Printer Friendly

Elective shareholder liability.

   A. The Dodd-Frank Approach
   B. Dodd-Frank's Bailout-Inducing Provisions
   C. The Impossibility of "Never Again"
   D. Fifteen to Twenty-Five Percent Capital Adequacy as a Supplement to

   A. Mechanics
      1. The shareholder election
      2. Governmental collection
      3. Two-year blackout
      4. Pro rata share
      5. Ability to nullify evasion efforts through rulemaking
      6. Treble damages, including litigation costs
   B. Benefits
      1. Taxpayer reimbursement
      2. Risk management
      3. Partial elimination of TBTF subsidies
      4. An active derivatives market for shareholder liability
      5. Tailoring corporate form and leverage structure
     A. Alternative Risk-Bearers Purportedly Superior to Shareholders
      1. Creditors
         a. Contingent capital
      2. Corporation and government: government participation in SIFI
      3. Industry and government: an insurance fund
      4. Directors and officers
      5. The taxpayers
     B. Where Are the Shareholders?
      1. Poor shareholders
      2. The sacred cow of limited liability
      3. Banks and limited liability in history
     C. General Arguments Against Non-Limited Liability
      1. Easy evasion and enforceability
      2. Difficulty of determining who would be on the hook
      3. Endogeneity problem
      4. Exogeneity problem
      5. Too much government discretion


Taxpayer bailouts--that is, the deployment of public funds to prevent the rapid failure of a private institution--were the government's tool of choice during the fall of 2008. The resulting political and scholarly response has been nearly uniformly negative, with very few arguing in defense of bailouts, (1) and most advocating some version of the vision of "ending government bailouts as we know them." (2)

This reaction makes good sense at first blush. Bailouts are unjust, as private parties are able to pocket the benefits generated by their risky activities while forcing the costs onto the public. (3) They are massively unpopular and can dominate public discourse at the expense of other important policy issues. They are expensive, both directly through the cost of the bailout and indirectly through the general economic harm that the promise of bailouts can motivate. They represent extraordinary rule of law concerns, as governments bend and break laws that are aimed to prevent the utilization of taxpayer funds for private purposes. (4) And bailouts risk distorting the functioning of the entire free market financial system. As Larry Summers said, with ironic and unwitting prescience in 2000, "[I]t is certain that a healthy financial system cannot be built on the expectation of bailouts." (5)

But the problem with bailouts is not simply that they make for bad economics and bad politics. The problem, in addition to this parade of horribles, is that they are sometimes essential to the prevention of complete financial collapse, with its accompanying consequences of economic and human misery felt far beyond the financial markets themselves. Any policy reaction to bailouts must deal with this complexity--bailouts are almost always the wrong policy approach, except when they are not.

The Dodd-Frank Wall Street Reform and Consumer Protection Act fails to address this complexity. (6) Instead, Dodd-Frank seeks to end government bailouts--forever. (7) But this promise is implausible objectively, inaccurate subjectively, and, regardless, is undesirable as a matter of policy. The complex problem of bailouts after Dodd-Frank therefore remains unresolved.

This Article proposes to solve the problem of bailouts by means of a legal mechanism called "elective shareholder liability," which is both far less intrusive and far more effective than the regulatory apparatus that Dodd-Frank creates. (8)

Elective shareholder liability gives shareholders of the world's most important financial institutions--vailed systemically important financial institutions (SIFIs) in the clunky vernacular of banking regulation--the opportunity to choose how best to address their own too-big-to-fail (TBTF) problems themselves. (9) Elective shareholder liability gives SIFI shareholders the choice between two alternatives: either (1) dramatically limit the firm's leverage, consistent with a consensus proposal from leading economists--what we will call the "increased capital" option; (10) or (2) create a bailout exception to the SIFI's limited liability status, such that the government can recoup the losses associated with any taxpayer bailout from the SIFI shareholders directly.

The first of these options comes from a proposal entertained during the debates that produced Dodd-Frank, and still hotly debated as this Article goes to press. (11) Anat Admati, Peter DeMarzo, Martin Hellwig, and Paul Pfleiderer-among the world's leading financial economists--have proposed dramatically increasing the capital adequacy requirements to more than twice the level ultimately accepted under the Basel III Accords. (12)

Admati et al.'s proposal is simple, extremely effective at reducing TBTF problems, and has been, until now, fought vociferously by the banks it would affect. (13) The banks have argued, and their political supporters have embraced, the notion that, effective though increased capital requirements may be at avoiding future bailouts, the costs that such requirements inflict on banks and society are simply too great. (14) The banks assert that higher capital requirements mean less lending and, consequently, put a brake on economic recovery when recovery should be the government's exclusive focus. (15) Admati et al. make very little of these assertions in every respect. (16) Other economists agree. (17)

Elective shareholder liability resolves the impasse by allowing SIFIs to make those costs concrete: should their own internal assessments of increased capital requirements show that there are efficient benefits from leverage that outweigh the taxpayer costs of bailouts--both direct and indirect--then SIFIs can appropriately dismiss increased capital without forcing the costs of their private failure on to taxpayers. If that internal cost-benefit analysis comes out differently, and banks conclude that the only benefits of increased capital are those taken inefficiently from the pockets of taxpayers in the form of TBTF subsidies, they can opt into the alternative regimes and maintain their limited shareholder liability.

This is not a Hobson's choice--there may well be real economic value in preferring shareholder liability to increased capital requirements. For instance, elective shareholder liability is effectively a bet on a bank's ability to avoid a taxpayer bailout, and thus the shareholder liability that would result. The upside would be that banks could continue to use the subsidies of leverage--subsidies that exist through the tax code and implicit government guarantees.

Taken on its own merits, elective shareholder liability provides important cost internalization at critical junctures both ex ante and ex post. Ex ante, elective shareholder liability requires directors and officers to increase self-monitoring when, as is often the case, these officers and directors are themselves significant shareholders. Furthermore, when shareholders and management do not significantly overlap, shareholders have heightened incentives either to demand more ownership for their shareholder dollars to reflect bailout reimbursement, or, alternatively, to monitor management's own risk measurement practices such that bailouts become less likely because of internal controls. Ex post, elective shareholder liability creates a fund--until now unavailable--that can serve to reimburse taxpayers at least partially for the costs of bailouts. Elective shareholder liability is thus the only proposal that provides mechanisms of monitoring and reimbursement using the same legal apparatus.

To introduce and assess elective shareholder liability, the Article proceeds as follows. Part I provides the context of the Dodd-Frank legislation, explaining what the Act attempts to do to prevent taxpayer bailouts and why it will fail to deliver on that promise. Part I also explains increased capital requirements as an alternative to the Dodd-Frank structure. This Part is synthetic; readers familiar with these arguments should feel free to skim or skip this Part as needed.

Part II.A presents the six elements of elective shareholder liability. They are: (1) SIFI shareholder election of either a bailout exception to limited liability and a fifteen to twenty-five percent capital requirement; (2) if shareholder liability is elected and a bailout occurs, the government shall issue an obligatory assessment against shareholders, with a lookback period of one year; (3) collections are subject to a blackout period--similar to the automatic stay in bankruptcy--of two years, under which the collection cannot be assessed; (4) shareholders are liable pro rata; (5) the government shall have authority to declare efforts to evade liability through corporate ownership--using limited liability entities to own shares in non-limited liability SIFIs--null and void; and (6) shareholders who challenge these collections and lose will pay treble damages and the government's litigation costs.

Part II.B presents several benefits of elective shareholder liability, including better incentives for shareholders and officers for internal risk management; the partial elimination of the economically deleterious consequences of limited liability, namely shareholder risk shifting, debt overhang, and, more specifically, what one scholar has called "correlation seeking"; (18) creation of an active market of derivatives, the prices of which would contain more information regarding the potential for taxpayer bailouts--and their potential costs--than is presently available in the markets; and freedom for SIFI shareholders to tailor their corporate form and leverage structure to best provide for cost internalization, profit making, and economies of scale.

Part III.A summarizes a number of challenges to the proposal, beginning in some detail with the argument that the shareholders are not in the best position to control these risks or absorb these losses. Other candidates for risk control include creditors, the corporation itself through liquidation or government equity participation, and company directors and officers. However, out of all the possible alternatives, shareholders are for various reasons the best candidates for risk management and cost absorption.

Part III.B then explores the question of why shareholders have been ignored in this analysis. Some of that explanation is that scholars and policymakers may initially feel squeamish about exposing shareholders to liability beyond the value of their equity investment, either because the shareholders have suffered enough already, or, even if not, because corporate law forbids such liability. Part III.B provides the context that challenges both assumptions, while also highlighting other areas of law that similarly illustrate exceptions to prevailing rules of profit allocation. Part of that challenge is historical: elective shareholder liability draws on the partnership structure of investment banking that dominated the field throughout its history, until the late twentieth century. That structure provided for unlimited personal liability in the event of bank failure. The complexity and especially the size of modern investment banks make a simple return to partnerships infeasible; elective shareholder liability is thus a rule that is adapted to modern financial institutions but that can accomplish some of the same goals.

Part III.C then addresses more substantively a variety of other critiques that might be--indeed, have been--directed at any regime of non-limited liability generally. The Article's arguments come from the last sustained critique of limited liability, in the mid-1990s, by Henry Hansmann and Reinier Kraakman. (19) The critiques on the other side, principally from Joseph Grundfest, focus on the idea that recovering from shareholders will be a complicated, uncertain, and logistically and procedurally messy proposition. (20) This Article concedes that argument entirely. Instead, Part III.C argues that even with efforts to evade liability, elective shareholder liability can be structured such that the complexity and uncertainty of recovery are made to be strengths, not weaknesses. In the face of such uncertainty, shareholders are more likely to opt out of shareholder liability and into increased capital, thereby avoiding the complexity and uncertainty; to take measures to actively monitor risk accumulation in other ways; or to insure themselves against such loss, through the shareholder liability swaps described in Part II.B.4. Part III.C also addresses a variety of other arguments, including that elective shareholder liability would be inferior to alternatives such as contingent capital or would represent an improper extension of government discretion.

The Article concludes by noting that the burden is on those who would oppose elective shareholder liability to explain why SIFIs should not be subject to higher capital-adequacy requirements. In other words, an argument against elective shareholder liability is an argument against increased capital: the SIFIs must therefore explain the value of high leverage beyond the value of the TBTF subsidies that, Dodd-Frank notwithstanding, SIFIs continue to enjoy.

Some have argued that the time for reform has passed, and that Dodd-Frank and Basel Ill have fixed the problems that the financial crisis of 2008 exposed. (21) To this analysis, Thomas Hoenig, the former President of the Federal Reserve Bank of Kansas City, has the best answer:
   [Dodd-Frank and Basel III] do not solve the fundamental flaw in the
   system: there are highly complex and opaque banking organisations
   engaged in a variety of non-core, high-risk activities while backed
   by a public safety net. The problem is not that banks take risk,
   but that some are too complex for anyone to assess and control that
   risk. (22)

Elective shareholder liability forces the consequences of this complexity back to their origins. It ultimately presents the best hope of solving the problem of bailouts by making them less likely and less predictable, but still available if necessary as a regulatory response to avoid truly cataclysmic financial collapse.


On July 21, 2010, President Barack Obama signed the Dodd-Frank Act. (23) During the signing ceremony, the President asserted the consequences of the legislation in bold terms:
   [B]ecause of this law, the American people will never again be
   asked to foot the bill for Wall Street's mistakes. There will be no
   more tax-funded bailouts--period. If a large financial institution
   should ever fail, this reform gives us the ability to wind it down
   without endangering the broader economy. And there will be new
   rules to make clear that no firm is somehow protected because it is
   "too big to fail".... (24)

In expressing this view, the President echoed the words of the bill itself: the statute's long title tells us that it is designed to "end 'too big to fail'" and "to protect the American taxpayer by ending bailouts." (25) Unsatisfied with even these general exhortations, Congress added section 214 ("Prohibition on Taxpayer Funding"), declaring as binding law of the United States, "Taxpayers shall bear no losses from the exercise of any authority under this title." (26)

Despite the clarity of these convictions, there is wide consensus among legal scholars, (27) economists, (28) and even, perhaps surprisingly, members of the President's own administration (29) to the opposite effect: while Dodd-Frank makes many significant and substantial changes to American banking and securities laws, it will not end taxpayer bailouts. The Act may even make such bailouts more likely, not less. (30) This is deeply problematic: such bailouts will spawn political movements, lead to anti-incumbent backlash, (31) and if some commentators are to be believed, may even result in dramatic social and political upheaval. (32) The consequences of this reality--that bailouts continue to be a part of the regulatory response to financial crisis--prompt the need for proposals that are more finely tuned to bailout mitigation. Elective shareholder liability is such a proposal.

In order to understand why elective shareholder liability is necessary, it is important to understand what Dodd-Frank does, what it does not do, and the benefits that other regimes might offer. In Part I.A, I describe the regime that Dodd-Frank establishes to prevent bailouts. I then explain in Parts I.B and I.C why this effort will largely fail--both for reasons specific to Dodd-Frank, and for the broader conceptual reason that bailouts are impossible to eliminate completely and forever. The best hope for bailout avoidance is thus to make them less frequent and less predictable. Part I.D explains how dramatically increased capital adequacy requirements, a much-debated though until now largely rejected alternative, would prove the most effective at bailout reduction, and why the rejection of this alternative should prompt policy proposals such as elective shareholder liability.

A. The Dodd-Frank Approach

Dodd-Frank represents an attempt to prevent taxpayer bailouts in two ways: first, by making financial crises less likely to occur, using ex ante regulation; and second, when such crises do occur, by resolving them in a way that prevents them from eventually requiring taxpayer participation or ex post resolution.

The most prominent innovation, for the purposes of bailout prevention, is Dodd-Frank's creation of the Financial Stability Oversight Council (FSOC), a council of regulators that consists of the Secretary of the Treasury as Chairperson of the Council, and the heads of various finance-related federal agencies as other voting members. (33) The FSOC is given the authority first, to identify those organizations that become systemically important, and second, for such institutions, to determine what further regulation is required in order to prevent such SIFIs from destabilizing the financial system and requiring a taxpayer bailout.

A note of terminology is important here. Dodd-Frank envisions two classes of entities that will come under its regulatory regime. First, there are the entities regulated by the FSOC under direct authority of Dodd-Frank. Those are bank holding companies with assets of $50 billion or more. (34) Second, two-thirds of the voting members of the FSOC may determine that nonbank institutions pose systemic risks, based on a number of factors. (35) Both categories of institutions--which have both been labeled "SIFIs," though not in those words by Dodd-Frank itself (36)--are subject to heightened prudential regulation, and to Dodd-Frank's liquidation authority.

Others have described in lucid detail the nature and structure of DoddFrank's main features. I refer readers to those sources for a more thorough treatment than I offer here. (37) However, a brief summary of the prudential regulation and liquidation authority is important for the purposes of understanding elective shareholder liability.

The prudential regulation to which SIFIs are subjected focuses almost exclusively (38) on the concept of mildly increased capital-adequacy requirements, which have been central to banking regulation for decades. (39) The resolution authority, on the other hand, represents something new, at least in the context of SIFIs: Dodd-Frank is an effort to impose on SIFIs the resolution process used at present by the FDIC to resolve failed commercial banks. Under the Dodd-Frank version of this resolution, the FDIC, upon an affirmative vote of two-thirds of the FSOC, including the Treasury Secretary, will "receive" the failed institution, and use the Orderly Liquidation Fund (OLF) to ensure that the liquidation occurs without making taxpayer funds available. (40)

In an ideal version of a Dodd-Frank liquidation, a SIFI will be subjected to increased regulatory oversight that will mitigate the need for bailout because it has significantly reduced the risk of failure. In the event that such failure does occur, however, the SIFI can be taken over by the FDIC, backed by the administration, in an effort to keep the failure from destabilizing the rest of the financial system. The government takeover of the SIFI is temporary: as with the FDIC resolution process in general, the Dodd-Frank approach implies the availability of a willing and solvent acquirer to purchase some or all of the failed SIFI from the government. Under that ideal scenario, Dodd-Frank uses the government/taxpayer funds only as a temporary stopgap. If that ideal is realized, taxpayer bailouts become a thing of the past.

B. Dodd-Frank's Bailout-Inducing Provisions

The reality, however, is far different from the ideal just described. This is true for both practical and conceptual reasons.

Practically, several of the Act's features make taxpayer bailout more likely in the event of SIFI failure, especially cascading SIFI failure. (41) First, although the FDIC is ostensibly barred from tapping the public fisc during its liquidation process, Dodd-Frank still allows the FDIC to give creditors preferential treatment under limited circumstances (42)--that is, some creditors will get access to government funds on a preferential basis. The OLF is designed to meet whatever needs such creditors, and others, may have in the liquidation process. However, the FDIC can gain access to taxpayer funds, beyond the value of the OLF, through a statutory provision that allows the FDIC, "in its judgment," to borrow up to $100 billion from the Treasury for "insurance purposes." (43) Given the Federal Reserve's willingness, fully supported by the Bush and Obama Administrations, to stretch the bounds of legal power in order to save the system from systemic collapse, (44) it is fully plausible to imagine a future administration, through the FDIC and Treasury Secretary, authorizing a $100 billion bailout under the auspices of [section] 1824(a). The result is a de facto bailout.

In the event that the FDIC gives some creditors preferential treatment that exceeds the value of the fund, the Act does contain a clawback provision by which the FDIC will ostensibly recover those funds. (45) However, there is a gaping exception: if the FDIC finds that "serious adverse effects to the financial system" would result from such a clawback, it need not, at its election, pursue it. (46) As such, the FDIC--or, by proxy, the administration (47)--may make a Treasury loan into a bailout in intent, effect, and fact, the stiff prohibition encountered in section 214 of the Act notwithstanding.

