We Are Shocked
Eliot Spitzer's year-long criminal investigation of Merrill Lynch, concluded on May 21, 2002, with a $100 million settlement, has uncovered persuasive evidence of misconduct. A flood of Merrill Lynch office e-mails demonstrates that superstar analyst Henry Blodget publicly hyped stocks that he described, internally, as "pieces of junk," "crap," and "POS" (shorthand for an epithet unfit for a family newsletter). Merrill Lynch's nominal five-point stock ranking scale effectively contracted to three points—"Hold," "Accumulate," or "Buy"; the firm's Internet analysts never issued a "Reduce" or "Sell" recommendation. It appears that Merrill Lynch's misrepresentations were driven by a massive conflict of interest: analysts issued favorable recommendations to retain, or obtain, lucrative investment business from the covered companies. Contrary to the firm's representations to investors, no "Chinese wall" separated analysts and retail sellers from investment operations. Research analysts were compensated in part on the basis of their contributions to the firm's investment business.
While none of this should surprise reasonably informed investors, the Spitzer investigation and its political momentum have rattled the investment community itself. Merrill Lynch and its competitors are accustomed to dealing with federal regulators (in particular, the Securities and Exchange Commission), rather than ambitious state attorneys general. That, they say, is as it should be. National and international securities markets, and investors operating in those markets, need uniform, reliable rules. Markets must not be placed at the tender mercies of parochial state officials and be subjected to burdensome, duplicative state regulations.
Buoyed by his discovery of scandalous practices that eluded federal regulators, Mr. Spitzer dismisses the industry's self-serving complaints. His investigation, he says, "is a consequence of federalism. The whole new federalism approach vaunted by the Bush administration and the Reagan administration was designed to empower state securities regulators. That's what I'm doing." Throughout his investigation, the empowered Spitzer advertised his intention to seek much more than just punishment for Merrill Lynch. His objective is a "restructuring" of the nation's financial industry.
In media interviews Spitzer has elaborated on his commitment to federalism. The federal Securities and Exchange Commission, he maintains, is beholden to the very interests—securities exchanges, brokerage firms, investment houses, accounting firms, and corporate interests—that it is supposed to regulate. The New York attorney general, in contrast, represents the people, or at least the people of New York. Confronted with SEC abdication, New York will shake the agency out of its lethargy. In 2001, after small investors who relied on the analysts' glowing recommendations had lost their shirts in the Internet and telecom meltdown, SEC chairman Harvey Pitt—a former corporate lawyer with close ties to his former clients in the regulated industries—did next to nothing, except to mumble vaguely about the need for more informative disclosure statements. Only after Mr. Spitzer's dramatic discoveries did the SEC hop on the regulatory bandwagon and issue somewhat stricter rules concerning analysts' compensation, disclosures, and trading practices.
Mr. Pitt may indeed lack the appropriate distance from the regulated industries. Still, the notion that a state attorney general should push the national government into action perverts federalism. Putting aside that even captured federal bureaucrats may have independent, respectable reasons for doing nothing, our system of checks and balances is intentionally designed to impede federal action. Spitzerian policy entrepreneurship leaves Congress and the SEC no choice but to intervene, lest the securities markets be regulated into the ground by fifty ambitious state attorneys general whose agendas conflict in all respects but one—headline hunting. A state official's power to drive national action in this officious fashion is a power to preempt the national government. It is federalism upside-down.
Harvey Pitt has for now managed to rope Spitzer and regulators from eleven additional states into a federal-state "task force" to develop a broadly acceptable regulatory scheme. Investors, Pitt implores, need uniformity above all else. But the odds that the task force will devise a sensible regulatory arrangement are nil. In any event, the task force's handiwork will provide at most a temporary respite from state regulation on top of the federal rules.
Pimps and Policies
Regulatory options range from robust analyst disclosure requirements to none at all; from a government-mandated separation of investment banking and analysis to industry self-regulation. Plausible arguments exist for all those positions—and for pretty much anything in between. Almost certainly, however, a uniform regime, designed in an environment of demagogic one-upmanship, will come out wrong.
