Governing State Capitalism: The Case of Brazil
Brazil has become a main exemplar of contemporary pursuit of state capitalism by a Western economy. Major state-owned enterprises (SOEs) have survived the prior wave of privatizations, but they are by no means the only avenue for state influence over corporate governance in Brazil. As scholars have recently highlighted, the state increasingly acts as a minority, rather than majority, investor in Brazil as elsewhere. 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 (SCIIs).
This chapter provides a picture of the various instruments for state influence in Brazilian corporate governance, and examines the role of law in enabling and constraining this evolving variety of state capitalism. Despite their economic significance, the role of legal institutions in the governance of state capitalism has only recently begun to receive scholarly attention. With a few notable exceptions, case studies have focused primarily on transition and Asian economies. The influential OECD Guidelines on Corporate Governance of State-Owned Enterprises are, as their name suggests, prescriptive rather than descriptive in nature.
Our analytical framework posits that state capitalism entails a significant degree of governmental discretion in guiding economic activity. Consequently, the governance of state capitalism calls for a particularly intricate balance. On the one hand, the very choice for government ownership is premised on the benefits of discretionary state action; otherwise, arm’s-length regulation and subsidies could arguably achieve similar results in curing market failures or promoting developmental objectives. On the other hand, excessive state discretion raises the specter of abuse and cronyism, and, if taken to an extreme, may not only undermine performance but also pose a threat to the rule of law itself.
The analysis that follows examines how the legal conformation of state capitalism in Brazil strikes this balance. It begins to approach this theme by outlining the basic legal framework for state involvement in corporate governance. It then shifts gears to examine the relationship between law and state capitalism in action by scrutinizing two high-profile transactions involving state actors.
The Petroquisa case uncovers the rich legal and governance developments following the privatization of the chemical subsidiary of oil giant Petrobras in the 1990s: from the outcome of a shareholder derivative suit claiming billions of dollars in damages for controlling shareholder abuse to Petrobras’s subsequent 49% reinvestment in the private conglomerate to whom it had sold most of Petroquisa’s assets. The CBD dispute, in turn, reveals how Brazil’s development bank came to play a key, though ultimately unsuccessful, role in a control contest between a well-connected local businessman and a foreign investor.
This effort sheds light on the legal underpinnings of the observed (if partial) transition from majority to minority state ownership of enterprise, and the corresponding shift in the mechanisms for the exercise of state discretion in Brazil. First, state capitalism in Brazil is embedded in a particular legal infrastructure that makes it viable. Brazil’s corporate law is particularly amenable to the interests of the state both as controlling and as minority shareholder. In particular, both the strength of shareholder agreements and the board representation rights afforded to minority shareholders under the Corporations Law support the exercise of corporate influence by SCIIs.
Second, the choice between majority and minority government ownership implies a different legal and economic tradeoff in shaping state discretion. Uncontested and explicit government control of enterprise under majority ownership attracts a public law regime that limits the state’s discretionary power. When the state is a minority investor, by contrast, full corporate control over the firm is sacrificed in favor of a private law regime that is less rigid and, therefore, more amenable to discretionary action. The more governmental control of enterprise is uncontested and explicit, the more it is subject to countervailing legal controls – and vice versa.
Finally, as the CBD episode made clear, the law is not the only available check on the exercise of state discretion. In a democratic regime, media coverage and popular pressure can serve as antidotes against at least the most egregious instances of governmental abuse. Still, the promise of democratic accountability hinges critically on the availability of information on state action. So far an unfortunate byproduct of the shift to governmental influence through minority positions has been a corresponding reduction in the amount of information available to the public.
The remainder of this chapter is structured as follows. Part I defines state capitalism in terms of expanding the scope of the government’s discretionary power over the economy, and the role of legal institutions as ensuring the optimal level of discretion. Part II outlines the current legal framework for the state’s direct involvement in corporate governance in Brazil. Part III then turns to the relationship between law and state capitalism in action by examining two major transactions involving government actors. Part IV concludes by drawing policy lessons from Brazil’s recent experience.
I. Governing State Capitalism: the Analytical Framework
State capitalism is a concept that is often used but ill-defined. Understood as a system in which the government plays a key role in promoting economic activity, the term is too encompassing: all capitalist economies would so qualify. There is no such thing as “state-free” capitalism. On the contrary, if there is one chief lesson from the burgeoning literature on neoinstitutional economics, it is precisely that state-provided legal institutions are essential for the proper functioning of capitalist systems.
Existing attempts to circumscribe the concept in view of its effects – as in Ian Bremmer’s definition of state capitalism as “the manipulation of market outcomes for political purposes” – have been equally unsatisfying. Every capitalist economy today is subject to laws aimed at altering market outcomes in view of politically determined ends – from minimum wage and other standard labor regulations to the bail-outs of financial institutions and the award of agricultural subsidies. The pursuit of political objectives is typical of democratic and autocratic regimes alike; it can hardly serve as the distinguishing trait of state capitalism vis-à-vis other systems.
At the other end of the spectrum, state capitalism is conflated with substantial government ownership of enterprise. Such a view has a long historical pedigree, dating back to the origins of the term. This conception may at times prove to be deceptively narrow, however, for the state can exercise discretionary influence over economic activity and corporate governance through mechanisms other than outright ownership of a controlling stake in the firm.
A more illuminating criterion to discern state capitalism from other “varieties of capitalism” is the relative scope of state discretion in shaping economic activity. Although the degree and contours of market interventions vary, in both liberal and coordinated market economies the state’s actions are, as a general matter, explicitly grounded in, and constrained by, law. For instance, the state might outlaw insider trading and cartels (as it does in the United States, but historically not in Germany) or impose employee board representation and tenure requirements (as it does in Germany, but not in the United States). The government may also extensively promote industrial policy and other public objectives through the tax system, as is the case in the United States. These interventions, however, are clearly enunciated ex ante and justiciable ex post. Instances of state ownership and the related exercise of state discretion certainly exist in all capitalist systems, but, viewed as a continuum, are comparatively circumscribed.
By contrast, we argue that the distinctive trait of state capitalism is the relatively unfettered exercise of discretion in the pursuit of political objectives. From a legal perspective, such expanded discretion is at least partially achieved through substantial (and appropriately structured) levels of state ownership. Ownership is, indeed, the ultimate legal and economic institution for the exercise of discretion. Thus, on face value at least, state ownership of enterprise achieves a formidable feat: it apparently reconciles the ample scope for state discretion with the central tenet of the rule of law – according to which state power is necessarily based on and limited by law. If this legal loophole is used too broadly, however, the standard conception of the rule of law risks becoming illusory.
We have known at least since Coase that the degree of discretion afforded by ownership (or, in Williamson’s terms, “hierarchy”) can offer distinct advantages. By simply using its voice or fiat power within the firm’s hierarchy, the state can economize on the transactions costs of writing and enforcing formal contracts or laws. Now canonical lessons from organizational economics suggest that state ownership is particularly appropriate when there is a market failure that cannot be cured by laws and regulations alone, be it because contract institutions are weak (hampering regulatory enforcement) or because non-contractible national security or sovereignty considerations are involved.