David Skeel reaches similar conclusions regarding Dodd-Frank's susceptibility to institutionalizing bailouts, but on a more conceptual basis. Skeel views the Dodd-Frank process, from the very nature of the FSOC to the FDIC's resolution system in the Orderly Liquidation Authority, as the epitome of government participation in the success and failure of free enterprise: what he calls the "corporatist" approach to banking regulation. (48) Under Skeel's formulation, corporatism is the "government partnership with the largest financial institutions." (49) That partnership consists of living, as an SIFI, under the Dodd-Frank umbrella of prudential regulation and a new liquidation process as either a bank with assets greater than $50 billion, or as a nonbank designated as a SIFI by the FSOC. While Wall Street is not famous for its preference for more government participation in its businesses, Skeel argues that the Dodd-Frank arrangement carries with it significant benefits for the banks themselves. Skeel writes:
   [T]here is no serious effort to break the largest of these banks up
   or to meaningfully scale them down. Because they are special, and
   because no one really believes the largest will be allowed to fail,
   they will have a competitive advantage over other financial
   institutions. They will be able to borrow money more cheaply, for
   instance, than banks that are not in the club. Dodd-Frank also
   gives regulators a variety of mechanisms they can use to channel
   political policy through the dominant institutions. The partnership
   works in both directions: special treatment for the Wall Street
   giants, new political policy levers for the government. (50)

The corporatist system that Skeel describes is wedded to the very fabric of Dodd-Frank's regulatory strategy. Dodd-Frank thus creates the potential for more taxpayer bailouts because there is no other plausible alternative under this regime when these banks fail. Conceptually, the Dodd-Frank system is a system of taxpayer bailouts.

C. The Impossibility of "Never Again"

It might be said that the bailout-inducing provisions just described are limited, at least to the $100 billion described above. That may well be true. But even so, the Act remains an ineffective mechanism for ending TBTF in every case, as it purports to do, largely because banning bailouts is an impossible goal, for several reasons. (51)

First, the regulatory structure created by the Dodd-Frank regime may well prove very useful in a one-off financial crisis not unlike the Long-Term Capital Management crisis in 1998. (52) But in the face of cascading SIFI failures--for example, the rapid failures from Bear Steams to Freddie Mac to Washington Mutual to Wachovia to Merrill Lynch to Lehman Brothers to AIG--it is exceedingly unlikely to function as predicted, simply because the $100 billion devoted to the enterprise will be quickly overwhelmed. Furthermore, since the Dodd-Frank regulatory structure relies on healthy banks to acquire the failing ones, a cascade of SIFI failures can ultimately lead to a scenario where there are no healthy banks left standing to help the FSOC implement its purposes. (53) In that event, the only balance sheet left standing is the government's, and the acquisition anticipated by this model is itself the bailout.

More importantly, though, scholars and policymakers must recognize that bailouts are not simply an economic calculation that focuses on the socially inefficient incentives associated with TBTF, but are a political decision made by those political actors who will confront the crisis as it is unfolding. In the event of pure financial terror, where, as in the fall of 2008, panic sweeps through a large number of financial institutions, bailouts are likely the only mechanism available to governments and regulators that could possibly work to stem the tide of failures. Financial crises are pathological and self-perpetuating in nature; they represent the perhaps initially panicked but ultimately rational assumption that there is a highly diminished possibility of recovering economic value from one's investments. In such instances, the need is for a lender of last resort; and that lender is, due to its sheer size, the government. (54) This was a motivating policy behind the creation of the Federal Reserve in 1913, (55) and remains a reality today. And the difference between a "bailout" using government funds and a loan of last resort is a highly difficult one to parse. (56) In light of that reality, banning bailouts--as Dodd-Frank does--does not address the problem, and is a commitment that cheapens the legislation. As Levitin colorfully argues:
   Law is an insufficient commitment device for avoiding bailouts
   altogether. It is impossible to produce binding commitment to a
   preset resolution process, irrespective of the results. The
   financial Ulysses cannot be bound to the mast. Although we may want
   Ulysses to be bound to the mast when the sailing is smooth to avoid
   the sirens' call of politically directed state intervention in the
   market, the situation changes once the ship has hit the rocks. Once
   the ship is foundering, we do not want Ulysses to be bound to the
   mast, lest go down [sic] with the ship and drown. Instead, we want
   to be sure his hands are free-to bail. The question, then, is not
   whether to have bailouts but how bailouts should be structured.

If we acknowledge, following Levitin, that bailouts are, at some level, ultimately inevitable, we can turn to the more productive question of identifying the party or parties who can best monitor SIFI risk to decrease bailout frequency, and also identify the party or parties who can make taxpayers whole in the face of the bailouts when they do occur. The rest of the Article proposes that the shareholders themselves--especially those shareholders who are also directors and officers of the corporation--can serve that dual function better than any other party, whether the government or the SIFI's other creditors.

D. Fifteen to Twenty-Five Percent Capital Adequacy as a Supplement to Dodd-Frank

The fact that the Dodd-Frank framework will fail to prevent taxpayer bailouts invites attention to the alternatives that were considered but ultimately dismissed. As mentioned, there have been scores of alternative proposals. (58) I focus, however, on one: dramatically increased capital adequacy requirements of fifteen to twenty-five percent. (59) I focus on increased capital because of its simplicity, ease of enforcement, and overinclusiveness.

The basic structure of banking capital regulation in the past three decades has been dictated in large part by three series of international proposals offered by the world's central bankers in their capacities as members of the Basel Committee within the Bank of International Settlements in Basel, Switzerland.

Collectively, these proposals are known as the Basel Accords; individually, they are Basel I, published in 1988; (60) Basel II, published in 2004; (61) and Basel III, proposed in 2010 and still subject to an ongoing approval process. (62) In all three instances, though in markedly different ways, the Basel Accords focus bank and systemic risk regulation on increased capital, or the amount of equity that the banks must retain as part of their capital structure. In other words, the Basel requirements represent limits on banks' ability to take on debt.

Basel III proposes to increase, gradually, the amount of common equity that banks maintain to seven percent (of which two and a half percent is a conservation buffer). (63) The banks themselves have fought these changes, and succeeded in forcing the phase-in to occur through 2018. (64) The banks' arguments are that increased equity requirements: (1) force banks to put capital aside that might be more profitably used for loans; (2) increase banks' funding costs, as equity requires a higher return than debt; (3) lower the return on equity, thus decreasing the overall value of the bank; (4) increase the banks' funding costs by making banks less able to borrow using government subsidies in the form of implicit guarantees and beneficial tax treatment; (5) intervene in the market to create a suboptimal solution; and (6) eliminate the important market discipline associated with borrowing. (65)

Anat Admati, Peter DeMarzo, Martin Wellig, and Paul Pfleiderer, leading financial theorists, have systematically debunked these and other arguments against dramatically increased capital requirements. (66) Addressing each in turn, these theorists show how a basic understanding of finance leads to the conclusion that arguments against increased capital are invalid, and sometimes completely specious. (67) Other economists, (68) regulators, (69) and market participants (70) have agreed.

First, the argument that banks must "set aside" liquid assets to fund capital adequacy requirements misunderstands the unfortunate vocabulary of banking. Increased capital refers to the percent of bank funding that comes from the residual equity shareholders, rather than from the banks' creditors. It has nothing to do with the level of liquidity reserves that banks must maintain. The two concepts have nothing to do with one another. Reserve requirements are important, and worthy of separate discussion. But capital requirements refer to the bank's capital structure, not the convertibility of its assets to cash.

The second and third arguments reflect a basic misunderstanding of what is referred to as the Modigliani-Miller Theorem, which garnered its authors the Nobel Prize in Economics in 1985. That theorem holds that, assuming away market and policy "frictions," the mix of a firm's equity and debt has no impact on the value of that firm's assets. Thus, any argument that assumes that an increased equity participation in banks means that the value of the firm will drop--explicit and implicit in arguments two and three above--are fallacious, unless pegged to an identifiable friction, such as tax subsidies, bailout guarantees, or bankruptcy costs.

The fourth argument is the most peculiar. The elimination of tax and TBTF subsidies is not a demerit of increased capital--such eliminations are the very point.

Fifth, there is the argument that because the market has created a preference for low equity participation for large banks, it must be the optimal, efficient solution. That argument ignores the fact that these banks are the recipients of both tax and implicit guarantees, which are decidedly suboptimal in that they privatize the benefits of bank behavior while socializing the costs. The argument effectively asserts that there will be transition costs as investors adjust the risk-profit balance that banks represent. These costs might be real and important, but they are not dispositive. Arguing to the contrary would defeat any effort to make universally approved changes to the status quo. Consider an analogy. When it became apparent that brick buildings do not fare well in even moderate earthquakes, most jurisdictions required that new constructions meet much higher standards and that old buildings be retrofitted, over time, at a reasonable pace. Subsequent earthquakes have proved that this retrofitting--costly though it was to the individuals who had to undertake it--has saved lives. The argument that the costs of this transition would be great, and that stakeholders--in this analogy, those who had to pay for the retrofitting--had come to expect a certain level of profitability that did not include this adjustment, would not have carried the day. The argument that investors will no longer support banks at a risk-adjusted, higher rate of capital is effectively the same argument.

Finally, although some scholars argued in the 1980s that leveraged buyouts allowed creditors to become a more effective monitor of corporate management than shareholders could be, (71) Admati et al. show that this disciplining effect only functions well "as long as the bank is able to satisfy these claims." (72) Even so, the argument in favor of the disciplining features of debt is an important one, and should be taken seriously. If debt diminishes agency costs consistent with the theory offered by Jensen and others, such that those savings outweigh the costs associated with excess leverage already discussed, then it is an argument worth considering at greater length though not here. (73)

Despite Admati et al.'s widely cited debunking of the arguments against capital adequacy requirements, the arguments against them continue with little variation. (74) The debate over capital adequacy is thus at something of an impasse. Elective shareholder liability seeks to resolve that impasse. To make that resolution plausible, the rest of the Article assumes two things: First, banks have as yet identified no specific social benefit to leverage; increased capital adequacy requirements are thus a superior regulatory alternative to resolve the problem of bailouts. Second, just because banks have failed to articulate the benefits of debt does not mean that those benefits do not exist. Therefore, an alternative structure that allows banks to use market forces to identify those benefits would be a superior regulatory regime. Elective shareholder liability is that regime.


A. Mechanics

Elective shareholder liability would include the following characteristics:

(1) SIFI shareholders shall elect between shareholder liability in the case of a bailout, or a fifteen to twenty-five percent capital adequacy requirement; (2) if shareholder liability is elected, and a bailout occurs, the government shall issue an obligatory assessment to be collected against shareholders; (3) the collection process is subject to a blackout period of two years under which the government cannot collect the debt; (4) shareholders are subject to pro rata liability based on the one year prior to the bailout; (5) the government shall have authority to declare efforts to evade liability through corporate ownership--using limited liability entities to own shares in non-limited liability SIFIs--null and void; and (6) shareholders will pay treble damages, including the government's litigation costs, should they force the collection into court or affirmatively sue to evade collection.

1. The shareholder election

One might argue that taxpayer bailouts are exogenous events that shareholders cannot predict and therefore should not be required to fund ex post. That argument has merit, particularly in light of the fact that it is impossible to separate the bailout decision from the political dimensions of the decisionmaking process. As argued above, bailouts are political decisions--and politics do not always play according to economic logic.

In light of that uncertainty, elective shareholder liability is an optional regime. Shareholders would vote to amend their charters, (75) such that the SIFI in question is identified as having a bailout exception to its limited liability status. In that way, just as some corporations are identified by their tiered shareholder structures, (76) some SIFIs would be identified as elective shareholder liability SIFIs.

Of course, there is absolutely no incentive for shareholders to elect to follow this regime if the alternative is the regulatory structure presently in place. Thus, in order to make the election plausible, shareholders would choose between expanded shareholder liability and the fifteen to twenty-five percent capital adequacy requirements addressed above. In that sense, banks' shareholders can put their money behind the arguments that have, so far, been routinely debunked.

One model in banking for this kind of optional regulatory regime was the provision in Dodd-Frank section 716 that almost survived the legislative process. Under the former section 716, banks would have to choose between maintaining their status as bank holding companies--and thus their access to the Federal Reserve's discount window and other liquidity mechanisms--and maintaining their derivatives business. (77) A similar structure would define elective shareholder liability. (78)

As mentioned in the Introduction, the election described here is not a false one: because shareholders could manage their liability through means described more fully below, some banks could prefer shareholder liability to restrictions on debt. In efficient markets, those banks with the best risk-management systems could presumably elect shareholder liability, comfortable in their ability to manage that liability to their shareholders' benefits.

As in any instance of an election of this kind, one should be mindful of the risk of adverse selection, such as when the seller of a differentiated product cannot determine important characteristics of the buyer and so misprices the product. (79) Put differently, there may be a risk of SIFIs with good risk-management practices all opting for increased capital, and those with poor risk management opting for elective shareholder liability. This, however, is not a concern, nor really an instance of adverse selection. For adverse selection to be a problem, there must be hidden costs to the selection that selectors can exploit. Because the quantitative terms of the election are variable--the shareholders in an elective shareholder liability regime will pay the variable costs of a taxpayer bailout, or the limits on the debt inherent to capital requirements can be adjusted--there is an easy mechanism by which adverse selection could be corrected in the event that it became an issue. Furthermore, the market disciplining function for those SIFIs seeking elective shareholder liability will be such that they will be unable to raise any capital at all if they do not have at least a basic measure of risk management, as shareholders will be unwilling to opt into such a system.

Two final mechanical questions remain regarding the dissenting shareholders of a bank that elects shareholder liability. Given the fact that shareholders of SIFIs are almost certainly public shareholders, appraisal rights--following the model of the Delaware General Corporation Law--would not be available, as appraisal rights are almost exclusively available only for limited transactions

for private companies. (80) Instead, the depreciated value of their shares on the public market would be their compensation. This should not trouble regulators at all, however, as any depreciation would likely be the consequence of the market adjusting for the risk of taxpayer bailouts. The new price, unless inefficiently priced in the markets, would reflect a truer value of that equity. (81)

2. Governmental collection

For purposes of taxpayer reimbursement, the most effective mechanism for securing unlimited shareholder liability for SIFIs is to introduce an obligatory collection against shareholders following a taxpayer bailout. Because this proposed system would recognize the inability to fully avoid future bailouts, the assessment imagined here would probably require the partial repeal of section 214 of Title II of Dodd-Frank. Section 214(a) presently reads: "Liquidation Required: All financial companies put into receivership under this title shall be liquidated. No taxpayer funds shall be used to prevent the liquidation of any financial company under this title." (82)

There are two approaches the government could take in order to introduce a regime of non-limited liability for taxpayer bailouts. First, it could leave section 214 as is, and require that taxpayer bailouts still eventually lead to the liquidation of bailed-out firms. The emphasis here is on "eventually." But that option is suboptimal. First, "eventually" is tedious legal sophistry, which is an awkward place to start in a world of policy proposals. Second, requiring eventual liquidation undermines the real value that government bailouts might offer. The best justification for a government bailout is as a lender of last resort to otherwise solvent and economically sound entities that have fallen victim to an unforeseeable collapse of their short-term funding sources. Requiring the liquidation of firms that are the beneficiaries of a government bailout defeats this purpose.

The second approach, consistent with the idea that bailouts may prove an essential tool to save otherwise viable firms caught in a financial maelstrom, would require the repeal of section 214(a). Even then, an additional section 214(d) would be added that would create an obligatory assessment against the shareholders of a financial company that receives taxpayer funds, regardless of whether those funds were granted under the exercise of authority of Dodd-Frank or otherwise.

Again, collection would be obligatory--in the event that there are taxpayer losses following a government bailout, the government must collect them. Collection cannot be waived, either by the government's own discretion after the blackout period (described below) or as part of a bailout negotiation package. The idea here is that the assessment is neither a cost that has to be litigated in order to be enforced, nor is it a chit the government might invoke during negotiations to avoid the appearance or reality of sweetheart deals for the government representatives' former colleagues.

3. Two-year blackout

A third element of elective shareholder liability is a two-year blackout period, similar to the automatic stay in bankruptcy, on the issuance of the assessment. This is important for several reasons. First, too much wrangling in the moment of a financial panic might render government intervention wholly useless by injecting more uncertainty into the marketplace at a time when market participants are least able to bear it. A blackout period will allow the government to make whatever interventions are necessary in the moment, stabilize the markets, and then seek recovery after the dust has settled.

A blackout period will also allow the government to take adequate stock of the losses that taxpayers have faced as a result of the intervention. If there are no losses to be borne--probably because other mechanisms to recoup costs were taken and were successful, as in the case of the 2008-09 bailouts (83)--there is nothing for the government to recover. The blackout period allows other mechanisms to be tried and fail before triggering liability.

Similarly, a blackout period will allow those shareholders whose firms have imploded to avoid facing a forced fire sale in the moment of crisis. A more orderly reimbursement period will therefore allow taxpayers to recover their losses without forcing the shareholders into selling positions that the crisis prompts.

The blackout period also affects government incentives during a financial crisis in two sequential ways. First, there should be a separation between the bailout decision and the ability to recover money from shareholders to pay for it, in order for the government officials to bear the political costs of bailouts, which constituents will almost certainly view with hostility and skeptic ism. (84) Hence the two years: there will be an intervening election between bailout and reimbursement to hold those political actors responsible for the bailout they approved. Disallowing the collection up front for two years will force the government to effectively ignore the recovery option of elective shareholder liability in the short term. Second, the government must therefore pursue other mechanisms that may decrease taxpayer exposure. If the government cannot count on recovery from shareholders as part of its politically accountable program, it will seek other ways to decrease taxpayer exposure--with creditor haircuts and liquidation among the most promising.

4. Pro rata share

In determining how to structure the terms of shareholder liability, we face essentially two models. First, there is the private general partnership model utilized by investment banks until the period from 1970 to 1999, when they all changed forms and became public corporations. In that model, the liability is joint and several, with each partner liable for the debts of the entire partnership. Second, there is the model utilized by commercial banks between 1865 and 1933, under which shareholders paid only their pro rata share.

