DEREGULATED SECURITIES MARKETS, LAX CORPORATE GOVERNANCE, AND CORRUPTION: EVIDENCE FROM THE NEVADA OTC EXPERIMENT

Transcription

1 DEREGULATED SECURITIES MARKETS, LAX CORPORATE GOVERNANCE, AND CORRUPTION: EVIDENCE FROM THE NEVADA OTC EXPERIMENT Parham Holakouee* Can we rely on retail investors in lightly regulated markets to discipline the corporate governance decisions of firm managers? Publicly traded firms in the over-the-counter ( OTC ) markets operate in relative opacity, avoiding the federal disclosure laws and listing requirements demanded of firms on formal U.S. exchanges (e.g. NYSE, Nasdaq). How do firms in this weakly regulated market respond to the option of a lax legal regime? Nevada amended its corporate law in 2001, effectively removing director and officer liability for breaches of well-established corporate fiduciary duties. This Nevada amendment is concurrent with a surge in the proportion of OTC firms incorporating in Nevada: 16% before 2001 vs. 59% after The extant literature on the impact of this legal change has focused solely on the major exchanges where Nevada incorporations have increased by 20%; the more striking rise in the OTC has been overlooked. This paper examines the rise of Nevada incorporations in the OTC. I empirically analyze Nevada-incorporated firms using a differences-in-differences empirical design with a novel, hand-collected dataset of SEC Trading Suspensions as my outcome variable. I find strong statistically and economically significant evidence that firms choosing post-2001 Nevada corporate law are the firms most likely to be engaging in corporate misconduct. Because the OTC markets lack the disclosure requirements and regulatory constraints that protect investors on the national exchanges and with OTC markets already a haven for insider abuse Nevada s lax corporate law may be removing a key deterrent to corporate corruption where it is needed most. *Parham Holakouee, University of California at Berkeley, Ph.D. Candidate 1

2 I. INTRODUCTION A. Change Amid Controversy In the summer of 2001, the Nevada legislature was facing a steep budget shortfall and in need of a fresh source of revenue. A proposed bill to increase incorporation revenues by amending the state s corporate laws sparked a divisive debate. Nevada s proposed strategy targeted a niche segment of the incorporation market: firms seeking the strongest director and officer liability protection. Because Delaware s longstanding dominance in the interstate competition for corporate charters had set up towering barriers to entry, competing directly with Delaware would be an ill-advised strategy. 1 The proposed Nevada amendment would target an underserved segment of the incorporation market by insulating managers from fiduciary duty liability. The bill faced fierce opposition by legislators who worried the liability protections were extreme and would appeal to the most corrupt firms. The legislators made impassioned pleas that the promised revenues would come at a terrible price and would attract rip off artists and corporate crooks. 2 After contentious debate, the bill s detractors ultimately relented and the controversial amendment passed; 3 managers of Nevada-incorporated firms were now insulated from liability for all but the most flagrant breaches of their fiduciary duties. The Nevada legislators less ominous predictions regarding the effect of introducing a lax option to the corporate law menu have come true. Nevada has significantly increased its share of incorporations while raising its charter fees 4 yielding substantial revenue for the state. 5 But the issue that most concerned legislators in the debates preceding the amendment remains unsettled: is Nevada s lax corporate law attracting firms engaging in fraud and insider misconduct? 1 Over 50% of all large, publicly traded firms in the U.S. and 64% of Fortune 500 firms are incorporated in Delaware. Delaware Secretary of State, Division of Corporations. Delaware s network externalities, expert business judiciary, and established body of business case law are among the factors that drive Delaware s difficult to replicate competitive advantages. 2 Senate Debate, S.B. No. 577, One Hundred and Eleventh Day (May 6, 2001) (statement of Sen. Terry Care and statement of Sen. Dina Titus). 3 Legislators opposing the bill ultimately gave in after receiving assurances that the promised revenues from incorporations would fund raises in teachers salaries. 4 In 2003, Nevada raised its maximum annual tax for domestic corporations from $85 to $11,100. Though this is a significant increase from the negligible tax it had prior to 2003, it is still substantially less than Delaware with a maximum annual tax of $180, Michal Barzuza, Market Segmentation: The Rise of Nevada as a Liability-Free Jurisdiction, VA. L. REV (2012). 2

3 B. Why Analyze the Nevada Effect in the OTC? Nevada s rise as an incorporation destination has recently spawned a literature that has debated whether the Nevada option is: (a) allowing managers to escape liability while expropriating shareholders, or (b) an efficient choice for the subset of firms incorporating in Nevada. 6 What has been ignored in this literature is the extraordinary rise of Nevada incorporations in the Over-the-Counter equity markets ( OTC ). This paper analyzes the impact of Nevada corporate law in the OTC markets. While OTC firms have generally been glossed over in the law, business and economics literature, I direct my analysis to this market for the following reasons: (1) Disclosure requirements and other disciplinary forces faced by exchange-traded firms render breaches of certain fiduciary duties infeasible even if permissible by corporate law. 7 OTC firms, on the other hand, are free from these sources of scrutiny and disclose to the public relatively scant information. The addition of a permissive corporate law option is likely to be a more meaningful addition to the choice set for firms lacking these other checks on their fiduciary duties; (2) The OTC markets exhibit a dramatic rise in Nevada incorporations that coincides with the 2001 Nevada law change (16% Nevadaincorporated before 2001 vs. 59% after). This striking rise dwarfs the relatively moderate increase on the major exchanges but has been unnoticed in the extant literature. While Nevada has received attention recently as a viable number-two on the major exchanges, its post-2001 dominance in the OTC has been overlooked; (3) The surge in post-2001 Nevada OTC incorporations provides a wealth of new data to empirically test hypotheses related to the impact of the Nevada amendment. On the national exchanges, the proportion of firms incorporating in Nevada remains small even after 2001, yielding only a small number of post-2001 Nevada incorporators. This lack of data 6 See Barzuza, supra Note 4; Bruce H. Kobayashi & Larry E. Ribstein. Nevada and the Market for Corporate Law, SEATTLE UL. REV. (2012); Michal Barzuza & David Smith, What Happens in Nevada? Self-Selecting into Lax Law, 27 REV. FIN. STUD (2014); Jens Dammann, Just How Lax is Nevada Corporate Law? A Response to Profess or Barzuza, 99 VA. LAW REV., In Brief 1 (2013). 7 In addition to the more expansive disclosure demands faced by firms on the major exchanges, these firms are also subject to the listing requirements of the exchange these include financial listing requirements (e.g. minimum assets) and governance requirements (e.g. independent auditors, independent directors). Also, shareholders on the exchanges include institutions and other sophisticated investors that more closely scrutinize managerial decision-making. Investors in the OTC markets are nearly always individual, retail investors institutions are almost entirely absent from the OTC. 3

4 has hindered the production of decisive empirical conclusions regarding the effect of the Nevada change; (4) Though the OTC markets comprise a small fraction of the market capitalization on the major exchanges, the level of fraud and shareholder misappropriation on these markets is substantial. 8 If excessive managerial laxity broadens opportunities for fraud and insider theft, OTC investors will likely be among its first casualties. The paper proceeds as follows. Part II describes the increase in Nevada incorporations and introduces the SEC Trading Suspensions data. Part III examines the OTC markets and introduces the theory underlying the paper s empirical approach. Part IV discusses the related literature. Part V explains the change in law, focusing on distinctions between the Nevada and Delaware exculpation statute. Part VI presents the efficient/inefficient account of Nevada incorporation and introduces the paper s empirical data. Part VII presents and discusses the empirical results. Part VIII concludes. II. INCREASE IN NEVADA INCORPORATIONS AND SUSPENSIONS A. Striking Rise in OTC Nevada Incorporations Nevada has increased its share of incorporations among exchange-traded firms by 20% since 2001 despite a sizeable increase in its incorporation tax. 9 Ignored in the literature thus far is the extraordinary rise of Nevada incorporations in the OTC markets. Among OTC firms that provide some level of disclosure and incorporated in the U.S. between June 15, 2001 through December 31, 2011, 59.34% are Nevadaincorporated. By comparison, 25.58% of disclosing OTC firms over the same period are incorporated in Delaware. This is a significant shift from the distribution of firms prior to Of 5,469 pre-2001 disclosing OTC firms incorporated in the U.S., 16.11% incorporated in Nevada, while 41.30% incorporated in Delaware (See Figure 1). 10 On the national exchanges, Delaware maintains the lion s share of out-ofstate incorporations (81.90%) and more than ten times the number of Nevada outof-state incorporations (6.66%). 11 The 20% increase in Nevada incorporations in 8 SIPC (Securities Investor Protection Corporation) estimates $1-3 Billion in annual losses due to microcap fraud. It further states that the vast majority of microcap fraud takes place in the OTC markets. 9 See Barzuza, supra Note 4 at Mergent Online, 11 See Barzuza, supra Note 4 at