The allotment of ample discretionary power to the state evidently is not free from inconveniences. To put it mildly, the overly broad exercise of discretion by the state as enterprise owner is in tension with the rule of law. Discretion can unsettle both predictability and equality – two fundamental attributes of the rule of law that have profound economic implications. The potential instability of governmental action, coupled with the specter of unequal treatment dispensed to different parties, discourages long-term investment and undermines a level playing field in which individuals and firms can compete. In other words: discretion may bring about perils ranging from excessive uncertainty to cronyism.
II. The Instruments of State Influence in Brazilian Corporate Governance
State involvement in Brazilian corporate governance takes place through ownership of a majority or a minority of a firm’s voting capital, as well as through the extension of long-term loans subject to conditionalities. Majority investments comprise the prototypical state-owned enterprises, with the government exercising uncontested corporate control over the firm by holding more than 50% of its voting stock. Minority state investments, in turn, can be channeled through a constellation of SCIIs, such as the development bank, SOE pension funds, and other public investment vehicles.
A. The State as Majority Shareholder
SOEs are a key instrument of state capitalism in Brazil. Created by law, they are technically termed mixed enterprises (sociedades de economia mista), when the state holds the majority of the voting stock, or public enterprises (empresas públicas), when the state is the sole shareholder. Although the absolute number of SOEs declined sharply following the wave of privatizations in the 1990s, their economic clout remains significant.
Listed SOEs currently include large firms such as Petrobras, one of the world’s chief oil companies, and Banco do Brasil, Latin America’s largest bank by assets. Taken together, listed SOEs accounted for nearly one-third of Brazil’s market capitalization in 2007, even if down from a staggering 70% in the 1970s. Wholly-owned public enterprises – which include the likes of Brazil’s Development Bank (Banco Nacional de Desenvolvimento Econômico e Social – BNDES), and CEF (Caixa Econômica Federal), a state-owned commercial bank – are also major economic players. At 6% of GDP, BNDES’s outstanding loan volume in 2013 was two times larger than that of the World Bank.
In Brazil, as elsewhere, a main driving force behind the establishment of government-controlled corporations was the greater degree of discretion afforded by a private law regime, thus increasing much-needed flexibility and expediency in government action. In 1933 President Franklin Delano Roosevelt exalted the creation of the Tennessee Valley Authority as “a corporation clothed with the power of government but possessed of the flexibility and initiative of a private enterprise.” The same goal of bestowing the government with the use of the corporate form to achieve “zero bureaucracy, as if it were a private company,” was also an explicit rationale behind the rise of government corporations in Brazil. Nonetheless, discretion invariably creates potential for abuse, which in turn triggers countervailing legal responses.
Brazil’s evolving legal treatment of contracting practices and employee recruitment at SOEs illustrate this evolution. Under a largely discretionary regime for the most part of the twentieth century, Brazilian SOEs experienced excessive politicization of employment practices, with job positions often filled based on favoritism and pork, rather than on firm needs and labor capabilities. This outcome led to a strong reaction: Brazil’s 1988 constitution subjects SOEs’ hiring to public examinations, the same mode of recruitment applicable to public servants. Public examination procedures are hardly the best method to select prime talent in employee hires (as hinted by their virtual absence from competitive private sector recruitment), but have the merit of ensuring equal opportunity and reducing favoritism.
For similar reasons, government contracting in Brazil is subject to highly strict and formalistic public procurement requirements, which in theory reduce the possibilities for corruption at the cost of expediency. State-controlled corporations created by law (sociedades de economia mista) are only partially subject to such restrictions, which apply only to ancillary agreements that are not central to their line of business. For contracts that are inextricably linked to their economic purpose – e.g., the extension of loans by a public bank – the state as owner operates with a degree of discretion akin to that of a private actor. Hence, although generally subject to public procedures for recruitment and contracting at large, BNDES operates like a private bank in its ability to choose whom, when, and under what terms it shall offer its (highly subsidized) lines of credit – which makes it arguably the most paradigmatic instrument of the Brazilian-style of state capitalism.
In order to appropriately circumscribe the state’s discretionary power as a controlling shareholder, the legal regime currently applicable to SOEs is at least partially enshrined in Brazil’s constitution. First, in view of Brazil’s embrace of a market economy based on “free enterprise”, SOEs face limitations as to scope. Art. 173 specifically restricts the direct undertaking of economic activity by the state to contexts in which it is “necessary” for “national security” or for “relevant collective interest” – even if the latter concept is arguably too vague to effectively restrain government expansion.
Second, the Constitution outlines the key features of SOE governance. Each SOE must be created by a statute that specifically defines its corporate purpose. As previously mentioned, SOEs are generally subject to public procurement requirements for contracting, as well as to formal public contests for employee recruiting. SOE operations are also exposed to the external oversight of Tribunais de Contas, administrative bodies constitutionally in charge of external control of government activities and expenditures. Finally, in order to level the playing field vis-à-vis their private competitors, SOEs cannot enjoy fiscal privileges that are not extensive to private sector firms, and must be governed by the same legal regime applicable to private companies with respect to civil, commercial, labor, and tax obligations.
As a general rule, the same body of corporate law applies to SOEs and private firms alike. Only a handful of special rules and exceptions apply – some of which designed to protect minority shareholders while others to ensure state prerogatives. An instance of the former is the rule granting appraisal rights to minority shareholders whenever the government takes full control of an existing private firm – for the assumption is that private shareholders had not consented to participate in a company that may pursue public objectives. Conversely, the main distinctive feature of the legal regime of a mixed enterprise resides in the rule provided by art. 238 of the Corporations Law: while making clear that the government is subject to the same duties and responsibilities applicable to a private controlling shareholder, it explicitly authorizes the state to “direct the activities of the company so as to attend to the public interest that justified its creation.”
Yet corporate law as a whole has provided only feeble constraints to the behavior of SOEs. First, corporate law typically grants managers substantial discretion in decisions that are untainted by conflicts of interest, by force of the “business judgment rule” doctrine of U.S. corporate law and its functional equivalent in Brazil. Second, as I have argued elsewhere, the state’s dual role as corporate governance player and referee means that the government is uniquely positioned to shape the applicable legal regime with its interests as shareholder in mind.
As vividly demonstrated by the narrative of Petroquisa’s privatization, this can happen in at least three ways: (i) the original corporate law regime is made particularly amenable to the interests of the state as shareholder; (ii) the state sponsors legal reforms to change the applicable rules when the latter prove to be inconvenient; and (iii) courts and regulatory agencies are uniquely reluctant to enforce legal restrictions against state actors. All of this, in turn, contributes to increase the degree of discretion enjoyed by the state as controlling shareholder above and beyond that afforded by ownership alone to private actors.
B. The State as Minority Shareholder
SOEs – be they public or mixed enterprises – can and do serve as minority shareholders in a number of corporations in Brazil. As described in Part III.A below, Petrobras is a large shareholder in major chemical firm Braskem. Particularly significant are the equity stakes of BNDESPAR, a wholly-owned subsidiary of BNDES that operates as its equity arm. With equity stakes in 74 closely-held firms and 71 publicly traded corporations (which account for nearly 60% of Brazil’s total market capitalization), BNDESPAR plays a central role in supporting Brazil’s national champions. Because these investments involve formally minority positions – narrowly defined as representing less than 50% of a firm’s voting capital –, the recipient firms continue to be de facto qualified as wholly private companies.
b. Public pension funds
Any thorough account of the instruments of state influence over Brazilian corporate governance must encompass the role of public pension funds. In this respect, one caveat is necessary: unlike the other actors discussed here, public pension funds are not legally owned by the government. Consistent with international practice, public pension funds in Brazil are organized as separate legal entities that shall be managed for the benefit of SOE retirees.
