The Economic Consequences of Legal Origins Professor Andrei

The Economic Consequences of Legal Origins Professor Andrei Shleifer Harvard University and NBER Barcelona January, 2007

Modern Theory of Corporate Finance • Earlier research on CG focused on such problems as: • Managerial consumption of perquisites (Jensen and Meckling 1976), • Managerial effort (Holmstrom 1979), and • Over-investment in pursuit of growth (e. g. , Baumol 1959, Jensen 1986). • Last 20 years: Modern theory of corporate finance focuses instead on the ability of corporate insiders to divert corporate wealth to themselves or “private benefits of control” (Grossman and Hart 1988, Hart 1995, Zingales 1994). • Most firms are not “Bearle-and-Means” (1932) corporations: They have owners who control and run them (LLS 1999) • Those in control– controlling shareholders or managers– have the power to divert corporate wealth to themselves through self-dealing in various forms: 1. Executive perks, excessive compensation, and loans to officers

Modern Theory of Corporate Finance • Empirically, such diversion has been investigated in several contexts: • U. S. savings and loans crisis (Akerlof and Romer 1993), • Asian & Mexican financial crises (La Porta, Lopez-de-Silanes, and Zamarripa 2003, Johnson et al. 2000 a), • Legal disputes over tunneling (Johnson et al. 2000 b), • CG during transition from socialism (Glaeser et al, 2001) and reform in developing countries (Lopez-de-Silanes 2003). • Estimating private benefits of control from voting premia and the treatment of controlling shareholders in takeovers (Nenova 2003, Dyck and Zingales 2004). • Economists have followed Legal scholars (Clark 1986) pointing to the crucial role of law in the control of investor expropriation: Self dealing limits external finance and leads to financial

Modern Theory of Corporate Finance • Initial research argues and shows that differences in legal investor protection across countries shape the ability of insiders to expropriate outsiders, and thus determine investor confidence in markets and consequently their development (LLSV 1997, 1998, 2000). • “Law and Finance” made three contributions: 1. Investor protection explains the development of financial markets. 2. As a result of colonial transplantation, legal origin is exogenous. Moreover, legal origin is correlated with investor protection. 3. Legal origin is the explanation of the observed differences in investor protection. • The current debate is about:

The Regulation of Self-dealing • Goals: 1. Develop a measure of investor protection that is both theoretically grounded and describes how societies regulate the self-dealing problem. 2. Examine whether anti-self-dealing measures are linked to development of financial markets and which regulatory approaches work best. 3. Identify the determinants of the regulation of self-dealing; Test for alternative theories of what explains investor protection 4. Shed light on what makes common law different from civil law. • Two impractical approaches to self-dealing: 1. Laissez faire -- Leave self-dealing unregulated and let courts sort it out.

Methodology

The Proposed Transaction ü Simple transaction (purchase of equipment) between two entities (“Buyer” and “Seller”) controlled by the same shareholder (“Mr James”), who is on the board of both firms. Mr. James owns 60% of Buyer Co. shares Buyer Co. Mr. James owns 90% of Seller Co. shares Seller Co. Buyer Co. buys equipment from Seller Co. • Key Problem: • The proposed transaction may have a business purpose. • For example, purchasing the equipment may lead to expanded sales. • James is on both sides of the transaction and may benefit if Buyer acquires overpriced equipment from Seller.

The Regulation of Self-Dealing 1. Replicate conditions in an arm’s-length transaction: • Disclosure and approval requirements by law before Buyer acquires Seller’s trucks • Immediate disclosures after the decision to enter the transaction has been made. 2. Empower shareholders to seek remedy for expropriation through the courts: • Since even a duly approved and disclosed transaction may damage Buyer, litigation may be necessary to obtain restitution. • So, we keep track of how easy it is for shareholders to obtain redress through the courts when the transaction damages Buyer if all disclosure and approval requirements are met. Crucial, since the laws of most countries provide harsh penalties for breaking disclosure and approval requirements. • Factors that affect the odds that the plaintiff prevails in court include liability standards and the right to compel evidence.