Because there is extensive turnover in the shareholdings of public companies today, a pro rata liability system makes far more sense for a TBTF non-limited liability regime. (85)

There remains the difficult problem of the temporal dimension of share ownership. A backstop must therefore be determined to establish a relevant window of ownership. I propose, arbitrarily, one year--that is, the relevant shareholders for purposes of this proposal are those who owned shares within one year of the taxpayer bailout. The pro rata share for which each shareholder will be liable, then, is a function of the amount of ownership and the time of ownership. In a world of flash trading, where shareholdings may last only fractions of a second, we must round each shareholding period to another arbitrary period--say, the nearest month. (86)

Recovery rates under this regime will be far from perfect. Indeed, if they are anything like the double-liability regime that existed for commercial banks, the recovery rate will be roughly fifty percent. (87) And although the taxpayer reimbursement would be issued as an assessment, and not a judgment that would need to be litigated, the reality is that some parties will fight the assessment. Litigation costs will thus be a factor, and will likely arise in efforts to identify shareholders who are seeking to hide behind other corporate entities. Nevertheless, as we will discuss further, there are still significant benefits to this modification, even if recovery is low and litigation costs are high.

5. Ability to nullify evasion efforts through rulemaking

This element is an effort to give the government the blanket authority to declare, through rulemaking, that any efforts to evade liability through use of the corporate form are null and void. This is a far-reaching authority, but an important one. As will be discussed below, some of the principal objections to non-limited liability in the past have been on the basis that shareholders will have trivially easy ways of avoiding liability through the use of the corporate form. In that case, non-limited liability becomes a shell: if all shareholders are themselves limited liability entities, then any effort to collect more than the equity investment is immediately thwarted.

The nature of such rulemaking may seem, at first glance, quite sweeping, but there is good support for such anti-evasion mechanisms. It is essentially an administrative effort to systematize the judicial doctrine of piercing the corporate veil, a largely pilloried judicial doctrine that, "[l]ike lightning, ... is rare, severe, and unprincipled." (88) The anti-evasion mechanism here would be dramatically more predictable: efforts to use the corporate form in all instances to evade liability for taxpayer bailouts will fail. The uncertainty inherent in veil-piercing cases would be eliminated. Of course, the exact contours of such regulation would depend on a more thorough assessment of the rulemaking authority than is undertaken here, but there is no reason why the FSOC could not engage in this sort of rulemaking--particularly as it is consistent with the FSOC's overall mission "to identify risks to the financial stability of the United States that could arise from the material financial distress or failure, or ongoing activities, of large, interconnected bank holding companies." (89)

6. Treble damages, including litigation costs

Another anti-evasion element of the proposal is that while shareholders are welcome to challenge the collection efforts, just as they might challenge their taxes, they must pay treble damages, including litigation costs, in any case they lose. This element serves two purposes: First, it discourages such suits in the first place. Determining the identity of the shareholders should be a straightforward enterprise, as should the determination of the taxpayer losses. Efforts to evade straightforward collections will thus become very expensive for shareholders, as they will pay both their own and the government's litigation costs. Second, the entire point behind the proposal is to keep the taxpayers from subsidizing a SIFI's TBTF status. If SIFIs can evade the assessment while running up the government's litigation costs during the collection process, then much of this value is lost.

The concept of treble damages would not be unique to this context. Treble damages already exist in antitrust, (90) racketeering, (91) trademark, (92) patent, (93) and environmental actions. (94) Indeed, treble damages have a long historical pedigree within which liability evasion in banking would fit comfortably. (95)

B. Benefits

An elective shareholder liability regime would be the most effective of any proposal previously considered at both bailout prevention and bailout reimbursement, and would provide a number of benefits to both markets and regulators.

1. Taxpayer reimbursement

Elective shareholder liability makes money available for taxpayer reimbursement without requiring government participation in equity and its concomitant problems. (96) Until now, there has been no conversation about recovery from shareholders, beyond the provision in Dodd-Frank that shareholders cannot recover via FDIC liquidation until all money has been restored to the taxpayer. (97) Elective shareholder liability changes this. The product of these assessments is, after government recovery, money that was not previously availa-available.

As mentioned, the expected recovery value in this instance may well be quite small. Under the model of commercial banks under double liability, recovery was, historically, just fifty percent. (98) Taxpayer reimbursement, then, is not the chief benefit, nor is the workability of the recovery mechanism the most important aspect of the proposal. But the amount of recovery here is not nothing--if shareholders elect liability, and prepare accordingly, then the triggering mechanism for liability will result in a payout to the "winner" of the bet, just like any bet that sophisticated investors make and lose. Thus, equity investments may come to look more like derivatives investments rather than traditional equity investments. Derivatives are frequently pegged to fluctuations with theoretically uncapped limits. Equity is not. Taxpayer reimbursement is thus an expected benefit simply because investors will have bargained ahead of time for the bet and would, conceivably, prepare to pay out when the bet ends against them.

2. Risk management

Elective shareholder liability also dramatically increases shareholder incentives to monitor risk in several ways. First, shareholders who elect to hold non-limited liability equity will have increased monitoring incentives, as they will have a serious financial stake in reducing the risk that would lead to a taxpayer bailout. Under the traditional model of corporate governance, this oversight is effectuated through the election and reelection of the board.

Of course, the extent to which shareholders can and/or should effectively participate in corporate governance is probably the most hotly debated issue in corporate law of the last twenty years. (99) But we can sidestep that fraught debate. Elective shareholder liability benefits do not rely on shareholder activism so much as changed costs and incentives for management and their SIFI firms. The very structure of elective shareholder liability forces firm management to monitor risk, since the failure to do so would result in prohibitive equity costs. That is, shareholders would, by virtue of refusal to participate in the equity markets, signal the essential need to contain risks well beneath the level at which bailouts would occur. Shareholders will thus provide the monitoring that may be otherwise absent in TBTF firms. And even more directly, SIFIs have historically used equity compensation of varying kinds to reward employees. If that continues--and it is so built into the legal structure of public firms that elective shareholder liability is unlikely to change it--then the shareholders most interested in risk management will be the directors and officers themselves. (100)

Although elective shareholder liability contains common elements with the historical partnerships that characterized investment banks, the proposal is not simply a return to those non-limited liability structures. Instead, it is an effort to increase management and shareholder incentives to manage risk such that bank activity contemplates all of the costs of doing business, and not just those costs that the shareholders cannot pass through to the taxpayers. Bebchuk and Spamann make a similar argument in their effort to reform executive compensation for the SIFIs by advocating for compensation linked to total return on assets, as opposed to simply return on equity. (101) Bebchuk and Spamann's efforts, complementary in several ways to elective shareholder liability, address the same phenomenon: under the current regulatory regime, SIFI shareholders' incentives are aimed at displacing as much of the cost of their risky behavior to the public as is possible. Elective shareholder liability would realign those incentives and increase the probability of more adequate risk management.

Another way in which elective shareholder liability influences the SIFI's risk management for the better is through the partial elimination of what Richard Squire has called "correlation seeking," or the practice of seeking "to incur contingent debts that correlate, or that through asset purchases can be made to correlate, with the firm's insolvency risk." (102) Because these contingent debts are correlated with insolvency risk, they often become due only after shareholders have already lost their entire interest in the firm. (103) Correlation seeking is a nonefficient transfer of wealth to shareholders (who enjoy the upside of contingent debts) from creditors (who ultimately foot the bill when things go wrong). But correlation seeking is much more than that--it destabilizes financial systems, increases borrowing costs for other parties, and inefficiently creates overinvestment in those activities where, contrary to finance theory, risk stays the same but returns increase. (104)

Elective shareholder liability addresses this problem both directly and indirectly. It addresses correlation-seeking directly because if correlation seeking results in an insolvency followed by a taxpayer bailout, the benefits that accrued to shareholders will no longer be beyond the bounds of recovery. Thus, the risk-return balance that is disturbed by correlation seeking is partially restored. But even if the correlation seeking does not cause an insolvency followed by a bailout, elective shareholder liability still provides value indirectly. It is impossible to know with certainty which firms will or will not be bailed out. Thus, any SIFI that has an opportunity to seek the correlated risks that Squire describes will have a check on that incentive: the mere possibility of bailout and elective shareholder liability will render that strategy less effective. (105)

3. Partial elimination of TBTF subsidies

There are many benefits that come with the SIFI status, not least of which is the perception in the marketplace that the firm's debts will always be paid regardless of its level of solvency. Indeed, Standard & Poor's, the credit rating agency, has made this implicit government support central to its analysis of SIFI creditworthiness. (106) The lowered borrowing costs are, themselves, the TBTF subsidies.

Elective shareholder liability changes these subsidies. While firms may still receive preferential treatment in credit markets in light of their SIFI status, the fact that these firms' shareholders will be on the hook to refund the cost of the bailout means that the costs of failure will be shared between creditors and shareholders, rather than foisted upon the government. (107) But the shareholders would pay for them--not the taxpayers.

4. An active derivatives market for shareholder liability

To understand the next benefit, consider a simple analogy. There are two options that one might face when considering a home purchase in an area known for having hurricanes. First, do not move to a place where there are hurricanes. Second, if you do move there, buy insurance.

I have already discussed the analogue to avoiding shareholder liability completely: simply vote for capital adequacy, and if the rest of the firm disagrees, sell the shares. For insurance, though, elective shareholder liability presents a fascinating opportunity--I would argue its greatest feature--in the form of the creation of a new derivative. Call it the shareholder liability swap (SLS). The issuer would guarantee to pay the holder of equity in an elective shareholder liability SIFI enough to cover any losses following a taxpayer bailout. In this sense, an SLS is similar to a credit default swap (CDS), which pays a bondholder the value of a bond in the event the issuer of the bond defaults.

At present, there is significant informational content gleaned from the CDS market about expectations of failure. Regulators making snap decisions in the fall of 2008 depended on the CDS markets for information concerning likelihood of failure. (108) And the present apparatus for making these kinds of credit decisions is dependent on the credit rating agencies themselves, who have come under significant criticism during the recent crisis. Indeed, the SEC has just announced its intention to expunge references to the credit rating agencies throughout the securities laws. (109)

Some scholars, including leading derivatives law scholar Frank Partnoy, have argued that CDSs should replace credit ratings agencies in their evaluations of likelihood of default, (110) as this kind of market information can react more quickly than the ratings agencies possibly could.

This presents the opportunity for a derivative that would be used to determine the likelihood of a government bailout--an SLS. Whatever the benefits that Partnoy et al. have identified from CDSs as a mechanism for rating credit risk, an SLS derivative would be superior. The present CDS market reflects the likelihood of debt default for a specific company. This was recently made explicit in a report by Standard & Poor's, which described the proposed rating methodology to be used in assessing a bank's creditworthiness. (111) The report addressed the fact that banks' credit is logically tied to "an institution's support framework," including "potential government support." (112) The very nature of TBTF, however, indicates that SIFI creditors will not bear the full cost of failure given the firms' systemic importance. Accordingly, the CDSs of SIFIs will also include the likelihood of a bailout--and in the case of a bailout, SIFI creditors will be made whole, or partially so. In this sense, it is difficult if not impossible to separate out the informational content in CDSs from the market's instinct about a forthcoming bailout. The higher the likelihood of a bailout, the lower the CDS spread in the market.

Such would not be the case for SLSs. In that case, the issuer would provide a guarantee to the purchaser that, in the event of a government suit against the shareholder to recover the partial value of the taxpayer bailout, the SLS issuer would front the cost. As the likelihood of a government bailout increased, the premiums paid on those SLSs--due, again, even after the equity investor divested from the SIFI's equity--would increase. The informational content, of great benefit to the marketplace and regulators alike, would more purely reflect the likelihood of bailout. The traditional CDS market would therefore more purely reflect the likelihood of default, unclouded by the noise associated with a potential bailout.

Furthermore, the presence of taxpayer reimbursement derivative insurance would not necessarily eliminate the monitoring benefits of the elective shareholder liability regime. The fact that the premiums start to increase means only that purchasers of such insurance would be in a position to agitate for changes in the management's risk strategies, as increased premiums would directly affect the value of their equity investments.

Of course, any proposal, post-AIG, that incorporates CDS-like derivatives into the architecture of systemic risk reduction requires a bit more explanation. After all, CDSs were the AIG crisis.

But there are reasons why SLSs after Dodd-Frank would not look like CDSs before it. Title VII of Dodd-Frank provides an institutional framework for derivatives regulation that will go a long way toward making the kind of opacity that occurred in the AIG debacle much less likely. And although much derivatives reform is still ongoing through the rnlemaking process at the SEC and CFTC, the architecture is already in place. Standardized derivative contracts like SLSs would be exchange traded, allowing the issuers of SLSs to simply take the form of any member of the exchange. And while the exchanges themselves may become TBTF in their own right, (113) that is the case regardless of whether the SLS market thrives or not. Adding SLSs will increase the presence of derivatives, but not the strength or weakness of the present derivatives regulatory architecture.

5. Tailoring corporate form and leverage structure

The last benefit of elective shareholder liability comes from its most important feature: optionality. Shareholders are given the option, via shareholder vote, to determine whether the benefits of size and leverage (through capital adequacy) are truly worth the cost of increased likelihood of taxpayer bailout. If they are, then shareholders can make that determination and proceed under the theory that taxpayer bailouts will be funded on the backs of the shareholders themselves. This ability to tailor corporate form, leverage structure, and asset size makes elective shareholder liability a relatively bank-friendly regime-it identifies the cost of public bailouts and allows banks to choose how that cost is to be internalized.


This Part engages the many objections and counterarguments against elective shareholder liability, beginning with the argument that shareholders are not the best group to manage systemic risk. This Part also engages some of the arguments lodged against the last academic critique of limited liability, offered by Henry Hansmann and Reinier Kraakman in 1992, by showing in some cases why arguments defending limited liability failed in the first place, and, where they succeeded, illustrating why elective shareholder liability does not suffer from the same weaknesses as general corporate non-limited liability.

A. Alternative Risk-Bearers Purportedly Superior to Shareholders

One class of objections to elective shareholder liability concentrates on the identity of the residual bearers of risk. Shareholders, it may be argued, are not best situated to cover losses beyond their equity investment, both because of attributes unique to shareholders, and because other parties would manage that risk better. In particular, scholars and commentators have addressed five sometimes-overlapping sources that might diminish the extent of taxpayer bailouts or provide some mechanism for taxpayers to recoup bailout costs in general. They are: (1) creditors, in the form of haircuts; (2) the government and the corporation, by allowing government participation as common or preferred shareholders or creditors; (3) industry, through a liquidation insurance fund; (4) directors and officers of the corporation, who might face liability for the consequences of their risky decisions that result in the need for a taxpayer bailout; and (5) no one at all--that is, the taxpayer bears the cost of the failed firm. In each case, the alternative is inferior to shareholders in an elective shareholder liability regime.

The first is effectively the bankruptcy proposal championed by David Skeel. The second is the story of the bailouts of 2008-09. The third is the approach taken by Dodd-Frank. And the fourth is a novel proposal by Bill Cohan, columnist for the New York Times, developed more fully by legal scholars Claire Hill and Richard Painter. (114) Each of these proposals has its virtues, but each ultimately fails. The first three fail either at the task of reimbursement or, more importantly, at changing market incentives in a way that would make these bailouts less likely to occur. The fourth proposal, the Cohan-Hill-Painter proposal, fails simply because it is unworkable. Elective shareholder liability does not make the same mistakes, and thus accomplishes what no other considered proposal does: it targets SIFI incentives to avoid bailout ex ante while also providing for some means of taxpayer reimbursement ex post.

1. Creditors

The first group to whom the government, in a crisis situation, may turn is the SIFI's creditors. Indeed, the creditors are the crisis--their clamor for repayment creates the selloffs that trigger the crisis in the first place. A bailout, it has been said, is a creditor bailout. (115) Requiring creditors to take lower return--a "haircut," in market parlance--is one effective way to diminish the size of the eventual pool provided by taxpayers to support a failing institution.

The argument in favor of institutionalized creditor haircuts is reminiscent of the proposals to make the regulation of systemic risk more like the bankruptcy process. Indeed, institutionalized creditor haircuts undergird the entire bankruptcy system. In bankruptcy, creditors are generally prevented from pressing their claims against an insolvent institution during a specified period following the bankruptcy filing. (116) If the institution is no longer economically viable, an orderly liquidation of the firm's assets will allow creditors, according to the seniority of their debt, to maximize their recovery. That maximized recovery, however, may still be less than the face value of the debt, depending on the size of the asset pool, the number of creditors, and the nature of the creditors' contracts with the debtor. (117) Creditors receive so many cents on the dollar for their debts, and are thus forced, by economic reality, to "take a haircut" on their debt. This process is orderly, (118) the rules clear, and its strengths and weaknesses well established. It also guarantees that creditors bear the cost of the failed entity, by limiting their recovery to the size of the bankrupt entity's as sets and nothing more. As noted, Skeel and others have argued that the existing apparatus of bankruptcy should be applied, with some modifications, to the resolution of financial crises. His basic argument is that "giv[ing] the bankruptcy laws a chance," (119) with some adjustments to the present Bankruptcy Code, (120) would end the need for what he calls the "institutionalized bailouts" created under Dodd-Frank.

The arguments in favor of institutionalized creditor haircuts have a persuasive logic for both economic and political reasons. Economically, the reality is that the failure of the SIFI has occurred precisely because it cannot pay its debts as they have come due--its debts, tautologically, to creditors. Forcing creditors to take a haircut as a matter of crisis regulation has important implications for the taxpayer commitment. Indeed, the two are in many ways reciprocal--the larger the haircut, the smaller the bailout. Additionally, institutionalized creditor haircuts--whether through bankruptcy or otherwise--would function effectively in crisis prevention. Creditors who face the guarantee of haircuts in the case of a failure will be more likely to impose more financial discipline on their counterparties. It is implausible to assume that after the AIG experience, counterparties to sellers of CDSs will make assumptions about the sellers' creditworthiness if they know that there is a very real risk that they will not recover the face value of the CDS contracts.