Nevertheless, the governance structure of SOE pension funds is particularly conducive to political interference. By law, the SOEs shall appoint half of board members, as well the board president of SOE pension funds, with the remaining directors being elected by the fund’s beneficiaries. As a result, the CEOs of SOE pension funds are typically appointed by the CEO of the respective SOEs, whom, in turn, are directly or indirectly appointed by the country’s president. The upshot is that political interference in pension fund investment and governance decisions is not only theoretically feasible but also commonplace in practice.
The well-known role of Previ (together with BNDES) in ousting the chief executive of mining giant Vale do Rio Doce, a privatized firm, illustrates the use of pension funds to further the government’s economic agenda. Despite Vale’s stellar financial performance, its CEO Roger Agnelli attracted the ire of President Lula for what he viewed as unnecessary employee firings and underinvestment in the country. Through the use of “minority” holdings by state actors (and especially Previ’s holding of nearly 15% of Vale’s equity), Agnelli was replaced by an executive presumably more sympathetic to the government’s policies priorities.
Indeed, pension funds of SOEs play a particularly dominant role in Brazil’s equity markets. Previ, the pension fund of Banco do Brasil’s employees, is the largest such fund in Brazil and holds equity stakes in 43 publicly traded companies, which together account for over one-half of Bovespa’s market capitalization. In 2006, Previ alone held the equivalent of 5% of Brazil’s market capitalization. Taken together, public pension funds in Brazil account for 64.6% of the country’s pension fund industry as of 2013.
The corporate governance clout of public pension funds is significantly magnified by their peculiar investment strategy. Private pension funds in Brazil have typically allocated less than 30% of their investment portfolio in corporate stocks, given the historically high interest rates paid on sovereign debt. Previ, by contrast, apportions over 60% of its portfolio to equity investments “in the main Brazilian corporate groups.” Counter to international practice, its equity positions are made of very large stakes – with a median of 6.79 % and an average of 15.87% of the equity capital in only 45 firms in strategic sectors, such as energy, banking, tourism, and telecom.
These sizable stakes translate into significant governance clout. Every year, Previ alone elects over 100 members of boards of directors and supervisory board (Conselho Fiscal). Previ is also known for promoting an activist agenda. The fund’s latest priority has been to defend that Brazilian firms, whenever possible, adopt a “global” strategy so as to improve Brazil’s balance of payments. Such an agenda, needless to say, has more to do with the current government’s goal of fostering national champions than with the pursuit of shareholder value.
c. New investment vehicles
In recent years, Brazil has established two new SCIIs that have reinforced the government’s participation in corporate governance. The FI-FGTS, an investment fund of FGTS, was created in 2007 to finance investments in energy and infrastructure. It is funded by the preexisting FGTS, a fund instituted in 1966 to compensate for the elimination of stability in employment contracts. Financed by employers’ mandatory monthly contributions of 8% of wages, FGTS provides workers with individual accounts whose amounts can only be withdrawn in specific circumstances dictated by law, such as involuntary termination of employment, major illness, retirement, or housing acquisition. By law, FGTS’s resources are to be channeled to housing and urban infrastructure; workers’ accounts, in turn, are remunerated according to a predetermined rate – which, in recent years, has been significantly lower than actual inflation.
With net assets at approximately US$13 billion in 2013, FI-FGTS thus employs FGTS excess resources in debt and equity investments in a variety of infrastructure and energy companies. Such investments include bond holdings in the former port-operations subsidiary of LLX, a publicly-traded company controlled by (recently disgraced) Brazilian magnate Eike Batista, and a 30% equity stake in Odebrecht TransPort S.A., a transportation firm of the powerful Odebrecht group. Odebrecht TransPort S.A. – which was a chief player in the recent “privatization” of Brazilian airports during the Dilma administration – alone accounts for 7.6% of the FI-FGTS’s assets, and also boasts BNDESPAR in its ranks of shareholders.
In 2008, following the lead of other resource-rich countries, Brazil created its sovereign wealth fund (Fundo Soberano do Brasil – FSB). The purpose of the fund is to “promote the investment in assets in Brazil and abroad, constitute public savings, mitigate the effects of economic cycles, and further strategic projects located abroad.” The fund has since invested in Brazilian government debt as well as in stocks of SOEs, such as Petrobras and Banco do Brasil. At first, the state’s minority investments in SOEs might be puzzling. It would seem as though the government has little to gain in terms of additional influence by accumulating stakes through SCIIs in firms that are already under state control, but this is not always the case. Until loud protests by minority shareholders eventually led the government to reverse course, SCIIs had a decisive voice in electing the representatives of minority shareholders in the board of directors of Petrobras, thereby ensuring an even greater degree of dominance by the state in the company’s governance.
d. The legal regime of minority state investments
As hinted by the previous section, the fact that state actors hold minority (rather than majority) positions in the firm does not mean that their governance influence is muted. In fact, the reverse is true. Brazilian corporate law confers significant rights on minority shareholders holding a large stake in the firm. For instance, the Corporations Law allows holders of at least 15% of the company’s voting stock or 10% of the company’s nonvoting preferred stock to appoint one board member each. Board representation of minority shareholders can be even greater, however, since the statute also contains a mandatory rule providing for cumulative voting upon request. Moreover, a closely-held corporation’s charter may grant special board appointment rights to holders of certain classes of preferred stock – a provision that, according to the official “Exposition of Motives” of the corporate statute, was included in the statute to endorse a practice then habitually employed by BNDES.
Still, focusing on the mandatory provisions of the corporate statute alone would greatly understate the clout of SOEs as minority shareholders. The main reason for this is the widespread use of shareholder agreements – a mechanism that has become a staple of modern Brazilian corporate governance. A 2001 legal reform gave the enforcement of shareholder agreements a boost: it explicitly recognizes the availability of specific performance as a remedy for their breach, and prevents votes cast in violation of duly filed agreements from being counted in shareholder as well as in board meetings (in the latter case, despite criticism by corporate governance advocates, who claim that such a mechanism compromises the exercise of directors’ independent business judgment).
SCIIs have also made ample use of shareholder agreements to influence corporate strategy and governance. BNDESPAR alone is a party to 20 shareholder agreements with publicly-traded companies, and SOE pension funds are party to several others. Although their clauses vary from case to case, it is common for shareholder agreements to have a voting component, according to which its parties (including government-controlled actors) have director appointment rights as well as veto rights over major corporate decisions.
But even though the state can have substantial voice and influence over corporate governance through what are formally “minority” holdings, outside investors enjoy little formal protection from political interference. First, although appraisal rights apply in view of a government taking of corporate control, this remedy is unavailable if state actors become minority investors through market purchases. Second, the fiduciary duties of controlling shareholders and managers under Brazilian law are sufficiently broad so as to encompass nonshareholder constituencies.
