, 1983 on the Democratization of the Public Sector], JOURNAL OFFICIELDELAR´EPUBLIQUE FRAN¸CAISE [J.O.] [OFFICIAL GAZETTEOF FRANCE], 2326 (July 27, 1983). 36. 37. 38.For instance, Gaz de France and Snecma followed this pattern after their

Id. art. 5. Id. art. 6.

privatization.

39.See, e.g., Benjamin Mojuy´e, French Corporate Governance in the New Millenium:

Who Watches the Board in Corporate France?, 6 COLUM. J. EUR. L. 73, 75 (2000).

40.Christophe Boillon, La gouvernance des entreprises `a participations publiques:

L’´Etat comme actionnaire pri´e, m´emoire, Universit´e Paris-Sud, 17 (2013–2014).

41. 42.See Loi 2014-1 du 2 janvier 2014 habilitant le Gouvernement `a simplifier et s´ecuriser la vie des entreprises [Law 2014-1 of Jan. 2, 2014 on Empowering the Government to Simplify and Secure the Life of Enterprises], JOURNALOFFICIELDELAR´EPUBLIQUE FRAN¸CAISE [J.O.] [OFFICIAL GAZETTEOF FRANCE], 50 (Jan. 3, 2014); see also Order 2014948 of Aug. 20, 2014, supra note 27.

43.Order 2014-948 of Aug. 20, 2014, supra note 27, at art. 1 (Fr.). 44.

Id. art. 6.

Seq: 12

45. 46.

Id. art. 4. Id.; D´ecret 2014-949 du 20 aoˆut 2014 portant application de l’ordonnance 2014948 [Decree 2014-949 of Aug. 20, 2014 on Implementing Order 2014-948], JOURNAL OFFICIELDELAR´EPUBLIQUE FRAN¸CAISE [J.O.] [OFFICIAL GAZETTEOF FRANCE], 14018 (Aug. 23, 2014).

47.See D´ecret no. du 30 octobre 1935, art. 2 (repealed). 48.

Id. art. 6. If the meeting does not approve the state’s nominees, the state can then make such temporary appointments unilaterally, subject to confirmation at the subsequent shareholder meeting. If the number of board seats is greater than ten, the state is entitled to appoint at least two board members.

49.Compte rendu du Conseil des ministers [Minutes of the Council of Ministers], Gouvernance et op´erations sur le capital des soci´et´es `a participation publique [Governance and Capital Transactions of Publicly Owned Companies] 2014 (Fr.), http:// www.gouvernement.fr/conseil-des-ministres/2014-08-20/gouvernance-et-operationssur-le-capital-des-societes-partic [https://perma.cc/73GX-DH43].

Id. art. 11. Id.; Order 2014-948 of Aug. 20, 2014, art. 5 (Fr.).

50.Law 83-675 of 26 July 1983, art. 5 (Fr.). 51. 52. 53.D´ecret 2012-915 du 26 juillet 2012 relatif au contrˆole de l’Etat sur les r´emun´erations des dirigeants d’entreprises publiques [Decree 2012-915 of July 26, 2012 on State Control Over the Remuneration of Managers of Public Enterprises], JOURNAL OFFICIELDE LA REPUBLIQUE FRAN¸CAISE [J.O.] [OFFICIAL GAZETTEOF FRANCE], 12283 (July 27, 2012).

54.Order 2014-948 of Aug. 20, 2014, art. 6 (Fr.).

Seq: 13

55.For an early analysis of this pattern, see Pargendler, State Ownership, supra note

21, at 2953–54.

56. 57.Loi 2014-384 du 29 mars 2014 visant `a reconqu´erir l’´economie r´eelle [Law 2014-384 of Mar. 29, 2014 on Aiming at Reconquering the Real Economy], JOURNAL OFFICIELDE LA REPUBLIQUE FRAN¸CAISE [J.O.] [OFFICIAL GAZETTEOF FRANCE], 6231 (Apr. 1, 2014).

58.APE Report of 2014–2015, supra note 26, at 19. 59.See M. Ren´e Barbier de La Serre et al., L’´Etat actionnaire et le gouvernement des entreprises publiques (Rapport `a M. Francis Mer, Ministre de l’´economie, des finances et de l’industrie) [The State Shareholder and the Government of Public Enterprises (Report to Mr. Francis Mer, Minister for the Economy, Finance and Industry)], 1 (2003) (Fr.), http://www.ladocumentationfrancaise.fr/var/storage/rapports—publics/ 034000095.pdf.

60.D´ecret 2004-963 du 9 septembre 2004 portant cr´eation du service `a comp´etence nationale “Agence des participations de l’Etat” [Decree 2004-963 of Sept. 9, 2004 Establishing the Service of National Competence “State Shareholding Agency”], JOURNAL OFFICIELDELA R´EPUBLIQUEFRAN¸CAISE [J.O.] [OFFICIAL GAZETTEOF FRANCE], 15929 (Sept. 10, 2004).

61.D´ecret 2004-963 du 9 septembre 2004 portant cr´eation du service `a comp´etence nationale “Agence des participations de l’Etat” (rectificatif) [Decree 2004-963 of Sept. 9, the Service of National Competence “State Shareholding 2004 Establishing

Seq: 14

Agency”(addendum)], JOURNAL OFFICIELDELA R´EPUBLIQUEFRAN¸CAISE [J.O.] [OFFICIAL GAZETTEOF FRANCE], 16017 (Sept. 11, 2004).

62.Shareholding Structure July 12, 2017, AREVA, http://www.new.areva.com/EN/ finance-1166/shareholding-structure-of-the-world-leader-in-the-nuclear-industry-and-ma jor-player-in-bioenergies.html?XTCR=1,68&XTMC==CATITAL%20STRUCTURE?XTMC [https://perma.cc/N3HX-NBD8].

Id. art. 2. Id. Id. art. 1, art. 2.

63.Decree 2004-963 of Sept. 9, 2004, art.1 (Fr.). 64. 65. 66. 67.Agence des participations de l’Etat [APE] [State Shareholding Agency], The French State as Shareholder, 7 (2012) (Fr.), https://www.economie.gouv.fr/files/files/ directions_services/agence-participations-etat/Documents/Rapports-de-l-Etat-action naire/2012/Overview_2012.pdf [https://perma.cc/6KR3-DZGX].

68. 69.APE Report of 2014–2015, supra note 26, at 20–21 (Fr.). 70.

Id. at 32.

Seq: 15

71.Yann Le Guernigou, Les dividendes de l’Etat inqui`etent la Cour des comptes [State Dividends Worry about Court of Auditors], REUTERS (May 27, 2005) (Fr.), http:// fr.reuters.com/article/businessNews/idFRKBN0OC19S20150527 [https://perma.cc/ F233-C3GC].

72.APE Report of 2014–2015, supra note 26, at 9 (Fr.). 73.From the early days of the country’s founding, corporations were viewed with suspicion as dangerous concentrations of power, and their governance was closely bound up in debates about the proper allocation of power between the national government and the states. One important result of this historical legacy is that corporate law in the United States is principally the province of the states, not the U.S. Congress.

74.Pargendler, State Ownership, supra note 21, at 2927; see infra text accompanying

note 101.

75.See, e.g., Alyssa Katz, The Rise and Fall of Fannie Mae: A Timeline, AOL.COM (Dec. 24, 2010), https://www.aol.com/2010/12/24/the-rise-and-fall-of-fannie-mae-a-timeline [https://perma.cc/W85P-ZGNG]; Kate Pickert, A Brief History of Fannie Mae and Freddie Mac, TIME (July 14, 2008), http://content.time.com/time/business/article/0,8599,1822 766,00.html [http://perma.cc/QU25-FDX4].

Seq: 16

76.David Goldman, CNNMoney.com’s Bailout Tracker, CNN MONEY (Feb. 16, 2009),

http://money.cnn.com/news/storysupplement/economy/bailouttracker/index.html [https://perma.cc/U6XT-9GJN].

77.See, e.g., Lloyd D. Musolf, American Mixed Enterprise and Government Responsi-

bility, 24 WESTERN POL. Q. 789 (1971).

78.See Robert A. Lively, The American System: A Review Article, 29 BUS. HIST. REV. 81, 92 (1955) (finding that although “Public losses were probably no greater than those of early private investors . . . [,] Massachusetts suffered a $9,500,000 loss on [its] $28,856,896 Hoosac Tunnel expenditures . . . .”).

79.

John Joseph Wallis, Richard E. Sylla & John B. Legler, The Interaction of Taxation and Regulation in Nineteenth-Century U.S. Banking, inTHE REGULATED ECONOMY: HISTORICAL APPROACHTO POLITICAL ECONOMY 121, 142 (Claudia Goldin & Gary D. Libecap eds., 1994). For a more general review of these conflicts, see Pargendler, supra note 21, at 2928–31. 80. Id. 81.Lively, supra note 78, at 83. 82. 83.LAWRENCE M. FRIEDMAN, A HISTORYOF AMERICAN LAW 121 (3d ed. 2005). 84.See id. 85.

Id.; John Joseph Wallis, Constitutions, Corporations, and Corruption: American States and Constitutional Change,1842 to 1852, 65 J. ECON. HIST. 211, 232, 251 (2005). 86.ALEXANDER HAMILTON, Final Version of the Second Report on the Further Provision Necessary for Establishing Public Credit: Report on a National Bank, inTHE PAPERSOF ALEXANDER HAMILTON236, 305–10 (Harold C. Syrett & Jacob E. Cooke eds., 1963).