Politically, creditor haircuts may also reduce the backlash associated with creditors' failure to share in the losses of a failed SIFI. The AIG Bailout is, again, Exhibit A. The first iteration of the AIG bailout occurred on October 18, 2008, when the government announced the first $85 billion loan to AIG. (121) Of this initial loan--which subsequently grew to an authorized amount of $182.5 billion (122)--$30 billion went directly into the hands of Goldman Sachs, Credit Suisse, Societe Generale, and others of AIG's banking counterparties at the face value of their contracts, causing a firestorm of controversy. (123) This peculiar decision--to pay AIG's counterparties at full value--was made by the Federal Reserve Bank of New York and the U.S. Treasury. Goldman Sachs, the investment bank that stood to make billions from an AIG bailout, has asserted through its CEO, Lloyd Blankfein, that it never received any request that Goldman Sachs take a haircut on its debts. (124) Indeed, it was reported that AIG began efforts to attempt to negotiate a haircut with its creditors, but that the Federal Reserve Bank of New York insisted that it pay its debts at full value. (125)

The benefits of creditor haircuts notwithstanding, there are two reasons why creditors can neither prevent taxpayer bailouts, nor subsequently reimburse taxpayers once those bailouts occur. First, when looking at the role of ex ante bailout avoidance, creditor haircuts are plausible, but imperfect. The idea that creditors will serve as effective monitors in helping firms avoid failure is a familiar concept. (126) But the recent crisis shows its weakness. Setting aside the example of Lehman Brothers--which John Taylor and David Skeel view as overstated (127)--the example of AIG stands starkly against the notion that creditors will prevent SIFI collapse through their own self-interest. AIG's counterparties did little to discipline the firm in its risk concentration--the best they did was, in some cases, hedge against that loss. (128) Hedging may well be the best approach for the counterparty itself, but it hardly qualifies as the monitoring that defenders of creditor haircuts have claimed.

Second, creditor haircuts will, in the extreme cases, be insufficient unless haircuts become creditor wipeouts. Contingent claims could so massively overwhelm a SIFI's assets as to render the creditors nearly completely wiped out. (129) AIG eventually needed $182 billion of taxpayer aid, only $30 billion of which would have come from its CDS counterparties. (130) The remaining $152.5 billion was due to other losses and obligations. Only if all such counterparties were wiped out completely would taxpayers be spared. Advocating for this sort of extreme position--bankruptcy at all costs to society--is precisely the weakness in the bankruptcy-for-SIFIs approach. The wrong kind of financial crises can occur such that full creditor wipeouts cause otherwise stable institutions to teeter, then collapse, creating an endless cascade of failure and panic that eventually results a Great Depression-style economic landscape. (131)

Moreover, as Secretary Geithner has said on numerous occasions in defense of the AIG bailout, creditor haircuts are not always the best mechanism for restoring confidence in the financial system in the heat of crisis, as fears of counterparty exposure continue to speed investors in their flight to safety. (132) When counterparties to a failing SIFI are facing the same systemic risks, forced haircuts may only extend the crisis, rather than eliminate it. The fact that the very nature of creditor haircuts forces decision making the eye of the financial storm also makes such haircuts a less palatable backstop to preventing taxpayer bailouts.

Institutionalized creditor haircuts therefore offer a partial solution to problems of both SIFI failure prevention and taxpayer reimbursement. But again, that solution is not complete. Even under a creditor-haircut regime, taxpayers remain the residual bearers of risk for SIFI failure. And while that risk may never be fully eliminated, elective shareholder liability provides exactly the mechanism to reduce such risks further than a bankruptcy and/or institutionalized creditor haircut approach could. (133)

a. Contingent capital

Another variation of the argument that creditors should be made to bear the losses comes in the form of a specific kind of debt instrument, called "contingent capital" or "coco" bonds. Coco bonds create a system whereby debt is converted, "in a timely fashion," into equity, replacing lost capital before a crisis leaves SIFIs depleted and in need of taxpayer bailout. (134) John Coffee, the leading proponent of coco bonds among legal scholars, argues that the debt conversion that undergirds coco bonds "makes sense because it avoids the costs, delay and uncertainty of a bankruptcy proceeding, while (1) scaling down the firm's debt burden when necessary and (2) still permitting the firm to use tax-advantaged debt financing." (135)

Coco bonds and elective shareholder liability share the same policy motivation: the reduction of moral hazard and TBTF by making equity more plentiful relative to the SIFI's risk profile. And to the extent that coco bonds have already gained traction among some industry participants--indeed, the first ever issuance of coco bonds, by Credit Suisse, was oversubscribed by eleven to one (136)--they represent an easier path to implementation than the non-limited liability securities proposed here.

There are, however, two problems with coco bonds that make them inferior to elective shareholder liability. The first is the relationship between market contagion and the "triggers" that signal the conversion from debt to equity. The question of when the conversion should take place is still debated. (137) To the extent that the conversion occurs on the depletion of capital, a coco bond is no different from a mandatory capital requirement, and in that sense is a worthwhile proposal and complementary to capital requirements. But if the trigger is set by reference to other factors--the downgrading of debt by a credit rating agency, spikes in CDS spreads, or other market-sensitive information--the conversion itself can be a source of the very contagion and panic that the bonds are designed to avoid. This is not just a question of irrational market jitters. When the conversion itself is a function of opaque assets suddenly losing value, panic can rationally take hold as counterparties rush to assess who else is similarly, but still secretly, exposed to the offending assets. This narrative should sound familiar: the reaction to massive, iterative writedowns of mortgage-backed securities in 2007-08 produced this same dynamic.

The second problem for coco bonds is existential: they simply do not do a better job of crisis avoidance than a higher baseline capital requirement, precisely because they allow banks to lose market value to a crisis point before the conversion takes place. A stronger, more consistent capital requirement is superior because the crisis point is much less likely to occur. Similarly, if a SIFI elects shareholder liability, the incentives for avoiding that crisis point increase.

The argument in favor of coco bonds, at a theoretical level--decidedly the plane on which elective shareholder liability engages in this debate--is strong, but not conclusive. The virtue of elective shareholder liability, regardless of the election made, is the ability of SIFIs and their trailing regulators to avoid both the definitional and decisional elements of the conversion trigger that could, by its very triggering, invoke panic.

2. Corporation and government: government participation in SIFI equity

The second way that taxpayers can be made whole is through government participation in the equity, or to a lesser extent, the debt, of a bailed-out SIFI. This is the approach used in the bailouts that occurred under the Troubled Asset Relief Program (TARP), an approach that has been facially successful. (138) In this scenario, and contrary to the original TARP proposal, the government injects capital into failing SIFIs through equity participation, rather than--as the TARP name suggests--through the purchase of the "troubled assets" that infected the SIFIs' balance sheets.

The motivating policy behind government participation in SIFI equity is precisely the interest in making taxpayers whole. The effort under TARP was to "save" these institutions--a salvation that meant, by definition, that they would be operable and functional again in the future. In such a case, government participation in equity meant that the taxpayer could benefit from the growth of these institutions and eventually recover the initial payment from the Treasury.

This has been the case in the recent bailouts. With the exceptions of Freddie Mac and Fannie Mae, the Treasury has recouped its initial payments of $700 billion. A report from the quasi-independent Congressional Budget Office estimates that the cost of the bailout, originally expected to be $350 billion, will be $25 billion. Even this figure represents, to Secretary Geithner, too large an estimate; as he has stated, "The cost of TARP is likely to be no greater than the amount spent on the program's housing initiatives. The remainder of the investment programs under TARP--in banks, AIG, credit markets, and the auto industry--will likely, in the aggregate, ultimately yield a positive return for taxpayers." (139)

In this instance, government participation in private corporations' equity appears to be a stunning success: a bailout occurred, and the taxpayer was made whole. If that is the goal, why look elsewhere for a mechanism of reimbursement if government equity participation is all that is needed?

We look elsewhere for three reasons. First, the government may not always be so lucky. While the "bailouts" for Goldman Sachs, JP Morgan, and Wells Fargo were profitable from the very beginning, (140) and those for Citigroup, Bank of America, and AIG ultimately paid off (141)--to the surprise of many-Freddie Mac and Fannie Mae stand as stark examples of bailouts that may not pay off. (142) And even if no tax dollars are lost in the government-sponsored-enterprise bailouts, there is no guarantee that such will always be the case. The government takes a risk--sometimes not a very well-calculated risk--when it participates in the recovered health of bailed-out institutions, and sometimes that risk means that taxpayers lose.

Second, as Nobel Prize-winning economist Joseph Stiglitz has explained, the actual economic cost of bailouts is far more than the amount of money handed over from the government to the failed banks. That dollar amount is a "drop in the bucket" compared to the damage that such failures and bailouts do to the overall economy. Stiglitz estimates the cost of the bailouts in the trillions, not billions, of dollars. (143)

Third, government equity participation carries with it a host of problems for the way we structure and regulate capital markets. Three scholars have taken up this issue recently, with overlapping conclusions. J.W. Verret argues that government shareholder ownership undermines corporate theory, largely because the new controlling shareholder--the federal government--is immune from federal securities lawsuits and state corporate lawsuits by virtue of its sovereign immunity. (144) Moreover, as Verret argues, the government explicitly negotiated for the entrenchment of this immunity by requiring TARP participants to waive claims on liability under the Securities Act of 1933 and the Securities Exchange Act of 1934. (145) Marcel Kahan and Edward Rock reach similar conclusions, arguing further that because existing law is insufficient to protect minority, noncontrolling shareholders in corporations where the government is the controlling shareholder, we face a choice: we must either "develop[] new structures of accountability [or] bring[] this anomalous era of government control to a speedy conclusion." (146) Kahan and Rock further argue that government interests are going to be intensely conflicted in their roles as politicians, shareholders, and policymakers. What is best for the corporation may not be best for the country, as the administration would define it; and what may be best for the corporation may not be in the President's best political interests. When those interests collide, the presumption that shareholders are united in their interest to maximize the value of the corporation--the assumption that undergirds most of corporate law--no longer holds true.

Even though government participation in equity through the TARP process proved successful, it is difficult to dismiss the Verret and Kahan-Rock criticisms of the role of the state as a shareholder. In other words, government participation in the equity of the corporation comes with a cost that may make the question of taxpayer reimbursement seem quaint by comparison.

Furthermore, the promise of government equity participation does nothing, ex ante, to force firms to internalize the costs of failure. Indeed, counting on the willingness and availability of the government to participate in a SIFI's equity as part of a bailout structure is the very definition of a taxpayer bailout. That guarantee is precisely the kind of regulatory action that banking regulation should avoid.

3. Industry and government: an insurance fund

Individual SIFIs may also be bailed out by their competitors, through an industry fund. Under this model, the taxpayer is only responsible for covering the failed SIFI's losses beyond the extent covered by an industry fund. This approach is significantly expanded and developed by Jeffrey Gordon and Christopher Muller, (147) and by Arthur Wilmarth. (148)

There are three inputs that determine the cost of the taxpayer bailout: (1) the SIFI's liabilities, (2) the assets of the insurance fund designed to prevent taxpayers from paying for the bailout, and (3) the SIFI's assets.

When the insolvency is relatively modest, and the industry fund is sufficient to cover the difference between the insolvency and the institution's liquidation or resale, then there will be no taxpayer bailout. But when that fund is bankrupt, then the taxpayer is on the hook. The point, then, is that under a liquidation model, the taxpayer is still the residual lender, with only the industry fund to protect them. Thus, the value of this model is at best the deferral of taxpayer bailouts in some cases.

The Gordon-Muller and Wilmarth models anticipate this residual loss in two ways. First, they propose to significantly expand the size of the fund such that the likelihood of bailout is even more remote. (149) Second, Wilmarth proposes to make mandatorily issued contingent capital a significant component of bank director and officer compensation. (150) This proposal, particularly the version of it that these scholars present, has significant merit.

However, history has not been kind to this model of banking regulation in the event of systemic crises, at least if the criterion of success is taxpayer loss avoidance. The last systemic banking crisis that the country faced was the savings and loan crisis of the 1980s and early 1990s. In that crisis, savings and loan institutions faced mass failures, especially in the South and Southwest. To respond to this collapse, the government passed the Financial Institutions Reform, Recovery, and Enforcement Act of 1989 (FIRREA). (151) FIRREA established the Resolution Trust Corporation to wind down failed thrifts using government resources--in intent and effect, a taxpayer bailout, wherein the government became the owner of the failed institution. Indeed, one contemporaneous scholar called FIRREA "one of the more dramatic, controversial, and expensive pieces of federal legislation in our history." (152) And while some would argue that the Resolution Trust Corporation was necessary and helpful, it was not costless. Private estimates put the taxpayer cost of the Resolution Trust Corporation between $140 and $150 billion; (153) the government's own estimate is roughly the same. (154) And the Federal Deposit Insurance Fund, the mechanism used by the FDIC to cover the costs of liquidation of small retail banks, is itself insolvent and has been for some time. As of November 2010, the net worth of the fund was negative $8 billion. (155)

Even if that fund were completely and adequately capitalized, there are real problems that might be anticipated from the perspective of political economy. First, there will be an enormous temptation to use that money to fund government debt, as was the case with the Social Security fund under the Greenspan Commission in the early 1980s. (156) There are real problems with a plan that would stabilize teetering banks by destabilizing the credit of the United States through a massive sell-off of Treasury bonds. Second, and even more likely, an idle trillion dollars in a bailout insurance fund is going to be a prime target for deregulation when times are good. Arguments in favor of releasing the brake on the economy in the form of idle reserves are likely to gain traction in good times, making the preservation of these funds for times of financial stress unlikely.

Furthermore, deposit insurance funds provide little incentive for the industry to monitor itself and others. Whether the industry is healthy or ill, all SIFIs pay into the fund regardless of how well-managed their risk. (157) While it is true that the funds historically have been pegged to the riskiness of asset classes, such a system is always backward-looking and usually pegged to the riskiness as identified by regulators or legislators (158) rather than the more dynamic risk management of private market participants. There is no diminished assessment for avoiding bailout, nor heightened assessment in the event of bailout. Thus, deposit insurance funds are only a mechanism for providing liquidity in the resolution process. And once that liquidity is gone, as will be the case in any serious financial crisis, the residual risk bearer is still the taxpayer.

In spite of the reality of the precariousness of this model, Dodd-Frank institutionalizes the liquidation model for SIFIs. The criticism I raise here is simply that such liquidations do not provide for taxpayer reimbursement in the event of bailout beyond the value of the fund itself. If the liquidation model works well in some crises, then there is no need to wonder about taxpayer reimbursement. When the right crisis does occur, however, the liquidation model will not protect taxpayers.

4. Directors and officers

One proposed alternative for reimbursement is to make the bankers themselves pay. The best articulation of this sentiment comes from Bill Cohan, on the opinion pages of the New York Times, and from law professors Claire Hill and Richard Painter in a more fully developed account. Cohan, a former investment banker and author of the first true financial crisis book, (159) decries the "irresponsible, accountability-free incentives [bankers and traders] have had for the past 40 years to generate as much revenue as they possibly can each year, regardless of the consequences." (160) Cohan blames Wall Street compensation practices and argues that such practices lead to "excessive risk-taking." (161) Cohan views Dodd-Frank as having failed to rein in such practices, and instead argues for a system based on personal liability:
   [E]ach of the top 100 executives at Wall Street's remaining
   'systemically important' firms [must] be personally liable for the
   risks they take. Not just their unexercised stock options or
   restricted stock, but every asset they have in their possession:
   from their cars to their fancy homes to their bulging bank
   accounts. (162)

There is an intuitive appeal to requiring directors and officers to face the cost of their risk-taking. First, these executives are those who have most aggressively taken advantage of the value of inflating the very bubbles whose explosions require taxpayer bailouts. For example, Dick Fuld, the former CEO of Lehman Brothers, who directly supervised the institution's descent into bankruptcy, took home $484 million in the eight years preceding the firm's implosion. (163) Stan O'Neal, the similarly situated CEO of Merrill Lynch, received $320 million during an even shorter period of similar failure to oversee his firm's risk-taking: roughly $160 million from 2002-07 as CEO, and roughly $160 million as severance after his termination in 2007 for the very risk-taking that forced his successor to sell the firm to Bank of America. (164)

The directors and officers are also, by far, the best situated of any party-including shareholders--to avoid the risks that could eventually cause the firm to implode. In terms of ex ante prevention, to the extent that any party can predict the need for a bailout, no one would do better than directors and officers; and liability for such losses would concentrate their minds wonderfully, much like the prospect of a coming hanging. (165)

The law already provides a mechanism for recovery against wealthy officers of corporations that prove to be less financially successful than initially reported. Under section 304 of the Sarbanes-Oxley Act, the SEC can sue to seek reimbursement from any CFO or CEO whose company must "prepare an accounting restatement due to the [company's] material noncompliance" with financial-reporting securities laws. (166)

Dodd-Frank embraces and extends the Sarbanes-Oxley clawback approach. Dodd-Frank requires that all publicly traded companies adopt some clawback policy that allows the SEC to recover incentive-based compensation from any executive who received such pay based on financial information that had to be restated due to material noncompliance with the federal securities laws. In the case of such restatement, the executive has to disgorge the difference between what she would have received under the restated information and what she did receive under the incorrect, originally disclosed information. (167)

The Dodd-Frank clawback is stiff medicine, but it is only helpful by analogy to the Cohan proposal. The clawbacks are based on material restatements of reported earnings, not on the actual costs of risks taken. But the analogy is helpful all the same: the idea is that executives internalize the cost of their decisionmaking by catching both the upside, which they have long received, and the downside, which has been effectively pushed on to taxpayers under the present model of TBTF. In that sense, the clawback provisions in both Sarbanes-Oxley and Dodd-Frank perform, partially, the function that Cohan encourages.

The difference between the clawback provisions and the Cohan proposal, though, is that Cohan and those who agree with him would have officers be responsible for a much broader set of losses than either of the current clawback provisions. Cohan wants risks to be absorbed by executives; the clawback provisions only address fraud. Thus, the Cohan proposal would be much more expansive, and force firms--and, pointedly, their executives--to internalize the costs of their risk-taking in a way that the clawback provisions simply do not.

The problem with Cohan's and Hill and Painter's proposals is simply workability. In the flow of business decisions, it can be difficult to pin the losses from any one decision--for example, to issue CDSs without a corresponding hedge, as AIG did--on any one person. In that event, who is on the hook? The trader who initiated the first deal? The executives who approved it? The lower-level employee who made the hundredth such deal, following previously established protocol? Hill and Painter deal with this problem by apportioning liability strictly, (168) an implausible suggestion in light of the amount of money in play. Although executives earning eight figures may not fit the popular image of "judgment-proof," they would quickly become judgment-proof if liabilities several orders of magnitude higher than their net worth were introduced. Elective shareholder liability deals with this problem much more effectively through pro rata division of liability.