Controlling shareholders may breach their fiduciary duties when they direct the company to pursue a goal that is foreign to its purpose to the detriment of minority shareholders or even the “national economy.” According to the Corporations Law, managers shall discharge their duties in view of the company’s purpose, but with due regard to the “requirements of public good and of the social function of enterprise.” Consequently, the government may be able to exercise nontrivial influence through its minority stakes – and, with respect to closely-held corporations, with very little transparency to the public.
The main constraint on governmental interference through minority shareholdings is of an economic rather than legal nature. Because the government is unable to control the firm on its own, it must necessarily garner support from private investors. Some private shareholders specifically negotiate for contractual protection against value-destroying political interference. Their shareholder agreements with SCIIs overtly specify the firm’s purpose of maximizing the financial returns to investors, and mandate that shareholders exercise their voting rights accordingly. But even in the majority of cases where such a provision is absent, the lack of voting control requires the state-controlled institutional investors to broker deals with private investors, who are presumably more interested in maximizing shareholder value than in furthering the government’s agenda.
III. State Capitalism in Action
Any overview of ownership structures and the “law on the books” applicable to both SOEs and SCIIs would mean little if it did not reflect the operation of state capitalism in action. The discussion that follows offers a glimpse into the relationship between state ownership, the legal infrastructure, and corporate governance through the analysis of two salient cases: one involving Petroquisa, once a subsidiary of Petrobras, and another featuring CBD, a private firm in the supermarket and retail industry.
A. Petroquisa: The Shifting Modes of Governmental Influence
Perhaps the most prominent lawsuit involving a state-controlled enterprise in Brazil concerned the alleged abuse in the asset sales by Petrobras Química S.A. – Petroquisa, as part of the country’s large-scale privatization program. The controversy drew significant attention, not only because derivative actions are rare in Brazil, but also due to the sheer sums involved. When the Superior Court of Justice (Superior Tribunal de Justiça – STJ) issued a final decision on the case in 2012, the plaintiff’s damages claim amounted to roughly US$12 billion after compound interest and inflation adjustment. Nonetheless, as will become apparent, the case achieved even greater notoriety for the precise content of the decision – and the lessons it offered about the relationship about law and state capitalism in Brazil.
Petroquisa was established in 1967 as the first subsidiary of Petrobras, Brazil’s state-owned giant. Petroquisa was in charge of Petrobras’s major chemical business, ending up holding substantial stakes in 34 companies and to account for roughly one-third of Brazil’s chemical industry. In 1989, Petroquisa went public, raising a total of US$40 million from private minority investors. Following the IPO, Petroquisa remained a subsidiary of Petrobras, as part of a pyramidal structure that is frequently used by Brazilian SOEs.
Only two years after the offering, however, the government decided to privatize Petroquisa’s subsidiaries. Law 8,031 of 1990 (the “Privatizations Law”) authorized the government to receive junk government bonds (moedas podres) in consideration for the sale of privatized assets – a provision that allegedly served the government’s interest in reducing Brazil’s mounting public debt. Moedas podres were so dubbed because they traded at a substantial discount due to illiquidity, but, under the new law, the government would accept them as payment at face value. Petrobras, as the controlling shareholder of Petroquisa, approved the sale of almost all of Petroquisa’s holdings in exchange for only US$ 1,000 in cash and over US$940 million in moedas podres, culminating in the “virtual liquidation” of the company.
Anticipating complaints by minority shareholders, Petroquisa initially contemplated the possibility of repurchasing its minority shares and delisting before implementing its asset sales. The plan was, however, abandoned, based on the conclusion that such a move would only increase the legal risks by transferring the controversy to Petrobras and its minority shareholders. Following the asset sales, minority shareholders complained of expropriation, but Petroquisa was unwilling to settle the dispute by agreeing to pay back the amounts it raised during its IPO.
Porto Seguro Imóveis Ltda., a minority shareholder in Petroquisa, filed suit. It claimed that Petrobras, as the controlling shareholder of Petroquisa, acted in abusive fashion in approving the sale, which privileged the interests of the federal government over those of the company and its minority shareholders. Both the lower court and the state court of appeals agreed with plaintiff’s argument, holding Petrobras liable to Petroquisa for damages in the original amount of US$ 2.37 billion. Throughout the lawsuit, Petroquisa’s lawyers emphasized the importance of the case for the credibility of Brazil’s institutional environment.
In 2012, however, no less than 20 years after the claim was filed, the dispute took a surprising turn. The Superior Court of Justice sitting in Brasilia, which is Brazil’s court of last resort for the interpretation of federal legislation, reversed the prior decisions and dismissed the lawsuit for lack of standing. The Court’s decision rested primarily on the argument that a subsequent merger between Petroquisa and Petrobras in 2011 led to the identity – or, in technical terms, the “confusion” (confusão) – between debtor (Petroquisa) and creditor (Petrobras), which, under the Brazilian Civil Code, conduced to the extinction of any outstanding obligations between the parties.
Albeit based on an apparent technicality, such a legal conclusion bears profound implications for Brazilian corporate law applicable to both private and public firms. If broadly applied, it permits a corporate controlling shareholder to erect an absolute shield from liability for abuse of minority shareholders of a controlled company by later promoting a merger between both entities. The outcome of the Petroquisa lawsuit thus offers a textbook illustration of a case driven by the interests of the state as controlling shareholder imposing negative externalities on the legal regime applicable to private firms as well, hence decreasing the protection afforded to minority shareholders as a whole.
Curiously, even though the procedural argument alone was sufficient to dismiss the suit, the Court’s sympathies to the interests of the state as shareholder led it to further elaborate on the merits of the case. Confirming the general unwillingness to use corporate law to constrain state action, it affirmed the absence of abuse, given what it viewed as the state’s right to change the rules of the game through subsequent statutes. It argued that, in view of the Privatizations Law, the state’s representatives in Petrobras were under a legal duty to comply with its provisions even if they conflicted with the Corporations Law. The Court concluded that, in light of SOEs’ “strategic role for the state,” the government is entitled “for strategic reasons and with legal basis to adopt decisions that are quite different from those that would be made by a private shareholder.”
Following the Petroquisa decision, minority shareholders of both private firms and SOEs had greater reasons to worry: the former, in view of the use of a subsequent merger as an absolute defense against a derivative suit for abuse by a controlling shareholder; the latter, as the court made clear that special statutory changes could do away with the protections apparently afforded to SOE shareholders by the corporate statute. Apart from the lawsuit, however, the developments following Petroquisa’s privatization are emblematic of the changing contours of state capitalism in Brazil.
In the privatization auctions, Odebrecht, a large diversified conglomerate that is one of Brazil’s major business groups, emerged as the acquiror of the lion’s share of Petroquisa’s subsidiaries. Yet such a change in ownership structure did not entail a retreat of state support to the chemical industry. As one member of the House of Representatives (Câmara dos Deputados) put it in 2001, Brazil had transitioned from a “state monopoly to a private firm oligopoly” generously funded by BNDES.
Both state involvement and sector concentration would intensify in subsequent years. Ironically, after having sold most of its chemical business to Odebrecht in exchange for arguably too low a price paid in junk government bonds in the 1990s, Petroquisa held 11% of Braskem (the vehicle of Odebrecht’s chemical business) when the company went public in 2002. At the time, a shareholders agreement was in place granting Petroquisa significant veto rights. Following Petroquisa’s merger with Petrobras in 2011, the latter became a direct shareholder holding a substantial stake in Braskem.