5 the major exchanges has justifiably garnered the attention of scholars and commentators; the more than three-fold increase in the proportion of OTC Nevada incorporations after 2001 has been overlooked. FIGURE 1: PERCENTAGE OF TOTAL OTC INCORPORATIONS IN NEVADA AND DELAWARE BEFORE AND AFTER JUNE 15, % % OF OTC INCORPORATIONS 50.00% 40.00% 30.00% 20.00% 10.00% NEVADA DELAWARE 0.00% BEFORE JUNE 2001 AFTER JUNE 2001 B. Overview of SEC Trading Suspensions Data To empirically examine whether the firms engaging in fraud and shareholder expropriation are choosing post-2001 Nevada corporate law, I use a novel, handcollected dataset of SEC Trading Suspensions ( Suspension ) as my outcome variable. The Securities and Exchange Commission ( SEC ) will enforce a Suspension to protect investors from trading a stock suspected of gross insider misconduct. Factors that can trigger a Suspension include the spread of false or misleading information about the company, stock price manipulation, and suspicious insider transactions. 12 In almost every case, the suspended firm is 12 SEC Office of Investor Education and Advocacy. Investor Bulletin: Trading Suspensions (May 2012) SEC Press Release

6 permanently removed from the OTC marketplace and its share price drop to zero or dramatically below its pre-suspension price. A detailed discussion about SEC Trading Suspensions and the motivation for using Suspensions in this paper s empirical design is presented in Part VI. The raw data indicates a substantial shift in the likelihood of a Suspension after the Nevada law change. Among firms incorporated in the decade before the Nevada amendment, the probability of a Suspension is slightly higher for Delaware-incorporated firms than for Nevada-incorporated firms. 13 After the law change, the probability of a Suspension for Nevada-incorporated firms is nearly three times the probability that a Delaware-incorporated firm will face a Suspension. Similar results hold for comparing Nevada-incorporated firms to the probability of a Suspension firms incorporated in all other states combined. A year-by-year analysis shows a substantial jump in the spread between Nevada and Delaware Suspension probabilities for firms incorporated immediately after the change in Nevada law (See Figure 2). 14 A thorough discussion of the empirical design and empirical analysis for this paper is provided in Part VII. 13 Part VII(E) discusses why Delaware is the appropriate control group and why out-of-state incorporations are used in the principal empirical analysis. 14 These measures indicate increases in the probability of a Suspension. Because the share of Nevada incorporations has also significantly increased, the share of total Suspensions is significantly more pronounced in the post-period. 6

7 FIGURE 2: SUSPENSION PROBABILITY FOR NV INC. FIRM MINUS SUSPENSION PROBABILITY FOR DE INC. FIRM BY YEAR OF INC III. AN OVERVIEW OF THE OTC MARKETS A. Corruption in the OTC The OTC markets have long been a haven for rampant insider abuse; microcap stock fraud costs investors an estimated $1-3 Billion per year. 15 The SEC and Financial Industry Regulatory Authority, Inc. ( FINRA ) provide repeated warnings to investors regarding the dangers of investing in the O TC markets due to the lack of transparency and widespread corruption in the OTC marketplace. 16 If sunlight is the best of disinfectants, the over-the-counter markets are the dark, murky underworld of U.S. securities markets The Securities Investor Protection Corporation (SIPC) See SEC Press Release: SEC and FINRA Warn Investors About Penny Stock Scams Hyping Dormant Shell Companies; at: 7

8 The vast majority of the more than 10,000 firms trading in the OTC markets are lawful companies that simply fall short of the major exchanges listing requirements or are unable or unwilling to meet the high costs of regulatory compliance. These legitimate companies pay a significant price for the fraudsters they trade alongside. Manager-owners who use the opacity of the OTC markets to steal from investors raise the cost of capital to law-abiding OTC firms and left unchecked threaten the viability of this source of capital. B. Exchange-Traded Firms Managers of companies trading on the major, national exchanges (e.g. NYSE, Nasdaq, American Stock Exchange) face several disciplining mechanisms in their relationship vis-à-vis shareholders: (1) The SEC requires detailed, mandatory disclosures that provide a wealth of information to investors; (2) The major exchanges maintain strict listing requirements that not only require minimum financial thresholds but also demand specific governance standards independent audit committees, independent directors and other criteria that deter managerial abuse; (3) The shareholders of companies trading on the major exchanges, too, provide an important source of discipline. While institutional shareholders almost never trade on the OTC markets (many institutions are prohibited from trading OTC stocks) these and other sophisticated investors provide an important source of discipline to managers of firms trading on the national exchanges. Active monitoring by large-scale, sophisticated investors combined with the abundance of information demanded by regulators ensures a watchful eye on any manager action taken at the expense of shareholders. C. OTC Firms Managers of OTC firms are far less scrutinized in their interactions with investors. The SEC does not require disclosures from stocks trading in the OTC Microcap Fraud at SEC Investor Alert: Investor Alert Don t Trade on Pump-And-Dump Stock s at: 17 The full quote by Justice Brandeis is the following: "Publicity is justly commended as a remedy for social and industrial diseases. Sunlight is said to be the best of disinfectants; electric light the most efficient policeman." Louis D. Brandeis, Other People's Money and How the Bankers Use It, 92 (1914). 8

9 Markets. 18 Though many OTC firms voluntarily provide disclosures, these disclosures are generally far less than what is demanded of exchange-traded firms. While each of the national exchanges demand specific listing criteria before firms may trade on their exchange, there are no listing requirements to trade in the OTC markets; a firm can have no assets, zero revenue and still trade in the OTC markets. Shareholders of OTC stocks are nearly always retail investors lacking both the financial incentive and knowledge to properly discipline directors and officers. Managers of exchange-traded firms must justify any key decision to institutions armed with a wealth of information. OTC managers are disciplined by little more than the distant gaze of a transient retail shareholder with neither the information nor the sophisticated understanding to discern how the manager s actions impacts their investment. D. State Corporate Law as a Source of Managerial Discipline The key remaining piece in the patchwork of forces that deter abuse by corporate managers is the system of law that governs the relationship between shareholders and managers: state corporate law. Under the internal affairs doctrine, the laws of the corporation s state of incorporation govern its internal affairs the relationship between the corporation s shareholders and management. 19 On June 15, 2001, firms were provided with a new option in this choice of law regime: by incorporating in Nevada, managers could be free from liability even after breaching well-established corporate fiduciary duties to their shareholders. 1. Heterogeneous Effect of Lax Corporate Law Adopting an exceedingly lax corporate law does not impact all firms equally. For exchange-traded companies, certain managerial actions now permissible under Nevada s loose corporate law would likely remain infeasible under the web 18 After 1999, the SEC requires all firms listed in the OTC Bulletin Board to provide disclosures. The SEC does not require firms trading in the OTC Markets (previously Pink Sheets) to provide disclosures unless subject to other criteria, separate from their listing status. However, starting in 2007, the OTC Markets has taken independent action to distinguish firms trading in its marketplace based on their level of disclosure and other factors related to issuer quality. The OTCQX requires the highest levels of disclosure and a more thorough review process. The OTCQB tier requires a less stringent disclosure that was most recently become stricter in June The remaining firms in the OTC Pink have no disclosure requirements imposed by the OTC markets; nevertheless, some firms in this market choose to voluntary provide varying levels of disclosure See RESTATEMENT (SECOND) OF CONFLICT OF LAWS, 304, 307 (1971). 9