5. The taxpayers

Whatever the strength of the preceding four proposals, taken together they do not shield the taxpayer from residual liability. This is, as I argue above, to be expected. Bailouts cannot be retired as a regulatory mechanism. The only hope for financial regulation is to make them as few and far between as possible. When they do occur, though, there should be some mechanism to guarantee the maximum recovery possible for taxpayers. Under the present model-liquidation of SIFIs with an industry fund not unlike the Deposit Insurance Fund, with the potential for creditor haircuts and perhaps government equity participation--the taxpayer's potential liability is not minimized to the fullest extent possible. The taxpayer remains on the hook. To bridge that gap, we must return to the forgotten corporate participants in this discussion: the shareholders.

B. Where Are the Shareholders?

No one has addressed whether shareholders should join the list of parties above and reimburse taxpayers following a bailout. (169) Shareholders are probably omitted for two reasons. First, the assumption is that the shareholders have already paid. In the AIG case, for example, shares were trading in February 2008 at roughly $50 per share. (170) By June 2009, the price had fallen to under $1 per share. (171) The shareholders who rode the bubble through to its implosion have already lost their shirts. Alternatively, perhaps the terms of the bailout itself already unceremoniously wiped out the shareholders. In the Bear Stearns case, for example, the government forced JP Morgan to offer a mere $2 per share to the Bear Steams shareholders. Treasury Secretary Hank Paulson was insistent that the shareholders could not be seen as receiving a bailout. (172) But he certainly did not insist that the shareholders pay the full cost of bailing out others. Indeed, the shareholders were extraordinarily angry at the lowball offer, and felt comfortable holding out for the $10 per share they eventually received-even though the firm's assets were literally worth nothing at that point.

To be sure, there has been some hostility to the idea that shareholders will survive bailouts with some part of their equity stake intact. Dodd-Frank confronts this question by prohibiting shareholders from recovering any "payment until after all other [creditors'] claims and the [Orderly Liquidation] Fund are fully paid." (173) Thus, unlike the Bear Steams shareholders, shareholders under Dodd-Frank cannot receive a payout until after all other claims have been paid, including the OLF's claims. Perhaps, then, the administration and Congress's view in light of these restrictions is simply that shareholders of failed institutions have already lost enough. Why kick them while they are down? Even Joseph Stiglitz, no friend of big banks, offhandedly echoed this prevailing sentiment. (174)

Second, attempting to pursue shareholders for losses beyond their equity contribution violates one of the central tenets of corporate law--limited shareholder liability. Limited liability is a nearly universal feature of corporate law throughout the world, (175) and is credited with providing investors with the appropriate incentive to allow socially important, high-risk, high-yield activities to flourish in the first place. (176) Forcing shareholders of financial institutions to participate beyond that initial equity value violates this precept, and (one might argue) (177) would unleash consequences that could undermine the entire capitalist system. (178)

1. Poor shareholders

Neither argument is persuasive. In the first place, it is not clear that the shareholders will already have lost their shirts through their equity participation. This is an empirical question, and the topic of another paper. How many of the shareholders of bailed-out institutions were long-term shareholders? How many sold out before the firm's collapse? Beyond the market value of the share, how much did shareholders receive in dividends during the previous three years, for example? It may well be that shareholders were handsomely rewarded through dividends and market exit before the firm's implosion, or even as distressed investors took advantage of what might be called a bailout bump in the markets. Indeed, Bebchuk et al. find that in the cases of Lehman Brothers and Bear Steams, some of the largest shareholders of these firms--the top executives--cashed out a significant portion of their performance-based equity in the months prior to each firm's implosion. (179) If such is the case across the board, it is a losing argument to say that taxpayers, more than shareholders, should bear the residual risk of failure.

Nor have shareholders already paid enough. The firm has not paid enough to cover its own losses, and has therefore succeeded in milking the moral hazard-the firm has privatized benefits and socialized costs. Shareholders have won that game of chicken against taxpayers, by relying on their own limited liability. Additionally, the question of identifying shareholders is, of course, an ephemeral one: shareholders change constantly. Some shareholders lost nothing as a result of the bailout, because they timed their exit well enough to avoid the losses. For these reasons, it is erroneous to assume without analysis that shareholders have suffered from their firm's near-failure.

2. The sacred cow of limited liability

A variation of the argument in favor of limited liability is that any ex post adjustment to the allocation of profits violates the legal protection afforded to shareholders in the first place. There are, however, other examples in the law where profit allocations are revisited, even in the face of otherwise legally binding contracts. Doctrines such as unjust enrichment, (180) fraudulent conveyance, (181) clawbacks, (182) and, especially, voidable preferences (183) all point to the same kind of ex post reevaluation of legally contracted profit allocations. Elective shareholder liability does the same.

But the argument against elective shareholder liability is also more existential. By its very nature, opening the discussion of shareholder liability beyond an equity contribution violates limited shareholder liability. It also harkens back to the early 1990s, when Henry Hansmann and Reinier Kraakman prompted a debate on the theoretical justification for limited liability in the context of involuntary tort creditors, which I will refer to hereafter as the "tort debate."

The tort debate was important and interesting in its own right, and shows up again later in this Article. But for our present purposes, it is beside the point. One of the strongest arguments against Hansmann and Kraakman was simply that limited liability for corporations, even as to claims by involuntary tort creditors, is how corporations have been organized for much of recent history. (184) Changing that practice would be, at the very least, seriously disruptive to the way in which we conceive of corporations and their roles in a capitalist economy. Regardless of whether limited liability is justified vis-a-vis tort creditors, the question of whether limited liability makes sense for financial institutions is equally a question for history. History supports the institution of limited liability for most corporations. For banks, however, it does not.

3. Banks and limited liability in history

Whatever the virtues of the historical basis for limited liability for general industrial corporations, the historical basis for limited liability for banks-whether commercial or investment--is far weaker. From 1865 to 1933, many commercial banks were not limited liability entities. (185) The norm at the time was that, in the event of a bank failure, bank shareholders--usually the bank's own management--would reimburse depositors at double the value of shareholders' equity. In some places, that liability was even unlimited. The view was that the shareholders themselves were better situated to control the bank's risks and that the depositors could thus look to the shareholders for reimbursement in the case of failure.

The end of the double-liability regime occurred as part of the Banking Act of 1933, more famous for its institution of deposit insurance (186) and the Glass-Steagall provisions that separated commercial and investment banking. The Act occurred on the heels of the greatest banking crisis the country has ever known, the crisis of 1932-33. At that point, the newly inaugurated Franklin Roosevelt called for a national banking holiday in the ten states that had not already shuttered their banks. (187) The view, in the face of such bank closures, was simply that the banking shareholders were in just as sad a spot as the depositors and that forcing the bankers to pay the depositors in the face of such failures would only deepen the crisis. (188) Besides, the innovation of deposit insurance, which Roosevelt reluctantly accepted in the Glass-Steagall Banking Act, (189) made the central tension between bankers and depositors less relevant. (190)

A more defensible claim is that today's banks have changed so dramatically that the idea of general non-limited liability is impossible or implausible. But that argument does not address elective shareholder liability, as elective shareholder liability is best understood as an exception to limited liability rather than a regime of non-limited liability. The exception is only for taxpayer bailouts, not general debts. And because of the many efficient ways discussed above in which shareholders can avoid liability, one cannot summarily conclude that the presence of elected, non-limited liability securities in public markets is economically impossible.

Investment banks present an even more recent example of non-limited liability. Until the middle of the last century, almost all investment banks were organized as general partnerships, meaning that the partners were personally liable for all of each firm's losses. Prior to 1970, investment banks were prohibited from being listed as public corporations. (191) This changed slowly and in 1999, with Goldman Sachs's conversion to a public corporation, the private partnership model of a "bulge bracket investment bank" was rendered extinct. (192)

Many have mused about the possible deleterious consequences of this conversion from partnership to corporation. (193) My point here is not to raise those same issues, as interesting as they are. It is, instead, to illustrate that the sacred cow of corporate law--limited shareholder liability--is of relatively recent vintage for banks, both investment and commercial. The argument that we cannot require shareholder participation in taxpayer reimbursement because the model of limited shareholder liability is too central to the way that banks operate is simply ahistorical.

To be clear, although any mention of introducing shareholder liability in the context of banking is certainly reminiscent of the partnership era, elective shareholder liability is not a reversion to that model. Elective shareholder liability is about a partial exception to the corporation's (or bank holding company's) status as a limited liability entity. The traditional debts to creditors are still limited by the entity's limited liability status. The partial exception applies to taxpayer bailouts that end up costing the government money at the end of the designated period. Thus, the regime imagined under elective shareholder liability is distinct from the partnership era in investment banking, or double liability in retail banking. The point in this Subpart is simply to state that the banking system has functioned under alternative liability models. Claims that banks cannot function under such alternative models are ahistorical.

C. General Arguments Against Non-Limited Liability

Many of the arguments that might be raised against elective shareholder liability are similar to those raised against other arguments to relax limited liability. This debate is a long one, and was most recently advanced in the tort debate of the 1990s, mentioned above. I address some of these general arguments below, focusing on enforceability and evasion. (194)

1. Easy evasion and enforceability

The most significant critique against non-limited liability in the tort debate comes from Joseph Grundfest. His argument proceeds in two parts, addressed separately here. First, he argues that an non-limited liability world would encourage equity investors who are essentially judgment-proof to "specialize in holding equity that is susceptible to third party claims under a proportionate liability regime." (195) Using a variety of derivatives, more attachable investors could still gain the economic exposure to SIFI equity without the corresponding risk of liability. In this model, it becomes trivially easy for investors to eliminate additional pools of assets from the availability of later claimants, and removes any monitoring that might otherwise exist.

Grundfest's objections are the strongest against non-limited liability regimes, and represent real problems for elective shareholder liability. Without question, evasion efforts along these lines--and along other lines that the elective shareholder liability proposal does not and cannot anticipate--would be attempted, perhaps successfully, by SIFIs that adopt elective shareholder liability. But there are two reasons why, despite that reality, these objections are not fatal. First of all, employees and shareholders in financial institutions are frequently one and the same. A significant portion of the shares of Goldman Sachs, for example, are owned by employees of the firm. (196) Prohibiting bank employees from owning these shares in the name of regulatory arbitrage seems to be a quixotic goal, given how firmly entrenched shareholding as compensation has been and continues to be for large banks. (197)

To the extent that shareholders might try to use the corporate form to evade liability, the government will have the tools to nullify such efforts. One of the six elements of elective shareholder liability is the ability to declare efforts to hide behind limited liability of the corporate form null and void as a matter of law. Any effort to use that form is thus not available. Disregarding legal relationships on public policy grounds is hardly new territory in American jurisprudence. For example, there is a long history of declaring certain financial contracts permissible, but legally unenforceable. (198) Similarly, because share holders who evade collection and force litigation will have to pay treble damages and litigation costs, there is a dramatic disincentive for evasion.

Grundfest is skeptical of regulatory efforts at stamping out regulatory arbitrage: "Prior attempts," he writes, "to discover such regulatory El Dorados have invariably ended in frustration and expense, and there is no reason to believe that this expedition would meet a happier fate. (199) 1 concede this point entirely. But elective shareholder liability, including the anti-evasion elements, is no regulatory El Dorado. What it does instead is introduce complexity and uncertainty into the calculus of regulatory arbitrage. It is precisely this complexity and uncertainty--features not normally heralded in the formulation of financial regulatory policy--that make elective shareholder liability so appealing. In light of their uncertain ability to evade liability, many shareholders would simply either opt out of the system entirely, through capital adequacy or share sales, or purchase an SLS to insure against such liability. As already discussed, both of these scenarios overcome Grundfest's objections while simultaneously causing better cost internalization, TBTF subsidy reduction, and increased incentive for risk management.

Hansmann and Kraakman's original response to Grundfest also casts doubt on one of Grundfest's central premises: that a stable of judgment-proof investors exists to provide the function that Grundfest discusses. (200) Given the equities at stake, it may be helpful, in an iterative way, to run the experiment and determine its effect rather than to presume futility from the beginning.

Another reason exists to differentiate Grundfest's objection in the context of the tort debate from a potential objection in the non-limited liability regime for SIFIs. The scenario in which a firm becomes subject to bankruptcy-inducing tort liability is not difficult to imagine. Indeed, in a world with a healthy stable of plaintiffs' lawyers constantly scanning for the next asbestos or Dalkon Shield litigation, it is a stretch to categorize corporate bankruptcy at the hands of involuntary tort creditors as a fat-tail risk, a risk of a high-impact event that has a low probability of occurring. These kinds of risks are real and present for firms of all kinds, but particularly for those that engage in the manufacture of potentially toxic products.

Such is not the case for the beneficiaries of taxpayer bailouts. The incidence of nonreimbursed taxpayer bailouts is extremely low. And because there are so many other mechanisms at the government's disposal to protect against taxpayer loss, the likelihood that taxpayer reimbursement will be sought through an elective shareholder liability collection is extremely small. The idea, in the face of such a small risk, that shareholders would sacrifice the benefits of shareholdings, or run the risk of using a transparently obvious regulatory arbitrage, is unlikely.

To be sure, the elective shareholder liability proposal does not share Grundfest's certainty in Wall Street's ability to perfectly evade liability. Nor does it espouse the opposite certainty in the government's ability to exact it. Instead, elective shareholder liability will encourage enough shareholders to either buy insurance or move out of the tornado zone, as it were, and thereby take steps toward solving the bailout proposal without ever exposing their shareholders to personal liability.

Another variation of the evasion argument is the issue of international coordination. Policymakers have continually lamented the reality that international variation in banking regulation produces opportunities for evasion and regulatory arbitrage. (201) This is the reality of an uneven international banking regulatory framework, the politics of which makes the American banking reform process look like a game of Candy Land by comparison. This version of the evasion argument is a serious one, perhaps even insurmountable. But it is not obvious why the international coordination problem would be any more of a barrier in the context of elective shareholder liability than it is in any but the most banal of banking policy proposals. This is not to say that international coordination is not a serious argument, but only that it is a problem that confronts the entire effort of banking regulation wherever and however it occurs.

A related problem is that of enforceability. In response to the original thrust against limited liability in the tort debate, Janet Alexander responded with a proceduralist's take on the obstacles confronting non-limited liability. (202) In her response, Alexander contends that personal jurisdiction and choice of law requirements prohibit private litigants from collecting on state corporate law actions, or even recovery of judgment actions, which are also questions of state law.

Her arguments--persuasive, though contested (203)--are inapplicable here, for two reasons. First, Alexander concedes that the procedural arguments she identifies are based in state law. The regime imagined in this Article is explicitly a federal procedure. Second, Alexander describes a recovery regime based in litigation, which is not the case with elective shareholder liability. Instead, elective shareholder liability is structured as a governmental collection, similar to a general tax assessment. Thus, personal jurisdictional requirements are not applicable; the same is true for the choice of law objections Alexander makes elsewhere.

That said, there will be enforceability problems. There always are, and the costs of litigation, when it occurs, will lead to evasion, the anti-evasion efforts hardwired into the proposal notwithstanding. Even so, evading shareholder liability remains an uncertain proposition for those who would first elect it and then seek to avoid it. Investors who hope to evade the costs of the liability that they have elected will only be able to do so if they are right in anticipating that their evasive methods will be successful. In other words, as the likelihood of successful evasion diminishes, the likelihood of either capital adequacy election or improved risk management increases.

2. Difficulty of determining who would be on the hook

There are two versions of this argument. First is a recordkeeping argument, which is valid and would require changes to the way we keep records of shareholder participation. The second is an evasion argument, which is met by the anti-evasion elements of elective shareholder liability already discussed.

As to the first argument, access to current shareholder records is not required under the securities laws for any shareholder below the threshold requiring disclosure under the Williams Act. (204) Because the Williams Act is motivated by an effort to make structurally coercive corporate takeovers more difficult, (205) that threshold will be unlikely to apply to almost any shareholder against whom recovery would be sought here. Thus, one significant obstacle to elective shareholder liability is that it would require the identities of shareholders for the relevant period to be disclosed to the government.

Of course, the mechanisms for disclosure are already in place. Corporations pay dividends, and dividends are paid to shareholders. The corporations presumably have access to the names of those institutions and individuals that own their shares. Gaining access to that information would be a question, perhaps a complicated one, of requiring the SIFIs to share that information.

The second argument is a variation of the Grundfest regulatory arbitrage argument. In light of the fact that shareholder identity is straightforward to ascertain, this critique must refer instead to the likelihood of equity investors using other LLPs to evade liability.

Grundfest makes a compelling claim that one must not underestimate capital market participants' enthusiasm for cost evasion. Nevertheless, the "LLP Russian doll" argument still runs thin. Arguing that elective shareholder liability is unenforceable because investors will be very clever about evading responsibility is like arguing that we should give up on pursuing wrongdoers because some of them are very clever at evading capture. First, not all equity investors will resist payment. Many, perhaps most, investors will take calculated risks, enjoying the upside when the risk cuts their way and footing the bill when the risk does not. Every day in the capital markets, investors admit losses and pay out losing trades. Liability for taxpayer bailouts is simply a lost trade. Elective shareholder liability would not require perfect recovery to be effective; indeed, partial recovery should be expected. The point is to have that partial recovery, both for the pool of resources that it opens up and more importantly for the effects that it will have on the ex ante incentives of shareholders (as well as officers and directors).

3. Endogeneity problem

The biggest weakness in elective shareholder liability may be that the decision to bail out a firm and the ultimate amount of a required bailout cannot be independently determined: the government will make the TBTF determination simultaneously with the evaluation of a failed firm's financial needs. In that sense, the market discipline required to make informed equity investment decisions-the fundamental background assumption underlying elective shareholder liability--is undermined because market participants cannot discern the ultimate shareholder liability exposure before the government has decided what that exposure will be. In that case, it would seem a dubious enterprise to force shareholders to be liable for a figure that they had no hand in determining.