Braskem’s current ownership and governance structure is illustrative of the new modes of government involvement in business corporations in Brazil. Braskem is a publicly traded firm whose shares are listed on the São Paulo Stock Exchange (BM&F Bovespa) and whose ADRs are traded on the New York Stock Exchange. As of mid 2013, Odebrecht and Petrobras held 50.1% and 47% of Braskem’s voting capital and 38.3% and 36.2% of its total capital, respectively. BNDESPAR held and additional 12.8% of its preferred non-voting stock. Finally, BNDES is a major creditor of the firm, which in 2012 had over US$ 1 billion of debt outstanding at the usual subsidized rates.
Petrobras, a key supplier of Braskem, and Odebrecht are party to a shareholders agreement, which gives Odebrecht the upper hand: it appoints the majority of the board, the CEO, and is solely responsible for approving the business plan. Petrobras’s rights under the agreements are far from negligible, however. It appoints four (out of 11) members of Braskem’s board of directors and its vice-chairman, as well as two members and the chairman of the supervisory board. The agreement also grants Petrobras veto rights with respect to mergers, spin-offs, and large investment and asset sales.
Even more illuminating of the current style of state capitalism in Brazil are the “guidelines for the exercise of voting rights” in Braskem by Odebrecht and Petrobras, which strive to reconcile the private shareholders’ special emphasis on profitability with the government’s promotion of national champions. Under the agreement, Odebrecht and Petrobras “undertake to exercise their voting rights in the Companies so that Braskem and Braskem’s Controlled Companies have a professional, efficient and productive management, preserving and incrementing their profitability, in order to maximize their shareholders’ compensation.” The parties also agree to adopt a dividend policy that “seeks to maximize the distribution of profits” (subject to the firm’s financial health), and to only pursue new projects that are “explicitly profitable” in view of the weighted average cost of Braskem capital. Yet the voting guidelines also expose Braskem’s status as a national champion by mandating the “permanent pursuit of Braskem’s growth in Brazil and abroad.”
Nor is Petrobras the only state actor involved in Braskem’s governance. According to its securities filings, the company in effect has two other relevant corporate agreements in place: one with BNDESPAR and yet another with SOE pension funds Previ and Petros. Although BNDESPAR holds only nonvoting preferred shares, its shareholders agreement with Odebrecht grants it the right to appoint a board member for as long as it holds at least 5% of Braskem’s total capital. And under a memorandum of understanding with Previ and Petros, the latter enjoy special tag-along rights in control sales.
If government involvement in the chemical industry persists, the applicable legal framework has changed dramatically due to the different choice of ownership and governance structures. Prior to privatization, Petroquisa was a mixed enterprise under the uncontested direct voting control of Petrobras and, indirectly, of the federal government. Such control rights were accompanied by important legal checks and mechanisms of accountability aimed at mitigating abuses of state discretion. As a mixed enterprise, Petroquisa was legally required to hire employees through public examinations and to choose suppliers through formal public procurement procedures. Its operations also fell under the purview of Tribunais de Contas.
Because the government’s major support and influence takes the form of “minority” stakes, Braskem remains formally a private venture subject to a wholly private legal regime. If, on the one hand, the government relinquishes discretion by sharing control with private business groups, on the other hand it increases the scope for discretionary action by evading the restrictions imposed by public law. Functional discretion thus persists – with the public-private alliance arguably leading not only to potentially greater efficiency, but also to cronyism and opacity. For instance, escaping supervision by Tribunal de Contas was a key argument in the proposal for Braskem to takeover the major government project for the construction of an integrated refinery and petrochemical complex.
B. Control Contest for CBD: The State as Supporting Actor
State influence over corporate governance in Brazil extends far beyond SOEs and firms in which SCIIs hold significant equity stakes. The control contest over Companhia Brasileira de Distribuição (CBD) between businessman Abílio Diniz and French group Casino Guichard-Perrachon shows that the state can make an influential appearance in unexpected contexts. The episode also reveals that, under special conditions, appropriately pitched media campaigns can operate as a potent extralegal check on state discretion.
CBD is a large Brazil consumer retail company that, among other things, operates the popular supermarket chain Pão de Açucar. Founded as a local bakery in 1948 by a Portuguese immigrant, Pão de Açucar grew into a large business through a combination of organic expansion and multiple acquisitions. In the mid-1990s, its founder’s eldest son, Abilio Diniz, assumed control of the business and took the company public in both São Paulo and New York, raising hundreds of millions of dollars.
By the late 1990s, however, CBD faced high levels of indebtedness, and was on the lookout for new investors. It eventually agreed to join forces with France’s leading retailer Casino Guichard-Perrachon SA, which became a large minority shareholder in the company in 1999. In 2005, Casino increased its voting stake in CBD to 50% and executed a shareholders agreement with Diniz family members to regulate their joint control over the firm. The agreement granted Casino an option to take over control of CBD in 2012. Brazil’s Securities Commission (Comissão de Valores Mobiliários – CVM) understood that Casino had effectively paid for the company’s control as early as 2005, and mandated a public bid for minority shares, as provided under Brazilian corporate law.
As the 2012 deadline approached, Diniz grew increasingly uneasy about relinquishing control of CBD. After several negotiation attempts aimed at extending his tenure as CBD board chair failed, he orchestrated a bold plan: a merger between CBD and the Brazilian division of Carrefour, a French supermarket chain with a significant presence in Brazil and Casino’s chief rival at home. Under the proposed transaction structure, Diniz would retain control of the combined entity. Controversially, though, Diniz’s stratagem was made feasible through significant state support: the transaction proposal advanced by BTG Pactual, an investment bank, was premised on an equity investment of over US$ 2 billion by BNDESPAR.
BNDESPAR’s conspicuous role in the planned merger between CBD and Carrefour drew substantial scrutiny. The prospect of consolidation between two of Brazil’s largest supermarket chains enraged consumers and sent off red flags from a competition law standpoint. Observers soon began to raise questions about the rationale for BNDESPAR’s support to the transaction, whose developmental benefits were far from evident.
Unsurprisingly, Diniz praised BNDES’s effort of preventing the sector from being “totally denationalized,” which he characterized as a “service to society.” Development Minister Fernando Pimentel underscored the transaction’s “strategic importance” for Brazil, as it would “for the first time, through a large international retail chain, place Brazil’s industrialized products abroad.” Nonetheless, it was the statement by Civil Cabinet head Gleisi Hoffman that most irked commentators. Even though BNDESPAR is a wholly-owned subsidiary of BNDES – which is, in turn, wholly owned by the federal government –, she claimed that BNDESPAR’s equity investment “was a market action” that “had nothing to do with the government’s decision,” since “there were no public funds involved.”
In response to these developments, Casino launched an aggressive communications campaign accusing Diniz of devising a value-destroying subterfuge in order to “frustrate the provisions of the shareholder agreements that regulates CBD and, indirectly, expropriate Casino of its control rights acquired and paid for in the year 2005,” which it described as disrespectful of “the rule of law and of the respect for private property that characterize Brazil.” Casino’s attorney (and former Minister of Justice under a previous administration) qualified the developments as a “corporate coup,” for “they made a secret agreement and they are using BNDES to pressure the French.” Jean-Charles Naori, Casino’s chief and controlling shareholder, then flew to Brazil and met with BNDES managers to discuss the matter. As he subsequently reported in an interview with a major local magazine, he asked the Bank not to “compromise its credibility by tolerating a fraud and an expropriation.” His not-so-implicit message was that a government-owned bank was assisting a Brazilian oligarch in repudiating a valid agreement with a foreign investor.