Clearly, brokerage and investment firms have a conflict of interest. Since investment activities (such as initial public offerings and secondary market issues) are highly profitable and research and analysis lose money, the former will eventually compromise the latter. Chinese walls will mask rather than preclude abuses, and even the most conscientious analysts will in the end surrender to the rainmakers' dictates. A press release-style document on Mr. Spitzer's website, egregiously mischaracterized as his investment protection chief's legal "affidavit" and calculated to cast Merrill Lynch and its employees in the worst possible light, actually illustrates the analysts' efforts to maintain professional standards. One exasperated analyst complained that she did not "want to be a whore for f-ing management" of a Merrill Lynch investment client. "We are losing people's money and I don't like it." Sophisticated investors may be aware of, or at least suspect, the industry's conflicts of interest. But they may also have believed that Henry Blodget and his employer would care about their reputation and thus play it reasonably straight. The fact that that safeguard—essential to rational market operations—seems to have failed may well warrant some form of regulation.
On the other hand, mandatory disclaimers ("We are all whores.") or even a forced divestiture of investment banking may prove counterproductive. While Spitzer has zeroed in on the industry's in-house conflicts, analysts also operate under "buy-side" pressures to tout stocks—for example, from portfolio managers. Regulation that fails to capture and suppress those or still more subtle "conflicts" and biases (such as an analyst's genuine fondness for "his" industry) may easily create a moral hazard and imbue investors with a false sense of security.
Mandatory disclosure, moreover, often suppresses information. The alternative to disclosing everything to everyone all at once is not to disclose anything to anybody, at least not before the information has become worthless. Amidst the hype and touting, Merrill Lynch analysis contains valuable information, and it comes to light only and precisely because of the firm's conflict of interest. Now that stocks are being "offered," 24/7, to an ever broader public, fewer and fewer investors (relatively speaking) may be capable of separating the information from the hype. That problem may warrant a governmental response—for example, public information services. But regulation that makes everyone an ignoramus cannot possibly enhance market transparency.
In short, we know very little about the optimal level of financial disclosure regulation. As part of its $100 million bargain with Spitzer, Merrill Lynch has agreed to implement certain policies to limit in-house conflicts of interest. (For example, analysts will no longer receive payments directly from investment-banking revenues.) But we do not know whether this arrangement reflects Spitzer's sense of sound public policy. Over the course of his investigation, he flirted with far more drastic demands, up to a full institutional separation between analysts and investment bankers. In truth, Spitzer's idea of efficient regulation is whatever deal Merrill Lynch would accept for a dismissal of criminal charges. Mr. Pitt has repeatedly revised his proposed approach in response to the latest Spitzer press release; his idea of optimal regulation is whatever deal Mr. Spitzer will accept.
Legislators who as little as a year ago sponsored bills for the careful study of analyst disclosure regulation are now contemplating criminal penalties for misleading investment advice—on what basis, they will not say. The House of Representatives would courageously delegate further regulatory responsibility to the SEC. In the Senate, Illinois Republican Peter G. Fitzgerald has proposed substantive disclosure standards. His bill, however, excludes from coverage anybody who has anything to do with the sale of securities (including Merrill Lynch)—a drafting error that, while inadvertent and curable, inspires little confidence about the senator's comprehension as to why or how his proposal might benefit investors and markets.
Preemption, Forward and Reverse
Cluelessness is a powerful argument against uniform federal rules. The strongest argument for federal intervention is that it constitutes the only alternative to regulatory balkanization. Merrill Lynch and its competitors have customers in all fifty states, and since each state applies its own fraud and consumer protection laws to securities transactions, the investment firms will be subject to fifty different state laws. The law of the most aggressive, regulation-minded state will govern the firms' obligations—until another grandstanding attorney general decides to impose even stricter demands. So (the thinking goes) we must have a federal rule, even at the price of stupidity, to ensure uniformity and to preempt future, more aggressive state regulation. That aspiration, though, is a mirage.
The potential for conflicts between national regulation and state regimes has existed since the enactment of federal securities statutes in the 1930s. States regulate and prosecute securities violations under the common law of fraud and under "blue-sky" laws (so called because they were enacted to clamp down on frauds who would promise investors the blue sky). Federal securities laws explicitly preserve those state laws and actions. As noted, state officials and courts generally enforce their own laws against any company—including companies domiciled elsewhere—that does business within the state. Still, for almost half a century, state courts and regulators by and large contented themselves to operate in the interstices of federal regulation.
