Limitations of Legal Transplants and Convergence to Corporate Governance Practices in Emerging Markets: The Brazilian Case

Abstract

Little is known about the outcomes of legal transplants carried out by emerging economies aiming to develop their securities markets. This chapter contributes to fill this gap by analyzing the effectiveness of investor protection rules of the Novo Mercado, a self-regulatory transplant created in 2000 in Brazil to distinguish companies committed to higher governance standards. Employing the institutional autopsy approach through an in-depth qualitative analysis of two controversial cases, I conclude that the legal transplantation generated mixed results. On the one hand, Brazil greatly developed its capital markets after the Novo Mercado’s creation. On the other hand, the Novo Mercado did not work as expected regarding investor protection due to deficient enforcement and rule’s interpretation. I argue that these problems may derive from conflicts of interests of the stock exchange, the formalistic interpretation of rules and a lack of adequate oversight structure. The Brazilian case demonstrates that cultural, political, economic, and institutional elements permeate entire markets and that the transplant did not alter the main agency problem between controlling and minority shareholders that still characterizes Brazilian companies, even those listed on the Novo Mercado. I believe that this result should be considered for any legal transplantation.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

The world seems to be undergoing a convergence in investor protection rules following EU harmonization aiming at positively influence the development of securities and capital markets. Such a convergence should especially aid developing markets by adding certainty, predictability, and control for foreign investors.¹ The advocates of the convergence thesis argue that this phenomenon can take place through legal transplants, in which a single rule or even an entire legal system is borrowed from another country (Watson 1993), or by functional convergence, in which different rules or institutions generate the same result (Gilson 2001).

There is little consensus in the literature on legal transplants and much debate regarding their very existence (e.g., Santos 2006; Legrand 1997; Siems 2008), their definition and occurrence (e.g. Watson 1978; Kanda and Milhaupt 2003), and their potential outcomes and effectiveness (e.g., Bebchuk and Roe 1999; Coffee 2000; Gilson 2001; Berkowitz et al. 2003). Despites the debates about legal transplant theory, the adoption of one’s state legislation, as well as self-regulations, by another is commonplace (Watson 1978). In this sense, the convergence of investor protection rules would take place through the spread of self-regulation, which could be more effective than the local rules, especially when a country’s legislation is insufficient to protect investors (Coffee 2000). Coffee (2000, p. 55) describes the evolving relation between strong self-regulation and regulation after legal transplants as follows: “If strong self-regulation can first bring about the appearance of deeper, more liquid securities markets, such legislation will predictably follow.”

The National Association of Securities Dealers Automated Quotations (NASDAQ), the world’s first electronic stock market focused on technology companies, is a well-known example of self-regulation. European stock exchanges, aiming to stimulate capital market-driven listings of younger and smaller companies with high growth potential, have started to replicate the NASDAQ model by transplanting self-regulation rules to improve investor protection through high standards of transparency² (Rasch 1994; Bottazzi and Da Rin 2002). This chapter highlights Germany’s Neuer Markt, a higher-quality market established through self-regulatory standards that were more stringent than required by German law (Coffee 2000).³ The Neuer Markt inspired a self-regulatory legal transplant into the Brazilian stock market: In 2000, the BM&FBovespa, the Brazilian stock exchange, launched a listing segment called the Novo Mercado (New Market). This premium listing segment aimed to distinguish companies committed to higher governance standards. Besides being a case of self-regulation transplanted from a developed country to an emerging country, the Novo Mercado has an additional peculiarity that makes it unique to researchers and practitioners: It can be considered a second-order legal transplant, since the German Neuer Markt is already a legal transplant from the U.S. NASDAQ.

This chapter aims to investigate investor protection outcomes deriving from the legal transplant carried out for the BM&FBovespa’s Novo Mercado. Specifically, I analyze the effectiveness of corporate governance practices adopted by companies listed on the Novo Mercado. In principle, the Novo Mercado Listing Rules (NMLR)⁴ ensure the effective protection of investors in the companies listed in this segment. However, certain cases throughout the 2000s cast doubt on whether this goal was achieved. I provide an in-depth and qualitative analysis of two controversial cases to assess the effectiveness of the NMLR and the role of the institutions responsible for its enforcement. My methodology employs an institutional autopsy approach, developed by Milhaupt and Pistor (2008).

This Brazilian case study is important for several reasons. First, Brazil is one of the largest emerging countries and a member of the BRICs. Second, the Brazilian Novo Mercado has been considered a success abroad, as well as an inspiration to be copied by other markets. Third, although certain negative experiences in terms of investor protection have occurred with Novo Mercado companies, studies providing in-depth analyses of such cases are lacking. Fourth, this research makes a broad contribution to the literature, since there is a lack of qualitative studies about the effectiveness of legal transplants in emerging markets.