JAMES W. ELY, JR., RAILROADSAND AMERICAN LAW 16–17 (2001).

Seq: 17

Id. at 332–33. Id. at 329. Id. at 331. Id. Id. at 333.

87. 88. 89. 90. 91. 92.Musolf, supra note 77, at 796. 93. 94. 95. 96. 97. 98.Herman Schwartz, Governmentally Appointed Directors in a Private Corporation—

Id. Id. Id. at 797. Id. at 798. Id.

The Communications Satellite Act of 1962, 79 HARV. L. REV. 350, 359–60 (1965).

Seq: 18

99.C. Herman Pritchett, The Government Corporation Control Act of 1945, 40 AM.

POL. SCI. REV. 495, 508 (1946).

100.Until a 2008 charter amendment, the President had the right to appoint five out of 18 directors at Fannie Mae and Freddie Mac. Kathleen Day & David S. Hilzenrath, Presidential Pals Populate Fannie, Freddie Boards, WASH. POST(June 27, 2003), https:// www.washingtonpost.com/archive/business/2003/06/27/presidential-pals-populatefannie-freddie-boards/47a3d1bf-2392-40f6-92f9-55ed0ced57e3/?utm_term=.Cb8b3ca3 bd3e [http://perma.cc/W6CM-39RG].

101.Pickert, supra note 75. 102.U.S. GOV’T ACCOUNTABILITY OFFICE, GAO-09-782, FANNIE MAEAND FREDDIE MAC: ANALYSISOF OPTIONSFOR REVISINGTHE HOUSING ENTERPRISES’ LONG-TERMStructures 14–16 (2009).

103.THOMAS H. STANTON, GOVERNMENT-SPONSORED ENTERPRISES 23 (2002). 104.

Seq: 19

105.Gretchen Morgenson, Mortgage Giants Leave Legal Bills to the Taxpayers, N.Y.

TIMES, Jan. 24, 2011, at A3.

106.STAFF H.R. COMM. ON OVERSIGHT & GOV’T REFORM, 111TH CONG., THE ROLEOF GOVERNMENT AFFORDABLE HOUSING POLICYIN CREATINGTHE GLOBAL FINANCIAL CRISISOF 2008 16 (Comm. Print 2009).

Id. at 24.

107. 108.See Steven M. Davidoff & David Zaring, Regulation by Deal: The Government’s Response to the Financial Crisis, 61 ADMIN. L. REV. 463, 526–27 (2009); Paul Krugman, Gordon Does Good, N.Y. TIMES, Oct. 12, 2008, at A29.

109.See generally Barbara Black, The U.S. as Reluctant Shareholder: Government, Business and the Law, 5 ENTREPRENEURIAL BUS. L.J. 561 (2010) (discussing the U.S. government’s role as shareholder in several companies following the 2008 financial crisis).

110.Steven M. Davidoff, Uncomfortable Embrace: Federal Corporate Ownership in the

Midst of the Financial Crisis, 95 MINN. L. REV. 1733, 1735–36 (2011).

111. 112. 113. 114.

Id. at 1733. Id. at 1752. Id. at 1748. Id.

Seq: 20

115.See Starr Int’l Co. v. United States, 121 Fed. Cl. 428, 431 (2015). 116.FED. RES. BANKOF N.Y., AIG CREDIT FACILITY TRUST AGREEMENT AMONG FED RES. BANKOF N.Y., AND JILL M. CONSIDINE, CHESTER B. FELDBERHAND DOUGLAS L. FOSHEE, AS TRUSTEES 2 (2009). Id. Id. §2.04(d).

117. 118. 119.Davidoff & Zaring, supra note 108, at 530. 120.Exchange Agreement dated June 9, 2009 between Citigroup Inc. and United States Department of the Treasury, §4.18, available at http://www.sec.gov/Archives/ edgar/data/831001/000119312509128765/dex103.htm [https://perma.cc/LB8SQMGV].

121.History of Fannie Mae & Freddie Mac Conservatorships, FED. HOUSING FIN. AGENCY, https://www.fhfa.gov/Conservatorship/pages/history-of-fannie-mae—freddieconservatorships.aspx [https://perma.cc/LFL6-TYMQ].

122.U.S. DEP’T TREASURY, THIRD AMENDMENTTO AMENDEDAND RESTATED SENIOR PREFERRED STOCK PURCHASE AGREEMENTBETWEEN U.S. DEP’T TREASURY & FED. NAT’L MORTGAGE ASS’N 1 (2012).

123.See, e.g., Richard A. Epstein, The Government Takeover of Fannie Mae and Freddie Mac: Upending Capital Markets with Lax Business and Constitutional Standards, 10 N.Y.U. J.L. & BUS. 379, 408–09 (2014); Jonathan Macey & Logan Beirne, Stealing Fannie and Freddie (Apr. 27, 2014) (unpublished working paper) https://papers.ssrn.com/sol3/ papers.cfm?abstract_id=2429974 [https://perma.cc/9S5H-DQB3].

Seq: 21

124.See generally Starr, 121 Fed. Cl. 428 (concluding that the plaintiff shareholders suffered no economic loss, because without the government’s loan the company would have declared bankruptcy).

125.Chris Isidore, U.S. Ends TARP with $15.3 Billion Profit, CNN MONEY (Dec. 19, 2014, 11:27 AM), http://money.cnn.com/2014/12/19/news/companies/governmentbailouts-end/index.html [https://perma.cc/TP2M-AU7H].

126.Davidoff, supra note 110, at 1757–58. 127.SeeMARIANA MAZZUCATO, THE ENTREPRENEURIAL STATE 328–29 (2011). 128.See, e.g., Alberto Mingardi, A Critique of Mazzucato’s Entrepreneurial State, 35 CATOJ. 603, 613 (2015) (arguing that the SBIR program “adds up to little more than forcing some public bodies to sign checks”).

129.See Starr, 121 Fed. Cl. at 472 (“The manner in which FRBNY controlled AIG with its handpicked CEO, carefully selected board members, and its hundreds of onpremises advisers belies any conclusion that the operations of the trust were independent”).

130.See, e.g., Black, supra note 109; Davidoff, supra note 110.

Seq: 22

131.See Pargendler, State Ownership, supra note 21, at 2965–66 (discussing how the acquisition of Bear Stearns by J.P. Morgan, as engineered by the government, likely ran afoul of corporate law rules).

132.See Marcel Kahan & Edward B. Rock, When the Government Is the Controlling

Shareholder, 89 TEX. L. REV. 1293, 1317–19 (2011).

Id. at 1318–19. Id. at 1318. IMF, Norway: Selected Issues, IMF Country Report No. 14/260, at 25 (Aug.

133. 134. 135. 2014).

136.ORG. FOR ECON. CO-OPERATION & DEV. [OECD], Regulatory Reform in Norway: Marketisation of Government Services—State-Owned Enterprises 7 (2003), https:// [hereinafter www.oecd.org/regreform/32682052.pdf OECD, Regulatory Reform in Norway].

[https://perma.cc/NT25-3Z25]

137.Ola Honningdal Grytten, The Economic History of Norway, EH.NET, https://

eh.net/encyclopedia/the-economic-history-of-norway [https://perma.cc/HM3B-CLTM].

138.See generally Lars Jonung, Lessons from the Nordic Financial Crisis, 2011 ANN. MEETING ALLIED SOC. SCI. ASS’N 12 (2011), https://www.aeaweb.org/conference/2011/ retrieve.php?pdfid=413 [https://perma.cc/R7UV-SUYX].

139.NORWEGIAN MINISTRYOF TRADE, INDUSTRYAND FISHERIES, DIVERSEAND VALUE CREATING OWNERSHIP 89 (2013–2014), https://www.regjeringen.no/contentassets/ 899ac257df2648d788942b78c6d59787/en-gb/pdfs/stm201320140027000engpdfs.pdf [https://perma.cc/XXG7-BN3K] [hereinafter 2013–2014 WHITE PAPER].

Seq: 23

140.OECD, Regulatory Reform in Norway, supra note 136, at 7. 141.See, e.g., Mark C. Thurber et al., Exporting the “Norwegian Model”: The Effect of

Administrative Design on Oil Sector Performance, 39 ENERGY POL’Y 5366, 5366 (2011).

142.NORWEGIAN MINISTRYOF TRADE, INDUSTRYAND FISHERIES, ACTIVE OWNERSHIP 24– 25 (2010–2011), https://www.regjeringen.no/contentassets/63b6e091e7e84effbb0e7e 2083fc21a8/en-gb/pdfs/stm201020110013000en_pdfs.pdf [hereinafter 2010–2011 WHITE PAPER].

143. 144.See2013–2014 WHITE PAPER, supra note 139, at 61 (Nor.). 145. 146.OECD, Regulatory Reform in Norway, supra note 136, at 17. Overall centralization of SOE shareholdings (including non-commercial SOEs) has been less extensive, however, with sixty-five percent of SOEs remaining under the supervision of sectoral ministries as of 2005. Stine Ludvigsen, State Ownership and Corporate Governance: Empirical Evidence from Norway and Sweden 44–45 (Mar. 2010) (unpublished Ph.D. dissertation, BI Norwegian School of Management) (on file with BI Norwegian Business School).