That arrangement, though, collapsed roughly a decade ago under the onslaught of aggressive plaintiffs' lawyers—including that modern nightmare, the trial lawyer with a badge (formally known as "attorney general"). Proceeding under state and federal legal theories, those constituencies subjected the regulated industries to obligations well in excess of the federal requirements. Congress responded with a series of increasingly ham-fisted, but nevertheless ineffective, attempts to preempt state law and regulation.
In 1995, for example, Congress enacted the Private Securities Litigation Reform Act. That statute effectively banned class-action "strike suits," in federal court, against companies whose stock price dropped precipitately (while company officials, allegedly, withheld vital information from unsuspecting investors). Class-action lawyers then filed strike suits in state courts, under state laws. Just as promptly, Congress responded, in 1998, by imposing the restrictions it had imposed on federal courts in 1995 on the state courts.
The 1998 act represents the most comprehensive federal preemption in just about any area of federal law. In purporting to preclude any state actions, in federal or state court, over any activity "in connection with the purchase or sale" of a security, the statute pushes the outer limits of the federal government's constitutional authority. And yet state courts have continued to entertain strike suits—now under a different state law cause of action called "holding." Companies and investment houses, the theory goes, may fraudulently induce investors to hold a security, and that conduct is not connected to its sale. Ergo, no preemption. Strike suits have escalated both in number and in monetary value to the point of prompting the personal finance section of the redesigned, reader-friendly Wall Street Journal to provide shareholders with helpful advice on how to collect the rewards of such actions.
Eliot Spitzer's campaign is a direct result of failed federal preemption over state fraud litigation. In June 2001 Merrill Lynch settled a case, brought by Mr. and Mrs. Debasis Kanjilal, who claimed to have relied to their detriment on one of Henry Blodget's fabulous "buy" recommendations for the shares of a company with which Merrill Lynch had an undisclosed underwriting relationship. (The stock went from their purchase price of over $122 to $11—which certainly proves detriment, though not necessarily reliance.) What was de basis of de suit? A supposedly nonpreempted state fraud claim. When Merrill Lynch settled for $400,000 (even before arbitration, let alone discovery and litigation), Eliot Spitzer smelled blood and initiated his investigation—first against Merrill Lynch, then against a half-dozen other investment houses. Spitzer's investigation has prompted both prosecutions by other attorneys general and a raft of private class-action lawsuits against Merrill Lynch. The trial lawyers may well succeed in turning the one-time bank of the future into the bull of the past.
Federal regulators, legislators, and the regulated industries naturally hope to arrest those dynamics by means of federal preemption. They ignore, however, James Madison's elementary insight that federal preemption does not operate on states—period. It operates on states that have every incentive, and countless ways, to evade and eviscerate the federal rules. An effective federal rule would not only have to placate Eliot Spitzer and to preempt the state law theories on which the current initiatives of the trial lawyers and attorneys general are based. It would have to anticipate and preempt the next evasion—and the one after that. That is not doable.
As the ingenious "holding" actions and Eliot Spitzer have shown, state fraud or blue-sky actions cannot be contained by federal regulation. They would have to be abolished. Any law student, though, knows how to couch a fraud as a breach of contract or a violation of some general duty of fair dealing. To prevent that foreseeable maneuver, a federal law would have to wipe out the state law of contract. It would also have to bar state causes of action that we have never heard of but surely will hear of once more expansive federal preemption kicks in. In short, effective federal preemption would have to abolish state common and statutory law. Congress has never gone to that extreme and will not do so—least of all now that Eliot Spitzer is tooting his horn as the little investor's last best hope.
Since effective federal preemption is impossible, whatever federal regulation may emerge will simply be an opening bid for the next round of state impositions in excess of a supposedly uniform baseline. That unappealing prospect warrants a look at a more promising solution: Decentralize securities regulation. Make it work like corporate law.
The Delaware Model
Securities law and corporate law both govern the relations between the managers of a firm and its investors. The principal distinction lies in not the subject matter but rather the operative choice-of-law principle. Under corporate law, a company's choice of a particular state's law governs its transactions. If a company chooses Delaware law (as most do), its charter under Delaware law, and the rights and obligations that travel with it, govern the rights of management and shareholders, wherever they may live. Securities law operates on the opposite principle: legal relations are governed by the law of the purchaser's or customer's home state (unless they are preempted by federal law).