As the main result, I argue that the legal transplantation did not alter the main agency problem between controlling and minority shareholders that still characterizes Brazilian companies, even those listed on the Novo Mercado. Perhaps this is the result of any legal transplantation. Even though, Brazil greatly developed its capital markets after the Novo Mercado’s creation and passed through significant legislative reforms to improve shareholder rights and disclosure to investors. On the other hand, the Novo Mercado did not work as expected regarding investor protection rights due to two main factors: the lack of enforcement of Novo Mercado listing rules and rule interpretation. The cases analysis showed that these factors are related, in turn, to conflicts of interests, the formalistic interpretation of rules and a lack of adequate oversight structure.

The chapter is organized as follows. Section 2 describes the global convergence of investor protection rules and the influence of the German Neuer Markt on the Brazilian Novo Mercado. Section 3 addresses the motivation behind BM&FBovespa’s creation of the Novo Mercado. Section 4 presents the implementation and evolution of the Brazilian capital markets since the Novo Mercado’s creation, analyzes two cases involving companies listed on this premium segment and assesses potential positive effects of Novo Mercado on Brazilian regulation. Section 5 concludes.

2 Global Convergence of Investor Protection Rules: The German Neuer Markt and Its Influence on the Brazilian Novo Mercado

The 1990s were a period of significant changes in the European capital markets, particularly due to deregulation, implementation of a single currency, demutualization, and stock exchange mergers (Rasch 1994; Bottazzi and Da Rin 2002). This time was also characterized by high-tech innovations. The German stock exchange could not include small and medium-sized innovative company issues at that time and German companies were then cross-listing in the NASDAQ (Burghof and Hunger 2004). In support of the emergence of these entrepreneurial companies, the Deutsche Börse, among other stock exchanges, created a self-regulated listing platform in 1997, the Neuer Markt, considered the most demanding in terms of disclosure and accountability (Bottazzi and Da Rin 2002).

The Neuer Markt was the most successful of the European new markets, with the largest number of listed companies (more than 330, with over €234 billion of capitalization). However, with the bursting of the dotcom bubble, the segment collapsed to just two companies in 2000, with €29 billion of market capitalization (Burghof and Hunger 2004). However, the bubble bursting was only one element of that crisis; there were also problems such as insider trading, audit firms manipulating reports, and biased prospectuses resulting from conflicts of interests with banks hired to carry out the IPOs of unprepared companies (Hess et al. 2001 apud Burghof and Hunger 2004). Some of these problems were due to failures of compliance and enforcement of Neuer Markt rules. Burghof and Hunger (2004) note that noncompliance with Neuer Markt rules was rarely punished. Given the lack of enforcement of its rules, the fraud and irregularities, and the dotcom bubble bursting, weakened investor credibility in the companies listed in this segment ultimately led to the termination of the Neuer Markt in 2003.

Nonetheless, as detailed in the following sections, the BM&FBovespa (then called Bovespa) created the Novo Mercado by emulating the Neuer Markt. The similarities between the two listing segments were so great that it can be argued that a legal transplant indeed occurred.⁵ In addition to the great similarity in the listing requirements, there were similarities in the segments’ implementation, involving public and private institutions willing to change the culture of the business community.⁶ It is also interesting to note that the German and Brazilian stock markets had been diagnosed with similar problems: weak equity markets in relation to the gross domestic product, low liquidity, high concentrations of control, and little issuer interest in new public offerings.⁷

3 Motivation for the Creation of the Novo Mercado

The Brazilian stock market’s prospects in the late 1990s appeared particularly grim. On the one hand, there was not a significant interest in new initial public offerings (IPOs) since the domestic market was small and illiquid and had a low savings rate.⁸ On the other hand, large Brazilian companies were increasingly interested in cross-listing their shares in the U.S. market, attracted by stock exchanges that would provide them greater visibility and lower cost of capital. As a result, part of the already scarce liquidity in the Brazilian market moved to international stock exchanges, especially the New York Stock Exchange. The trading volume at the Brazilian stock exchange fell from US$191 billion in 1997 to US$65 billion in 2001. This drop was followed by a reduction in the number of listed companies, from 550 in 1996 to 440 in 2001.

In addition to the macroeconomic problems and several crises involving emerging economies (including Brazil) in the late 90s, most of the factors that undermined the attractiveness of Brazilian stock market were associated with a lack of fairness and inequality in the treatment of minority shareholders. These problems were partly due to the 1976 Corporate Law, which allowed the issuance of up to two-thirds of its share capital in preferred (non-voting) shares, which allowed full control of a company with a reduced percentage of its shares (Nenova 2006). Additionally, opaque disclosure enabled indiscriminate related party transactions (OECD 2003) and the prosecution of controlling shareholders was probably a slow and onerous process for minority shareholders in addition to the lack of expertise by Brazilian courts at that time.