The Questionnaire & its Process • In cooperation with 9 law firms, we designed an tested a questionnaire covering: 1. who approves the transaction; 2. what needs to be disclosed, when and to whom (e. g. , Board, Shareholders, etc); 3. what are the duties of officers, directors, and controlling shareholders; 4. how the transaction could be rescinded, and by whom; 5. what causes of action are available to recover damages; 6. what needs to be proven under each cause of action; 7. who has standing to sue under each available cause of action; 8. availability of direct and derivative suits; 9. access to information and discovery rights; and 10. fines and criminal sanctions. • We received answers, and back-up laws or precedent, from 102 of the 115 law firms We read the laws and coded the respondents’ answers. •

Key Clarification: The Scope • We consider garden-variety self-dealing transactions, in which the controllers of companies make choices that may benefit them at the expense of other investors, but follow the law regarding disclosure and approval procedures. All legal approvals and required disclosures were met • We are not dealing with cases of corporate crime such as Enron or Parmalat: To stop such cases, every country uses harsh criminal punishments. • We are interested in a very different situation: • If a controlling shareholder wants to enrich himself but also follow the law, how difficult is it for minority shareholders to

Data 1. Approval and Ex-Ante Disclosure requirements 2. Ex-Post Disclosure and Burden of litigation 3. Public enforcement 4. Measures of stock market development.

Legal Origin Distribution Legal Origins = English = French = German = Scandinavian = Socialist

Ex-Ante Ex-Post Public Enforcement Regulation and Income Outcomes Results Ex-Ante Control of Self-Dealing

Ex-Ante Ex-Post Public Enforcement Regulation and Income Outcomes Results Ex-Post Control of Self-Dealing

Ex-Ante Ex-Post Public Enforcement Regulation and Income Outcomes Public Enforcement Results

Results Development of Stock Markets

Economic Effects: Stock Market Cap and Self-Dealing

Anti-Self-Dealing Index and Block Premium • Figure VI: Partial-regression leverage plot of Block Premium against the index of anti-self-dealing, controlling for Log GDPpc and efficiency of the

Anti-Self-Dealing Index and Ln Firms / Pop • Figure VII: Partial-regression leverage plot of Log listed firms per million people against the index of anti-self-dealing, controlling for Log GDPpc

Anti-Self-Dealing Index and IPOs/GDP • Figure VIII: Partial-regression leverage plot of IPOs-to-GDP against antiself-dealing in regressions controlling for Log GDPpc and efficiency of the

Anti-Self-Dealing Index and Ownership Concentration • Figure IX: Partial-regression leverage plot of Ownership Concentration against ex-ante (left) & ex-post (right) anti-self-dealing, controlling for

Table X Instrumental Variables Regressions

Public Enforcement and Stock Market Capitalization • Figure VIII: Partial-regression leverage plot of stock market capitalization & index of public enforcement, controlling for Log GDPpc and efficiency of

Public Enforcement • Table XI and Figure X show that public enforcement is not associated with more developed stock markets. • Advocates of public enforcement may dismiss our findings by: 1. Arguing that what deters self-dealing is the probability that criminal sanctions will actually be imposed (rather than their mere existence). Unfortunately, we lack data on actual enforcement practices to test this. We have used Bhattacharya and Dakou (2002) but does not work. Divided Rich versus Poor countries, but does not work. Conjecture that a proxy for actual enforcement would have a hard time fitting the data since criminal sanctions are simply unavailable in roughly half the sample (32 countries) and this group exhibits enormous variation in the size of stock markets (e. g. Hong Kong vs. New Zealand).