147.OECD, Regulatory Reform in Norway, supra note 136, at 14. 148.2013–2014 WHITE PAPER, supra note 139, at 89 (Nor.).

Seq: 24

149.Thurber et al., supra note 141, at 5367. 150.NORWEGIAN MINISTRYOF TRADE, INDUSTRYAND FISHERIES,AN ACTIVEAND LONGTERM STATE OWNERSHIP70 (2006–2007), https://www.regjeringen.no/contentassets/ 01527a83111e45639d5dbb0d84882a44/en-gb/pdfs/stm200620070013000en_pdfs.pdf [https://perma.cc/7GRX-PZ4Z] [hereinafter 2006–2007 WHITE PAPER].

Id. Id. Id. Id.

151. 152. 153. 154. 155.2013–2014 WHITE PAPER, supra note 139, at 67 (Nor.). 156. 157.See Beate Sj˚afjell, Sustainable Companies: Possibilities and Barriers in Norwegian

Id. at 68.

Company Law, 11 INT’L & COMP. CORP. L.J. 1, 32 (2013).

158.SeeSTATOIL ASA,2014 BOARD STATEMENTON CORPORATE GOVERNANCE 4 (2014) [hereinafter 2014 BOARD STATEMENT]; Mark C. Thurber & Benedicte Tangen Istad, Norway’s Evolving Champion: Statoil and the Politics of State Enterprise 32 (Program on Energy & Sustainable Dev., Working Paper No. 92, 2010), https://pesd.fsi.stanford.edu/ sites/default/files/WP_92%2C_Thurber_and_Istad%2C_Statoil%2C_21May2010.pdf [https://perma.cc/NWN3-XJ99].

Seq: 25

159.OECD, Regulatory Reform in Norway, supra note 136, at 23. 160.2013–2014 WHITE PAPER, supra note 139, at 64 (Nor.). 161.Sj˚afjell, supra note 157, at 6. 162.See 2014 BOARD STATEMENT, supra note 158, at 7–8. 163.See id. at 8. 164.ORG. FOR ECON. CO-OPERATION & DEV. [OECD], Peer Review 6: Risk Management and Corporate Governance, at 32, DAF/CA/CG(2013)5/FINAL (Dec. 12, 2013), http:// www.oecd.org/officialdocuments/publicdisplaydocumentpdf/?cote=DAF/CA/CG(2013) 5/FINAL&docLanguage=EN [https://perma.cc/4B8L-CW2H]. While some listed SOEs such as Statoil and Telenor have a corporate assembly, there are others which have opted out of this requirement.See 2013–2014 WHITE PAPER, supra note 139, at 65 (Nor.).

165.See 2014 BOARD STATEMENT, supra note 158, at 9. 166.See 2006–2007 WHITE PAPER, supra note 150, at 21 (Nor.). Listed companies are not technically required to have a nomination committee, as the Public Companies Act treats nomination committees as voluntary bodies. Norsk utvalg for eierstyring og selskapsledelse [NUES] [Norwegian Corp. Governance Board], Norsk anbefaling EIERSTYRING OG SELSKAPSLEDELSE [The Norwegian Code of Practice for Corporate Governance] 27 (2014), http://wpstatic.idium.no/nues.no/2017/06/2014-10-30Anbefaling2014NOR web.pdf [https://perma.cc/4B8L-CW2H] [hereinafter Code of Practice].

167.Code of Practice, supra note 166, at 26 (Nor.). 168.2014 BOARD STATEMENT, supra note 158, at 4.

Seq: 26

169.OECD, Regulatory Reform in Norway, supra note 136, at 23. 170.Thurber & Istad, supra note 158, at 20. 171. 172.OECD, Regulatory Reform in Norway, supra note 136, at 14. 173.Ludvigsen, supra note 146, at 39. 174.See id. at 122 (finding that twenty-three percent of the chairmen of Norwegian SOEs fall in the category of political representatives); Jenni Maria Nossum, Corporate Governance in Oil-Lubricated Norway: Regulation, Practice, Ethics and Incoherence 11 n. 59 (Nordic & Eur. Company Law Working Paper Series, Working Paper No. 2015-11, 2015), http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2622072 [https:// perma.cc/8NG7-84BJ].

175.2006–2007 WHITE PAPER, supra note 150, at 22 (Nor.). 176. 177. 178.Norway: Selected Issues, supra note 135, at 35. 179.2013–2014 WHITE PAPER, supra note 139, at 67 (Nor.). 180.See id.

Seq: 27

Id. at 65. The Norwegian state’s principles of corporate governance are as

181. follows:

1.All shareholders shall be treated equally. 2.There shall be transparency in the State’s ownership of companies. 3.Ownership decisions and resolutions shall be made at the general meeting. 4.The board is responsible for elaborating explicit objectives and strategies for the company within the constraints of its articles of association; the state sets performance targets for each company.

5.The capital structure of the company shall be appropriate given the objective

and situation of the company.

6.The composition of the board shall be characterized by competence, capacity and diversity and shall reflect the distinctive characteristics of each company.

7.The board assumes executive responsibility for administration of the company, including performing an independent supervisory function vis-`a-vis the company’s management on behalf of the owners.

8.The board should adopt a plan for its own work, and work actively to

develop its own competencies and evaluate its own activities.

9.Compensation and incentive schemes shall promote value creation within

the companies and be generally regarded as reasonable.

10.The company shall work systematically to safeguard its corporate social

responsibility.Id. at 68 box 8.1.

182.See 2006–2007 WHITE PAPER, supra note 150, at 63 (Nor.). 183.NORWEGIAN MINISTRYOF TRADE, INDUSTRYAND FISHERIES, GUIDELINESFOR REMUNERATIONOF SENIOR EXECUTIVESIN COMPANIES WITH STATE OWNERSHIP 2–3 (2015) (Nor.), https://www.regjeringen.no/contentassets/4391143c1f0a472faa0b3975e00e3c78/ guidelines-for-remuneration.pdf [https://perma.cc/4DHM-X8BL]. 184.2013–2014 WHITE PAPER, supra note 139, at 74 (Nor.). 185.2006–2007 WHITE PAPER, supra note 150, at 30 (Nor.). 186.

Seq: 28

Id. at 102–27.

187.2013–2014 WHITE PAPER, supra note 139, at 91–102 (Nor.). 188. 189.2010–2011 WHITE PAPER, supra note 142, at 5–6 (Nor.). 190.2013–2014 WHITE PAPER, supra note 139, at 10 (Nor.). 191.See id. at 10–11. Id. at 40. 192. 193. Id. at 41. 194.See id. at 93. 195.Members and Partners, ORG. FOR ECON. CO-OPERATION & DEV. [OECD], http:// www.oecd.org/about/membersandpartners/ [https://perma.cc/G2BR-6ZFF].See World Bank Group [WBG], Corporate Governance of State-Owned Enterprises in Latin America: Current Trends and Country Cases 142, 89468-LAC (2014), http://documents.world bank.org/curated/en/2014/07/20183864/corporate-governance-state-owned-enterpris es-latin-america-current-trends-country-cases [https://perma.cc/G2BR-6ZFF] (illustrating that Mexico owns PEMEX) [hereinafter World Bank Group, Corporate Governance of State-Owned Enterprises in Latin America].

196.SeeORG. FOR ECON. CO-OPERATION & DEV. [OECD], The Governance of MixedOwnership Enterprises in Latin America: Discussion Paper 4 (2012), http://www.oecd

Seq: 29

.org/daf/ca/LA-SOE-Mixed-Ownership-2013.pdf [https://perma.cc/N7VZ-UMN4] [hereinafter OECD Discussion Paper].

197.Corporate Governance of State-Owned Enterprises in Latin America: Current Trends and Country Cases, supra note 195, at 3. Although the federal government has been the controlling shareholder of Ecopetrol, ISA and ISAGEN, ETB is owned and controlled by the capital district of Bogot´a.Bogota City Government Seeks to Sell Controlling Stake in ETB, REUTERS(Apr. 29, 2016), https://www.reuters.com/article/colombia-etb/ bogota-city-government-seeks-to-sell-controlling-stake-in-etb-idUSL2N17W1GW [https:/ /perma.cc/A9ZX-P923].

198.OECD Discussion Paper, supra note 196, at 4; IMF, Chile: Selected Issues, Country Report No. 02/163, at ¶ 160 (Aug. 2002).SeeORG. FOR ECON. CO-OPERATION & DEV. [OECD], Ownership Oversight and Board Practices for Latin American State-Owned Enterprises 5 (2012), https://www.oecd.org/daf/ca/2011LatinAmericanCorporateGovern anceRoundtableSOEOwnership2013.pdf [https://perma.cc/7LKY-FL86]. Although Peru formally has nine listed SOEs, almost all of them are either wholly owned by the state or lack meaningful trading activity.

199.See Corporate Governance of State-Owned Enterprises in Latin America: Current

Trends and Country Cases, supra note 195, at 96.

200.SeeORG. FOR ECON. CO-OPERATION & DEV. [OECD], OECD Review of the Corporate Governance of State-Owned Enterprises: Colombia, at 23, 33–35 (2015), http:// www.oecd.org/daf/ca/OECD-Review-Corporate-Governance-SOE-Colombia.pdf [https:/ /perma.cc/WZ5S-SAM9] [hereinafter OECD Review]; see also REUTERS supra note 197.