In April 2000 Brazil held the First Latin American Corporate Governance Roundtable in Sao Paulo organized by the OECD. The event was co-organized by BM&FBovespa getting support from the CVM, the Brazilian Institute of Corporate Governance (IBGC) ⁹ and the International Finance Corporation (IFC). It produced a “Synthesis Note” and a “White Paper on Corporate Governance in Latin America” aimed at promoting the development of the region’s countries capital markets. Based on the typical characteristics of these countries’ corporate governance practices, the following reform priorities were defined: (i) the fair treatment of shareholders (in cases of control change or delisting), (ii) the encouragement of investor activism, (iii) the improved quality and integrity of financial reporting, (iv) the improved disclosure of related party transactions and conflicts of interests, (v) the development of effective boards and independent directors, (vi) improvement of the quality of the legal structure (through better enforcement), and (vii) the increased effectiveness of regulators and better corporate dispute resolution.

Based on the OECD’s analysis and recommendations, the BM&FBovespa decided to increase the protection of minority investors through contractual mechanisms. As a result, it created two related mechanisms: the BM&FBovespa special listing segments in 2000 (Novo Mercado, Level 2 and Level 1)¹⁰ and the Market Arbitration Panel in 2001 to ensure the effectiveness of self-regulation. The migration to the Novo Mercado listing segment required the exclusive issuance of ordinary (voting) shares, ensuring the adoption of the one share – one vote principle.

4 Implementation and Evolution of the Novo Mercado

Although the Novo Mercado was created in late 2000, it was only after 2004 that the number of companies listed on it significantly increased. The evolution of the number of companies listed and the volume traded suggests that the Novo Mercado project contributed to the takeoff of the Brazilian stock market, as Fig. 1 shows.

Fig. 1

Number of companies in the BM&FBovespa in special listing segments (Level 1, Level 2, and the Novo Mercado) from 2000 to 2011 (Source: BM&FBovespa)

The Novo Mercado seems to have helped the BM&FBovespa achieve many of its objectives, such as improving the corporate governance practices of listed companies, raising the market’s attractiveness for domestic and international investors, increasing its institutional relevance and regional leadership, broadening the base of domestic investors, and increasing the market’s competitiveness and efficiency.

Most companies that went public between 2000 and 2010 opted to list on the Novo Mercado directly. In addition, during that time no IPOs occurred in standard level. As expected, foreign investors boosted the Novo Mercado, accounting for around 70 % of the IPO volume. This high acceptance indicates that the BM&FBovespa was able to establish trust among investors and issuers through the Novo Mercado.

An IPO wave occurred in Brazil from 2004 to 2007, with 113 IPOs during this period, about 20 times more than in the eight previous years. As a result, in early 2008 the Novo Mercado reached 100 listed companies. The growing interest of foreign investors, as well as the participation of domestic investors, substantially increased the volume traded and the liquidity of the shares, as Fig. 2 shows.

Fig. 2

Average monthly volume (millions of Brazilian reals) traded on the BM&FBovespa from 2000 to 2009 (Source: CVM Informative (through September 2009))

This growth brought complexity and a sense of euphoria to the market that led to the emergence of new problems, such as equity kicking, the issuance of Brazilian Depositary Receipts (BDRs) by Brazilian companies disguised as foreign companies, control acquisitions without mandatory tender offers or appraisal rights, and the implementation of new control-enhancing mechanisms like poison pills and pyramidal structures of control. Interestingly, some of these problems were also observed at the German Neuer Markt.

The emergence of controversial cases involving companies listed on the Novo Mercado have been criticized as violating the rights of minority investors. Considering the wide-ranging, negative publicity of certain cases, the BM&FBovespa would be expected to act in a way to verify if there was any violation of the NMLR.¹¹ As a result, the next two subsections analyze two cases of Novo Mercado-listed companies that potentially violated investor rights to verify if any violation of the NMLR really occurred, as well as to compare the problems of the Novo Mercado today with those of the Neuer Markt before its collapse. Using Milhaupt and Pistor’s (2008) approach, I analyze the two cases and their institutional responses, specifically regarding self-regulatory (BM&FBovespa) and regulatory (CVM) activities.

4.1 The Cosan Case

Cosan is one of the largest producers of sugar and ethanol worldwide. Founded in 1936, its revenues were R$6.3 million (around US$3.1 billion), with earnings before interest, taxes, depreciation, and amortization of R$718 million in 2009. In 2005, Cosan conducted its IPO directly on the Novo Mercado. Then, in 2007, the company announced a controversial reorganization plan: Its controlling shareholders, two firms controlled by Rubens Ometto Silveira Mello, decided to implement a complex change of Cosan’s shareholding structure.