In face of the ensuing public backlash – and despite several prior statements denying political interference in BNDES’s operational decisions – President Dilma’s response was to explicitly condition support by the public Bank on Casino’s consent to the transaction. Such a decision is best understood from a political rather than a purely legal or technical standpoint. If the deal were indeed beneficial to the company and the national economy, as its proponents claimed, any existing contractual arrangements between the controlling shareholders would have to give way. Brazilian corporate law explicitly subordinates the effects of shareholder agreements to the fiduciary duties of controlling shareholders.
Casino never agreed to the proposed merger with Carrefour, which it regarded as flawed for multiple strategic and financial reasons. The deal thus fell apart, culminating in various arbitral claims and counterclaims under the shareholder agreement before the International Chamber of Commerce. Casino alleged that Diniz breached the shareholders agreement by procuring Carrefour’s bid without its prior consent, while Diniz argued that Casino’s disloyally thwarted a transaction that was beneficial for the company due to its conflict of interest as a competitor of Carrefour and future controlling shareholder of CBD.
The incident is instructive of the mechanics of government intervention in corporate governance in Brazil. First, the state as incarnated in BNDESPAR could play a pivotal role in a control contest of an entirely private firm by offering to finance a transaction that could not be exclusively funded by private lenders – as evidenced by the deal’s instant demise after the withdrawal of BNDESPAR’s support. Second, although the existing legal framework confers ample discretion to BNDESPAR in its operational decisions, its choices are not immune to political and democratic controls. Given the suspicion with which the public received the proposal’s announcement, a direct message from Brazil’s president rapidly derailed the merger.
Casino had won the battle but not yet the war. Even though it remained unclear for a while if further obstacles would emerge to the control transfer scheduled for June 2012, the agreement was eventually performed as planned, and Casino took over the group. Yet, the partner-turned-foes continued to share the company’s management despite their ongoing legal disputes. According to the terms of the same shareholders agreement, Diniz would continue to chair CBD’s board of directors for as long as his physical and mental condition permitted and the company kept a good track record in performance.
An additional ground for disagreement – also influenced by state actors – would appear in early 2013 when news surfaced that Diniz had been appointed board chair at processed meat firm Brasil Foods (BRF), another national champion fueled by government support. The company resulted from a 2009 merger between processed meat competitors Perdigão and Sadia, after the latter experienced outsized financial losses when its exchange rate derivative bets turned sour during the 2008 financial crisis. BRF was widely identified as a national champion following the merger. It enjoyed the support of various pension funds that had significant stakes in Perdigão, as well as generous funding from BNDES. Public pension funds Previ and Petros each account for over 12% of BRF’s capital.
Previ, the public pension giant, and Tarpon, a private investment fund, took credit for Diniz’s nomination, which failed to receive unanimous approval. SOE pension fund Petros was among the dissidents, arguing that the speed of the appointment prevented consensual deliberation. This new development led Casino to initiate yet another arbitral proceeding against Diniz claiming that his dual role as board chair of both CBD and BRF posed a conflict of interest, since BRF was one of CBD’s principal suppliers. It was not until September 2013 that Casino and Diniz settled all legal proceedings, with Diniz agreeing to resign from CBD’s board and to exchange his voting shares in the company for preferred nonvoting shares at a 1:1 ratio.
The dispute between Diniz and Casino for control of CBD proves that the influence of state actors in Brazilian corporate governance looms large. The episode is also revealing of the different manners by which law can constrain the behavior of SOEs. Although controversial from a social welfare standpoint, there was nothing illegal in BNDESPAR’s proposed financing of the transaction, since the absence of a detailed mandate gives the bank substantial leeway in reaching its investment decisions. The role of law in constraining state action was mostly rhetorical. Casino’s loud protests claiming that the transaction defied the “the rule of law” and the “respect for private property” in Brazil were likely a factor in its delegitimization in the eyes of the public and, ultimately, its loss of political support.
Nonetheless, it would be premature to conclude that extralegal mechanisms are sufficient to restrain the harmful excesses of state capitalism in Brazil. In the CBD controversy, Casino emerged as a clear loser of the government’s support to the merger, and had, in turn, every incentive to launch a campaign against it. Whenever the harm of misguided policies is more diffuse – as is often the case –, collective action problems serve as an obstacle to democratic accountability.
State capitalism is fundamentally about the degree of state discretion in economic affairs. In its present form, state involvement in Brazilian corporate governance relies on a combination of traditional government-controlled SOEs and the formally minority positions of SCIIs. From a legal standpoint, these different instruments reflect a tradeoff between corporate control and countervailing legal controls. Majority ownership confers greater corporate control, but at the cost of a stricter legal regime molded by public law. Minority stakes free state actors from legal constraints, but requires them to broker deals with private actors.
The association between the growing use of a private legal regime and the reduction in the amount of information available to the public is however worrisome. Requests by journalists and academics for material agreements entered into by BNDES have been systematically denied, based on dubious legal arguments. By sharing control with private actors, state power has plunged into obscurity. This is particular challenging, given that extralegal constraints on state power depend crucially on the availability of information. Reversing this trend – through stricter enforcement of existing laws or the enactment of new regulations – is therefore imperative for both efficiency and legitimacy reasons.
Yet addressing the shortcomings of state capitalism requires a better and more nuanced understanding of its nuts and bolts. The precise relationship between majority and minority state investment, for instance, remains unclear. Scholars have posited that these two modes of ownership operate as substitutes, but they might as well work as complements. It may be that the discretion afforded by fully-controlled SOEs enhances the government’s leverage over companies under shared public and private control. We leave this and other questions for future research.
* Professor of Law, Fundação Getulio Vargas School of Law at São Paulo (Direito GV).
 See, e.g., Ian Bremmer, The End of the Free Market (2011).
 Aldo Musacchio & Sérgio G. Lazzarini, Leviathan in Business: Varieties of State Capitalism and their Implications for Economic Performance (working paper, 2012), available at http://ssrn.com/abstract=2070942.
 For excellent case studies on China, Russia, and South Korea, see Curtis J. Milhaupt & Katharina Pistor, Law & Capitalism (2008); Li-Wen Lin & Curtis J. Milhaupt, We Are the (National) Champions: Understanding the Mechanisms of State Capitalism in China, 65 Stan. L. Rev. 697 (2013). A notable exception is provided by David M. Trubek, Diogo R. Coutinho & Mario G. Schapiro, New State Activism in Brazil and the Challenges for Legal Institutions, in Law and the New Developmental State (David M. Trubek et al. eds., 2013) (focusing on the new functionalities of the legal system in Brazil).
 OECD Guidelines on Corporate Governance of State-Owned Enterprises (2005).
 Mariana Pargendler, Aldo Musacchio & Sérgio G. Lazzarini, In Strange Company: The Puzzle of Private Investment in State-Controlled Firms, 46 Cornell Int’l L. J. 569, 574-5 (2013).