10 of disciplinary forces constraining managers. Enhancing the private benefits of control through lax corporate law is less valuable to the subset of managers obligated to disclose to their vigilant shareholders in abundant detail what they have appropriated from them. For OTC managers intent on expropriating investors, corporate law that permits wider latitude to line their personal pockets at the expense of shareholders is a meaningful addition to the corporate law menu. In addition, the decision to incorporate in a more lax legal regime will itself be less closely scrutinized by retail OTC investors than for managers that must continually justify corporate decisions to discerning institutional shareholders. IV. RELATED LITERATURE Legal scholars have long debated the merits of federalism in US corporate law: managers choose the state law that governs the internal affairs of their firm and states compete to attract incorporations. Some have argued this system leads to a race-to-the-bottom. Self-serving managers will choose state law that best meets their private needs; states will cater to the managers preferences at the expense of shareholders leading to a socially inefficient outcome. 20 Race-to-thetop advocates argue that because managers are ultimately accountable to their shareholders, managers will make decisions that best serve the firm s interests overall. States will then be competing to offer a body of law that best serves the interests of shareholders. 21 More recently, the debate regarding the desirability of state competition lay dormant as only one state appeared to be competing to attract incorporations: Delaware. Network externalities, established case law and a sophisticated 20 William L. Cary, Federalism and Corporate Law: Reflections upon Delaware, 83 YALE L.J. 663, (1974); The Race to the Bottom Revisited: Reflections on Recent Developments in Delaware's Corporation Law, 76 NW. U.L. REV. 913 (1982); Merrit B. Fox, Retaining Mandatory Securities Disclosure: Why Issuer Choice is Not Investor Empowerment, 85 VA. L. REV (1999). 21 See e.g., Frank H. Easterbrook & Daniel R. Fischel, The Economic Structure of Corporate Law (1991); Roberta Romano, The Genius of American Corporate Law (1993); Ralph K. W inter, Jr., The Race for the Top Revisited: A Comment on Eisenberg, 89 COLUM. L. REV (1989); Ralph K. W inter, Jr., State Law, Shareholder Protection, and the Theory of the Corporation, 6 J. LEGAL STUD. 251 (1977); Stephen M. Bainbridge, The Creeping Federalization of Corporate Law, REGULATION, Spring, 2003, at 31 (asserting that under state competition, oppressive regulation is impractical as firms preserve the option to exit). 10

11 judiciary with expertise in corporate law set high barriers to entry for any wouldbe competitors. 22 Delaware had cornered the incorporation market. In 2012, Michal Barzuza revived the race debate by exposing how Nevada has increased its share of out-of-state incorporations over the past decade by targeting a segment of the incorporation market with a preference for law more lax than what is offered by Delaware. 23 By offering an almost liability-free jurisdiction, managers with a strong preference for shockingly lax law may now choose Nevada rather than what had commonly been a choice between only Delaware and their home state. Barzuza (2012) raises concerns about the prospect of firms most in need of tighter constraints having the option now to choose exceedingly lax law. 24 Barzuza & Smith (2012) show empirically that Nevada-incorporated firms are more likely to have accounting restatements and aggressive accounting. After providing evidence that restatement frequency is correlated with poor governance, they assert that Nevada is disproportionately attracting firms with high private benefits and high agency costs. Kobayashi & Ribstein (2012) and Dammann (2013) have a different take on the consequences of the surge in Nevada incorporations. They remain steadfast to the rationale underlying the race-to-the-top view. If Nevada incorporation were an inefficient choice, shareholders would appropriately discount shares of Nevada-incorporated firms; managers would be deterred from making an incorporation choice that goes against the interests of the firm s shareholders. They conclude, therefore, that firms are selecting Nevada because its corporate law best fits the preferences of its shareholders. Donelson & Yust (2013) conduct a differences-in-differences analysis to measure the impact of the Nevada law change on firms already incorporated in Nevada. They find an economically significant decrease in the value of Nevada firms relative to other firms after the change in law. This decline in value among Nevada firms is concentrated in firms with the highest pre-2001 valuations and highest expected private benefits. 25 No clear consensus has emerged in the literature regarding the policy impact of Nevada s rise as an incorporation destination. What has not yet been analyzed in this literature and what may be a notable aspect of the impact of Nevada corporate law is the striking rise in Nevada incorporations in the OTC markets. 22 See, e.g., Lucian A. Bebchuk & Assaf Hamdani, Vigorous Race or Leisurely Walk: Reconsidering the Competition over Corporate Charters, 112 YALE L.J. 553, (2002); Marcel Kahan & Ehud Kamar, The Myth of State Competition in Corporate Law, 55 STAN. L. REV. 679, (2002). 23 See Barzuza, supra Note 4 at Id. 25 See Dain C. Donelson & Christopher G. Yust, Litigation Risk and Agency Costs: Evidence from Nevada Corporate Law, Working Paper (2013). 11

12 V. CHANGE IN LAW On June 15, 2001, Nevada significantly limited the scope of liability for its corporations directors and officers; this change was intended to drive new incorporations to Nevada while significantly increasing Nevada s franchise taxes. In a significant departure from Delaware, a director or officer of a Nevada corporations is liable only if: (a) his act or failure to act constituted a breach of his fiduciary duties as a director or officer; and (b) his breach of those duties involved intentional misconduct, fraud or a knowing violation of law. 26 Directors and officers of Nevada corporations face no other default liability. This sweeping liability protection became the default standard for all Nevada corporations; firms could only impose liability for other fiduciary duties after obtaining management approval. 27 In Delaware, default rules provide for the standard director duties of care and loyalty. Section 102(b)(7) of the Delaware Code permits Delaware companies to opt out of liability for breach of the duty of care conditioned on shareholder approval. 28 However, Delaware corporations may not opt out of the following: (1) Breach of the duty of loyalty; (2) Behavior that is not in good faith; (3) Improper personal benefits; and (4) Intentional misconduct, fraud, or a knowing violation of law. 29 After the Stone v. Ritter case, duties (1) (3) fall under a broader conception of the duty of loyalty. 30 The most relevant legal distinction is that while the duty of loyalty may not be contracted out of in Delaware, there is by default no liability for breaches of the duty of loyalty under post-2001 Nevada corporate law. While the Delaware exculpation requires shareholder approval and an amendment to the corporate charter to become effective, the Nevada exculpation statute is automatically applicable to all Nevada-incorporated firms. Another notable difference between post-2001 Nevada corporate law and Delaware law is that Nevada s Section (7) exculpation provision extends to officers as well as directors; (b)(7) of the Delaware code only exculpates directors from liability. 32 In addition, Nevada permits firms to indemnify directors for legal expenses for even bad faith acts as long as there is no intentional 26 See NEV. REV. STAT (7). 27 Id. 28 DEL. CODE. ANN. 102(b)(7). 29 Id. 30 Stone v. Ritter, 911 A.2d (Del. 2006). 31 NEV. REV. STAT (7). 32 DEL. CODE. ANN. 102(B)(7). 12

13 misconduct or knowing violation of law; 33 bad faith acts may not be indemnified in Delaware. 34 Along with the legal changes, Nevada also initiated an aggressive marketing campaign emphasizing its lenient corporate law and highlighting manager actions that would be legal in Nevada but actionable in Delaware. In 2003, Nevada followed through with its plan to raise incorporation revenue by raising its maximum annual tax from $85 to $11, While this is far below Delaware s $180,000 maximum annual tax, it is significantly above the negligible incorporation fees charged by all other states. 36 VI. EMPIRICAL TEST OF EFFICIENCY/INEFFICIENCY THEORY A. Efficient vs. Inefficient Nevada Incorporation The decision to incorporate in Nevada post-2001 reveals something new beyond the comparatively benign revelation of pre-2001 Nevada incorporation: a strong managerial preference for lax corporate law. This preference is not presumptively bad for shareholders. Lax corporate law may be more efficient for a subset of firms. Monitoring and enforcement by principals is itself a component of agency costs; in particular, the costs of litigation will be largely borne by the shareholders. Moreover, an excessive threat of litigation may inefficiently hamper managers from taking actions that enhance shareholder value. Shareholders of a firm with moderate agency costs and relatively high litigation costs may gain more from reducing litigation risk than what they lose by permitting wider latitude to their managers. Nevada s permissive corporate law may be moving some firms closer to their optimal balance between vigilance and laxity. If markets are efficient and shareholders are correctly pricing the incorporation choice, our presumption should be that post-2001 Nevada incorporators are choosing law that maximizes shareholder value. An alternative story is that for a subset of post-2001 Nevada incorporators, the managers are choosing Nevada because it expands their private benefits of control. Managers will accrue the personal benefits of adopting Nevada corporate law if shareholders do not discount share price to a level that exceeds the managers expected private benefits. Because managers principally decide the 33 NEV. REV. STAT (7). 34 DEL. CODE. ANN See Barzuza, supra Note 4 at Id. 13