The restructuring plan consisted of a corporate reorganization comprising three stages. The first would involve a global offering of Cosan Ltd. at the New York Stock Exchange (NYSE) issuing American Depositary Receipts Level III. The second would be a corporate restructuring, with Cosan Ltd. becoming Cosan’s new controlling shareholder. The last step would be the migration of shareholders from Cosan to Cosan Ltd., incorporated in Bermuda and listed on the NYSE. The second stage of the process was considered the most controversial for Cosan’s minority shareholders because it required the conversion of Cosan’s voting shares into two different classes of Cosan Ltd.: class A shares, each of which would be entitled to one vote, and class B shares, entitled to 10 votes each. This conversion was controversial because Brazilian Corporate Law (Article 110, Law 6.404/1976) forbids multiple voting shares. Additionally, class B shares would be exclusively held by Mr. Ometto and would automatically be converted into class A shares when sold. As a result, Ometto would control both Cosan Ltd. and Cosan with less than 10 % of the cash flow rights via a pyramidal structure (besides remaining as CEO and Chairman of the Board of both of them).

Unsurprisingly, investors did not welcome the proposed restructuring plan.¹² Ometto initially presented three options to minority shareholders: (i) to not exchange their shares, thus risking Cosan’s delisting from the Novo Mercado (if Cosan’s free float dropped below 25 %), (ii) to exchange Cosan shares for class A Cosan Ltd. shares traded on the NYSE, or (iii) to exchange Cosan shares for BDRs of Cosan Ltd. class A shares traded on the BM&FBovespa. Options (ii) and (iii) would submit the company and its investors to Bermuda’s laws.

Due to the negative impact of the first proposal, a month later Ometto proposed another share class called B2. This new class of shares would also provide 10 votes per share but would not be traded on any stock exchange due to a 3 years lock-up period. After that, if negotiated for any purpose, it would be automatically converted into class A shares. In any case, Ometto would ultimately indirectly hold 51.6 % of the total capital and 91.1 % of the voting rights of Cosan Ltd. This plan seemed to not only harm minority shareholders’ rights protected by the NMLR, such as the one share, one vote principle, but also violate the Brazilian legislation.

The first regulatory problem in this case deals with the request by Cosan Ltd. to CVM to be registered as a foreign company. Despite the fact that such a registration could set a precedent in unfair trade practices – since it involved a company subject to less stringent corporate rules than those for other companies (forbidden by Art. 4^°, VII, Law 6.385/1976) – the Brazilian regulator granted the request. A second, related problem was the absence of an agreement between the CVM and the regulatory authority of Bermuda for monitoring and sharing information (or even a multilateral agreement).¹³ The absence of monitoring is a relevant issue, since it could impede mechanisms presented in Art. 4^°, Law 6.385/1976, which requires the CVM to protect investors and maintain the market’s health.

Those two macro issues could have avoided violating Brazilian law if the CVM had not permitted an exception to a clear legal determination. There were two other regulatory problems. Cosan Ltd. requested an exemption from a legal device (Art. 33, CVM Instruction 361/2002) that prevents the offering of securities which are not admitted to negotiation in the Brazilian stock markets. Since Cosan Ltd. class A and B2 shares were not listed on a Brazilian organized market (and Cosan Ltd. was formally considered a foreign company), the company requested the CVM to let it adopt different procedures to register a voluntary tender offer (even when no onerous transfer of control would occur). Although the possibility exists for exceptional situations when listing shares (CVM Instruction 361/2002, Art. 34), it is clear that special circumstances are required, as well as the equitable treatment and adequate flow of information to shareholders. In response to the request, the CVM’s technical department understood¹⁴ that the exemption would be possible only for the exchange of class B2 shares and would not be legal or fair¹⁵ for class A shares. Opposing its technical department, however, the CVM’s board decided on January 8, 2008, to grant full exemption to Cosan Ltd. class A and B2 shares.

Finally, CVM exempted Cosan Ltd. from delivering an economic and financial feasibility study, a requirement for any company willing to launch its IPO when younger than 2 years (Art. 32, II, CVM Instruction 400/2003). The numerous subsequent investor complaints about compliance with corporate law reinforces this argument, with at least 23 formal complaints lodged directly with the CVM. Nonetheless, the CVM did not formally prevent the operation.¹⁶ Thus, the CVM ultimately allowed a company governed by tax haven laws to exchange securities, which are forbidden in the Brazilian market, even without an agreement on cooperation and the exchange of information with the regulatory authority of the other country.