201.See Ryan Dube & Daniela Ramirez, Colombia Sells Stake in Power Generator Isagen to Canadian Investment Fund, WALL ST. J. (Jan. 13, 2016), http://www.wsj.com/ articles/colombia-sells-stake-in-power-generator-isagen-to-canadian-investment-fund-14 52702800 [https://perma.cc/PBJ2-CPCD].

202.See Corporate Governance of State-Owned Enterprises in Latin America: Current Trends and Country Cases, supra note 195, at 96. Near the peak of its valuation, Ecopetrol alone accounted for over forty-six percent of the local market capitalization in 2012. See OECD Review, supra note 200, at 23.

203.See Ed Crooks et al., Ecopetrol Overtakes Petrobas by Market Cap, FIN. TIMES (Jan. 27, 2013), https://www.ft.com/content/818379d6-68a4-11e2-8c20-00144feab49a [https://perma.cc/47GX-GXED].

Seq: 30

204.See U.S. DEP’T ST., COLOMBIA INVESTMENT CLIMATE STATEMENT 18 (2015). 205.CONSTITUCI´ON POL´ITICADE COLOMBIA[C.P.] art. 150. 206.

Id. at art. 60; L. 226/95, diciembre 21, 1995, [42.159] DIARIO OFICIAL [D.O.]

(Colom.).

Id. at 32.

207.OECD Review, supra note 200, at 31. 208. 209.C´ODIGODE COMERCIO [C. COM] [COMMERCIAL CODE] arts. 461 & 464 (Colom.). 210.OECD Review, supra note 200, at 36. 211. 212. 213.MINISTRODE HACIENDAY CR´EDITO P´UBLICO, DECLARACIONDELA NACI´ONENSU CALIDADDE ACCIONISTA MAYORITARIODE ECOPETROL S.A., 3 (2007), http:// www.ecopetrol.com.co/documentos/40316_Declaracion_del_Accionista_Mayoritario_26—07—07.pdf [https://perma.cc/3SVV-GGM8].

Id. Id. at 53.

214.See OECD Review, supra note 200, at 53.

Seq: 31

Id. at 74.

215. 216.DEPARTAMENTO NACIONALDE PLANEACI´ON [DNP] [NAT’L PLANNING DEP’T] & CONSEJO NACIONALDE POL´ITICA ECON´OMICAY SOCIAL [CONPES] [NAT’L COUNCILON ECON. & SOC. POL’Y], POL´ITICAGENERALDEPROPRIEDADDEEMPRESASESTATALESDELORDENNACIONAL [GENERAL OWNERSHIP POLICYOF STATE-OWNED ENTERPRISES] 13 (2015) (Colom.) [hereinafter GENERAL OWNERSHIP POLICY].

217.OECD Review, supra note 200, at 53. 218.ADR-American Depositary Receipt, ECOPETROL (Oct. 26, 2014) (Colom.), http://

www.ecopetrol.com.co/wps/portal/web_es/ecopetrol-web/investors/equity/adr-ameri can-depositary-receipt/!ut/p/z0/04_Sj9CPykssy0xPLMnMz0vMAfIjo8ziLf0N3d09gg283 [https://perma.cc/S4H5-B96 Y2cTA0cPb2NLQxNPU0DXM31g1Pz9AuyHRUBLl7_Zw N].

219.Ecopetrol S.A. informa que aplicar´a para deslistarse voluntariamente de la Bolsa de Valores de Toronto [Ecopetrol S.A. Informs That It Will Apply to Voluntarily De-list from the Toronto Stock Exchange], ECOPETROL (Feb. 18, 2016) (Colom.), http://www.ecopetrol. com.co/wps/portal/es/ecopetrol-web/nuestra-empresa/sala-de-prensa/boletines-deprensa/Boletines/Boletines/Ecopetrol-deslistar-voluntariamente-Bolsa-Valores-Toronto [https://perma.cc/PW4R-NSFN].

220.Helena Garcia Romero & Laura Calderon Etter, The Political Economy of Fuel Subsidies in Columbia, at 7, OECD Environment Working Paper No.61 (Dec. 10, 2013), http://dx.doi.org/10.1787/5k3twr8v5428-en [https://perma.cc/PY6R-EL59].

221. 222.The subsidy is calculated as the “difference between the producer’s regulated revenues and the parity price; the latter is equivalent to the opportunity cost of the local product or to the import cost of gas or ACPM at the U.S. Gulf Coast reference price.” Ecopetrol S.A., Annual Report (Form 20-F) F-18 (Apr. 28, 2015) [hereinafter Ecopetrol 2014 20-F].

223.See Ecopetrol S.A., Annual Report (Form 20-F) 9 (June 30, 2009). 224.Compare Ecopetrol S.A., Annual Report (Form 20-F) (Apr. 30, 2012), with Ecope-

trol S.A., Annual Report (Form 20-F) (Apr. 29, 2013).

225.See Historical Dividends, ECOPETROL (June 5, 2017) (Colom.), http://www.eco petrol.com.co/wps/portal/web_es/ecopetrol-web/investors/equity/historical-dividends/

Seq: 32

!ut/p/z0/04_Sj9CPykssy0xPLMnMz0vMAfIjo8ziLf0N3d09gg283Y2cTA0cPb2NLQxNPX 38_Az0C7IdFQHKCMUU[https://perma.cc/QX46-JL69].

Id. at 120.

226.Ecopetrol 2014 20-F, supra note 221, at 14. 227. 228.OECD Review, supra note 200, at 17. 229.See, e.g., Georgina N´u˜nez & Andr´es Oneto, Corporate Governance in Brazil, Chile, Colombia, Mexico and Peru, Econ. Commission for Latin Am. and the Caribbean [ECLAC] 183, LC/W. 654 (2015) (“Ecopetrol offers a unique example of a company that has succeeded in combining majority State ownership with solid corporate governance, instilling confidence in the market with respect to its corporate management”); Andr´es Bernal et al., Corporate Governance in Latin America: Importance for State-Owned Enterprises —SOEs, at 28, Public Policy and Productive Transformation Series No. 6 (2012) (listing Ecopetrol and ISAGEN, together with Brazilian firms Petrobras and SABESP, as “successful listed SOEs” and “worldwide leaders”).

Id. at 32–33. Id. at 81.

230.OECD Review, supra note 200, at 32. 231. 232. 233.L. 489/98, diciembre 30, 1998, DIARIO OFICIAL [D.O.] (Colom.). 234.World Bank Group, Corporate Governance of State-Owned Enterprises in Latin

America, supra note 195, at 30; OECD Review, supra note 200, at 24–25.

Seq: 33

235.OECD Review, supra note 200, at 25. 236. 237.World Bank Group, Corporate Governance of State-Owned Enterprises in Latin

America, supra note 195, at 102.

238.For a detailed description of the plan, see GENERAL OWNERSHIP POLICY, supra note 25 (Colom.). Colombia had not yet obtained full OECD membership as of the date of publication.

Id. at 3. Id. Id. at 41. Id. at 42. Id. at 48.

239. 240. 241. 242. 243. 244.See generally Aldo Musacchio & Sergio Lazzarini (Consultants for OECD Secretariat), State-Owned Enterprises in Brazil: History and Lessons, 4–24, OECD Workshop on State-Owned Enterprises in the Development Process (Apr. 4, 2014).

Seq: 34

245.See Mariana Pargendler, The Unintended Consequences of State Ownership: The Brazilian Experience, 13 THEORETICAL INQUIRIES L. 503, 511 (2012) [hereinafter Pargendler, The Unintended Consequences].

246.Pargendler, State Ownership, supra note 21, at 2918–19. 247.B3, supra note 11, at 3. 248.CONSTITUI¸C˜AO FEDERAL [C.F.] [CONSTITUTION] art. 173 (Braz.). 249.CONSTITUI¸C˜AO FEDERAL [C.F.] [CONSTITUTION] art. 37, XIX (Braz.). 250.CONSTITUI¸C˜AO FEDERAL [C.F.] [CONSTITUTION] art. 71, III (Braz.). 251. 252.For the most recent rule in this respect, see Lei No. 11.101, de 15 de Fevereiro de

Id. art. 173, para. 1, II.

2005, DI´ARIO OFICIALDA UNI˜AO [D.O.U.] de 2.9.2005, art. 2 (Braz.).

253.CONSTITUI¸C˜AO FEDERAL [C.F.] [CONSTITUTION] art. 165, II (Braz.). 254.Decreto-Lei No. 200, de 25 de Fevereiro de 1967, DI´ARIO OFICIALDA UNI˜AO [D.O.U.] de 27.2.1967, art. 5, III (Braz.). SOE charters have historically included various provisions that abrogate the standard legal regime. Pargendler, The Unintended Consequences, supra note 245, at 506. Some of these special provisions exacerbated the influence of the state, such by granting the President of Brazil the right to directly appoint the firm’s chief executive. Other exceptions were protective of minority shareholders, for example, by granting them special board appointment rights. Petrobras, for instance, had such a system in place until the late 1990s.See Lucila Gabriel de Almeida, As duas faces da Petrobras: a persistente dinˆamica da empresa estatal no arranjo institucional brasileiro [The Two Facets of Petrobras: The Persistent Dynamics of the State Com-

Seq: 35

pany in the Brazilian Institutional Arrangement] 63, 74, 75, 104 (2011) (unpublished master’s dissertation, Funda¸c˜ao Get´ulio Vargas Escola De Direito) (on file with FGV Sistema de Bibliotecas FGV).