14 state of incorporation, they may be choosing socially inefficient law that provides them with distributional gains at the expense of their shareholders. 37 B. Empirical Test of Efficient/Inefficient Sorting Theories Thus far, the law and business literature has examined the efficient vs. inefficient sorting question by focusing on Nevada-incorporated firms on the major exchanges. 38 I examine the Nevada effect by conducting a differences-indifferences test comparing Nevada-incorporated and non-nevada-incorporated OTC firms. The principal empirical test compares Nevada-incorporated firms located outside Nevada to Delaware-incorporated firms located outside Delaware in the period before and after the legal change after accounting for firm-level controls. 1. SEC Trading Suspensions as Outcome Variable As my outcome variable, I hand-collect a unique dataset comprised of all firms that have faced an SEC Trading Suspension from January 1996 through June An SEC Trading Suspension is an outcome unequivocally revealing managerial abuse to the severe detriment of shareholders. Following the 10-day Suspension, firms generally face the following sequence of events: (a) termination of trading in the OTC marketplace; (b) trading possible only on the Grey Sheets where there is no market-makers or brokerdealers a firm demoted to the Greys will typically trade at a price close to zero, and at best a small fraction of its pre-suspension price; (c) the majority of Suspension firms will have their registration revoked shortly after the Suspension, and (d) suspended firms almost never return to normal trading on their pre- Suspension marketplace. In sum, a Suspension removes the company s stock from the OTC markets and renders the firm s shares either worthless or at a price substantially below its pre-suspension price These two accounts of the Nevada effect are not mutually exclusive. The fact that a subset of managers is inefficiently choosing Nevada to line their own pockets at the expense of shareholders does not rule out the possibility that other firms may be choosing Nevada because it is efficient and best serves the interest of its shareholders. Firms may vary in the effect Nevada corporate law has on their marginal efficiency. 38 See Barzuza, supra Note 4; Supra, Kobayashi & Ribstein supra Note 5; Barzuza & Smith supra Note 5; Donelson & Yust (2012) supra Note The SEC has released the names and filings related to SEC suspensions from 1996 through the present at: sec.gov/litigation/suspensions/suspensionsarchive.html. These dates are also an appropriate range within which to gather information on trading suspensions for companies incorporated after SEC Office of Investor Education and Advocacy. Investor Bulletin: Trading Suspensions (May 2012); and data provided by: Revocations. 14

15 Because of the harsh consequences of a Suspension and its impact on existing shareholders, the SEC only imposes this punishment if there is serious, unassailable evidence that the firm s share price is out of line with the true value of the firm. Ultimately, if the insider misconduct is severe enough, the SEC will take the punitive measure of removing the stock from the marketplace and significantly harming existing the stock s existing shareholders, in the interest of protecting future investors. 2. Overview of Empirical Results Among firms incorporated during the period from January 1990 to June 2001, the probability of a Suspension is slightly higher for firms incorporated in Delaware but not located in Delaware ( OS Delaware ) firms than for both firms incorporated in Nevada but not located in Nevada ( OS Nevada ), and firms incorporated in all other states ( Other States ). Among firms incorporated during the period before the law change, 15.7% of OS Delaware firms face a Suspension, while close to 12% of OS Nevada and 12% of Other Firms face a Suspension. This difference in Suspension probability between OS Delaware and OS Nevada in the pre-period is not statistically significant after covariates are added. Among incorporations during the period after June 2001, the probability of a Suspension is 8.5% for OS Nevada firms and nearly 3% for both OS Delaware and Other Firms. Comparing OS Nevada and OS Delaware firms in the period after the legal change, the difference in the probability of Suspension is significant after including covariates at a p-value below 1%. The overall differences-in-differences estimation between OS Nevada and OS Delaware is significant after including covariates at a p-value below 1%. This diff-in-diff outcome provides strong support for the hypothesis that the change in Nevada corporate law has led to a higher propensity for bad firms defined in my analysis as firms more likely to have a Suspension to incorporate in Nevada. C. Distribution of Firm Types The major exchanges and the OTC markets differ considerably in the distribution of firm types that populate each marketplace. Firms will have a different response to the introduction of a lax corporate law option depending on where they place along this distribution. Managers with a strong preference for extracting private benefits may have more to gain from adopting Nevada corporate law than firms in which manager/shareholder incentives are more properly aligned. 1. Firm Types 15

16 Within the OTC markets, at one end of the distribution is a subset of firms set up by manager-owners with the express purpose of expropriating investors. These manager-owners never intend to set up a legitimate business. One common strategy is to directly or indirectly spread false or misleading information to artificially raise share prices and later unload their shares to naïve investors at inflated prices. This set of manager-owners will also take advantage of any opportunity to use the corporate coffers to engage in lucrative self-dealing transactions. I will classify this subset of firms lying at the extreme tail of the distribution as Type 0. A second subset of OTC firms are comprised of manager-owners that have set up a legitimate business but are willing to use deceptive tactics to make money at the expense of their investors. These manager-owners are more likely to shift focus toward self-dealing transactions if their legitimate business plans go awry. I classify this subset of firms as Type 1. Firms fall within the range between Type 0 and Type 1 depending on the extent to which they run a true business versus the extent to which their chief objective is to make money through expropriating shareholders. The remaining OTC firms range between Type 2 and Type 3. Managers above the Type 2 threshold run legitimate businesses and will not utilize deceitful tactics to steal from shareholders. Firms will differ along the Type 2 to Type 3 spectrum based on the extent to which they value shareholder wealth relative to their personal wealth. At the extreme Type 3 end of the spectrum, managers obtain equal utility from a marginal dollar added to their personal wealth as they do a marginal dollar added to overall shareholder wealth. A Type 3 firm has zero agency costs. Critical to this analysis of firm types is that where a firm lies along the spectrum is only partly dependent on the managers individual preferences. Firm type is crucially influenced by the constraints to managerial power that externally enhance incentive alignment between shareholder and owner. The manager of a large, NYSE firm who would not hesitate to steal money from investors but is unable to do so due to SEC disclosures, regulator scrutiny and watchful investors will fall within the Type 2-3 range. 2. Firm Types on the Major Exchanges On the major exchanges, all but a negligible minority of firms are above the Type 2 cutoff. Disclosure requirements and disciplinary constraints deter Type 0/Type 1 managers from entering the national exchanges. A small minority of exchange-traded firms may devolve towards Type 1 status; even these firms were likely not at Type 0/Type 1 status at their inception. Disclosure requirements and listing requirements on national exchanges would immediately reveal the illegitimacy of Type 0/Type 1 firms; managers deceitful intentions would be laid bare. In addition to the lack of transparency, 16

17 corporate scammers rely on the small float, short selling restrictions and other characteristics of the OTC markets that enable a rapid, artificial rise in share price. 3. Investor Types What also distinguishes the OTC and major exchanges is the distribution of investor types in each marketplace. The major exchanges are saturated with institutional investors and sophisticated entities that thrive on arbitrage and market inefficiencies. Even if naïve investors were easily influenced to bid up the share price of a fraudster s stock, the smart money would quickly bid the price down to its appropriate price. In the OTC markets, institutions and other smart money are largely absent, and naïve, retail investors hoping to get rich quick abound. With an abundance of investors searching for a stock that can generate huge returns over a short period, the OTC markets provide ample prey for Type 0/Type 1 tactics. D. Nevada Corporate Law Across Firm Types 1. Incorporation Choice Across Firm Types Examining incorporation choices across firm types can inform our analysis of whether managers are choosing Nevada incorporation to the benefit or detriment of their shareholders. A Type 3 firm manager will always make choices that it expects will be in the best interest of its shareholders. The choice of Nevada incorporation by a Type 3 manager would strongly suggest that Nevada incorporation best serves the interests of this set of shareholders. A self-serving incorporation choice that harms shareholders is unlikely from a high-type manager. Managers in firms with the lowest agency costs will be most severely punished for the wrong incorporation choice since their principals closely scrutinize their decisions. In addition, closely monitored, high-type managers have less to gain by adopting a legal regime that permits private gains since they have the fewest opportunities to extract private benefits. 2. Incorporation Choice Among Low Type Firms On the other hand, Type 0/Type 1 firms can be expected to choose the state of incorporation that allows them to maximize the private benefits of control. These firms are concerned solely with maximizing their personal wealth while minimizing the expected risk of liability and punishment. Additionally, an incorporation decision that reduces shareholder value is less likely to be appropriately accounted for by the investor types that own shares of Type 0/Type 1 firms. A Type 0/Type 1 firm s choice of lax corporate law should be met with 17