In addition to problems with the Brazilian regulation, there were problems with the enforcement of the Novo Mercado’s self-regulation by the BM&FBovespa. The first was noncompliance with the NMLR, which requires BM&FBovespa to notify Cosan’s controlling shareholders about potential noncompliance when it was clear that the outcome of the restructuring would be the delisting of Cosan from the Novo Mercado. The BM&FBovespa should have acted according to items 12.1 and 12.6.1 of the NMLR, by warning managers and controlling shareholders to ensure compliance with the NMLR. These rules also require that the BM&FBovespa disclose these warnings at its website, which it did not; nor did it provide notification of any exemption (possible under item 3.1.2 of the NMLR) to Cosan or its controlling shareholder.

According to reports from the specialized media, the second proposal (class B2 shares) was the outcome of a meeting between BM&FBovespa representatives and Ometto.¹⁷ However, none of these supposed meetings (or notifications) were formally announced at the BM&FBovespa’s website, as required by the NMLR. This suggests the noncompliance of the BM&FBovespa itself with its own NMLR.

The second problem regarding self-regulation was the lack of enforcement of the NMLR with respect to the proposed restructuring, which would ultimately have led to Cosan’s delisting from the Novo Mercado. Since the listing level was created based on principles of investor protection, two related items dealing with mandatory tender offers (items 11.2 and 11.4 of the NMLR) could be applicable to the situation in order to avoid losses to minority shareholders, which did not happen. In addition, the BM&FBovespa could have used item 14.4, which allows its CEO to resolve cases not dealt with by the NMLR in the case of unforeseen events.

4.2 The Tenda Case

Tenda was a real estate company founded in 1994 by José Olavo Pinto (JO) and his son, Henrique Pinto (HP) that operated in affordable housing construction. The company went public in 2007, listing on the Novo Mercado. The controlling shareholder was HPJO Participações (HPJO), controlled by JO and his son. At its IPO, Tenda launched 47.7 % of its shares, reaching a price of R$9.00 per share. However, after the first quarterly report, shares devalued by 15 %. The company then presented reversed scenarios in 2008, dropping from a high of R$12.80/share (instead of an expected R$27.50/share) to R$3.75, its lowest value, by the end of August.¹⁸

On September 1st, 2008, JO and HP announced that they would incorporate a closed company, named Fit, a subsidiary of Gafisa (a real estate company also listed on the Novo Mercado). The transaction in fact resulted in Tenda’s transfer of control from HPJO to Gafisa without a control premium. This unusual operation was curiously called an “original acquisition”, since there was no previous controlling shareholder. As a result, Tenda shares were diluted by almost 60 %, with no appraisal rights for Tenda’s minority shareholders. The announcement of the transaction led to an abnormal value destruction of around R$ 800 million for Tenda (Silveira and Dias 2010) and shares reached their lowest price (R$1.02) in late November 2008. A year later, Tenda and Gafisa announced the last stage of the complex incorporation of Tenda shares by Gafisa. In December of 2009, Tenda became 100 % part of Gafisa and was finally delisted from the Novo Mercado. These operations seem to have violated not only the rights of minority shareholders stated in the Novo Mercado’s rules, but also the Brazilian legislation.

One can point out at least five major regulatory problems in the Tenda case. The first was the violation of the duties of diligence and loyalty by its controlling shareholder (Art. 153 and 155, Law 6.385/1976), since the negotiation benefited only Gafisa and not Tenda. These violations constituted an abuse of power since the controlling shareholders guided Tenda to favor another company to the detriment of its shareholders (Art. 117, Law 6.385/1976).

The second was that, despite the strong suspicion of the breach of the duties of diligence and loyalty and the nine formal claims filled by Brazilian and foreign minority shareholders, CVM did not investigate the operation or request any explanation from the controlling shareholders.

The third regulatory problem was the ratio of the stock prices used in the first transaction, the subject of another complaint made by foreign funds to the CVM. CVM Instruction 319/1999 forbids a company to set exchange ratios by using stock market prices unless its shares are part of broad stock market indices, such as Ibovespa. Although Tenda’s shares never integrated any stock market indices, the company used its stock market prices as the exchange ratio, another abuse of power according to CVM Instruction 319/1999.

The fourth problem was a case of insider trading confessed by JO. The CVM investigated the case but instead of meting out any punishment, it accepted a commitment term with a fee paid by JO to settle the case.

Fifth, there was not a disclosure of a material fact by Tenda’s director of investor relations after the unusual fluctuation of Tenda’s shares. The CVM investigated the case in Administrative Process 3278/2010 and concluded that the material fact should have been mandatory. The Tenda’s director of investor relations, JO, and HP then proposed another commitment term with a fee payment, also accepted by the CVM.