255.Lei No. 6.404, de 15 de Dezembro de 1976, DI´ARIO OFICIALDA UNI˜AO [D.O.U.] de 17.12.1976, art. 235, §1 (Braz.). Subsidiaries of SOEs, however, were subject exclusively to general corporate laws.Id. art. 235, §2.

256.CONSTITUI¸C˜AO FEDERAL [C.F.] [CONSTITUTION] art. 173, IV (Braz.). 257.Lei No. 6.404, de 15 de Dezembro de 1976, art. 236 (Braz.). 258.Lei No. 13.303, de 30 de junho de 2016, DI´ARIO OFICIALDA UNI˜AO [D.O.U.] de 1.7.2016, art. 4, §1 (Braz.) (providing that the legal entity that controls a mixed enterprise has the same duties and responsibilities of a controlling shareholder under the Lei No. 6.404, de 15 de Dezembro de 1976 and “must exercise the power of control in the interest of the company, subject to the public interest that justified its creation”).

259. 260.See, e.g., Lei No. 6.404, de 15 de Dezembro de 1976, art. 117, §1(a) (Braz.) (qualifying as abuse of control power the action of a controlling shareholder that “guides the company towards an objective that is foreign to its corporate purpose or damaging to national interest, or that leads it to favor another company, domestic or foreign, to the detriment of minority shareholders’ participation in the profits or assets of the company, or the national economy”).

261.For a discussion of the interests of the state in molding corporate laws in Brazil and beyond, see generally Pargendler, State Ownership, supra note 21; Pargendler, The Unintended Consequences, supra note 244.

Seq: 36

262.Musacchio & Lazzarini, supra note 243, at 5–16. Much of the international literature, however, has inaccurately depicted the rise of listed SOEs in Brazil as a strategy first initiated in the last two decades.See, e.g., World Bank Group, Toolkit for StateOwned Enterprise, supra note 4, at 44; Flores-Macias & Musacchio, supra note 4.

263. 264.See supra note 4 and accompanying text. 265.See supra notes7–13 and accompanying text. 266.See, e.g, Pargendler, State Ownership, supra note 21, at 2936. 267.Mariana Pargendler, Governing State Capitalism: The Case of Brazil, inREGULATING THE VISIBLE HAND? THE INSTITUTIONAL IMPLICATIONSOF CHINESE STATE CAPITALISM 377, 377 (Benjamin L. Liebman & Curtis J. Milhaupt eds., 2016).

268.See id. at 385–89.

Seq: 37

273. 274.

269.See id. at 384–85. 270.See id. at 385–89. 271.B3, supra note 11, at 3 (Braz.). 272.For a discussion of the Novo Mercado experiment, see Ronald J. Gilson, Henry Hansmann & Mariana Pargendler, Regulatory Dualism as a Development Strategy: Corporate Reform in Brazil, the United States and the European Union, 63 STAN. L. REV. 475, 494–501 (2011). Id. Id. The six requirements which form the core of the program are: (1) enhanced disclosure in securities filings, especially with respect to the SOE’s actions in enforcing public policies and their impact on the company’s financial performance; (2) establishment of a Compliance & Risk Department; (3) existence of an Internal Audit Department and a Statutory Audit Committee, under independent leadership and comprised of a majority of independent members; (4) enactment of a policy on related-party transactions, which shall provide, among other things, for the analysis by an independent corporate body; (5) determination of minimum criteria for the composition of the board of directors, the board of officers and the board of supervisors, including limitations on the appointment of political appointees and a ban on the appointment of representatives of the SOE’s regulator, directors of political parties and holders of elective office; and (6) compliance with requirements for the appointment of managers.Id.

Seq: 38

275.B3, supra note 11. 276.See Lei No. 13.303, de 30 de junho de 2016 (Braz.). 277.CONSTITUI¸C˜AO FEDERAL [C.F.] [CONSTITUTION] art. 173, §1 (Braz.). 278.Lei No. 13.303, de 30 de junho de 2016, art. 8 (Braz.). Id. art. 8, V, VII. 279. 280. Id. art. 9. 281.See id. art. 13, I. 282.See id. art. 17. In addition to mandating several years of experience in government or the private sector in the same or related area in which the SOE operates, the statute includes a long list of individuals who may not serve as directors or officers, including ministers and secretaries of state, regulators, legislators, leaders of political parties, union leaders, and contractual counterparties of the SOE.See id. The experience requirements, however, may be overly rigid, and may detract from the objective of promoting strong, independent and diverse boards.

283. 284. 285.

Id. art. 22. Id. art. 20. Id. art. 27. Art. 27 provides:

The public enterprise and the mixed enterprise will have the social function of fulfilling the collective objective or serving the national security imperative defined in the legal instrument of its creation.

Seq: 39

§1°x. The fulfillment of the collective interest addressed by this article must be geared towards the achievement of economic welfare and the socially efficient allocation of resources administered by the public enterprise and the mixed enterprise, as well as toward the following:

I – the economically sustainable expansion of consumer access to the products and services of the public enterprise and mixed enterprise; II – the development or utilization of Brazilian technology for the production and supply of products and services by the public enterprise and the mixed enterprise, always in an economically justifiable manner.

§2° The public enterprise and the mixed enterprise must, as dictated by statute, adopt practices of environmental sustainability and corporate social responsibility that are compatible with the market in which they operate. §3° The public enterprise and mixed enterprise may enter into a sponsoring agreement or contract with a natural or legal person for the promotion of cultural, social, sports, educational, and technological innovation activities, provided that they are proved to be linked to the strengthening of its brand, in compliance, as applicable, with the rules on procurement and contracts of this Law.Id.

286.See Decreto No. 84.128, de 29 de Outubro de 1979, DI´ARIO OFICIALDA UNI˜AO

[D.O.U.] de 30/10.1979, art. 3–4 (Braz.).

287.See Decreto No. 2.923, de 1 de Janeiro de 1999, DI´ARIO OFICIALDA UNI˜AO

[D.O.U.] de 1.1.1999, art. 7 (Braz.).

288.See Decreto No. 6.021, de 22 de Janeiro de 2007, DI´ARIO OFICIALDA UNI˜AO

[D.O.U.] de 22.1.2007, art. 1 (Braz.).

289.See Decreto No. 8.818 de 21 de julho de 2016, DI´ARIO OFICIALDA UNI˜AO [D.O.U.]

de 22.7.2016, art. 40 (Braz.).

Seq: 40

290.See Koichiro Fukui, Japanese National Railways Privatization Study: The Experience of Japan and Lesson for Developing Countries 6, Work Bank Discussion Paper No. 172 (Aug. 31, 1992), http://documents.worldbank.org/curated/en/25622146875235 0809/pdf/multi-page.pdf [https://perma.cc/7296-QDNJ].

291.

If the government continues to hold a portion of the shares even after shares are offered to the public, in addition to the Companies Act, the corporation will still be subject to the special law that specified its permitted activities and governance features prior to the privatization effort. In 2005, Japan removed the company law provisions from the Commercial Code and created a stand-alone Companies Act. Privatizations taking place before 2005 thus involved subjecting the newly created corporations to the Commercial Code. For simplicity, in this narrative we refer only to the Companies Act. 292.YOSHIYUKI KASAI, JAPANESE NATIONAL RAILWAYS: ITS BREAK-UPAND PRIVATIZATION 3

(2001). 293.

Seq: 41

294.Fukui, supra note 290, at 6–7. The Cabinet appointed a governor to chair the board of directors; the governor in turn appointed the remainder of the board with the approval of the Minister of Transportation. The Minister of Transportation appointed the members of the board of audit.Id.

295.MITSUHIDE IMASHIRO & TATSUJIRO ISHIKAWA, THE PRIVATISATIONOF JAPANESE

NATIONAL RAILWAYS: RAILWAY MANAGEMENT, MARKETAND POLICY 2 (1998).

296.East Japan Railway Company History, FUNDING UNIVERSE, http://www.funding [https://

universe.com/company-histories/east-japan-railway-company-history perma.cc/MX9W-M5C3].

297.Fukui, supra note 289, at 65. 298.Technically, the requirements of Japan’s Companies Act are framed in terms of “outside” rather than “independent” directors and auditors, with the former classification being defined in somewhat less demanding fashion than the latter. Some Japanese firms thus describe a specific member of the board as an “independent outside director.” This distinction is not significant for the purposes of our analysis, so we simply use the term “independent.”

Seq: 42

299.

JR East has a board of seventeen directors, three of which are independent, and a five-member board of corporate audit, four of which are independent.EAST JAPAN RAILWAY CO., ANNUAL REPORT57 (2015) (Japan).JR Central has sixteen directors (three independent) and five corporate auditors (three independent).CENTRAL JAPAN RAILWAY CO., ANNUAL REPORT 30 (2015) (Japan). JR West has fourteen directors (five independent) and four auditors (three independent).WEST JAPAN RAILWAY CO., ANNUAL REPORT42 (2015) (Japan).