Two things have occurred to serious students of the subject. First, the corporate model of decentralized, competitive regulation is very likely preferable to centralized regulation, including the partially centralized model of securities regulation. Second, the lessons learned from the corporate "Delaware" model are applicable to at least some areas of securities regulation. Yale Law School professor Roberta Romano, a leading authority and principal advocate of the Delaware model, has proposed its extension to the SEC's corporate disclosure regime and the accompanying antifraud provisions. The SEC regime, Romano argues, should remain in place. But companies should be given a choice between SEC regulation and the regime of any of the fifty states. Each state's regime would be exclusive of every other state's and of the SEC's. In other words, the SEC would effectively become a fifty-first state.
By way of illustrating the difference between the securities and the Delaware-Romano model, consider the Spitzer campaign. Its targets are New York-based companies that may have defrauded New York consumers. The Wall Street Journal's editorialists have asked whether Spitzer is running for governor of New Jersey—the suggestion being that the hounded companies might leave New York for the Garden State. But why would they, under existing jurisdictional rules? Even if Magic Rat, Inc., were to drive its sleek brokerage machine over the Jersey state line, Eliot "Maximum Lawman" Spitzer would continue to exercise jurisdiction so long as the firm retained New York customers. He has in comparable cases brought his authority to bear on companies with webservers that, while located in Antigua, were accessible from New York, and the New York courts have countenanced his extraterritorial forays.
Under the Delaware model and its jurisdictional principle, in contrast, brokerage houses would have a choice. They could accede to Spitzer's regime and wear his oversight as a badge of honor and integrity or else migrate to a more hospitable jurisdiction. The move across the border would terminate New York's jurisdiction and subject the firm to the rules of whatever state it had chosen.
Not all firms would make the same choice. Some might decide that the integration of investment banking, research, and retail brokerage would be a viable business model; others might decide that it would not be worth the candle. In each case, though, the firms and their customers would bear the risks, and the rewards, of subjecting themselves to one among many competing jurisdictions. Each state would attempt to match the largest number of buyers and sellers. That incentive would eventually produce efficient regulation.
Romano's detailed proposal for extending corporate law principles to securities regulation specifically exempts the regulation of brokerage and investment services. Corporate disclosure regulation, Romano argues, aims to protect investors, and competing state regulators have powerful incentives to adopt rules preferred by investors. But that is not necessarily true of retail brokerage. States might be tempted to attract the owners of brokerage houses by allowing them to prey with impunity on unsuspecting investors. Competition among states might spark a "race to the bottom." In light of the political and intellectual resistance to her proposals, Romano's scruples are sensible. Still, Romano leaves open the possibility that the competitive model might well and profitably be applied to brokerage, too, and a strong case can be made for that bold approach.
First, while the market constraints that discipline brokerage houses are less than perfect, they are complemented by political constraints. Since each state's rules would apply to each firm's in-state as well as "foreign" clients, states would be unlikely to compete for brokerage business by sanctioning rank exploitation. It is true that state legislatures are beholden to special interests. Those interests, however, include both investment firms—which might bend the rules to their own advantage—and trial lawyers, who would push in the opposite direction. A presumption of rough equality among those interests is not unreasonable. Moreover, a state accommodation to brokerage firms would probably require a highly conspicuous exemption from generally applicable laws against fraud, and no state legislature would easily entertain such a proposal.
Second, while the corporate law literature focuses on the question whether the Delaware model generates optimal regulation (or a "race to the top"), decentralized regulation should be compared, not to an economist's blackboard model but rather to the set of centralized, uniform rules that are likely to emerge. In the case at hand, the alternative to an admittedly uncertain and possibly suboptimal world of state competition is not a uniform federal rule (efficient or not). Given the impossibility of preemption, the alternative to competitive federalism is an absurd race—among states and between the states and the national government—toward ever higher levels of regulation.
Third, the Delaware model is often defended on the grounds that fully informed ("marginal") shareholders discipline corporate choices of jurisdictions. But even when we cannot be certain that marginal investors are fully informed, "buyer beware" is generally a better rule than "beware of the regulator." Caveat emptor reduces the buyers' moral hazard and provides incentives for the discovery and disclosure of information.