In addition to problems with the enforcement of Brazilian legislation, five other problems with the BM&FBovespa’s enforcement of the Novo Mercado’s self-regulation can be raised. The first self-regulatory problem was the non-application of items 8.4 and 8.6 of the NMLR by BM&FBovespa in potential disagreements relating to the transfer of control, which suggested the use of the Market Arbitration Panel to decide about controversial mandatory tender offers as well as issuing supplementary rules. On the contrary, the Stock Exchange ignored those items of NMLR, despite an unusual operation clearly planned to elude a tender offer rule designed to protect investors. That operation was constructed to avoid a tender offer,¹⁹ turning shareholders into “prisoners,” since the liquidity was extremely low after the operations’ announcement.

The second problem was the violation of items 3.1, 3.1.1, and 3.1.2 of the NMLR, which determine that a company must maintain a minimum free float of 25 % while its securities are listed on the segment. Though there were exceptions to this free float percentage, they did not apply in Tenda’s case. Therefore, there was a breach of the NMLR because Tenda had less than the minimum 25 % free float throughout 6 months and did not even ask the BM&FBovespa for an exception. If this exception had been requested and granted, the BM&FBovespa should have disclosed it on its website, according to item 3.1.2 of the NMLR.

The third problem was the violation of the lock-up period mandatory for controlling shareholders due to item 3.4 of the NMLR. The lock-up is 75 % in the first 6 months after a company’s IPO, so JO and HP were obliged to hold a minimum of 15.15 % of Tenda’s shares. Despite this, reports show that the percentage they held (12 %) was below this value, suggesting that they sold more shares than allowed in the period.

The fourth problem was that, after the operation involving Tenda and Fit, HP (its former CEO and controlling shareholder) was considered an independent director of Tenda, disregarding the concept of independent director established by item 2.1 of the NMLR.

The last issue was BM&FBovespa’s lack of notification of Tenda’s controlling shareholders about the breaches to the NMLR during the period analyzed. As mandated by item 12.1 of the NMLR, the stock exchange should have sent a written notice to Tenda to enforce the NMLR, but it did not. By not having used item 14.1 of the NMLR to decide on unforeseen events, the BM&FBovespa lost an opportunity to show its willingness to reduce problems caused by the operation.

4.3 Cases Results

The two cases analyzed demonstrate violations of investor protection rights established in the NMLR and Brazilian Corporate Law. Regarding the NMLR, both cases show at least five reasons why investor protection was not effective. The first concerns the relation between regulation and self-regulation in the Brazilian legal system, something crucial for effective investor protection. The second concerns the usual limitation of any rule: The NMLR’s inherent incompleteness is a characteristic of objective rules, since it is impossible for a rule to embrace all situations yet to arise. The third is the BM&FBovespa’s omission, once it could have interpreted the NMLR according to its founding principles when there were doubts about the fairness of the operations. The fourth reason highlight the need for additional specific procedures to apply the rules in accordance with the principles that guided the creation of the listing level segment. Finally, the fifth reason for the failure to ensure investor protection was the BM&FBovespa’s lack of enforcement of the NMLR and the lack of punishment for its clear violations.

The BM&FBovespa’s failure to perform its monitoring duty may have been due to conflicts of interest. While the stock exchange must ensure compliance with the NMLR, its listed companies contribute substantially to its revenue and any sanctions against them could negatively affect its financial outcome. Specifically, the BM&FBovespa could have acted as foreseen in items 12.1 and 12.6.1 of the NMLR. These rules do not need extensive interpretation or application; that is, they simply require that the BM&FBovespa notify companies and their controlling shareholders to guarantee compliance with the NMLR, as well as disclose the names of these companies on its website.

As the Cosan case shows, according to the specialized press, the stock exchange itself may have violated its own rules by not informing investors about a possible violation of the NMLR. As a consequence, if the BM&FBovespa (informally) tried to convince Ometto about changing his restructuring plan, it violated its own NMLR. In the Tenda case, however, the BM&FBovespa did not act at all, formally or informally (even after receiving two letters from foreign investment funds complaining about the operations), in violation of the mandatory free float, lock-up period, and concept of independent directors.

Thus, the lack of investor protection evidenced in both cases corroborates the argument that effective protection depend not only on a set of rules but, instead, on strict compliance through enforcement. In addition, since one could consider the NMLR a public self-regulation,²⁰ then there can be an intrinsic problem based on the reasons cited by Coffee (2001) of a self-regulator’s weak incentives to enforce rules.