18 skepticism; this decision is unlikely to be in the best interest of shareholders and is likely motivated by the manager s desire to enhance her freedom to extract the private benefits of control. 3. Incorporation Choice of Firms Between Type 1-3 For firms within the Type 1 through Type 3 range, the choice of incorporation sends an ambiguous signal. These managers are neither concerned solely with how their decisions impact their ability to extract private benefits, nor are they deciding solely on the basis of how the decision impacts its shareholders. The closer firms are to a Type 3, the more likely the former consideration drives their decision. Firms closer to a Type 1 will be more selfish, and less likely to be punished, for a decision that benefits them personally at the expense of shareholders. E. Interpreting Incorporation Trends in the Exchanges and in the OTC Firms trading on the major national exchanges comprise the vast majority of firms in the highest tiers of the Type distribution. Relative to OTC firms, the exchange-traded firms have significantly lower agency costs and are thus more likely to incorporate in a state that benefits their shareholders. We can draw no firm conclusions regarding whether the 20% increase in Nevada incorporations after the legal change is in the best interest of exchange-traded shareholders. OTC firms occupy the lower end of the Type distribution including a small but important minority of Type 0/Type 1 firms. The more than three-fold increase in the rate of post-2001 Nevada incorporation among firms with the highest agency costs raises alarms but remains empirically inconclusive. A milder increase in Nevada incorporation among low agency cost firms relative to higher agency cost firms is consistent with the hypothesis that managers are choosing Nevada to enhance their private benefits of control at the expense of shareholders. But this outcome is also consistent with a different story. There may be characteristics common to OTC firms and their shareholders that cause them to differentially benefit from lax corporate law. For example, if the risk of litigation entails a higher relative cost to OTC firms than exchange-traded firms, a higher proportion of OTC firms selecting Nevada would be an efficient outcome that serves the interests of OTC shareholders. A rise in Nevada incorporations among OTC firms above the Type 1 cutoff leaves unsettled the efficient vs. inefficient sorting debate. However, OTC firms within the Type 0 to Type 1 range provide a more equivocal signal with their incorporation choice. These firms are solely concerned with extracting as much as possible from naïve OTC investors. If a significantly higher proportion of Type 0/Type 1 firms are choosing post-2001 Nevada incorporation, we can reasonably 18

19 assert that for this subset of firms, Nevada incorporation is likely to be socially inefficient and harmful to OTC investors. Through revealed preference, a meaningful shift towards post-2001 Nevada incorporation by Type 0/Type 1 managers signals that exceedingly lax corporate law allows them to more effectively achieve their objectives. If the objectives of scamming manager-owners are socially undesirable, a legal change that enhances these objectives is socially undesirable. A sizeable shift towards post-2001 Nevada corporate law among the managers unconcerned with shareholder wealth also tips the scales in favor of less benign explanations for the movement of OTC firms above the Type 1 cutoff towards post-2001 Nevada incorporation. 1. Trading Suspensions and Firm Type The outcome variable in my empirical analysis is intended to detect firms in the Type 0/Type 1 category. A firm will never face an SEC Trading Suspension due to mere carelessness or inadvertent error. A firm will face a Suspension only if its insiders are believed to be intentionally stealing from their investors. Therefore, while not all Type 0/Type 1 firms will face a Suspension, only firms in the Type 0/Type 1 category are at risk for a Suspension. Firms that may have chosen Nevada for benign or shareholder wealth-enhancing reasons will never face a Suspension. VII. EMPIRICAL ANALYSIS A. Objectives of the Empirical Tests The overarching goal of my empirical analysis is to shed some light on the normative implications of adding a lax liability option to the menu of corporate law choices in the OTC markets. The extant literature on Nevada incorporation suggests that if high agency cost firms are disproportionately choosing Nevada, this provides strong evidence for the pessimistic Nevada effect: managers are choosing lax law to expropriate their shareholders. Applying this logic, the striking rise in the rate of post-2001 Nevada incorporation in the OTC markets firms with the highest agency costs is itself strong support for the inefficiency hypothesis. I assert that high agency costs firms disproportionately opting for Nevada laxity is itself inconclusive but does provide a clue that can direct us towards a more convincing empirical test. The collection of firms choosing Nevada law and sharing in the attribute of high agency costs are also likely to share other key characteristics in common. These other unifying characteristics could conceivably make Nevada incorporation an efficient choice for this subset of firms. To rule out the plausibility of efficiency-based Nevada incorporation, I must refine the group of high agency cost firms further. A subset of firms in which 19

20 managers incorporation decision is not guided by efficient, shareholder valuemaximization will have the following characteristics: (1) Its managers are concerned exclusively with maximizing their personal gains; (2) Its managers have demonstrated disregard for the interests of shareholders; (3) Its shareholders will be unlikely to appropriately discount an incorporation choice that is against their interests. If among post-2001, Nevada-incorporated firms, a significantly higher proportion of firms fulfill criteria (1) (3) above, there is strong support that at least for this subset of firms these firms are unlikely to be choosing Nevada for reasons motivated by enhancing shareholder value. B. Outcome Variable With no readily available data that meets criteria (1) - (3), I have constructed a dataset of firms that have faced an SEC Trading Suspension between A discussion of why this outcome variable satisfies the criteria above is provided in Part VI.B(1) of this paper. To demonstrate the severity and permanence of an SEC Trading Suspension, I present the following data: 1,246 firms have had their trading suspended during the period from January 1, 2010 through April 3, Of these, 774 (62.1%) have had their registration revoked; the remaining 472 (37.9%) all continue to trade on the Grey Sheets. In Appendix A, I have included a list of all companies that have had an SEC Trading Suspension between September 17, 2012 and May 31, This list includes information regarding each firm s current state (e.g. Grey Sheets, Revoked) and their trading price before and after the suspension. C. Shell Suspensions The SEC suspends the trading of one additional group of firms but with a different objective. On a few occasions, the SEC will suspend in bulk e.g. 379 firms on May 14, 2012 a large group of shell firms that the SEC is concerned may be a vehicle for future scams. This set of trading suspensions is in furtherance of what the SEC has termed Operation Shell Expel. 41 I have not included these suspensions in my analysis since this is an outcome intended to deter potential future fraud rather than to punish existing manager expropriation. Shell companies do not meet my criteria for a bad firm. As an 41 See 20

21 additional safeguard, by including only SEC-filing OTC stocks, shell companies that have their trading suspended will not show up in my universe of firms. 42 D. Universe of Firms The universe of firms I use to compare the probability of suspension across groups is crucial to the reliability of my empirical test. Any form of selection bias would be problematic. Through Mergent Online, I was able to construct a dataset that contains every firm in the Mergent database that: (a) has traded in the OTC markets, and (b) has provided disclosures to the SEC. What concerned me most in constructing this dataset was the potential for survivorship bias. Inactive firms comprise a critical part of my analysis; firms that have had their trading suspended will no longer be actively trading in the OTC marketplace. In addition, if firms could drop out of the database upon dissolution, this too could skew my results. By including firms in Mergent s Inactive database, all firms that have at any time traded in the OTC marketplace regardless of their current status are included. Repeated assurances from the individuals who help produce the Mergent Online data provided me with confidence that firms will never drop out of its database; a set of firms I was looking at for a particular incorporation year would indeed comprise every SECdisclosing firm that had incorporated in that year. Mergent also lists the year and state of incorporation for every firm in its database. If any state changes its state of incorporation, the year and state of incorporation will be listed for all reincorporations. Because the state and year of incorporation is crucial to my analysis, I hand-checked a random sample of 50 firms by locating their incorporation filings. The data provided by Mergent was consistent with the state and year of incorporation in the documents I reviewed on EDGAR for each of the 50 firms. E. Incorporation State Across Time My dataset comprises all SEC-disclosing OTC firms incorporated in the United States and formed and initially incorporating between I divided these firms into two overall groups cutoff at a date of incorporation before and after June 15, 2001 (the effective date of the Nevada law change). Because only the year is provided in the data output from Mergent, I individually checked the date of incorporation for all firms incorporated in 2001 grouping them into the pre- or post- group depending on the precise incorporation or reincorporation 42 My empirical results are robust to including all shell companies. In fact, the empirical results providing even stronger results supporting the increase in the probability of firms having a trading suspension if incorporated in Nevada post