Alternative Theories Politics

Politics and Investor Protection • Investor protection may be determined by politics rather than legal origin [Pagano & Volpin 2005, Perotti & von Thadden 2006, Roe 2000, Rajan & Zingales 2003)]. • Proportional electoral systems are conducive to weaker investor protection than majoritarian systems. • Table XVII Panel A: • Common law countries have sharply higher anti-self-dealing scores. • Proportional representation is associated with lower anti-self-dealing scores. • When both are included, only the latter is statistically significant But multicollinearity makes it difficult (corr -. 46). • So, we run univariate regressions for common and civil law countries separately using proportional representation to explain anti-self-dealing regulation (Panel B). Proportional representation is insignificant in both regressions. • Split into countries above and below the median competitiveness of the legislature (Panel C). If common law is a proxy for electoral rules, it should not predict the development of securities markets in non-

Proportional Representation in different Legal Origins • Figure XI: Partial-regression leverage plot of anti-self-dealing index against proportional representation in common law countries (left graph) and civil law ones (right graph) in regressions that control for (log) income

Legal Origin in different Democratic Regimes • Figure XII: Partial-regression leverage plot of anti-self-dealing against legal origin for countries with high (left graph) and low (right graph) competitiveness of the legislature in regressions that control for (log)

Conclusions • Constructed a new index of shareholder protection for 72 countries (so far). • Addresses specifically the protection of minority shareholders against self-dealing transactions benefiting controlling shareholders. • Better grounded in theory than index of anti-director rights (LLSV 1997, 1998) • Anti-self-dealing index exhibits same properties as both the antidirector rights index, and the indices of shareholder protection through securities laws (LLS 2006). • It is sharply higher in Common law than in French civil law countries. • Statistically significant and economically strong predictor of stock market development across countries. • Results support findings in earlier work, but also show that theoretically -grounded measures of investor protection are closely tied to financial development.

Implications: Progress in the last 10 years 1. Measurement of Investor Protection 2. Interpretation of Legal Origin 3. Regulatory Strategies and Policy Reform

Implications for the Measurement of Shareholder Protection • 4 measures of shareholder protection with different methodology and different situations • So, what is “the best” measure for researchers to use? No clear answer measures are highly correlated with each other. • But some differences: 1. Measures from securities laws: (49 countries) • “Work” best in terms of predicting stock market outcomes • Appropriate for studies of protection of investors buying securities, as opposed to corporate governance per se. 2. Anti-director rights index: (72 countries) • Advantage of continuity with many previous studies; 3. Anti-self-dealing index: (72 countries) • Greater conceptual clarity in general, and relevance to the pervasive problem of corporate self-dealing (or tunneling) in particular. crucial • If self-dealing is the central problem of corporate governance in most countries, the law’s effectiveness in regulating this problem is the

Correlations

Implications for the Interpretation of Legal Origin I. Examination of legal rules gives further insight in pronounced differences in the way Common and Civil law protect investors: 1. Confirmation of Johnson et al. (2000) conjecture that common law is more suspicious of conflicted transactions, and subjects them to closer scrutiny: Greater disclosure and more arms-length approval. 2. Central difference between common and civil law: Ex-ante transparency in self-dealing. 3. Approaches appear to derive from long-standing legal principles which over time are incorporated into the statutes that we actually observe. II. Broader vision: • Results are consistent with Djankov et al. (2003) that common law is distinguished from civil law by its encouragement of private solutions to problems of “disorder. ” • Statutory law aims to reduce costs of private solutions, not replace w/ public ones.

Implications for Regulatory Strategies & Policy Reform • Taken at face value, our work has implications for corporate governance improvement: 1. The Public sector has a role to play as designer of the rules: • Countries with successful stock markets give shareholders the information they need and the power to act – including both voting and litigation -- on this information. • No evidence that successful countries rely heavily on fines and criminal sanctions (LLS, 2006). 2. Specific Recommendations for policy reform: • Combine full disclosure of self-dealing transactions with the requirement of approval by disinterested shareholders – inexpensive and straightforward to implement. Need to set a lower bound, particularly because of concentrated ownership But, no real reason for it not to work in rich or poor or civil law countries