Regarding the role of the Brazilian regulator on Brazilian legislation, the evidence of both the Cosan and Tenda cases shows that the CVM’s behavior was at least questionable. Among other duties, CVM should (i) promote the expansion and smooth and efficient functioning of the stock markets and encourage continued investments in the capital of publicly held enterprises; (ii) protect securities holders and market investors against irregular issues of securities and illegal acts of public corporations; and, (iii) ensure compliance with fair trade practices in the securities market.²¹ However, in the Cosan case, the CVM allowed a company governed by tax haven laws whose shareholders and operations were all located in Brazil to exchange securities forbidden in the Brazilian market. Despite the CVM’s technical department having decided that class A shares of Cosan Ltd. would not be permitted on Brazilian stock markets, the CVM’s board opposed this decision and decided to make an exception for Cosan Ltd. The CVM then allowed the company to trade securities strictly forbidden by Brazilian legislation in the Brazilian market. The CVM also authorized the trading of depositary receipts of Cosan Ltd., which resulted in its lack of authority over the company since it was foreign based and without a prior agreement of cooperation or information exchange with its foreign regulator. The Brazilian regulator could have refused to accept the registration of a pseudo-foreign company established abroad only to avoid Brazilian laws.

In the Tenda case, the CVM favored the merger over a tender offer for economic reasons. However, this merger overrode the essential rights of minority shareholders, such as appraisal rights or a mandatory tender offer. The case also evidences breaches of CVM Instruction 319/1999, as in the exchange ratio of shares (Art. 11) and the CVM’s own violation constitutes an abuse of control (Art. 15). In this sense, the letters of foreign investment funds to the CVM bore no results. The CVM also omitted about the second operation (taking place 1 year after the first) possibly being an illegal indirect transaction. The CVM only acted when the controlling shareholder confessed committing insider trading and when the atypical fluctuation of shares was obvious. These cases were concluded without a clear judgment due to a consent decree (which controlling shareholders simply paid a fine to settle).

The high number of formal complaints to the CVM in the two cases²² and the wide dissemination of negative news by the specialized media²³ reinforce the view that both operations harmed investors. These cases show that investor protection depends on the regulator properly fulfilling its role (Kraakman et al. 2004) and suggest that the mere creation of a special listing segment is not sufficient to provide an institutional infrastructure or to ensure the enforcement of and compliance to its rules. In conclusion, analysis of these cases show that the typical Brazilian agency problem between controlling and minority shareholders may persist even for companies listed on the Novo Mercado.

4.4 Potential Positive Effects on Brazilian Regulation

Despite the CVM’s lack of timely responses in both cases analyzed, it is important to note that the Brazilian legal framework developed substantially throughout the 2000s due to the enactment of new laws, as well as improvement of the CVM’s activities regarding investor protection.

Regarding legal reforms, Brazil has faced major changes since 2000, with the enactment of Laws 10.303 of 2001, 10.411 of 2002, and 11.638 of 2007.²⁴ The CVM, in turn, has clearly increased its normative activity. For instance, it published 163 instructions between 2000 and 2010 regarding investor protection. In particular, CVM Instruction 480/2009 seems to have been an indirect response to the Cosan case, establishing new criteria for the registration of foreign companies.

One also observes a substantial increase in the number of consent decrees (with fines). While from 1997 to 2005 the average number of consent decrees was around 5 per year, this figure grew to around 60 per year from 2005 to 2009.

The CVM’s activities have grown consistently with the increasing number of companies listed and trading volume. This evidence clearly indicates that, even if the Novo Mercado did not guarantee effective investor protection in the cases analyzed, the increased number of companies and complexity of the Brazilian market has improved the CVM’s performance. In conclusion, it is clear that the Novo Mercado’s transplant led to legislative and regulatory changes. The improvement of investor protection through more restrictive rules, along with other elements of the country’s macroeconomic environment, resulted in a paradigm shift for Brazil’s capital markets.

5 Conclusions

In this chapter I analyze the creation of the Novo Mercado, a self-regulated listing segment of the BM&FBovespa that was transplanted from the German Neuer Markt. I consider the context of the Novo Mercado’s emergence, the details of its implementation, and its evolution during its first 10 years of operation. I also focus on two controversial cases involving companies listed on the Novo Mercado to analyze a regulator’s responses, those of the CVM, and a self-regulator’s responses, those of the BM&FBovespa.

The overall analysis indicates that the Brazilian legal transplant generated mixed results in terms of investor protection. On the one hand, the Brazilian capital markets developed significantly throughout the 2000s in terms of the number of newly listed companies, volume traded, investors and market intermediates. It is also clear that relevant legislative reforms have taken place since 2000 as well as CVM’s regulatory activity, both aiming at improving shareholder rights and disclosure for investors. Even if the Novo Mercado did not ensure effective investor protection in the cases analyzed, one can affirm that development of a legal framework fostering better investor protection occurred in parallel with the development of the legal transplant.