22 date. In total, there are 2,150 firms in my pre-period group and 1,509 firms in my post-period group. Within these groups, I identified firms based on their state of incorporation; I grouped Nevada, Delaware and all other states into three sub-groups within each of the two periods (six total subgroups). I distinguish Delaware because the choice of Delaware incorporation provides a more appropriate control group for the Nevada-incorporated treatment group. While the vast majority of firms in the Other States category have simply chosen the default option of incorporating in their home state, firms choosing Delaware or Nevada and paying the applicable higher charter fees and incorporation taxes are expressing a relatively stronger preference for their incorporation state. Table 1 and Figure 3 display the proportion of overall OTC incorporations by state: TABLE 1: PERCENTAGE OF TOTAL INCORPORATIONS BY STATE IN PERIOD BEFORE AND AFTER NEVADA LAW CHANGE DELAWARE NEVADA OTHER PRE- PERIOD POST- PERIOD % (986) 25.58% (385) 24.52% (528) 59.34% (878) 29.68% (639) 16.08% (242) 22

23 FIGURE 3: PERCENTAGE OF TOTAL INCORPORATIONS BY STATE BEFORE AND AFTER NEVADA LAW CHANGE 100% 90% 80% OTHER STATES OTHER STATES % OF OTC INCORPORATIONS 70% 60% 50% 40% 30% 20% DELAWARE DELAWARE NEVADA OTHER STATES DELAWARE NEVADA 10% NEVADA 0% Focusing the Analysis on Out-of-State Incorporations I narrowed the Delaware and Nevada groups further by identifying those not incorporating in their home state defined earlier as OS Delaware and OS Nevada. This refinement was important for two reasons. First, Nevada/Delaware firms will be more likely to incorporate in their home state irrespective of the change in Nevada law. This may confound my identification strategy since the Nevada or Delaware incorporation decision is not necessarily being made through a channel consistent with my research question. Firms incorporating in Nevada or Delaware but outside their home state are expressing a strong preference that cannot be explained by convenience or geography; this allows me to more carefully distinguish the characteristics common to firms choosing a particular state of incorporation. Firm location influences the incorporation state through a different channel that can potentially confound the empirical analysis. The second reason to compare only out-of-state incorporations is that there may be characteristics that distinguish firms choosing to incorporate at home 23

24 versus out-of-state. For example, smaller and less developed firms may, as a group, be more likely to incorporate in their home state because of the higher fees and incorporation taxes of incorporating out-of-state. These smaller firms may also be more (or less) likely to have a Suspension. Because more firms are based in Nevada than in Delaware, the group of overall Delaware firms is likely to contain a higher proportion of out-of-state firms. Comparing only out-of-state incorporators allows me to more cleanly identify incorporation decisions influenced by the Nevada law change. Focusing exclusively on out-of-state incorporations, Table 2 below lists the proportion of firms in each state as a percentage of all firms incorporated in the pre- and post- periods. TABLE 2: OUT-OF-STATE NV/DE INCORPORATIONS AS A PERCENTAGE OF ALL OTC INCORPORATIONS OS OS DELAWARE NEVADA PRE- PERIOD POST- PERIOD % (981) 25.25% (380) 22.30% (480) 52.56% (791) F. Differences-in-Differences Estimation The empirical question I investigate is whether the change in Nevada law has caused a higher proportion of bad firms to incorporate in Nevada. My definition of a bad firm is one in which its managers expropriate their shareholders for private gain. I use firms that have suffered a Suspension as a proxy for this definition of bad firm. Looking solely at the difference in my outcome variable across states does a poor job of identifying a causal effect since there are differences other than the law change across states. Similarly, looking solely at the changes in the outcome variable in the treated location (Nevada) before and after June 2001 could be caused by other changes in Nevada over the relevant time period. Using a diff-in-diff design, I control for some of these concerns. Assuming no major differences in trends between states over time and in the absence of the Nevada law change the difference in Suspension probability between Nevada and non-nevada firms in the pre-period would be expected to persist in the post period. With this counterfactual in hand, the difference in the difference 24

25 between Nevada and non-nevada after the law change is presumed to be caused by the law change (the treatment). In its most basic form, the estimate of the treatment is the following equation: with: [(Pr(Susp.) 11 - Pr(Susp.) 01 ) - (Pr(Susp.) 10 - Pr(Susp.) 00 )] Pr(Susp.) it = probability of the binary outcome Suspension for group i at time t. i = 0 for the control group (non-nevada), i = 1 for the treatment group (Nevada). t = 0 for the pre-period (before June 2001), t = 1 for the post period (after June 2001). For the OLS estimate, using Yit as the outcome (Y it = 1 if Suspension, Y it = 0 if no Suspension) for firm i in period t (with t defined again as t = 0 for preperiod, t = 1 for post-period), the model is the following: Y it = β 0 + β 1 X i + β 2 T t + β 3 X i * T t + ε it In the model above, X i is a dummy variable taking the value of 1 if the firm is in Nevada, and T t is a dummy variable taking the value 1 if in the period after the law change. β 3 is the D.I.D. estimator the coefficient on the interaction between X i and T t and our key variable of interest; β 3 takes a value of 1 for Nevada firms incorporated after the law change. When I include covariates, with CV = vector of all covariates, the model is the following: Y it = β 0 + β 1 X i + β 2 T t + β 3 X i * T t + β 4 CV + ε it 1. Differences-in-Differences Before Adding Covariates Armed with our complete universe of SEC- filing OTC firms from separated by incorporation state and incorporation period, we can conduct our baseline difference-in-differences estimation. By including no controls at this stage, this estimation will provide only a crude measure of the changes across states in the pre- and post- periods. The tables below display the results: 25

26 TABLE 3: BASELINE DIFFERENCES-IN-DIFFERENCES ESTIMATION BETWEEN OUT-OF-STATE NEVADA AND OUT-OF-STATE DELAWARE WITHOUT COVARIATES DELAWARE PRE-PERIOD NEVADA PRE-PERIOD DIFFERENCE PRE-PERIOD DELAWARE POST-PERIOD NEVADA POST-PERIOD DIFFERENCE POST-PERIOD SUSPENSION STD. ERROR T-STATISTIC P > T ** *** 0.000*** *significance at 0.1 level, **significance at 0.05 level, ***significance at 0.01 level D.I.D. TABLE 4: BASELINE DIFFERENCES-IN-DIFFERENCES BETWEEN OVERALL NEVADA AND OVERALL DELAWARE WITHOUT COVARIATES DELAWARE PRE-PERIOD NEVADA PRE-PERIOD DIFFERENCE PRE-PERIOD DELAWARE POST-PERIOD NEVADA POST-PERIOD DIFFERENCE POST-PERIOD SUSPENSION STD. ERROR T-STATISTIC P > T *** *** 0.000*** *significance at 0.1 level, **significance at 0.05 level, ***significance at 0.01 level D.I.D. TABLE 5: BASELINE DIFFERENCES-IN-DIFFERENCES BETWEEN OUT-OF-STATE NEVADA AND ALL OTHER STATES WITHOUT COVARIATES OTHER STATES PRE-PERIOD NEVADA PRE-PERIOD DIFFERENCE PRE-PERIOD OTHER STATES POST-PERIOD NEVADA POST-PERIOD DIFFERENCE POST-PERIOD SUSPENSION STD. ERROR T-STATISTIC P > T *** 0.045** *significance at 0.1 level, **significance at 0.05 level, ***significance at 0.01 level D.I.D. The principal baseline estimations comparing OS Nevada and OS Delaware in Table 3 show a sizeable shift in the probability of suspension in the pre- and post- period. OS Nevada firms incorporated in the period prior to the legal change were 24.2% less likely to have a Suspension than OS Delaware firms incorporated prior to the legal change; OS Delaware firms had a 15.7% probability of a Suspension while OS Nevada firms had a 11.9% probability of a Suspension. The relative probability of suspension is considerably different in the period after the legal change. OS Nevada firms incorporated in the period after the legal change are 193% more likely to have a Suspension than OS Delaware firms 26