300.See, Fukui, supra note 289, at 91; KASAI, supra note 292, at 160. 301.Nihondenshin Denwa Kabushikikaishatou Nikansuru Houritsu [Law Concerning Nippon Telegraph and Telephone Corporation, Etc.], Law. No. 87 of 2005, art. 4 (Japan) [Hereinafter Nippon Telegraph and Telephone Corporation].

302.Nippon Telegraph and Telephone Corporation, supra note 301, at 111 (Japan). 303.Yoshiro Takano, Nippon Telegraph and Telephone Privatization Study: Experience of Japan and Lessons for Developing Countries viii (Work Bank, Discussion Paper No. WDP 179, 1992).

304.

JAPAN TOBACCO INC., ANNUAL REPORT 142 (2014) (Japan).

Seq: 43

Id. at 51. Id.

305. 306. 307.Mark A. Levin, Smoke around the Rising Sun: An American Look at Tobacco Regula-

tion in Japan, 8 STAN. L. & POL’Y. REV. 99, 120 (1997).

308.Nippon Telegraph and Telephone Corporation, supra note 301 (Japan); Takano,

supra note 302; JAPAN TOBACCO INC.,supra note 303, at 352 (Japan).

309.Maintaining some level of government ownership was probably a means of dealing with political opposition to privatization of these industries, and a reflection of the perceived public interest of these industries. While the same may have been true of the railway business, JNR was in sufficiently dire economic straits that complete privatization and reorganization along regional lines may have seemed like the only alternative. Hiromi Tamamura, The Actual State and Effect of Privatization in Japan 4–5 (2004) (Japan), http://www.jftc.go.jp/eacpf/03/privatization.pdf [https://perma.cc/RY3UXGJH].

310.Levin, supra note 306, at 100. 311.Thus, in addition to offering standard postal services, post offices in Japan provide customers the opportunity to deposit savings and to purchase various types of insurance policies and annuities.History, JAPAN POST HOLDINGS (Japan), https:// www.japanpost.jp/en/corporate/changes [https://perma.cc/QR34-YPRV].

312.

Seq: 44

313.NOBUYUKI KINOSHITA, THE ECONOMICSOF JAPAN’S POSTAL SERVICES PRIVATIZATION 6

Id. at 5.

(2008). 314. 315.Patricia L. Maclachlan, Two Steps Forward, One Step Back: Japanese Postal Privatization as a Window on Political and Policymaking Change, in POLITICAL CHANGEIN JAPAN: ELECTORAL BEHAVIOR, PARTY REALIGNMENT, AND THE KOIZUMI REFORMS 157, 158 (Steven R. Reed et. al eds., 2009).

Id. at 159–60.

316. 317.See id. at 162–65. 318.Anthony Faiola, Japan Approves Postal Privatization, WASH. POST (Oct. 15, 2005),

http://www.washingtonpost.com/wp-dyn/content/article/2005/10/14/AR200510140 2163.html [https://perma.cc/UL3C-QFVM].

JAPAN POST CORPORATION, ANNUAL REPORT94, 113 (2004) (Japan).

319. 320.See Japan Post’s Murky Privatization, JAPAN TIMES (Oct. 30, 2015), https://www. japantimes.co.jp/opinion/2015/10/30/editorials/japan-posts-murky-privatization/#.W crLxGUYs7c [https://perma.cc/BU5B-K4ZT].

321. 322.See id. 323.Kinoshita, supra note 312, at 17.

Seq: 45

324.Michiyo Nakamoto, Japan Post Prepares for Privatisation, FIN. TIMES (Oct. 26, 2012), https://www.ft.com/content/c2d71a0c-1f23-11e2-b906-00144feabdc0?mh q5j=E6 [https://perma.cc/X5F7-AA3P].

325. 326.PASSAGEOFTHE AMENDMENTTO POSTAL PRIVATIZATION LAW, HOUSEOF COUNCILORS

6 (2012).

327.FED. HOUSING FIN. AGENCY, supra note 121. 328. 329.Atsuko Fukase, Japan Post and Financial Units Soar in Trading Debuts, WALL ST. J. (Nov. 4, 2015, 9:44 AM), http://www.wsj.com/articles/japan—post—and—financial— units—soar—in—trading—debut—1446600123 [https://perma.cc/8MUW-M7PG].

330.See Kinoshita, supra note 312, at 17. 331.See Ronald J. Gilson & Curtis J. Milhaupt, Choice as Regulatory Reform: The Case of Japanese Corporate Governance, 53 AM. J. COMP. L. 343, 344 (2005) (analyzing regulatory reforms in 2002 allowing Japanese companies to adopt a previously unavailable committee structure for the board of directors).

Id. at 352–53. JAPAN POST HOLDINGS CO., LTD., CORPORATE GOVERNANCE REPORT 12 (2017)

JAPAN POST BANK CO., LTD., ANNUAL REPORT18 (2015) (Japan); JAPAN POST INSUR-

ANCE, CO. LTD., ANNUAL REPORT32 (2015) (Japan).

332. 333. (Japan). 334.

Seq: 46

335.See Japan Post’s Murky Privatization, supra note 319. 336.See id. 337.Fukase, supra note 329. 338.See Fukui, supra note 290, at 128. 339.

Id. at 126, 128.

Seq: 47

340.See generally Tan Cheng-Han et. al, State-Owned Enterprises in Singapore: Historical Insights into a Potential Model for Reform, 28 COLUM. J. ASIAN L. 61, 77–91 (2015).

Id. at 79–80.

341. 342.Curtis J. Milhaupt, Chinese Corporate Capitalism in Comparative Context, in THE BEIJING CONSENSUS? HOW CHINA HAS CHANGED WESTERN IDEASOF LAWAND ECONOMIC DEVELOPMENT 275, 285 (Weitseng Chen ed., 2017).

343.FAQs: Why Was Temasek Established, TEMASEK (Sing.), http://www.temasek.

com.sg/abouttemasek/faqs# [https://perma.cc/A5TT-QTGT].

344.

Id. This is a company with no more than twenty shareholders and no corporate shareholders. It is exempt from filing audited financial statements with the public registry. Singapore Company Act (Act 42 of 1967) s 4(1). As discussed below, however, Temasek has made public its Group Financial Summary and portfolio performance since 2004.

345.See id. 346.Grant Kirkpatrick (Consultant for OECD Secretariat), Managing State Assets to Achieve Development Goals: The Case of Singapore and Other Countries in the Region, at 7, OECD Workshop on State-Owned Enterprises in the Development Process (Apr. 4, 2014), https://www.oecd.org/daf/ca/Workshop_SOEsDevelopmentProcess_Singa pore.pdf [https://perma.cc/SQ6X-MQMK].

Id. at 8. Id.

347. 348. 349.Dan W. Puchniak & Luh Luh Lan, Independent Directors in Singapore: Puzzling Compliance Requiring Explanation, in INDEPENDENT DIRECTORSIN ASIA: A HISTORICAL, CONTEXTUALAND COMPARATIVE APPROACH 311, 311–51 (D.W. Puchniak et al. ed., 2017).

Seq: 48

Id. at 14. Id. at 8.

350.Kirkpatrick, supra note 346, at 9. 351. 352. 353.TEMASEK, TEMASEK REVIEW 2017 12 (2017) (Sing.). 354.Puchniak & Lan, supra note 349, at 38–39. 355.Kirkpatrick, supra note 346, at 10. 356.TEMASEK, TEMASEK REVIEW 2006—MANAGINGFOR VALUE 25 (2006) (Sing.). 357.Puchniak & Lan, supra note 349, at 35. 358. 359.Constitution of the Republic of Singapore (1999 reprint) Arts. 22A, 22C. The President is the head of state, who cannot be a member of a political party at the time of his or her election and may not have served in the government for at least three years prior to his/her election.Id. Art. 19.

Id. at 35–36.

360.Puchniak & Lan, supra note 349, at 36.

Seq: 49

(cid:129)Temasek is an investment company. We own and manage our assets

based on commercial principles.

(cid:129)As an active investor, we shape our portfolio by increasing, holding or decreasing our investment holdings. These actions are driven by a set of commercial principles to create and maximize risk-adjusted returns over the long term.

(cid:129)As an engaged shareholder, we promote sound corporate governance in our portfolio companies. This includes the formation of high caliber, experienced and diverse boards.

(cid:129)Our portfolio companies are guided and managed by their respective boards and management; we do not direct their business decisions or operations.

(cid:129)Similarly, our investment, divestment and other business decisions are directed by our Board and management. Neither the President of Singapore nor our shareholder, the Singapore Government, is involved in our business decisions.363

361. 362.Constitution of the Republic of Singapore (1999 reprint) art. 22. 363.The Temasek Charter, TEMASEK REV. 2017(Sing.), http://www.temasekreview.

com.sg/overview/the-temasek-charter.html [https://perma.cc/Y6MP-E7FL].

364.An Active Investor, TEMASEK REV. 2017 (Sing.), http://www.temasekreview.com.sg

/investor/an-active-investor.html [https://perma.cc/4XR2-6QA7].

365.Puchniak & Lan, supra note 349, at 37. 366.

Seq: 50

367.See Linaburg-Maduell Transparency Index, SOVEREIGN WEALTH FUND INST., http:// www.swfinstitute.org/statistics-research/linaburg-maduell-transparency-index [https:// perma.cc/Y8YY-UTVW]. Norway’s sovereign wealth fund GPFG and the Alaska fund also achieved the highest rating.Id.