On the other hand, the Novo Mercado did not work as expected at least on the controversial cases involving Cosan and Tenda analyzed in this chapter. They provide clear evidence of the violation of investor protection rights, the primary reason for the Novo Mercado’s legal transplant. Two main factors appear to have led to investor losses in the two cases: the lack of enforcement by both the self-regulator (BM&FBovespa) and the regulator (CVM) and the interpretation of the rules. Several potential causes are associated with these factors.

The first one is conflict of interest. The BM&FBovespa must ensure that listed companies (including itself) comply with the NMLR but these same companies also contribute substantially to its own revenue. Any company sanction or even notification could negatively affect the BM&FBovespa’s financial outcome.²⁵ This is not surprising, regarding the problems of self-regulation enforcement stated by Coffee (2000). The CVM’s board, in addition, contradicted its own technical department in one of the cases analyzed. Although this would not be a problem per se, this decision allowed an exception to the enforcement of Brazilian corporate law when it was clear that it could potentially harm investors. Since the CVM’s board consists of politically nominated directors and its technical department consists of only career officials, one could suppose this fact to be related to some kind of regulatory capture, an event that tends to occur more often in environments characterized by crony capitalism, something usually associated with Brazil.²⁶

The second potential cause is a formalistic interpretation of the rules. The two cases analyzed provide evidence that both operations were thought to bypass the NMLR, as well as Brazilian legislation. The controlling shareholders used indirect operations that resulted in big losses for investors, in violation of their rights. The BM&FBovespa and the CVM did not interpret the self-regulatory and regulatory rules in their essence. Although Brazilian corporate law contains some of the same principles, the interpretation of rules is usually formalistic. Perhaps this is a clear example of a different outcome resulting from a legal transplant from a different legal environment. Thus, self-regulatory origins and traditions of a law’s interpretation can have important impacts on its enforcement. In this sense, rules are not enough to guarantee a transplant’s success, since its interpretation must adequately consider its legal environment.

The third is the lack of adequate oversight structure by the BM&FBovespa and the CVM. The rapid growth in the number of companies and in the complexity of the market probably was not matched by a corresponding increase in regulatory structure (e.g., in terms of budget and human and technical resources) to adequately monitor the market. In the case of the BM&FBovespa, a strong monitoring structure requires a larger budget, which directly affects its profits. Since the BM&FBovespa is a publicly listed company, its ultimate goal is to maximize profits by minimizing costs. Therefore, the lack of oversight may have occurred due to the stock market inherent conflict of interest between maximizing profits, minimizing costs and monitoring its compliance with its own rules.

These three specific causes naturally fall within a broader context of intrinsically linked elements: the institutional, cultural, economic and political aspects. In any market, institutions can be more or less plastic (Coffee 1999), but they always attempt to reinforce themselves without major modifications (North 1990). Emerging markets such as Brazil have, at least in theory, less efficient institutions, possibly the result of lower economic development. These cultural, political, economic, and institutional elements permeate the entire market²⁷ and affect the existence of investor protection rules. These elements indicate that the conflict between controlling and minority shareholders, the key problem before the transplant, remained the same for companies in the Novo Mercado afterwards.

There are five potential lessons for emerging markets, based on the Brazilian experience. First, the legal transplantation and mere adoption of rules are not sufficient to protect investor rights, contradicting the massive literature on legal reforms from World Bank based on studies from La Porta et al. (1997, 1998) among others. Second, conflicts of interest from both the regulatory and self-regulatory spheres can hamper the enforcement of rules. Third, legal reforms aimed at increasing investor protection based on different law systems may have unexpected outcomes, since they are interpreted according to the context of the country in which they have been transplanted. Thus, legal transplants cannot ignore the context of the environment in which they are transplanted. On the contrary, typical agency problems pervade the entire legal system and derive from the ownership structure that generates the main conflict of interest in companies.²⁸ Fourth, although legal transplants can be quickly implemented at low cost and signal the adoption of better practices for investors (Milhaupt and Pistor 2008), it is necessary to develop an efficient structure to oversee company compliance of the rules as well as enforce investor protection rights. Therefore, the enforcement of the transplanted rules depend on the institutions involved and the way they interpret the rules within their context (Berkowitz et al. 2003; Milhaupt and Pistor 2008). Fifth, the legal transplant can have a direct impact on the growth of the legal framework aimed at protecting investors.

This chapter adds to the lines of research that investigate legal transplants, the enforcement of investor protection rules, the drivers of legislative and regulatory changes, and the unforeseen consequences (externalities) of legal transplants.