That insight, to be sure, runs against the thrust of securities regulation, which—as the Supreme Court has hilariously put it—aims to "substitute a philosophy of full disclosure for the philosophy of caveat emptor."12 Arguably, though, that entire enterprise is misguided. When Aunt Millie starts day trading whatearnings.com on some "inside" tip—disclosed to 10 million television viewers—we have a problem, regardless of whether the tip came from the disinterested ghost of King Solomon or from conflicts-ridden Salomon Smith Barney. It is hard to say what role the endeavor to replace caveat emptor with disclosure may have played in the demand-driven Internet and telecom hype, but it obviously did not help matters.
Competitive federalism presents a rough medium between freedom of contract and caveat emptor on the one hand and regulatory ambitions on the other. Instead of a libertarian La-La Land of no regulation, competitive federalism would produce whatever regulation buyers and sellers agree on—by contract. Like a Delaware charter, state disclosure regulation would function as a kind of official Good Housekeeping seal of approval. Eliot Spitzer would function as a credible intermediary agent in a market where private, self-regulatory institutions, as well as federal regulators, seem to have failed. Not as exalted a role as that of Global Investors' Avenger—but a vastly more valuable and appropriate one.
The proposal to structure brokerage disclosure regulation along the lines of corporate law would no doubt encounter purportedly practical objections. It may seem radical and unconscionable, for example, to permit brokerage firms to pick their own domicile law. Even putting aside the economic and political constraints that would discipline those choices, though, extant securities regulation subjects investors (though of course not attorneys general) to a far more brutal choice of law regime: most disputes are subject to mandatory, binding arbitration before panels that are stacked with representatives of the regulated industries. Substituting competitive federalism for that regime would empower investors.
Ostensibly practical arguments are in the end a smoke screen for the political interests that occupy securities regulation. The competitive federalism option runs counter to the interests of state attorneys general and trial lawyers, which alone suffices to sink it. Its consideration would require an official acknowledgment on the part of federal regulators and legislators that they don't know what they are doing, which is true but also contrary to their interests. It would require a recognition on the part of the brokerage industry that federal preemption offers at most a temporary reprieve—which would render this particular sector the only American industry that does not deserve, on account of its lack of strategic sense, what the trial lawyers dish out.
Even so, some policy entrepreneur should roll out the competitive disclosure regulation proposal. (One plausible candidate for the task is Alabama attorney general Bill Pryor, who has a seat on his state's securities commission, which by some fluke is a member of the Spitzer-Pitt task force.) Doing so would greatly enrich a pathetic public debate—for example, by challenging the preposterous effort to "replace" caveat emptor with disclosure obligations and by raising the verboten subject of the appropriate balance among the analysts', the brokers', and the investors' responsibilities. More important still, it would direct attention to a central but unasked question: With the exception of Eliot Spitzer, who authorized Eliot Spitzer to "restructure" the U.S. financial markets?
Attorneys general have over the past decade managed to create a parallel national government on issues from product safety to antitrust law to tobacco regulation. In all those areas, some enterprising attorney general has reversely preempted the national government, typically in cahoots with trial lawyers and his fellow attorneys general. In all instances, the usurpation was made possible by the effectively unbounded extraterritorial reach of the usurpers' authority. What distinguishes the Spitzer campaign is its creator's enthusiasm in shouting his national ambition from the rooftops.
The absurdity of having a state official act as a national policy czar is in this case masked by the apparent plausibility of his step into the breach. Federal regulation has in fact failed investors. For that reason, among others, Harvey Pitt's whiny insistence on the need for national uniformity—coupled with sycophantic entreaties to state regulators—is politically futile and misguided as a policy prescription. A more promising way of opening an overdue debate is to thank Eliot Spitzer for his services—and to insist that federalism, properly understood, facilitates his initiative but also, and of needs, limits its appropriate reach. The jurisdictional principle Roberta Romano has called The Genius of American Corporate Law is the same principle that Justice John Paul Stevens has more broadly identified as the rock bottom of American federalism: citizens choose their state—not the other way around.