27 incorporated after the law change; OS Delaware firms had a 2.9% probability of a Suspension while OS Nevada firms had an 8.5% probability of a Suspension. A high t-statistic = 3.63 on the β 3 coefficient is strong evidence against equality of the differences. Since I have not yet controlled for confounding variables and alternative mechanisms that may be influencing this outcome, I can draw no firm empirical conclusions about the causal effect of the Nevada law change. The estimation comparing all Nevada- incorporated to all Delawareincorporated firms provides marginally stronger but nearly identical results to that for the out-of-state estimation discussed above with a t-statistic = The estimation comparing OS Nevada with all Other States also provides strong support against equality of the differences with a t-statistic = 2.00 and significance at a p-value = This result is not as strong as the estimations relative to Delaware. This outcome may be partly caused by the significantly smaller number of post-period observations for firms in the Other States category only 242 (16%) of firms in the post-period are not incorporated in Nevada or Delaware. G. Differences-in-Differences with Covariates To substantiate the results of the baseline diff-in-diff estimations, I must include covariates to control for alternative mechanisms that could be driving the results. If, for example, there are other changes in the composition of firms incorporating in Nevada/Delaware after June 2001 that make them more/less likely to have a Suspension, I would need to account for these alternative channels before concluding the Nevada law change is likely to be causing the results. 1. Description of Covariates As was touched upon earlier in the discussion of narrowing the empirical analysis to out-of-state incorporations, firm geography impacts incorporation choice. Firms incorporated in Nevada are far more likely to choose Nevada incorporation but should not have a higher probability of a Suspension based on my causal story. If, for example, firms headquartered in Nevada are more likely to have the qualities that lead to a Suspension, this could be confounding the results of my baseline estimation. Using a diff-in-diff estimation partly addresses this concern since the influence of Nevada-incorporated firms located in Nevada could be expected to affect both the pre- and post-period results. However, there may be changes over time in the composition of firms located in Nevada that make them more likely to face a Suspension in the post-period. In sum, Nevada-based firms are firms more likely to choose Nevada but not through the channel that is consistent with my hypothesis: firms choosing Nevada after 2001 due to a preference for permissive 27

28 corporate law. I include a dummy variable equal to 1 if the firm s home state is Nevada. Firms with a closer proximity to Nevada have also been shown to be more likely to choose Nevada incorporation. I follow Barzuza & Smith (2012) and include a dummy variable that takes a value of 1 if the firm s state of headquarters is located West of the Mississippi River. I also include a covariate for a firm s age: the log of the number of years that have passed since the company was founded. It turns out that within our universe of firms, older firms are significantly more likely to face a Suspension. This could simply be due to the longer time horizon during which the firm has faced the possibility of a Suspension. For example, a firm incorporated in 2008 will have a relatively short time period within which to engage in actions that could be grounds for a Suspension and get investigated and punished by the SEC. The longer time horizon for Suspension also may explain the lower overall probability of a Suspension in the post-period. Certain industries may be populated with a higher number of firms engaging in fraud and corruption. I identify firms based on the first two digits of their SIC code and separate the firms based on where their two-digit code falls within the ten SIC code categories set forth in the World Business Directory. 43 I construct dummy variables for each of the first nine of these industry categories. Foreign firms adopting U.S. corporate and securities law (what has been termed bonding ) have many characteristics that distinguish them from USbased firms. There is some literature indicating that such firms are more likely to engage in corrupt, fraudulent practices while other observers suggest foreign firms trading in OTC markets are less likely to be corrupt. 44 I include a dummy variable equal to 1 if the firm lists its headquarters outside the United States. Firms running illegitimate business often have very little to show in real assets. The SEC specifically mentions low assets as a red flag for firms likely to engage in microcap fraud. I include a variable for the log of the firm s total assets. 2. Results of Differences-in-Differences with Covariates The following tables display the results for my diff-in-diff analysis for the relevant treatment and control groups after all covariates have been included: 43 World Business Directory: 44 See Cross-Border Reverse Mergers: Causes and Consequences, Jordan Siegel and Yanbo Wang (2013). 28

29 TABLE 6: DIFFERENCES-IN-DIFFERENCES BETWEEN OUT-OF-STATE NEVADA AND OUT-OF- STATE DELAWARE WITH COVARIATES DELAWARE PRE-PERIOD NEVADA PRE-PERIOD DIFFERENCE PRE-PERIOD DELAWARE POST-PERIOD NEVADA POST-PERIOD DIFFERENCE POST-PERIOD SUSPENSION STD. ERROR T-STATISTIC P > T *** 0.001*** *significance at 0.1 level, **significance at 0.05 level, ***significance at 0.01 level D.I.D. TABLE 7: DIFFERENCES-IN-DIFFERENCES BETWEEN NEVADA AND DELAWARE WITH COVARIATES DELAWARE PRE-PERIOD NEVADA PRE-PERIOD DIFFERENCE PRE-PERIOD DELAWARE POST-PERIOD NEVADA POST-PERIOD DIFFERENCE POST-PERIOD SUSPENSION STD. ERROR T-STATISTIC P > T *** 0.000*** *significance at 0.1 level, **significance at 0.05 level, ***significance at 0.01 level D.I.D. TABLE 8: DIFFERENCES-IN-DIFFERENCES BETWEEN OUT-OF-STATE NEVADA AND ALL OTHER STATES WITH COVARIATES OTHER STATES PRE-PERIOD NEVADA PRE-PERIOD DIFFERENCE PRE-PERIOD OTHER STATES POST-PERIOD NEVADA POST-PERIOD DIFFERENCE POST-PERIOD SUSPENSION STD. ERROR T-STATISTIC P > T ** 0.055** *significance at 0.1 level, **significance at 0.05 level, ***significance at 0.01 level D.I.D. Our principal diff-in-diff estimation comparing the difference between the probability of a suspension between OS Nevada and OS Delaware in the pre- and post- periods and after including all covariates provides us with very strong results. Our D.I.D. coefficient yields a t-statistic = 3.34 and significance at a p- value of 0.001; this is strong empirical support for a positive difference in the differences. We obtain a similar outcome comparing all Nevada and all Delaware firms with a t-statistic = These results provide us with strong empirical support that the change in Nevada law has caused a greater proportion of bad firms firms more likely to have a Suspension to incorporate in Nevada. Our diff-in-diff results comparing OS Nevada with Other Firms yields a positive outcome with a t-statistic = 1.92 and a p-value of This result is not 29

30 as strong as our principal estimation comparing Nevada with Delaware. As was mentioned earlier, this outcome is partly due to the much smaller number of postperiod control observations. 3. Examining the Effect of the Covariates Across Periods To take a closer look at the effect of the covariates on the outcome variable in the period before and after the legal change, I display below the results of a logit regression with Suspension as the outcome variable and a dummy variable taking a value of 1 for OS Nevada incorporators and 0 for OS Delaware incorporators as the explanatory variable (column in bold). TABLE 9: PRE-PERIOD LOGIT REGRESSION WITH SUSPENSION AS DEPENDENT VARIABLE Independent Variable Coefficient Std. Error Z-Statistic P > z NV/DE Log Assets Foreign NV Proximity Log of Firm Age * Ind. Agriculture Ind. Mining Ind. Construction Ind. Manuf Ind. Transport Ind. Wholesale Ind. Retail Trade Ind. Financial Ind. Services Notes: This is a logistic regression that includes all Out-of-State Nevada incorporated firms and all Out-of-State Delaware incorporated firms incorporated from January 1, 1990 through June 14, Dependent variable takes a value of 1 if the firm has faced an SEC Trading Suspension. NV/DE variable takes a value of 1 if firm is incorporated in Nevada and 0 if incorporated in Delaware. Foreign variable takes a value of 1 if firm is headquartered outside the U.S. NV Proximity takes a value of 1 if firm is headquartered west of the M ississippi. Industry dummies use the first digit of the World Business Director s Standard Industry Classification Code. All firms in the Agriculture and Retail Trade industries had no suspensions; therefore, both industry dummies perfectly predicted No Suspension. *significance at 0.1 level, **significance at 0.05 level, ***significance at 0.01 level 30