This is a promising argument, but if taken seriously, it would undermine much of the activity within the capital markets. First, the sovereign debt markets are similarly subject to endogenous decisionmaking of governments, and yet those markets continue to function. Second, the currency-exchange markets move constantly on the news of government decisions. (206) In foreign exchange trading, decisions about the value of currency--whether announcements regarding Chinese renminbi management or the American Federal Reserve's efforts to stave off deflation through quantitative easing--are fundamentally government decisions that have immediate effects on the value of currency.

4. Exogeneity problem

The inverse of the endogeneity problem is that taxpayer bailouts may, in some cases, be fully exogenous events that shareholders can neither control nor affect. If that is the case, then the justification for elective shareholder liability is diminished if such recovery is viewed punitively. This is particularly true of nonbank SIFIs that are so designated after the fact.

This, again, is why election is the most important feature of the proposal. Shareholders can elect to be subject to elective shareholder liability, or change their balance sheets or leverage structures to avoid it. If they opt for the latter, even in the event of a taxpayer bailout, there will be no liability, and thus no concern for the exogeneity of a bailout.

Furthermore, the exogeneity of bailouts may be another illustration of why voidable preferences in bankruptcy are the best legal analogy for elective shareholder liability. Voidable preferences do not presume nefarious or culpable activity, only activity that taken as a whole is inefficient in light of the debtor's financial situation. Thus, elective shareholder liability--even in cases where the systemically important default occurs completely independently of the corporation's decisionmaking--restores the shareholders, and not the taxpayers, as the residual bearers of risk. This is not because they have behaved badly, but because they are better situated than the taxpayers to absorb the losses associated with systemic default.

5. Too much government discretion

Because under elective shareholder liability the government retains the full discretion to litigate claims against shareholders, another argument is that where there is discretion, there is abuse. Under that argument, elective shareholder liability would extend government discretion and thus extend the potential for abuse, politically motivated litigation, or some other distortion of the process. Some scholars have noted that the Treasury Department in particular, to whom the authority to litigate these cases might plausibly be extended, enjoys particular freedom from oversight among departments and agencies. (207)

There are a few inherent checks on this potential abuse of discretion. First, this litigation can only occur after the stay on collection has expired, two years after the bailout. This will mitigate the value of such litigation to any purpose other than taxpayer reimbursement.

Additionally, it is a stretch to think that a mandatory collection against shareholders does anything to meaningfully extend the power of the federal government. The entire IRS is built on the premise that the government can collect from citizens. Allowing another collection against SIFI shareholders is hardly a broad extension of government power.


This Article has made the case for elective shareholder liability, an obligatory governmental assessment against shareholders who opt into the regime, providing a mechanism by which taxpayers can be reimbursed in the arguably inevitable event that a bailout does occur. I have argued that elective shareholder liability complements other proposed regimes, and indeed, that the regime will be unnecessary if banks' shareholders adopt the more stringent structural changes passed over during the recent debates that produced the flawed Dodd-Frank Act.

Elective shareholder liability is not an elegant solution. Like any regulatory reform effort, the intended effects may be less or more than has been argued here; the unintended effects, tautologically, remain unaddressed. But though it is an inelegant regime, elective shareholder liability would to replace a far more deeply flawed one. Moreover, elective shareholder liability is a compromise proposal: it seeks to find a middle ground between the academics who find no social benefit to the extent of banks' leverage, and the banks who promise economic apocalypse in the face of increased capital requirements. To argue against elective shareholder liability, opponents must offer a more satisfactory explanation of why the rewards of excessive leverage should accrue to shareholders while the costs are borne by the taxpayers.

(1.) See Part I.C for a discussion of an important exception in the work of Adam Levitin.

(2.) See generally ENDING GOVERNMENT BAILOUTS AS WE KNOW THEM (Kenneth E. Scott et al. eds., 2009) (introducing various perspectives on ending government bailouts completely).

(3.) Various scholars have estimated the value of these too-big-to-fail (TBTF) subsidies. Two scholars estimate that TBTF subsidies that translate into just fifty basis points on debt yield a subsidy of $34.1 billion per year for the eighteen largest bank holding companies. See DEAN BAKER & TRAVIS MCARTHUR, CTR. FOR ECON. & POLICY RESEARCH, THE VALUE OF THE "Too BIG TO FAIL" BIG BANK SUBSIDY 2 (2009), available at Standard & Poor's has proposed incorporating the promise of government support into some of its computations of the creditworthiness of the largest banks. See STANDARD & POOR'S, BANKS: RATING METHODOLOGY 3 (2011), available at CriteriaFinanciallnstitutionsRequestforCommentBanksRatingMethodology.pdf.

(4.) Examples from the recent crisis are legion. For general arguments to this end, see DAVID SKULL, THE NEW FINANCIAL DEAL: UNDERSTANDING THE DODD-FRANK ACT AND ITS (UNINTENDED) CONSEQUENCES 8-12 (2011). For the case of Bear Steams, see Marcel Kahan & Edward Rock, How to Prevent Hard Cases from Making Bad Law: Bear Stearns, Delaware, and the Strategic Use of Comity, 58 EMORY L.J. 713, 744 (2009), which argues that the government-guaranteed JP Morgan Chase acquisition of Bear Stearns violated Delaware law. For the case of Chrysler, see Mark J. Roe & David Skeel, Assessing the Chrysler Bankruptcy, 108 MICH. L. REV. 727, 763-64 (2010), which argues that judicial resolution of the Chrysler bailout enabled a later violation of best practices in bankruptcy and potentially of the Bankruptcy Code itself.

(5.) Lawrence H. Summers, International Financial Crises." Causes, Prevention, and Cures, AM. ECON. REV., May 2000, at 1, 13. This pronouncement is ironic, as Summers was an architect of the Dodd-Frank Act, which as discussed in Part I.B below makes these bailouts more likely, not less. Summers is on record disagreeing with that assessment of the Dodd-Frank Act. See This" Week (ABC television broadcast Apr. 4, 2010) (transcript available at http:/ larry-summers/ story?id=10280914&singlePage=true).

(6.) See Dodd-Frank Wall Street Reform and Consumer Protection Act, Pub. L. No. 111-203, 124 Stat. 1376 (2010) (codified in scattered sections of the U.S. Code).

(7.) See infra Part I.A.

(8.) This is the first of two articles that aim to solve the problem of bailouts. The second is Anat R. Admati, Peter Conti-Brown & Paul Pfleiderer, Liability Holding Companies, 59 UCLA L. REV. (forthcoming Mar. 2012) (introducing a new financial institution, "liability holding companies," to guarantee the debts of SIFIs). Though the present Article should be internally comprehensible, the two papers are better read in conjunction with one another.

(9.) I use the shorthand "TBTF problems" to refer generally to taxpayer reimbursement, bailout prevention, and TBTF subsidies. When addressing one of these specifically, I use more precise language.

(10.) See Anat Admati et al., Letter to the Editor, Healthy Banking System Is the Goal, Not Profitable Banks', FIN. TIMES, Nov. 9, 2010, at 8, available at cms/s/0/63fa6b9e-eb8e-11df-bbb5-00144feab49a.html#axzzlQDLV5XER (representing the opinion of twenty leading financial economists that effective banking reform requires capital increases beyond the Basel III baseline).

(11.) The quantity of scholarship focused on these issues since the financial crisis, in both the legal and the economic/financial sphere, is staggeringly high. I engage what I view as the most important structural proposal in Part I.D below. For other examples of proposals with important implications for bailout prevention, see GARY B. GORTON, SLAPPED BY THE INVISIBLE HAND: THE PANIC OF 2007 (2010), which identifies the risks inherent in the overnight repo markets and advocates for a deposit-insurance style of protection for these markets; SKEEL, supra note 4, which makes the case for bankruptcy as an alternative regime to Dodd-Frank liquidation); and Lucian A. Bebchuk & Holger Spamann, Regulating Bankers' Pay, 98 GEO. L.J. 247 (2010), which focuses on regulation of banker pay as a mechanism of broader financial regulation. See generally SIMON JOHNSON & JAMES KWAK, 13 BANKERS: THE WALL STREET TAKEOVER AND THE NEXT FINANCIAL MELTDOWN (2010). For helpful work on the structure of systemic risk and its regulation, see generally Iman Anabtawi & Steven L. Schwarcz, Regulating Systemic Risk: Towards an Analytical Framework, 86 NOTRE DAME L. REV. 1349 (2011); and Steven L. Schwarcz, Systemic Risk, 97 GEO. L.J. 193 (2008).

(12.) See Anat Admati et al., Fallacies, Irrelevant Facts, and Myths in the Discussion of Capital Regulation: Why Bank Equity Is Not Expensive 1, 58 (Mar. 23, 2011) (unpublished manuscript), available at 86.pdf.

(13.) This debate is still unfolding. Switzerland has partially adopted the proposal. See Elena Logutenkova, UBS Says Swiss Capital Rules Should Be Aligned Globally, BLOOMBERG (Mar. 23, 2011, 11:05 AM), -says-swiss-capital-rules-should-be-aligned-internationally.html. At least one U.S. Federal Reserve governor, according to some reports, appears sympathetic to Admati et al.'s arguments, with one report of the governor considering a capital surcharge of seven percent above the level required by Basel III. See Editorial, Tarullo's Capital Idea, WALL ST. J., June 16, 2011, at A18, available at SB 10001424052702303848104576385782130686772.html.

(14.) These arguments and others are summarized in Admati et al., supra note 12, at 10-11.

(15.) See, e.g., Vikram Pandit, We Must Rethink Basel, or Growth Will Suffer, FIN. TIMES, Nov. 11, 2010, at 11.

(16.) See infra Part I.D.

(17.) See John H. Cochrane, The More Bank Capital, the Safer the Bank, WALL ST. J., July 15, 2011, at A15; see also David Miles et al., Optimal Bank Capital 6 (Bank of Eng. External MPC Unit, Discussion Paper No. 3 l, 2011 ), available at http://www.bankofengland

(18.) Richard Squire, Shareholder Opportunism in a Worm of Risky Debt, 123 HARV. L. REV. 1151, 1152-53 (2010).

(19.) See Henry Hansmann & Reinier Kraakman, Toward Unlimited Shareholder Liability for Corporate Torts, 100 YALE L.J. 1879, 1879-81 (1991).

(20.) See Joseph A. Grundfest, The Limited Future of Unlimited Liability: A Capital Markets Perspective, 102 YALE L.J. 387, 389 (1992). For Hansmann and Kraakman's response, see Henry Hansmann & Reinier Kraakman, Do the Capital Markets Compel Limited Liability? A Response to Professor Grundfest, 102 YALE L.J. 427 (1992).

(21.) See, e.g., Letter from James Armada, CEO, Clearing House Ass'n, and Paul Saltzman, President, Clearing House Ass'n, to Timothy F. Geithner, Sec'y of the Treasury, et al. (June 15, 2011), available at

(describing the many innovations of Dodd-Frank, and calling efforts to increase capital "premature" in light of these innovations).

(22.) Thomas Hoenig, Why the Sign Must Say: No UBS in the USA, FIN. TIMES, June 16, 2011, at 9.

(23.) See President Barack Obama, Remarks by the President at Signing of Dodd-Frank Wall Street Reform and Consumer Protection Act (July 21, 2010) (transcript available at frank-wall -street-reform-and-consumer-protection-act).

(24.) Id.

(25.) Dodd-Frank Wall Street Reform and Consumer Protection Act, Pub. L. No. 111203, 124 Stat. 1376, 1376 (2010).

(26.) 12 U.S.C. [section] 5394 (Supp. IV 2010).

(27.) The argument that Dodd-Frank fails to end TBTF is embraced by those with widely divergent views of the necessity of such bailouts and the virtues of Dodd-Frank's other provisions. Adam Levitin argues that Dodd-Frank does not end bailouts because it cannot, noting that "[b]ailouts are an inevitable feature of modern economies, in which the interconnectedness of firms means that the entire economy bears the risk of an individual firm's failure." Adam J. Levitin, In Defense of Bailouts, 99 GEO. L.J. 435, 439 (2011). Arthur Wilmarth argues that the specifics of Dodd-Frank make it likely to result in more bailouts in the future, since "[t]he [Federal Reserve Board] and the Federal Home Loan Banks retain authority to provide emergency liquidity assistance to troubled [large, complex financial institutions]" and "[t]he FDIC can borrow from the U.S. Treasury and can also use the 'systemic risk exception' to the Federal Deposit Insurance Act in order to generate funding to protect creditors of failed SIFIs and their subsidiary banks." Arthur E. Wilmarth, Jr., The Dodd-Frank Act. A Flawed and Inadequate Response to the Too-Big-to-Fail Problem, 89 OR. L. REV. 951, 956 (2011). David Skeel argues that not only does DoddFrank fail to prevent future bailouts, but also it institutionalizes bailouts, which he views as far inferior to requiring a bankruptcy regime for SIFIs. See SKEEL, supra note 4, at 8-9.

(28.) See John B. Taylor, The Dodd-Frank Financial Fiasco, WALL ST. J., July 1, 2010, at A19 (arguing that Dodd-Frank's structure gives the government more authority than it needs, and creates a situation where "any cozy relationship between certain large financial institutions and the government that existed before the crisis will continue"); Simon Johnson, Tunnel Vision, or Worse, from Banking Regulators, N.Y. TIMES ECONOMIX (Jan. 20, 2011, 5:00 AM), -banking-regulators (arguing that the administration's study on limiting the size of banks' balance sheets will only lead to banks that are de facto TBTF and will require bailouts in the event of failure).

(29.) Neil Barofsky, Special Inspector General for the Troubled Asset Relief Program, quotes Secretary Geithner, then President of the Federal Reserve Bank of New York, as saying the following: "In the future we may have to do exceptional things again if we face a shock that large. You just don't know what's systemic and what's not until you know the nature of the shock." OFFICE OF THE SPECIAL INSPECTOR GEN. FOR THE TROUBLED ASSET RELIEF PROGRAM, EXTRAORDINARY FINANCIAL ASSISTANCE PROVIDED TO CITIGROUP, INC. 44 (2011), available at %20Assistance%20Provided%20to%20Citigroup,%20Inc.pdf.

(30.) See SKEEL, supra note 4, at 8-9; Wilmarth, supra note 27; Taylor, supra note 28.

(31.) The ouster of one of the most conservative senators in the U.S. Senate, Utah Senator Bob Bennett, because of his support for the bailouts provides a good example of such backlash. See Kirk Johnson, Utah Delegates Oust Three-Term G.O.P. Senator from Fall Race, N.Y. TIMES, May 9, 2010, at A23. Of course, there were other factors in play, including hostility toward incumbents and a preference for Bennett's opponent, Michael Lee, the well-pedigreed son of a beloved figure in both the national Republican Party and Utah politics. See Jesse Zwick, "Senator Junior DeMint," NEW REPUBLIC (Dec. 23, 2010, 12:00 AM),

(32.) Robert Reich, former Labor Secretary under President Clinton, suggests that future bailouts may result in something close to social revolution. See ROBERT B. REICH, AFTERSHOCK: THE NEXT ECONOMY AND AMERICA'S FUTURE 8 (2010); cf BRUCE ACKERMAN, THE DECLINE AND FALL OF THE AMERICAN REPUBLIC 4 (2010) (exploring the presidency's potential to serve as "a vehicle for demagogic populism and lawlessness").

(33.) See 12 U.S.C. [section] 5321(b)(1) (Supp. IV 2010). The Council's members are: the heads of the Federal Reserve, the Securities and Exchange Commission, the Commodities Futures Trading Commission, the Consumer Financial Protection Bureau, the Comptroller of the Currency, the Federal Deposit Insurance Corporation, the National Credit Union Administration Board, the Federal Housing Finance Agency, and an independent member "having insurance expertise." Id. [section] 11 l(b)(1).

(34.) Id. [section] 5331(a).

(35.) Id. [section] 5323. This Article applies primarily to bank SIFIs, rather than nonbank SIFIs, because nonbank SIFIs might not be so designated until the moment of crisis, negating any ex ante monitoring value that may be gained through the shareholder election. As this Article goes to press, the FSOC continues to clarify the parameters it will use in identifying nonbank SIFIs. See Authority to Require Supervision and Regulation of Certain Nonbank Financial Companies, 76 Fed. Reg. 64,264 (proposed Oct. 18, 2011) (to be codified at 12 C.F.R. pt. 1310).

(36.) Dodd-Frank does not use the phrase "systemically important financial institutions," though the term has been a part of the parlance of financial regulation since at least 1999. See Alessandro Prati & Garry J. Schinasi, Financial Stability in European Economic and Monetary Union 19 (Princeton Studies in Int'l Fin., No. 86, 1999), available at

(37.) Wilmarth, supra note 27, is the most thorough, especially for lawyers. SKEEL, supra note 4, is the most accessible.

(38.) The famous Volcker Rule is an exception. The Volcker Rule is designed to limit the ability of investment banks to make bets using their own funds--or "proprietary trading'--as opposed to making the same bets on behalf of clients. See 12 U.S.C. [section] 1851. The same provision in Dodd-Frank also subjects nonbank SIFIs that sponsor hedge and private equity funds to increased capital requirements. Id. While the Volcker Rule has been compared to the Glass-Steagall Act, which separated commercial and investment banking, the reality is that the concept of proprietary trading is so difficult to interpret that much will come down to the interpretation of important gray areas, such as the definition of "market making" and "hedging." See FIN. STABILITY OVERSIGHT COUNCIL, STUDY & RECOMMENDATIONS ON PROHIBITIONS ON PROPRIETARY TRADING & CERTAIN RELATIONSHIPS WITH HEDGE FUNDS & PRIVATE EQUITY FUNDS 16 (2011), available at http://www %201% 2018%2011%20rg.pdf. According to Goldman Sachs, one of the largest investment banks in the world, with one of the largest proprietary trading desks, the Volcker Rule will have little effect on the firm's profitability. According to Goldman's CFO on February 9, 2011 (after the Volcker Rule study was released): "Whatever effects there have been [from the Volcker Rule], you've seen already .... We don't see that big of an effect from the Volcker rule on our revenues." Christine Harper, Goldman Sachs Says It Bought Too Many Illiquid Assets Before 2008 Crisis, BLOOMBERG (Feb. 9, 2011, 7:45 AM), many-illiquid-assets-pre-financial-crisis.html.