368.Puchniak & Lan, supra note 349, at 42. 369. 370.Code of Corporate Governance, Guidelines 2.1, 2.2, MONETARY AUTHORITY SING.,

(May 2012) (Sing.).

371.Puchniak & Lan, supra note 349, at 39. 372. 373. 374.Kirkpatrick, supra note 346, at 6. 375. 376. 377.

Id. at 10. Id. Id. at 10–11.

Seq: 51

378.See Puchniak & Lan, supra note 349, at 41. Most of these positions are or were with government bodies not connected to MOF, and are thus “far removed from anything to do with the regulation or governance of Temasek and/or its portfolio.”Id.

379.This is not to say that the Singapore government completely breaks the link between its involvement in commercial enterprise and its broader social policy goals. Rather, its strategy is to run the GLCs on an entirely commercial basis, seeking the highest returns possible, and using those returns to finance its social policies. In this way, Singapore-style state capitalism has a strongly redistributive strand, consistent with the PAP’s roots as a social democratic party.See Chua Beng Huat, State Owned Enterprises, State Capitalism and Social Distribution in Singapore, 29 PAC. REV. 499, 502 (2016).

380.Carlos D. Ram´ırez & Ling Hui Tan, Singapore Inc. Versus the Private Sector: Are

Government-Linked Companies Different?, 51 IMF STAFF PAPERS 510, 513 (2004).

381.Kirkpatrick, supra note 346, at 6.

Seq: 52

382.See Li-Wen Lin & Curtis J. Milhaupt, We Are the (National) Champions: Understanding the Mechanisms of State Capitalism in China, 65 STAN. L. REV. 697, 754 (2013). 383.Curtis J. Milhaupt & Wentong Zheng, Beyond Ownership: State Capitalism and the Chinese Firm, 103 GEO. L.J. 665, 671 (2015) [hereinafter Milhaupt & Zheng, Beyond Ownership].

384.

Jiahua Che & Yingyi Qian, Insecure Property Rights and Government Ownership

of Firms, 113 Q.J. ECON. 467, 467 (1998).

Id. at 154.

385.Yasheng Huang, How Did China Take Off?, 26 J. ECON. PERSP. 147, 152 (2012). 386. 387.Milhaupt & Zheng, Beyond Ownership, supra note 383, at 672. 388. 389.

Seq: 53

390. 391.Lin & Milhaupt, supra note 382, at 707. 392. 393. 394.

Id. at 714–15. Id. at 718. Id. at 717. Chinese Company Law recognizes two basic types of companies: a “company limited by shares” and a limited liability company, although the Chinese limited liability company bears little resemblance to an LLC in the United States. Gongsi Fa [Company Law] (promulgated by the Standing Comm. Nat’l People’s Cong., Dec. 28, 2013, effective Mar. 1, 2014), 2014 STANDING COMM. NAT’L PEOPLE’S CONG. GAZ. 53 (China); JIANGYU WANG, COMPANY LAWIN CHINA: REGULATIONOF BUSINESS ORGANIZATIONS INA SOCIALIST MARKET ECONOMY 53 (2014). Rather, it is modeled after the German corporate entity known as the GmbH. The WSOLLC is a subspecies of limited liability company, regulated by Chapter 2, Section 4 of the Company Law.

395. 396.Lin & Milhaupt, supra note 382, at 734–35. Prior to the creation of SASAC, SOEs were typically controlled by the specific line ministries from which they had been separated in the transition out of a centrally planned economy.Id.

397.SASAC is governed by the Law of the People’s Republic of China on State-Owned Assets of Enterprise.See Qiye Guoyou Zichan Fa [Laws on State-Owned Assets of Enterprise] (promulgated by the Standing Comm. Nat’l People’s Cong., Oct. 28, 2008, effective May 1, 2009) 2008 STANDING COMM. NAT’L PEOPLE’S CONG. GAZ. 636 (China).

398.Lin & Milhaupt, supra note 382, at 736. The number of SOEs under SASAC supervision has been gradually declining as part of government policy to streamline and enhance the profitability of the state sector.

399.

Id. at 735.

Seq: 54

Id. at 737–38. Id. at 738.

400. 401. 402.See Tan et al., supra note 340, at 79. 403.Milhaupt, supra note 342, at 285. 404. 405.Lin & Milhaupt, supra note 382, at 707, 740. 406.Milhaupt, supra note 342, at 289. 407.

Id. at 286.

In a recent example, the Chinese government instructed the SOEs under SASAC supervision to hire soldiers who would be laid off in the restructuring of the military.

Seq: 55

408.Milhaupt, supra note 342. 409.Lin & Milhaupt, supra note 382, at 711. 410.See id. at 700, 717. 411.SeeSEA-JIN CHANG & SANDY YUAN JIN, THE PERFORMANCEOF STATE OWNED ENTERPRISESIN CHINA: AN EMPIRICAL ANALYSISOF OWNERSHIP CONTROL THROUGH SASACS 10, 11, 14, 16 (2016) (Sing.), https://bschool.nus.edu.sg/Portals/0/docs/CGIO/soe-chinaresearch-report-2016.pdf [https://perma.cc/E8EY-MPN2].

412.See Guanyu Guoyouqiye Fazhan Hunhesuoyouzhi Jingji de Yijian [Opinions on Development of Mixed Ownership Economy by State-Owned Enterprises] (promulgated by the State Council, Sept. 23, 2015, effective Sept. 23, 2015) ST. COUNCIL GAZ., Oct. 20, 2015, at 13 (China). Chinese SOEs are usually known publicly by the listed firm. The public filings of the listed firms typically make rather cryptic disclosures relating to the state ownership structures in which they are nested, and almost no disclosure of the role of the CCP in China’s state sector, beyond a simple mention of the title of the party position held by a given executive in his or her bio.

413.BOSTON CONSULTING GROUP & CHINA DEV. RES. FOUNDATION, DEVELOPING MIXED (2014), http://

OWNERSHIP STRUCTURES www.bcg.com.cn/en/files/publications/reports_pdf/BCG_Developing_Mixed_Owner ship_Structures_and_Modern_Enterprise_Systems_Mar2014_ENG.pdf cc/L78T-5ZF2].

AND MODERN ENTERPRISE SYSTEMS 3

[https://perma.

Seq: 56

414.Opinions on Development of Mixed Ownership Economy by State-Owned Enter-

prises effective Sept. 23, 2015, para. 1 (China).

Id. para. 1. Id. para. 2. Id. paras. 9, 11, 12, 14, 15, 17, 18.

415. 416. 417. 418.Curtis J. Milhaupt & Wentong Zheng, Why Mixed Ownership Cannot Fix China’s State Sector, PAULSON INSTITUTE POL’Y MEMORANDUM 2 (Jan. 14, 2016), http://www.paul soninstitute.org/wp-content/uploads/2017/01/PPM_SOE-Ownership_Milhaupt-andZheng_English_R.pdf [hereinafter Milhaupt & Zheng, Why Mixed Ownership] [https:// perma.cc/JHD4-CGWP]. Id. at 4–5. Id. at 5. Id. at 20.

419. 420. 421. 422.Gabriel Wildau, China’s State-Owned Zombie Economy, FIN. TIMES (Feb. 29, 2016), https://www.ft.com/content/218a3710-adf0-11e5-993b-c425a3d2b65a [https:// perma.cc/63JE-QT9Q].

Seq: 57

423.See id. 424.Since the Third Plenum, the Chinese government has pushed through a merger between the largest two state-owned railroad rolling stock manufacturers and a merger between two giant state-owned electricity generating firms. Milhaupt & Zheng, Why Mixed Ownership, supra note 418, at 20. The government is reportedly planning to merge the largest SOEs in even more sectors, including the ship building and petroleum sectors.Id.

425.Lucy Hornby, China Rows Back State-Sector Reforms, FIN. TIMES (June 14, 2016), https://www.ft.com/content/92e52600-31f7-11e6-ad39-3fee5ffe5b5b [https://perma.cc /NPF3-EZWL].

Seq: 58

i t s i

y n a p m o C

o H

h s r e n w O

s r e g a n a

i t a s n e p m o C

h s r e b m

e

d r a o B

/ s n a i c i t i l

y b

s l a i c i f f

’

t v o G

y c i l

h s r e n w O

d e s o

l c s i

l c i l

b u P

—

s s o r

C

d e z i l a r t

n e C

s e l

u R

s n o

i t c i r t s e R

e s

k c o t S

d n

b n o N

d e t s i

C A S A S (

) k e s a

e T (

) s n o

p e c x e

r o n m h

w

(

—

p e c x e 0(

M G

t

u o

l i a b

) s n o

p e c x e

h

w

(

) s i s a b

i a l

p x e

p m o c (

) .

c t e

w e n (

w e n

; r e t t e l

l a u n n a (

w e n (

s e i r o g e t a c (

) s E O S

k e s a

e T (

) r e t r a h C

w a L

t e s s A E O S (

s e i r o g e t a C d n a

) s E O S

s t

h g

i t o V

d e t s i

r e d

o h e r a h S

s t

h g

e

g e R

v o G

p r o C

u s r u p (

c i l

b u p

) t s e r e t

e t a r a p s i

e c n a n r e v o G

e g n a h c x E

s e l

i c n

i r P

e c n a n r e v o G

t r a h C

b u S

l a r e n e G

e t a r o p r o C

s e i t i r u c e S

s

w a L

d n a

b u s (

w e n

;

e t

t a t s

e c n a t r o p m

l a n o

i t a

l a c i t i l

m o n o c E

l i z a r B

b u s (

) e t

t a t s

w o L

S U

e c n a r F

y a w

r o N

m u

d e

n a p a J

e r o p a g n

S

a n

h C

s e Y

= (cid:57)

o N =

a i

b m o

o C

Seq: 59

426.