31 TABLE 10: POST-PERIOD LOGIT REGRESSION WITH SUSPENSION AS DEPENDENT VARIABLE Independent Variable Coefficient Std. Error Z-Statistic P > z NV/DE *** Log Assets Foreign ** NV Proximity *** Log of Firm Age *** Ind. Agriculture Ind. Mining Ind. Construction *** Ind. Manuf Ind. Transport Ind. Wholesale Ind. Retail Trade Ind. Financial Ind. Services Notes: This is a logistic regression that includes all Out-of-State Nevada incorporated firms and all Out-of-State Delaware incorporated firms incorporated from June 15, 2001 through December 31, Dependent variable takes a value of 1 if the firm has faced an SEC Trading Suspension. NV/DE variable takes a value of 1 if firm is incorporated in Nevada and 0 if incorporated in Delaware. Foreign variable takes a value of 1 if firm is headquartered outside the U.S. NV Proximity takes a value of 1 if firm is headquartered west of the Mississippi. Industry dummies use the first digit of the World Business Director s Standard Industry Classification Code. All firms in the Agriculture and Retail Trade industries had no suspensions; therefore, both industry dummies perfectly predicted No Suspension. *significance at 0.1 level, **significance at 0.05 level, ***significance at 0.01 level A quick scan of the covariates during the pre-period shows that only firm age is predictive of the probability of Suspension (i.e. older firms within the range are more likely to have a Suspension). For the post-period, there is, of course, a strong result for Nevada incorporation predicting a higher probability of suspension. Among the covariates, age again increases the probability of a Suspension. In addition, firms headquartered in a foreign country and, surprisingly, headquartered in Nevada and closer to Nevada are less likely to have a Suspension; this result strongly suggests that Nevada location is not driving the higher proportion of Nevada-incorporated firms facing a Suspension Among industries, firms in the mining and construction are more likely to have a Suspension. Other industry variables had no significant effect on the probability of a suspension. Mining companies in the OTC are often used as vehicles for fraud since false and misleading information about potential mining discoveries has been an effective tactic to artificially raise the stock price. 31

32 H. Suspension Probability Year-by-Year The data presented thus far has displayed the change in the probability of a Suspension across states for firms incorporated over a multi-year block of time before and after the Nevada change. Figure 4 below presents the probability of suspension between OS Nevada and OS Delaware among firms incorporated in each specific year of my sample: FIGURE 4: SUSPENSION PROBABILITY FOR OUT-OF-STATE NEVADA AND OUT-OF-STATE DELAWARE FIRMS BY YEAR OF INC. Looking at firms incorporating in a particular year, we see that prior to the law change, there is only a modest difference in the probability of Suspension between OS Nevada and OS Delaware firms. The difference in Suspension probability widens in the years after the law change before trending downward for both states in The graph below was produced using the results of a year-by-year logit regression in which the explanatory variable is a dummy variable taking the value 1 for OS Nevada and 0 for OS Delaware and including all control variables. The graph displays the z-statistic corresponding to the coefficient for the explanatory variable comparing Nevada and Delaware firms by the year of incorporation. 32

33 FIGURE 5: Z-STATISTICS FOR OUT-OF-STATE NV INC. VS. OUT-OF-STATE DE INC. DUMMY VARIABLE AS A PREDICTOR OF SUSPENSION Z-score for Dummy Coefficient on OS NV Inc. Linear (Z-score for Dummy Coefficient on OS NV Inc.) H. Migrations Between Nevada and Delaware I have focused my empirical analysis on comparing out-of-state firms incorporated in Nevada and Delaware. Firms changing their incorporation state from Nevada to Delaware or vice versa are expressing a particularly strong preference for the corporate law of the state they are reincorporating into. In the pre-period, there are 19 reincorporations from Nevada to Delaware; 2 (10.53%) of these 19 reincorporations had a Suspension. In the pre-period, there are 17 incorporations from Delaware to Nevada; none of these 17 firms had a Suspension. In the post-period, 27 firms migrated from Nevada to Delaware; none of these 27 firms had a Suspension. In the post-period, 16 firms migrated from Delaware to Nevada; 3 (18.75%) of these firms had a Suspension. Though the Nevada-Delaware migrations are a small sample of firms, the differences-in-differences within this group is additional empirical support for the hypothesis that the Nevada law change has increased the proportion of bad firms incorporating in Nevada. The data for these migrations is provided in Table 11: 33

34 TABLE 11: REINCORPORATIONS BETWEEN NEVADA AND DELAWARE AND THE PROPORTION RECEIVING A SUSPENSION PRE-PERIOD NV TO DE PRE-PERIOD DE TO NV POST-PERIOD NV TO DE POST-PERIOD DE TO NV TOTAL FIRMS WITH SUSPENSION SUSPENSION PERCENTAGE 10.53% % Lastly, I focus my inquiry on reincorporations overall. Migrations between Nevada and Delaware most acutely distinguish my treatment from control groups since such migrations indicate both a strong preference for one state and a strong preference against the other. However, this comprises a relatively small sample of firms. Expanding to the set of all reincorporation to Nevada and to Delaware directs the analysis to a broader set of firms that have expressed a particularly strong preference for the state they are reincorporating into. The results for this inquiry are in Table 12 below: TABLE 12: OVERALL REINCORPORATIONS TO NEVADA AND DELAWARE AND THE PROPORTION RECEIVING A SUSPENSION Pre-Period Reinc. to DE Pre-Period Reinc. to NV Post-Period Reinc. to DE Post-Period Reinc. to NV Total Total Suspensions Percentage w/suspension Overall Suspension % % 2.82% 2.13% 13.51% 15.7% 11.9% 2.9% 8.5% Difference % 9.08% 0.67% +4.99% The table above shows that reincorporators to Nevada and Delaware more strongly exhibit the Suspension rates and spreads observed in the overall sub- 34

35 groups. Among reincorporations, there is a significantly higher Suspension rate for pre-period Delaware reincorporators and post-period Nevada reincorporators. This outcome supports the notion that bad firms were expressing a strong preference to reincorporate to Delaware before the law change but redirect this preference strongly to Nevada in the post-period. Reincorporators to Nevada before the law change are far below the pre-period Suspension rate; reincorporation to Nevada becomes a bad signal only after the Nevada law change. VIII. CONCLUSION Expropriating shareholders in the over-the-counter markets is a multi-billion dollar market funded by naïve, retail investors. Fraudsters seek refuge in these markets because the sources of light that can expose their crimes remain dim. Complementing the pervasive laxity in the OTC markets with the option of exceedingly lax corporate law takes away a vital layer of protection from the most vulnerable targets of corporate crime. With the most egregious perpetrators of microcap fraud increasingly incorporating in Nevada, the Nevada legislators most ominous fears have come true: the millions Nevada has gained in incorporation revenue may be coming at the price of the billions stolen from the victims of the corporate criminals incorporating in Nevada. 35

36 APPENDIX A List of all firms that received an SEC Trading Suspension from 9/17/12 through 5/31/13. Firms are separated based on their status as of 5/31/ GROUP 1: SEC SUSPENSION - NOT YET IN GREY SHEETS: 46 Data provided by: 36

37 GROUP 2: SEC SUSPENSION - SENT TO GREY SHEETS AND HAVE TRADED: 37

38 GROUP 3: SEC SUSPENSION - SENT TO GREY SHEETS AND HAVE NOT TRADED: 38

39 GROUP 4: SEC SUSPENSION - REVOKED A-F: 39

40 GROUP 4 (CONT D): SEC SUSPENSION - REVOKED G-Z: 40