Citizens lose that right if those with whom they wish to do business cannot escape the clutches of Eliot Spitzer or of any of his colleagues. If citizen investors are to have a choice, Merrill Lynch must have an exit. As for investors who like Mr. Spitzer's regime, let them choose it—by contract with a firm in his jurisdiction. And caveant cives!
1. Charles Gasparino, "Wall Street Has an Unlikely New Cop: Spitzer," Wall Street Journal, April 25, 2002.
2. See, for example, Nicole Duran and Barbara A. Rehm, "Policymakers Go Activist; Will They Overreach?" American Banker, May 6, 2002, p. 1; and especially a telling interview with Spitzer: "Spitzer: My Job Is to Protect Investors," Business Week Online, May 6, 2002 (available at www.businessweek.com/bwdaily/dnflash/may2002/nf2002052_2194.htm).
3. Empirical studies suggest that analysts' reputational concerns mitigate, but do not fully cancel out, the effects of in-house conflicts of interest. See Mathew L. A. Hayward and Warren Boeker, "Power and Conflicts of Interest in Professional Firms: Evidence from Investment Banking," Administrative Science Quarterly, vol. 43, no. 1 (1998), p. 1.
4. He has done so both because he must go with the evidence he has and because what he has looks comprehensible even to the New York Times, whose so-called news coverage consists substantially of lightly edited Spitzer press releases. See, for example, Gretchen Morgenson, "Requiem for an Honorable Profession on Wall Street," New York Times, May 5, 2002.
5. The Fitzgerald bill, S-1895, is cast as an amendment to the Investment Advisers Act of 1940. That statute regulates "buy-side" institutions that invest money on behalf of third parties (such as investment advisers and portfolio managers). "Sell-side" institutions such as Merrill Lynch are governed by the Securities Exchange Act of 1934.
6. Merrill Lynch's deal with Eliot Spitzer attempts to control this risk by requiring the consent of all fifty state attorneys general. To procure that consent, the settlement allocates $52 million of the $100 million settlement to forty-nine states and the remainder to New York. Needless to say, though, a one-time settlement cartel provides little protection against further policy escalation. It binds neither state legislatures nor private trial lawyers. Nor does it even dispose of pending investigations of other brokerage houses, which may easily generate additional policy impositions in the guise of legal settlements.
7. See Joshua D. Ratner, "Stockholders' Holding Claim Class Actions under State Law after the Uniform Standards Act of 1998," University of Chicago Law Review, vol. 68 (2001), p. 1035.
8. Michael Orey, "Cashing In on Shareholder Suits," Wall Street Journal, April 25, 2002. At the current rate, one quoted wag opined, class-action recoveries will soon top combined dividend payments.
9. Romano's pathbreaking work on the Delaware model appeared under the title The Genius of American Corporate Law (Washington, D.C.: AEI Press, 1993). She proposed the extension to securities regulation in an article entitled "Empowering Investors: A Market Approach to Securities Regulation," Yale Law Journal, vol. 107 (1998), p. 2359. In The Advantage of Competitive Federalism for Securities Regulation, an AEI Press volume scheduled for publication later this year, Romano persuasively defends her theory and empirical findings against critics.
10. The opaque reference is to "Senator" Bruce Springsteen's "Jungleland" (© 1975 Laurel Canyon Music Ltd.):
The Rangers had a homecoming
In Harlem late last night
And the Magic Rat drove his sleek machine
Over the Jersey state line
. . .
Well the Maximum Lawmen run down Flamingo [Lane]
Chasing the Rat and the barefoot girl
. . .
The ellipses cover warm beer, beautiful girls on the hood of a Dodge, and the part where "they all wind up wounded/Not even dead."
11. Under corporate law, companies may choose their state of incorporation without physically moving their operations. The question whether and under what circumstances the jurisdictional principle should be based on pure legal choice or rather on a company's actual domicile (as determined, for example, by the location of its headquarters) has important economic implications, but they are beyond the scope of this newsletter. Either principle would be preferable to the alternative of nonpreemptable, extraterritorial state regulation.
12. SEC v. Capital Gains Research Bureau, 375 U.S. 180, 186 (1963).
13. Saenz v. Roe, 526 U.S. 489, 510-11 (1999).
Michael S. Greve is the John G. Searle Scholar and the director of the Federalism Project at AEI.