 For a review of this literature, see Kevin E. Davis & Michael J. Trebilcock, The Relationship between Law and Development: Optimists versus Skeptics, 56 Am. J. Comp. L. 895, 945 (2008) (concluding that the empirical evidence generally supports the strong consensus that law matters for economic development, but precisely what types of legal institutions matter remains an open question).
 Ha-Joon Chang, 23 Things They Don’t Tell You About Capitalism (2010) (for a forceful articulation of this point).
 Quantitative measures of state size map poorly into conventional characterizations of state capitalism. Both government spending and consumption as a percentage of GDP are higher in Europe and the United States than in China. See Niall Ferguson, We're All State Capitalists Now, Foreign Affairs, Feb. 9, 2012.
 Friedrich Pollack employed the expression in the 1940s to downplay the differences between the capitalist regimes of the West and the state-owned and directed economy of the Soviet Union. State capitalism, in Craig Calhoun, Dictionary of the Social Sciences (2002).
 Musacchio & Lazzarini, supra note 2; Curtis J. Milhaupt & Wentong Zheng, Beyond Ownership: State Capitalism and the Chinese Firm (unpublished manuscript, 2013).
 For the classical formulation of the concept, see Peter Hall & David Soskice, Varieties of Capitalism: The Institutional Foundations of Comparative Advantage (2001).
 Id. (on the definition of liberal and coordinated market economies).
 See Anne Alstott, Gender Quotas for Corporate Boards: Options for Legal Design in the United States (working paper, 2013), http://ssrn.com/abstract=2360253, at 7 (“the tax system spends trillions of dollars of government money to shape business decisions – to craft what amounts to an industrial policy for the United States”).
 Mariana Pargendler, State Ownership and Corporate Governance, 80 Fordham L. Rev. 2917, 2925 et seq. (2012) (describing historical and contemporary instances of government ownership in the United States).
 Not all state-owned enterprises are alike. In particular, the legal regime can go a long way in reducing the scope for political influence in SOE governance. For instance, the legal framework affords the Norwegian government far less say over state-owned oil firm Statoil compared to the Brazilian state’s influence over Petrobras. Pargendler et al., supra note 5, at 593 and 600 (describing the ban on SOE board service by politicians and the stronger role of the industry regulator in Norway as mechanisms that ensure greater SOE independence vis-à-vis their Brazilian counterparts).
 In Roman law parlance, the owner enjoys broad rights (ius utendi, ius fruendi, ius abutendi) to employ her property as she sees fit. In economic terms, ownership reflects the “residual rights of control” over an asset that have not been contracted out. See Sanford J. Grossman & Oliver D. Hart, The Costs and Benefits of Ownership: A Theory of Vertical and Lateral Integration, 94 J. Pol. Econ. 691 (1986).
 For an excellent discussion of the tension between pervasive state ownership and the rule of law in transition economies, see Katharina Pistor & Joel Turkewitz, Copying with Hydra – State Ownership after Privatization, in Corporate Governance in Central Europe and Russia (Roman Frydman, Cheryl W. Gray & Andrzej Rapaczynski eds., 1996).
 Pargendler et al., supra note 5, at 574-5. Brazil’s swift government response to the financial crisis of 2008 brings home this point. The government resorted to its existing SOEs (in particular, state-owned banks BNDES and CEF) to significantly increase its credit to the private sector and lower market interest rates, hence circumventing the need for cumbersome and potentially contentious legislative action.
 Oliver E. Williamson, Public and Private Bureaucracies: A Transaction Cost Economics Perspective, 15 J. L. Econ. & Org. 306, 321–24 (1999); Andrei Shleifer, State Versus Private Ownership, 12 J. Econ. Persp. 133, 133 (1998).
 On these two features of the rule of law, see Stephen Holmes, Lineages of the Rule of Law, in Democracy and the Rule of Law (Adam Przeworski & José María Maravall eds., 2003).
 Sérgio G. Lazzarini, Aldo Musacchio & Mariana Pargendler, O Governo Como Acionista: Desafios para a Governança Corporativa no Brasil, at 80-1 in O Futuro da Governança Corporativa (Joaquim Fontes Filho & Ricardo Leal eds., 2013).
 Anthony Boadle, Brazil’s BNDES crowding out private banks from loans, OECD says, Reuters, Oct. 22, 2013.
 Pargendler, supra note 16, at 2926.
 The expression comes from Miranda Valverde in his Exposition of Motives of the bill that led to Brazil’s 1940 Corporations Law.
 Celso Antônio Bandeira de Mello, Curso de Direito Administrativo 197-198 (25 ed. 2007).
 To be sure, the optimal degree of discretion to be conferred on BNDES is the object of debate. See, e.g., Mario G. Schapiro, Making the Developmental State Work: How Does a Mandate Matter for the Brazilian Development Bank? (working paper, 2012), http://ssrn.com/abstract=1986850 (arguing that the adoption of a narrower mandate for BNDES in order to increase the Bank’s accountability).
 Constitution of the Federal Republic of Brazil (CFRB), art. 1o, IV, and 170.
 CRFB, art. 173, § 1º, III and art. 37, II.
 CRFB, art. 173, § 1º, II.
 Lei 6,404 of 1976, art. 236, sole paragraph.
 Brazilian law is not unique in this regard. See Marcel Kahan & Edward Rock, When the Government is the Controlling Shareholder, 89 Tex. L. Rev. 1293 (2011) (arguing that U.S. law fails to provide adequate protection to minority shareholders of government-controlled firms).
 For the application of the “business judgment rule” in Brazil, see Mariana Pargendler, Responsabilidade Civil dos Administradores e Business Judgment Rule no Direito Brasileiro (unpublished manuscript, 2013).
 See Pargendler, supra note 16; Mariana Pargendler, The Unintended Consequences of State Ownership: The Brazilian Experience, 13 Theoretical Inq. L. 503, 505–06 (2012).
 See Part III.A infra.
 Lei Complementar [Complementary Law] 108 of 2001, art. 8 (requiring SOE pension funds to be organized as the equivalent of a foundation or nonprofit corporation).
 Media coverage habitually attributes the choice of a pension fund’s chief to Brazil’s President. See Dilma muda comando de fundo de pensão, O Estado de S. Paulo, May 25, 2012.
 The use of public pension funds to further governmental objectives is by no means a novel phenomenon. In order to contribute to the government’s goal of promoting Brazil’s stock market development, SOE pension funds were required to invest a portion of their portfolio in local equity markets. Ary Oswaldo Mattos Filho, Prefácio, in Roberta Nioac Prado, Oferta Pública de Ações nas S.A. – Tag Along [Mandatory Bid Rule in Business Corporations - Tag Along] (2005) (referring to pension funds’ “unsual character of mandatory shareholder”). To be sure, Brazil is not unique in boasting pension funds that promote a political agenda. See, e.g., Roberta Romano, Public Pension Fund Activism in Corporate Governance Reconsidered, 93 Colum. L. Rev. 795 (1993) (describing instances of political pressure over public pension funds in the United States).
 For a more detailed description of this episode, see Lazzarini et al., supra note 22, at 84-7.
 Catherine Vieira, Fundação Previ tem cerca de 60% dos investimentos no mercado de ações, Valor Econômico, Jan. 18, 2006.