(39.) Dodd-Frank largely punts on the question of capital adequacy. In the relevant provision for SIFts, the Act gives the Federal Reserve authority to set "risk-based capital requirements and leverage limits, unless the Board of Governors . . . determines that such requirements are not appropriate for a company subject to more stringent prudential standards" because of the nature of its activities. 12 U.S.C. [section] 5365(b)(1)(A)(i) (emphasis added). There has been some reporting that Treasury officials sought to minimize Dodd-Frank's reliance on capital requirements in deference to Basel III. See Felix Salmon, Dodd-Frank vs Basel III, REUTERS (Jan. 20, 2011), 20/dodd-frank-vs-basel-iii ("When the big financial-overhaul bill was working its way through Congress, Treasury persuaded legislators to avoid passing rules on bank capital or liquidity. Leave all that to Basel, they said, so that there could be a global, unified system."). This is consistent with the approach taken under Basel, which focuses most of its attention on increased capital adequacy. See MACROECONOMIC ASSESSMENT GRP., ASSESSING THE MACROECONOMIC IMPACT OF THE TRANSITION TO STRONGER CAPITAL AND LIQUIDITY REQUIREMENTS 1 (2010), available at (discussing briefly the nature of the Basel III Accord's focus on capital adequacy).

(40.) See 12 U.S.C. [section] 5390(n).

(41.) This Subpart draws from Wilmarth, supra note 27, at 996-1003.

(42.) See 12 U.S.C. [section] 5390(h)(5)(E).

(43.) 12 U.S.C. [section] 1824(a).

(44.) Many of the structures used by the Federal Reserve to fund entities in crisis were designed to conduct regulatory arbitrage around restrictions on parties to whom the Federal Reserve could plausibly lend. Furthermore, in order to justify the provision of TARP funds to GM and Chrysler, the administration had to reach the strained conclusion that the car companies were "financial institutions" within the meaning of that term as used in section 101(a)(1) and as defined in section 3(5) of the Emergency Economic Stabilization Act of 2008. See 12 U.S.C. [section][section] 5201(5), 5211(a)(1). See CONG. OVERSIGHT PANEL, SEPTEMBER INSIGHT REPORT: THE USE OF TARP FUNDS IN THE SUPPORT AND REORGANIZATION OF THE DOMESTIC AUTOMOTIVE INDUSTRY 71-77 (2009), available at documents/cop-090909-report.pdf, for an exhaustive analysis of that strained interpretation.

(45.) See 12 U.S.C. [section] 5390(o)(1)(B), (D); see also Notice of Proposed Rulemaking Implementing Certain Orderly Liquidation Authority Provisions of the Dodd-Frank Wall Street Reform and Consumer Protection Act, 75 Fed. Reg. 64,173, 64,178 (proposed Oct. 19, 2010) (to be codified at 12 C.F.R. pt. 380) ("The Dodd-Frank Act also includes the power to 'claw-back' or recoup some or all of any additional payments made to creditors if the proceeds of the sale of the [failed SIFI's] assets are insufficient to repay any monies drawn by the FDIC from Treasury during the liquidation.").

(46.) 12 U.S.C. [section] 5390(a)(9)(E).

(47.) While the FDIC is designated an independent agency, and Sheila Bair, the FDIC Chairwoman during the financial crisis, was known for her willingness to buck her colleagues in the executive branch, it is also true that the FDIC and the administration reacted in the moment with remarkable consistency. The FDIC's designation as an independent agency is unlikely to make much of a difference in the middle of a financial crisis.

(48.) SKEEL, supra note 4, at 1.

(49.) Id. at 8.

(50.) Id. at 9.

(51.) See Levitin, supra note 27, at 439; Wilmarth, supra note 27, at 956 ("Dodd-Frank probably will not prevent TBTF rescues during future episodes of systemic financial distress."). Wilmarth was also one of the few scholars with the prescience to note the fault lines in the banking regulatory system years before the financial crisis. See Arthur E. Wilmarth, Jr., The OCC's Preemption Rules Exceed the Agency's Authority and Present a Serious Threat to the Dual Banking System and Consumer Protection, 23 ANN. REV. BANKING & FIN. L. 225, 229-30, 232 (2004) (arguing that the preemption rules of the Office of the Comptroller of the Currency exceed the agency's authority and present a serious threat to the dual banking system and consumer protection); see also Arthur E. Wilmarth, Jr., Too Big to Fail Too Few to Serve? The Potential Risks of Nationwide Banks, 77 IOWA L. REV. 957, 986-90, 1003-04 (1992) (arguing that the consolidation in the national banking sector had real implications for TBTF that could create problems for the functioning of the financial system).

(52.) See generally ROGER LOWENSTEIN, WHEN GENIUS FAILED: THE RISE AND FALL OF LONG-TERM CAPITAL MANAGEMENT 185-218 (2000) (describing the successful Federal Reserve-backed private bailout of the fund).

(53.) See SKEEL, supra note 4, at 132-35.

(54.) See CHARLES P. KINDLEBERGER & ROBERT Z. ALIBER, MANIAS, PANICS, AND CRASHES: A HISTORY OF FINANCIAL CRISES 10 (5th ed. 2005) ("Virtually every large country has established a central bank as a domestic 'lender of last resort' to reduce the likelihood that a shortage of liquidity would cascade into solvency crisis.").

(55.) See 1 ALLAN H. MELTZER, A HISTORY OF THE FEDERAL RESERVE 68-69 (2003) (stating that, although the various proposals considered in the debates preceding the enactment of the Federal Reserve Act were diverse and conflicting, "[a]ll proposals recognized that a central bank could serve as lender of last resort in a banking crisis").

(56.) See Cheryl D. Block, Overt and Covert Bailouts: Developing a Public Bailout Policy, 67 IND. L.J. 951, 968 (1992) (providing the first effort to define both "overt" and "covert" bailouts).


(58.) See supra note 11. See also Lawrence A. Cunningham & David Zaring, The Three or Four Approaches to Financial Regulation: A Cautionary Analysis Against Exuberance in Crisis Response, 78 GEO. WASH. L. REV. 39, 39-41 (2009), for a pre-Dodd-Frank assessment of the different approaches considered.

(59.) Admati et al., supra note 8, suggest this band of capital as the desirable solution, but offer no formal model for reaching that conclusion. David Miles, an economist who works for the Bank of England, estimated with coauthors the consequences for bank lending, cost of bank equity, and, most significantly, economic growth. Following Miles's models, the point at which capital requirements begin having a negative impact on economic growth occurs at roughly eighteen percent. See Miles et al., supra note 17, at 36.



(62.) Press Release, Basel Comm. on Banking Supervision, Group of Governors and Heads of Supervision Announces Higher Global Minimum Capital Standards (Sept. 12, 2010), available at

(63.) Id.

(64.) See id.

(65.) These arguments and others are summarized by Admati et al., supra note 12.

(66.) See id.

(67.) Their frequently revised paper is addressed to a general audience, and is thus worth reading in full for those for whom the arguments summarized below are insufficiently persuasive.

(68.) See Cochrane, supra note 17.

(69.) See Tarullo's Capital Idea, supra note 13.

(70.) See Robert Jenkins, Attention Investors. Capitalism Needs Your Help, FIN. TIMES (June 7, 2011, 3:22 AM), 9531-00144feab49a.html#axzz1k3mqG1MQ.

(71.) Michael Jensen, the father of agency theory, is the leading proponent of the view that debt is a solution to what he calls the "free cash flow problem," or the temptation that management will have to divert resources for private gain when there is unencumbered cash that flows through management before being returned to shareholders. See Michael C. Jensen, Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers, 76 AM. ECON. REV. 323, 323-29 (1986) [hereinafter Jensen, Agency Costs]; Michael C. Jensen, The Modern Industrial Revolution, Exit, and the Failure of Internal Control Systems, 48 J. FIN. 831 (1993); Michael C. Jensen, Eclipse of the Public Corporation, HARV. BUS. REV., Sept.-Oct. 1989, at 61.

(72.) Admati et al., supra note 12, at 30 (emphasis omitted).

(73.) See Admati et al., supra note 8, for an example of an alternative equity structure that includes greater attention to the theory of disciplining debt.

(74.) See Letter from James Armada & Paul Saltzman to Timothy F. Geithner et al., supra note 21; see also Financial Regulatory Reform: The International Context: Hearing Before the H. Comm. on Fin. Servs., 112th Cong. 12-18 (2011) (testimony of Hal S. Scott, Director, Committee on Capital Markets Regulation), available at

(75.) If that kind of federal interference in state corporate law is deemed too intrusive, an alternative is simply to allow shareholders to vote via changes to federal corporate law, as was the case in the corporate governance reforms that occurred in Dodd-Frank.

(76.) Examples of these kinds of companies include Ford Motor Company and Google. See FORD MOTOR CO., RESTATED CERTIFICATE OF INCORPORATION (2000), available at 01-000014-0002.txt (outlining Class B shareholders' voting power); GOOGLE INC., THIRD AMENDED AND RESTATED CERTIFICATE OF INCORPORATION OF GOOGLE INC. art. IV, [section] 2(a)(iii) (2011), available at (outlining Class B shareholders' outsized voting rights); id. art. IV, [section] 2(f)(i)(2) (identifying Class B shareholders).

(77.) See Matt Taibbi, Wall Street's War, ROLLING STONE, June 10, 2010, at 51-56, for a summary of the former section 716. Section 716 was scuttled at the insistence of the largest banks, who wanted to be able to enjoy both the profits from their derivatives business and also the government-sponsored liquidity from the Federal Reserve.

(78.) For a more general description of the architecture of "elections" in the law, see Heather M. Field, Choosing Tax: Explicit Elections as an Element of Design in the Federal Income Tax System, 47 HARV. J. ON LEGIS. 21, 24-25 (2010).

(79.) See Mark V. Pauly, Overinsurance and Public Provision of Insurance: The Roles of Moral Hazard and Adverse Selection, 88 Q.J. ECON. 44, 44-45 (1974).

(80.) See DEL. CODE ANN. tit. 8, [section] 262(b) (2011) (making appraisal rights available in the event of a merger or acquisition, but only when the consideration is not the stock of a publicly traded corporation).

(81.) Of course, in the event of a taxpayer bailout within one year of the shareholder election, dissenting shareholders should not be held liable as long as they sold their shares following the election. The alternative result would be inequitable.

(82.) Dodd-Frank Wall Street Reform and Consumer Protection Act [section] 214(a), 12 U.S.C. [section] 5394(a) (Supp. IV 2010).

(83.) See Lori Montgomery, U.S. to Pay $25 Billion for TARP, CBO Says, WASH. POST, Nov. 30, 2010, at A18. Although initial estimates of the financial crisis projected losses nearing $1 trillion, at present only Fannie Mae and Freddie Mac have yet to repay their taxpayer support. Even the notorious AIG has paid back the value of its taxpayer bailouts. See Robin Kwong, AIG Reaches Milestones in Paying Back US, FIN. TIMES (Jan. 12, 2011, 1:49 AM), html#axzz1k3mqG1MQ.

(84.) Michael Perino's recent account of the Pecora hearings exposes an early use of the term "bailout" that shows this discomfort. Perino records Pecora's cross-examination of a banker in these terms:

"[W]ould you say ... that the bank was bailed out of those loans under that process?"

[The banker] didn't care for the question any more than [his boss, cross-examined the day before]: "Whatever word you wish to use. The bank was relieved of these loans; yes, sir."

Pecora seemed to enjoy taunting the City Bank executives with that phrase, so he kept at it: "You have heard that term used before, 'bailed out,' haven't you? ... It is used in the common parlance of Wall Street, isn't it?"

"Well, I suppose so. But I do not use it."

"You think it has a harsh sound to the ear, is that it?"

[The banker] did not answer; he just smiled at Pecora. What could he say?


(85.) It is also the system that existed under the previous regimes of non-limited liability for commercial banks. See Jonathan R. Macey & Geoffrey P. Miller, Double Liability of Bank Shareholders: History and Implications, 27 WAKE FOREST L. REV. 31, 41 (1992).

(86.) Arithmetically, SHL = (SH/TSO) x (RP/TP) x (TB), where SHE = shareholder liability, SH = shareholding, TSO = total shares outstanding, RP = rounding period (in my example, one month), TP = total period of shareholding (in my example, one year, though both RP and TP must be expressed in the same units), and TB = taxpayer bailout.

To illustrate, suppose financial institution Megabank A is bailed out to the tune of $20 billion (TB = $20 billion). Shareholder Mutual Fund owned 10,000 of A's 100,000,000 shares for a period of two months, before selling the shares within one year before the bailout. Under this model, Shareholder Mutual Fund would thus be liable for $333,333: (10,000/100,000,000) x (2/12) x 20 billion.

(87.) See Macey & Miller, supra note 85, at 55.

(88.) Frank H. Easterbrook & Daniel R. Fischel, Limited Liability and the Corporation, 52 U. CHI. L. REV. 89, 89 (1985); see infra note 178.

(89.) 12 U.S.C. [section] 5322(a)(1)(A) (Supp. IV 2010).

(90.) 15 U.S.C. [section] 15 (2006).

(91.) 18 U.S.C. [section] 1964(c).

(92.) 15 U.S.C. [section] 1117.

(93.) 35 U.S.C. [section] 284 (amended 2011).

(94.) 42 U.S.C. [section] 9607(c)(3).

(95.) See Exxon Shipping Co. v. Baker, 554 U.S. 471, 507 (2008), for a fuller exploration of multiplying damages through history.

(96.) See discussion infra Part III.A.2.

(97.) See 12 U.S.C. [section] 5386 (Supp. IV 2010).

(98.) See Macey & Miller, supra note 85, at 55 (finding this rate "relatively high").

(99.) See Onnig H. Dombalagian, Can Borrowing Shares Vindicate Shareholder Primacy?, 42 U.C. DAWS L. REV. 1231, 1233-36 (2009), for a helpful summary of this literature. See also Stephen M. Bainbridge, Responses: Director Primacy and Shareholder Disempowerment, 119 HARV. L. REV. 1735, 1758 (2006) (arguing that "substantial efficiency benefits" flow from limiting shareholder participation); Lucian Arye Bebchuk, The Case for Increasing Shareholder Power, 118 HARV. L. REV. 833, 836 (2005) (presenting the case for allowing shareholders greater participation in corporate governance).

(100.) See Part III.A.4 below, which addresses proposals that would restore full personal liability to the directors and officers.

(101.) See Bebchuk & Spamann, supra note 11, at 249.

(102.) Squire, supra note 18, at 1152. For example, Freddie Mac and Fannie Mae were liable for guarantees on large numbers of mortage-based assets that were "highly likely to be triggered en masse under conditions when their shareholders would already be wiped out." Id. at 1153.

(103.) See id. at 1152-53.

(104.) See id.

(105.) Correlation seeking, as Squire describes it, is an offshoot of the well-established principle of risk-shifting, first proposed in the shareholder context by Jensen and Meckling and by Galai and Masulis. See Dan Galai & Ronald W. Masulis, The Option Pricing Model and the Risk Factor of Stock, 3 J. Fry. ECON. 53, 53-81 (1976); Michael C. Jensen & William H. Meckling, Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure, 3 J. FIN. ECON. 305, 331 (1976).

(106.) See STANDARD & POOR'S, supra note 3, at 48-60.

(107.) This assumes away litigation and recovery costs, no small thing when the actual recovery is sought.


(109.) See Press Release, SEC, SEC Proposes Rule Amendments to Remove Credit Rating References in Investment Company Act Rules and Forms (Mar. 2, 2011), available at

(110.) See Mark J. Flannery et al., Credit Default Swap Spreads as Viable Substitutes for Credit Ratings, 158 U. PA. L. REV. 2085 (2010).

(111.) See STANDARD & POOR'S, supra note 3, at 3-4.

(112.) Id. at 3.

(113.) See Thorsten V. Koeppl, Creating a "Too Big to Fail" Behemoth, FIN. POST (Feb. 24, 2011, 7:30 PM) %E2%80%98too -big-to-fail%E2%80%99-behemoth.

(114.) See Claire Hill & Richard Painter, Berle's Vision Beyond Shareholder Interests: Why Investment Bankers Should Have (Some) Personal Liability, 33 SEATTLE U. L. REV. 1173 (2010).

(115.) For references to bailouts as creditor bailouts, see GARY H. STERN & RON J. FELDMAN, TOO BIG TO FALL: THE HAZARDS OF BANK BAILOUTS, at ix (2004).

(116.) See 11 U.S.C. [section] 362(a) (2006).

(117.) See 11 U.S.C. [section] 506 (determination of security status); 11 U.S.C. [section] 507 (determination of priority in bankruptcy).

(118.) Arguably, it is so orderly a process that it is no longer necessary at all, at least for large firms. See Douglas G. Baird & Robert K. Rasmussen, The End of Bankruptcy, 55 STAY. L. REV. 751, 786-88 (2002) (arguing that for large corporations, prepackaged [section] 363 sales have made obsolete the original Chapter 11 paradigm of "saving" firms and protecting minority creditors).

(119.) See David Skeel, Give Bankruptcy a Chance, WKLY. STANDARD, June 29, 2009, at 27, available at 000/016/ 658hmvhc.asp.

(120.) See SKEEL, supra note 4, at 155-73 (arguing for several changes to the Bankruptcy Code to make SIFI bankruptcy possible).

(121.) See Press Release, Fed. Reserve Bd. (Sept. 16, 2008), available at http://www

(122.) William K. Sjostrom, Jr., The AIG Bailout, 66 WASH. & LEE L. REV. 943, 975 (2009) (tallying the various structures through which the government provided taxpayer funds to AIG).

(123.) And seemingly endless rounds of congressional inquiry into the A1G bailout. See Mary Williams Walsh, Drawing Fire, Geithner Backs Rescue of AIG, N.Y. TIMES, Jan. 28, 2010, at A3.

(124.) Hugh Son & Christine Harper, Blankfein Says He Wasn't Asked to Take AIG Haircut (Update2), BLOOMBERG (Jan. 13, 2010, 12:49 EST), apps/news?pid=newsarchive&sid=ayXyZExmn9sw.