John C. Coffee, Jr., Racing Towards the Top?: The Impact of Cross-Listings and Stock Market Competition on International Corporate Governance, 102 COLUM. L. REV. 1757, 1757 (2002).

427.See id. at 1830–31. 428.For a Chinese example, seeCURTIS J. MILHAUPT & KATHARINA PISTOR, LAWAND CAPITALISM: WHAT CORPORATE CRISES REVEAL ABOUT LEGAL SYSTEMSAND ECONOMIC DEVELOPMENT AROUNDTHE WORLD 124–28 (2008) (discussing governance failures in China Aviation Oil, an SOE listed on the Singapore Stock Exchange).

429.See Davidoff, supra note 110, at 1734. 430.Pargendler, State Ownership, supra note 21, at 2954.

Seq: 60

431. 432.

Id. at 2938. Id. at 2947.

Seq: 61

(cid:129)The state should act as an informed and active owner; (cid:129)The state should not intervene in management and should respect the inde-

pendence of SOE boards of directors;

(cid:129)The legal and regulatory framework for SOEs should ensure a “level playing

field” for SOEs and private enterprises;

(cid:129)Non-state shareholders should be treated equitably and granted full rights

as investors; and

(cid:129)SOEs should observe high standards of transparency and should adhere to the same disclosure, accounting and compliance standards as private listed companies.

433.See World Bank Group, Toolkit for State-Owned Enterprise, supra note 6; ORG. FOR ECON. CO-OPERATION & DEV. [OECD], Guidelines on Corporate Governance of State-Owned Enterprises (2015) [hereinafter OECD Guidelines].

Id. at 17.

434.See id.; OECD Guidelines, supra note 430, at 11. 435.See id. 436. 437.We note, however, that the most recent annual report filed by Petrobras with the SEC states that the government may use the company to pursue its macroeconomic objectives.Petrobras, supra note 12, at 20. Broad disclosures of this sort probably do not act as much of a constraint on the government.

Seq: 62

Id. at 18. Id.

438.See OECD Guidelines, supra note 433, at 7. 439. 440. 441.OECD, Regulatory Reform in Norway, supra note 136, at 7. 442.See Pargendler, State Ownership, supra note 21, at 2967, 2968–69; Pargendler,

The Unintended Consequences, supra note 264, at 503.

443.OECD Guidelines, supra note 433, at 18.

Seq: 63

Seq: 64

444.See Pargendler, Governing State Capitalism: The Case of Brazil, supra note 267, at 377–78 (“The particular variety of state capitalism prevailing in Brazil today reflects a combination of governmental control of traditional SOEs with the conspicuous exercise of shareholder activism by state-controlled institutional investors.”).

445.Tyler Durden, The Bank of Japan Will Be the Top Shareholder of 55 Companies by the End of 2017, ZEROHEDGE (Aug 14, 2016 6:41 PM), http://www.zerohedge.com/news/ [https:// 2016-08-14/bank-japan-will-be-top-shareholder-55-companies-end-2017 perma.cc/8UA6-DVBF].

446.Lucian A. Bebchuk, Reinier H. Kraakman & George G. Triantis, Stock Pyramids, Cross-Ownership, and Dual Class Equity: The Creation and Agency Costs of Separating Control from Cash Flow Rights, in CONCENTRATED CORPORATE OWNERSHIP 9_C22–23 (Randall Morck ed., 2000).

Seq: 65

Id. at 8.

447. 448.See OECD Review, supra note 200, at 23, 28. 449.See Order 2014-948 of Aug. 20, 2014, supra note 54, art. 4 (Fr.); see also Pargen-

dler, State Ownership, supra note 21, at 2953–54.

450.For instance, Petrobras adopts a dual-class structure that permits the federal government to exert uncontested control over the company despite holding only a minority of its equity capital. Other listed SOEs, such as BB Seguridade, are part of a pyramidal structure.

451.See OECD, Regulatory Reform in Norway, supra note 136, at 23. 452.See Lei No. 13.303, de 30 de junho de 2016 (Braz.). 453.This mechanism is favored by Brazil’s Constitution.CONSTITUI¸C˜AO FEDERAL [C.F.] [CONSTITUTION] art. 173, §1, IV (requiring the special statute on wholly-owned and mixed SOEs to regulate the “constitution and functioning of the board of directors and the board of supervisors, with the participation of minority shareholders”) (emphasis added).

Seq: 66

454.See id. 455.See Mariana Pargendler, How Universal Is the Corporate Form? Reflections on the Dwindling of Corporate Attributes in Brazil 27 (Working Paper, 2017). For instance, one prominent controversy in recent years concerned the renegotiation of the concession contracts between state-owned power company Eletrobras and the federal government, on terms that were allegedly unfavorable to the firm. This decision to approve the renegotiated agreement was submitted directly to the shareholders at their meeting, and approved with the vote of the state as majority shareholder—without any disclosure of the board’s position or recommendation on the transaction. For a discussion of this case, see Mario Engler Pinto Junior, Exerc´ıcio do Controle Acion´ario na Empresa Estatal: Coment´arios a Decis˜ao da CVM no Caso Eletrobr´as [Exercise of Shareholder Control Over State Own Enterprises: Comments About CVM’s Decision in Eletrobr´as Case], FGV Direito SP Research Paper Series, Paper no. 146, 2016 (Braz.).

456.Order 2014-948 of Aug. 20, 2014, art. 5 (Fr.).

Seq: 67

457.Lei No. 13.303, de 30 de junho de 2016, art. 20 (Braz.). 458.Milhaupt & Zheng, Why Mixed Ownership, supra note 418, at 680–81. 459.For a more detailed description of this argument, see Pargendler, Mariana & Salama, Bruno, A Contabilidade Paralela das Empresas Estatais [The Parallel Accounting of State Enterprises] VALOR ECONˆOMICO (May 11, 2016) at 1, 3 (Braz.).

460. 461. 462.

Id. at 2. Id. at 3. Id.

Seq: 68

463.See Black, supra note 109, at 563; Sewell Chan, A First Step on Fannie and Freddie, N.Y. TIMES (Mar. 23, 2010), http://www.nytimes.com/2010/03/24/business/24regulate .html?mcubz=3 [http:s//perma.cc/ND6J-UKDB].

464.OECD Guidelines, supra note 433, at 22, 49. 465.See OECD Review, supra note 200, at 46. 466.See Romero & Etter, supra note 220, at 6. 467. 468.Lei No. 13.303, de 30 de junho de 2016, art. 4 (Braz.). 469.See Lei No. 13.303, de 30 de junho de 2016, art. 20 (Braz.); B3, supra note 11. 470.See Milhaupt & Pargendler, supra note 17.

Id. at 7.

Seq: 69

471.Norway has made use of external oversight by Accounts Tribunals, Parliament, and the stock exchange. The Singapore Stock Exchange (SGX) launched an investigation into a Chinese SOE listed on the SGX, resulting in a $4.4 million settlement with the Monetary Authority of Singapore and criminal sanctions against the CEO and several members of the board.MILHAUPT & PISTOR, supra note 428, at 128. But the targeted SOE was foreign, and the actions of SGX can be viewed as a strategy to bolster Singapore’s international reputation as a rule-of-law market. Brazil’s CVM has imposed an administrative fine on the federal government for casting votes in the shareholders’ meeting to approve a conflicted transaction involving Eletrobras. Processo Administrador Sancionador CVM no. RJ2013/6635, decided on May 26, 2015. However, the decision was ultimately reversed on appeal to the Council of Appeals of the National Financial System (CRSFN), which argued that the government is entitled to guide the SOE toward the public interest, and is therefore not subject to the general regime on related-party transactions. Processo 10372.000246/2016-82, decided on June 28, 2017. For another decision in which the CVM fined the State of S˜ao Paulo as a controlling shareholder for abuse of control power, which is still subject to appeal, see Processo Administrativo Sancionador CVM no. RJ2012/1131.

472.For a discussion of this problem in China, see Zheng Lei, Benjamin L. Liebman & Curtis J. Milhaupt, SOEs and State Governance: How State-Owned Enterprises Influence China’s Legal System, inREGULATING THE VISIBLE HAND? THE INSTITUTIONAL IMPLICATIONS OF CHINESE STATE CAPITALISM 203, 212 (Benjamin L. Liebman & Curtis J. Milhaupt eds., 2016).

473.A loser-pays regime in Brazil illustrates this problem. 474.

Jenny E. Barbosa, Where Arbitration is the Only Game in Town, LEXOLOGY(Aug. 31, 2016), https://www.lexology.com/library/detail.aspx?g=3119df0d-5582-4dd9-867709b0582bddcd [https://perma.cc/L3PJ-BKZ6].

475.Eduardo Dami˜ap Gon¸calves & Flavio Spaccaqueche Barbosa, Arbitration Guide

Brazil,INTERNATIONAL BAR ASSOCIATION 1 (2013).

Seq: 70