 In the last 10 years, Brazil’s real interest rate (adjusted for inflation) averaged 3.7%, by far the highest among the BRICS. Russia, India, China, and South Africa offered real interest rates of -0.25%, 0.4%, -0.7%, and 0.4%, respectively.
 Previ Code of Best Corporate Governance Practices (2nd ed., 2012).
 See Bernard Black, Shareholder Passivity Revisited, 89 Mich L. Rev. 520, 553 (1990) (attributing the preference of U.S. pension funds for “broad diversification and passivity” to the regulatory environment).
 Graziella Valenti, Previ quer ampliar ativismo nas companhias, Valor Econômico, Aug. 30, 2013.
 Clarisse Spitz, Governo aumenta sua participação em Galeão já privatizado, O Globo, Dec. 24, 2013 (noting that, through FI-FGTS and BNDESPar combined, the government’s total stake in Odebrecht Transport exceeded 40%).
 This change in position took place after minority investors complained to CVM about the practices. Although CVM has not punished Petrobras, it issued a staff letter outlining its position on this issue, which ultimately led Petrobras to reverse course. Maria Helena Santana, Brazil: The corporate governance framework and practices relating to supervision and enforcement 58, in OECD, Supervision and Enforcement in Corporate Governance (2013).
 Law 6,404 of 1976, art. 141, § 4o., I and II.
 Law 6,404 of 1976, art. 141.
 Exposição de Motivos n. 197 pelo Ministro da Fazenda [Treasury Secretary] (June 24, 1976).
 Érica Gorga, Corporate Control & Governance after a Decade from “Novo Mercado”: Changes in Ownership Structures and Shareholder Power in Brazil (unpublished working paper, 2013) (on file with author).
 Law 6,404 of 1976, art. 117, § 1º, a (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 lead it to favor another company, domestic or foreign, to the detriment of minority shareholder’s participation in the profits or assets of the company, or the national economy”).
 Law 6,404 of 1976, art. 154.
 As part of this project, the author has formally requested the shareholder agreements entered into between BNDESPAR and closely-held corporations under Brazil’s Access to Information Law (Law 12,527 of 2011), which specifically covers SOEs (art. 1, II). Despite the constitutionally and legally guaranteed access, the request was systematically denied following [four] rounds of administrative appeals, based, inter alia, on the allegation that the agreements in question constituted “business secrets” whose disclosure would “weaken the State’s operation in stock markets, putting the financial performance of public resources at a disadvantage vis-à-vis private resources.”
 For instance, the Shareholders Agreement of América Latina Logística S.A. – ALL, dated February 1, 2011, which has BNDESPAR and SOE pension funds Previ and Funcef as parties, among others, provides that “the Company and its subsidiaries shall be managed with the primary purpose of generating profit and maximizing the return on the investment made by the parties in the shortest term possible, in view of the best interests of the Company and its subsidiaries” (Section 2.1(a)).
 Sabrina Valle & Mariana Durão, Petroquisa tinha um terço do setor químico, O Estado de S. Paulo, Aug. 25, 2012.
 See, e.g., Pargendler, supra note 16, at 2939 (describing the pyramidal structure of SOE telecom Telebras prior to privatization). The resort to pyramidal structures by Brazilian SOEs remain alive and well. BB Seguridade, a subsidiary of SOE Banco do Brasil, accounted for the world’s largest IPO in 2013. Cristiane Lucchesi & Francisco Marcelino, BB Seguridade Raises $4.25 Billion in World’s Biggest 2013 IPO, Bloomberg News, Apr. 25, 2013.
 André Rocha, Os riscos de investir em ações de estatais, Valor Econômico, Jun. 3, 2013.
 See note 62supra for the relevant statutory provision.
 Rio de Janeiro Court of Appeals (TJRJ), EAC 2003.005.00333, Fourth Civil Chamber, Reporting Judge Jair Pontes de Almeida, decided on May 18, 2004.
 See note 75infra and accompanying text.
 STJ, Special Appeal 745,739–RJ, Third Chamber, Reporting Justice Massami Uyeda, decided on Aug. 28, 2013.
 Memorial da Recorrida [Additional Brief of Appellee], Special Appeal 745,739-RJ, at 49 (arguing that the Petroquisa lawsuit amounted to a “leading case about Brazilian capital markets and abuse of control power” and that “saying to a minority shareholder that has litigated for over twenty years that its lawsuit will not have its merit appreciated due to a unilateral corporate maneuver by the defendant (…) represents not only a gross injustice, but also sends a catastrophic message to the market as to the dignity and reliability of the country”).
 On the impact of state ownership on the corporate governance environment, see note 38supra.
 Speech of Pedro Valadares, Câmara dos Deputados [Chamber of Deputies], May 28, 2001.
 Braskem’s Annual Report on Form 20-F for the year ended December 31, 2002, at 108-9.
 Petrobras’s Form 6-K, for the month of December 2011.
 Braskem’s Annual Report on Form 20-F for the year ended December 31, 2012, at 14.
 Shareholders’ Agreement of BRK Investimentos Petroquímicos S.A. and Braskem S.A., dated April 5, 2010, Section 2.1.
 Id., Sections 2.1.1 and 2.1.2.
 Braskem’s Annual Report on Form 20-F for the year ended December 31, 2012, at 135.
 Ricardo Rego Monteiro, Com a Braskem, Comperj sai da mira do Tribunal de Contas, Brasil Econômico, Mar. 16, 2010.
 Pão de Açucar completa 65 anos sem Diniz no comando, UOL Economia, Sept. 7, 2013.
 Guillermo Parra-Bernal, Brazil's Pao de Acucar agrees to investor buyout, Reuters, June 28, 2011.
 Márcio Falcão & Ana Carolina Oliveira, Fusão não terá dinheiro público, diz Gleisi, Folha de S. Paulo, June 30, 2011.
 Rafael Manzoni Jr. & Eliana Sobral, A batalha mais difícil de Abilio, IstoÉ Dinheiro, July 1, 2011.
 CBD Notice to the Market, dated June 28, 2011 (quoting Casino’s letter to CBD).
 Mario Cesar Carvalho, Casino lawyer accuses Abilio of “corporate coup”, Folha de S. Paulo, June 30, 2011.
 Beatriz Ferrari, BNDES não deve compactuar com uma fraude, diz Naori (interview to Veja website), Veja, July 6, 2011.
 Law 6,404 of 1976, art. 118, § 2o (“These [shareholder] agreements cannot be invoked to exempt the shareholder from the responsibility in the exercise of voting rights [art. 115] or power of control [arts. 116 and 117]”).
 These reasons ranged from an allegedly flawed strategic vision based on a declining model of hypermarkets to the overestimation of potential synergies and the dilution of CBD shareholders. CBD Notice to the Market, dated July 12, 2011 (quoting Casino’s press release).
 Wilkes Shareholders’ Agreement, dated November 27, 2006, Section 188.8.131.52.3(b).
 For a proposal of a more meticulous mandate for BNDES, see Schapiro, supra note 28.
 As reported by the international press, “Brasilia cannot be seen to have sided with a party in a dispute whose outcome risks undermining the rule of law -- a commodity critical to a country's ability to attract private investment.” Agnes T. Crane, Brazil steps in it by siding with Carrefour deal, Reuters, July 5, 2011.
 Mancur Olson, The Logic of Collective Action (1965) (for a classical articulation of this problem).