(125.) Hugh Son, New York Fed "Very Sensitive" on AIG, E-Mail Says (Update2), BLOOMBERG (Jan. 19, 2010, 6:53 PM), =newsarchive&sid=aX4pBpXXkoOU.

(126.) See Jensen, Agency Costs, supra note 71.

(127.) See Too Big to Fail--The Role for Bankruptcy and Antitrust Law in Financial Regulation Reform (Part I). Hearing Before the Subcomm. on Commercial & Admin. Law of the H. Comm. on the Judiciary, 111th Cong. 132-34 (2009) (statement of David A. Skeel, Jr.) (arguing that market evidence suggests that Lehman's bankruptcy had less to do with the subsequent crisis than has been presumed); SKEEL, supra note 4, at 19-40.

(128.) See Heidi N. Moore, Goldman-AIG: If Goldman Was Hedged, Why Take Government Money?, WALL ST. J. (Mar. 20, 2009, 5:44 PM), 2009/03/20/goldman-aig-if-goldman-was-hedged-why-take-government-money.

(129.) For a thorough discussion of the law and economics of these kinds of contingent liabilities, see Squire, supra note 18.

(130.) See Walsh, supra note 123.

(131.) This is the very contention that probankruptcy scholars argue is unfounded. See SKEEL, supra note 4, at 19-40 (explaining that the Lehman bankruptcy did not represent the crisis contagion that many have come to presume).

(132.) See, e.g., The Federal Bailout of A1G: Hearing Before the H. Comm. on Oversight & Gov't Reform, 111th Cong. 22-36 (2010) (statement of Timothy F. Geithner, Secretary of the Treasury).

(133.) Dodd-Frank is not silent on the question of creditor haircuts, but it is nearly so. Creditor haircuts are not mandated anywhere in the law, nor even mentioned beyond the need for the FSOC to conduct "a study evaluating the importance of maximizing United States taxpayer protections and promoting market discipline with respect to the treatment of fully secured creditors in the utilization of the orderly liquidation authority authorized by this Act." Dodd-Frank Wall Street Reform and Consumer Protection Act, Pub. L. No. 111-203, [section] 215(a), 124 Stat. 1376, 1518 (2010). Beyond that, any creditor haircuts will have to be negotiated--or not--as part of the overall bailout package.

(134.) See Charles W. Calomiris & Richard J. Herring, Why and How to Design a Contingent Convertible Debt Requirement 6 (Apr. 2011) (unpublished manuscript), available at

(135.) John C. Coffee, Jr., Bail-Ins Versus Bail-Outs: Using Contingent Capital to Mitigate Systemic Risk 10 (Columbia Law Sch. Working Paper Series, Working Paper No. 380, 2010), available at

(136.) Jennifer Hughes, Credit Suisse Cocos Issue Deluged, FIN. TIMES (Feb. 17, 2011, 6:59 PM), #axzzlk3mqGIMQ (describing Credit Suisse's $2 billion issuance, oversubscribed at $22 billion).

(137.) Coffee ducks the issue of triggers. See Coffee, supra note 135, at 34.

(138.) See Charles Riley, Treasury Close to Profit on TARP Bank Loans, CNNMONEY (Feb. 3, 2011, 11:18 AM),

(139.) Hearing with Treasury Secretary Geithner: Hearing Before the Cong. Oversight Panel, 111th Cong. 35 (2010) (statement of Timothy F. Geithner, Secretary of the Treasury).

(140.) According to Andrew Ross Sorkin's definitive reporting, these otherwise healthy banks were told to accept billions of dollars of government funding in the form of loans in order to remove the stigma for the loans to Bank of America and Citigroup. See ANDREW ROSS SORKIN, TOO BIG TO FAIL 524 (2009).

(141.) A footnote is appropriate here to challenge the idea that these SIFIs actually have paid off their debts to the government. Special Inspector General for TARP Neil Barofsky thinks that the government has underestimated its exposure to AIG by $40 billion. See Mary Williams Walsh, Treasury Hid AIG Loss, Report Says, N.Y. TIMES, Oct. 26, 2010, at B1.

(142.) See Chris Isidore, Fannie, Freddie Bailout: $153 Billion ... and Counting, CNNMONEY (Feb. 11, 2011, 1:48 PM), fannie_freddie_losses/index.htm.

(143.) Charles Scaliger, TARP Bailout Payback: "Drop in the Bucket," NEW AM. (Oct. 28, 2010, 7:00 AM), -bailout-payback-qdrop-in-the-bucketq.

(144.) See J.W. Verret, Treasury Inc.: How the Bailout Reshapes Corporate Theory and Practice, 27 YALE J. ON REG. 283, 286 (2010).

(145.) See id. at 312.

(146.) Marcel Kahan & Edward B. Rock, When the Government Is the Controlling Shareholder, 89 TEX. L. REV. 1293, 1293 (2011).

(147.) See Jeffrey N. Gordon & Christopher Muller, Confronting Financial Crisis: Dodd-Frank's Dangers and the Case for a Systemic Emergency Insurance Fund, 28 YALE J. ON REG. 151 (2011).

(148.) See Wilmarth, supra note 27.

(149.) Wilmarth advocates for an industry-funded liquidation fund, but does not specify its amount. See id at 1015-18. Gordon and Muller are more specific, advocating for a trillion-dollar fund scaled to the size of the economy. See Gordon & Muller, supra note 147, at 151.

(150.) Wilmarth, supra note 27, at 1008.

(151.) Pub. L. No. 101-73, 103 Stat. 183 (codified in scattered sections of 18 U.S.C.).

(152.) Block, supra note 56, at 952.


(154.) Timothy Curry & Lynn Shibut, The Cost of the Savings and Loan Crisis: Truth and Consequences, 13 FDIC BANKING REV. 26, 32 (2000), available at bank/analytical/banking/2000dec/brv13n2_2.pdf.

(155.) Press Release, FDIC, FDIC-Insured Institutions Earned $14.5 Billion in the Third Quarter of 2010, Up from $2 Billion a Year Ago (Nov. 23, 2010), available at http://www

(156.) See HOUSE COMM. ON WAYS & MEANS, 101ST CONG., BACKGROUND MATERIAL AND DATA ON PROGRAMS WITHIN THE JURISDICTION OF THE COMMITTEE ON WAYS AND MEANS 66 (Comm. Print 1989); see also Martin J. McMahon Jr., The Matthew Effect and Federal Taxation, 45 B.C.L. REV. 993, 1026-27 (2004).

(157.) See FDIC Assessment Rates, FDIC, assessments/proposed.html (last updated May 24, 2011).

(158.) See Risk Categories & Risk Assessment Rates: Key Provisions Pertaining to Risk-Based Assessments, FDIC, (last updated Nov. 2, 2011).


(160.) William D. Cohan, Op-Ed., Make Wall Street Risk It All, N.Y. TIMES, Oct. 8, 2010, at A27, available at -it-all. For the academic version of this argument, see Hill & Painter, supra note 114.

(161.) Cohan, supra note 160.

(162.) Id. The FDIC has adopted, to a very limited extent, a Cohan-esque policy. See Carla Main, FDIC Clawbacks, Mortgage Deal, Swipe Fees, UBS-Libor Probe: Compliance, BLOOMBERG (Mar. 16, 2011, 5:22 AM), -clawbacks-mortgage-deal-ubs-libor-probe-compliance.html.

(163.) See Brian Ross & Alice Gomstyn, Lehman Brothers Boss Defends $484 Million in Salary, Bonus, ABC NEWS (Oct. 6, 2008), &page=1.

(164.) See Eric Dash, The Price of Any Departure Will Be at Least $159 Million, N.Y. TIMES, Oct. 27, 2007, at A12.

(165.) According to Boswell, Johnson's now-axiomatic quote is: "Depend upon it, Sir, when a man knows he is to be hanged in a fortnight, it concentrates his mind wonderfully." JAMES BOSWELL, THE LIFE OF SAMUEL JOHNSON 612 (David Womersley ed., Penguin Books 2008) (1791).

(166.) Sarbanes-Oxley Act of 2002 [section] 304, 15 U.S.C. [section] 7243(a) (2006).

(167.) 15 U.S.C. [section] 78j-4.

(168.) Hill & Painter, supra note 114, at 1189.

(169.) There has been sustained discussion in the corporate governance literature about the role of shareholders in risk management. See RICHARD ANDERSON & ASSOCS., RISK MANAGEMENT & CORPORATE GOVERNANCE (2009), available at This was also the theme of some of the literature on corporate governance following the accounting scandals of the 2000s. See, e.g., Robert Eli Rosen, Risk Management and Corporate Governance: The Case of Enron, 35 CONN. L. REV. 1157, 1157-61 (2003). My question is slightly different, however, and has been summarily ignored up until now. I ask how shareholders' incentives can be changed to force cost internalization, rather than simply whether more corporate governance can provide for more shareholder participation in management decisionmaking. For a note of skepticism on increasing efforts to involve shareholders, via directors, in risk management, see Peter Conti-Brown & Ronald J. Gilson, The Limits of Independence in Institutional Design 5-6 (Jan. 27, 2012) (unpublished manuscript) (on file with author).

(170.) Sjostrom, supra note 122, at 943.

(171.) Id. at 945.

(172.) See SORKIN, supra note 140, at 36-37.

(173.) 12 U.S.C. [section] 5386 (Supp. IV 2010).

(174.) Stiglitz, in his condemnation of banks' efforts to influence

bank regulatory reform, asks, "The old system worked well for the bankers (if not for their shareholders), so why should they embrace change?" Joseph Stiglitz, America's Socialism for the Rich, GUARDIAN (June 12, 2009, 3:00 PM EDT), jun/12/america-corporate-banking-welfare (emphasis added).

(175.) John Armour et al., What is Corporate Law?, in THE ANATOMY OF CORPORATE LAW: A COMPARATIVE AND FUNCTIONAL APPROACH 9 (Reinier Kraakman et al. eds., 2d ed. 2009).

(176.) See Easterbrook & Fischel, supra note 88, at 89-93, for the legal-economic analysis of limited liability.

(177.) Again, it's hard to say exactly how the argument would go, since no one is making it.

(178.) Corporate law envisions an exception to limited liability, through the much-discussed doctrine of "piercing the corporate veil." This sometimes-defended but largely pilloried doctrine is, perhaps counterintuitively, of little use to the present analysis. First, the invocation of veil piercing, in the colorful words of one oft-quoted account, is "[l]ike lightning, [in that] it is rare, severe, and unprincipled." Easterbrook & Fischel, supra note 88, at 89. Second, the test used to determine when veil piercing is to be invoked, unpredictable though it may be in practice, depends almost exclusively to the extent to which shareholders have treated the corporate form as a "mere instrumentality." See, e.g., Albert v. Alex. Brown Mgmt. Servs., Inc., No. Civ.A. 762-N, 2005 WL 2130607, at *9 (Del. Ch. Aug. 26, 2005) (mem.) ("The corporate veil may be pierced where a subsidiary is in fact a mere instrumentality or alter ego of its parent." (footnote omitted)). See generally Peter B. Oh, Veil-Piercing, 89 TEX. L. REV. 81 (2010), for an excellent recent discussion of veil piercing and another heroic effort at bringing empirical clarity to an almost irrevocably muddied area of law.

(179.) See Lucian A. Bebehuk et al., The Wages of Failure: Executive Compensation at Bear Stearns and Lehman 2000-2008, 27 YALE J. ON REG. 257, 281 tbl.4 (2010) (summarizing the hundreds of millions of dollars that executives from Bear Steams and Lehman Brothers received through market exit in the years prior to their firms' collapses).

(180.) Unjust enrichment occurs when someone has retained "a benefit conferred by another, without offering compensation, in circumstances where compensation is reasonably expected"; it can similarly be defined as a "benefit obtained from another, not intended as a gift and not legally justifiable, for which the beneficiary must make restitution or recompense." BLACK'S LAW DICTIONARY 1678 (9th ed. 2009). Other definitions explicitly note the absence of an element of fraud: "Although unjust enrichment may arise from fraud or several other predicates, the element of fraud or tortious conduct on the part of a defendant is not necessary in an action for unjust enrichment." 66 AM. JUR. 2D Restitution and Implied Contracts [section] 11 (West 2011) (emphasis added) (footnote omitted). The lack of a fraud element is helpful here as an analogue to elective shareholder liability, as in the latter case fraud or malfeasance need not--and, indeed, probably did not--occur during the period preceding the taxpayer bailout.

(181.) Law governing fraudulent conveyance comes from two sources: the Bankruptcy Code, 11 U.S.C. [section][section] 544(b), 548 (2006), and the state applications of either the Uniform Fraudulent Transfer Act, or the older Uniform Fraudulent Conveyance Act. The basic structure in each is effectively the same: in each case, fraudulent conveyance occurs if the debtor makes any effort to transfer property with "actual intent to hinder, delay, or defraud any creditor," or if the debtor does not receive "reasonably equivalent value in exchange for the transfer or obligation," and the debtor was about to incur debts beyond her ability to pay. UNIF. FRAUDULENT TRANSFER ACT [section] 4(a) (1984). In such instances, courts--at the behest of creditors, or the bankruptcy trustee--can recover the proceeds of those transfers, with some exceptions, on behalf of the creditors or trustee.

(182.) According to Miriam Cherry and Jarrod Wong, a clawback is "a recovery device that is potentially draconian but justifiable under the triggering circumstances because of an inherent unfairness that would otherwise prevail." Miriam A. Cherry & Jarrod Wong, Clawbacks: Prospective Contract Measures in an Era of Excessive Executive Compensation and Ponzi Schemes, 94 MINN. L. REV. 368, 411 (2009). This obviously broad term has, so far, been applied in the contexts of executive compensation, see supra Part III.A.4, and Ponzi schemes. In the case of Ponzi schemes, clawbacks have been used to great effect in the Madoff case, generating headlines that suggest huge hits to entities as famous (or notorious, depending on one's perspective) as JP Morgan and the New York Mets. See Michael Rothfeld & Chad Bray, Madoff Trustee Buzzes Mets--Team Ignored Warnings, Says a Lawsuit That Demands up to $1 Billion, WALL ST. J., Feb. 5, 2011, at B1; Jonathan Stempel, JPMorgan Ignored Suspicions About Madoff: Lawsuit, REUTERS (Feb. 3, 2011, 5:42 PM EST), -idUSTRE7127RW20110203 (describing the $6.4 billion lawsuit filed by Irving Picard, the trustee of the Madoff case, against JP Morgan Chase for complicity with Madoff's fraud).

(183.) Under [section] 547(b) of the Bankruptcy Code, the trustee or debtor in possession can generally "void" any transfer of property that occurred under five conditions: the transfer (1) benefited another creditor; (2) was made on account of antecedent debt; (3) was made while the debtor was insolvent; (4) was made on or within 90 days, and (5) would put the creditor in a better position than it would otherwise have been. 11 U.S.C. [section] 547(b). Voidable preferences are especially relevant to elective shareholder liability because they imply no wrongdoing by those who must disgorge profits otherwise appropriately received. Voidable preferences serve a purpose similar to that of the automatic stay in bankruptcy, which is to keep creditors from jockeying against each other to the detriment of the common pool of the debtor's assets. In that case, even a garden-variety payment on a debt could be voided. The idea, relevant to elective shareholder liability, is that the value leaving the debtor may be appropriately given to the creditor in some circumstances, but not in all circumstances. What is legal in one context may not be when the shadow of insolvency begins to loom larger.

(184.) See Susan Pace Hamill, From Special Privilege to General Utility: A Continuation of Willard Hurst's Study of Corporations, 49 AM. U. L. REV. 81, 91 n.42 (1999) ("The assumption that limited liability protection automatically resulted when operating in the corporate form began to develop in the early nineteenth century and proceeded at an uneven pace across the states.").

(185.) See Richard S. Grossman, Double Liability and Bank Risk Taking, 33 J. MONEY CREDIT & BANKING 143, 144 (2001); Howell E. Jackson, Losses from National Bank Failures During the Great Depression: A Response to Professors Macey and Miller, 28 WAKE FOREST L. REV. 919, 920 (1993); Macey & Miller, supra note 85, at 31; John R. Vincens, On the Demise of Double Liability of Bank Shareholders, 12 BUS. LAW. 275, 275-76 (1957).

(186.) Banking Act of 1933 [section] 8, 12 U.S.C. [section] 1811.


(188.) See Macey & Miller, supra note 85, at 37.

(189.) See KENNEDY, supra note 187, at 153.

(190.) Interestingly, scholars have demonstrated that even through the Great Depression, the double-liability regime did not yield significantly less recovery for the depositors of failed banks. Macey & Miller, supra note 85, at 34; see also Berry K. Wilson & Edward J. Kane, The Demise of Double Liability as an Optimal Contract for Large-Bank Stockholders 5 (Nat'l Bureau of Econ. Research, Working Paper No. 5848, 1996). Howell Jackson contends that Macey and Miller miss the problem of recovery by focusing on ultimate recovery at the expense of risk of recovery, as recovery during the Depression took longer and was therefore more uncertain than during other periods under double liability. See Jackson, supra note 185, at 922.

(191.) See Alan D. Morrison & William J. Wilhelm, Jr., The Demise of Investment-Banking Partnerships: Theory and Evidence 1 (Jan. 2007) (unpublished manuscript), available at IBDemiseFinal%20Aug%2030%2006.pdf.

(192.) Id.

(193.) See, e.g., MICHAEL LEWIS, THE BIG SHORT: INSIDE THE DOOMSDAY MACHINE 263 (2010) (discussing some of the deleterious effects of the conversion from partnerships); Alan S. Blinder, Op-Ed., Crazy. Compensation and the Crisis, WALL ST. J., May 28, 2009, at A15; Steven M. Davidoff, A Partnership Solution for Investment Banks?, N.Y. TIMES DEALBOOK (Aug. 20, 2008, 10:24 AM), a-partnership-solution-for-investment-banks; Michael Lewis, The End, PORTFOLIO.COM (Nov. 11, 2008), The-End-of-Wall-Streets-Boom.

(194.) Of course, another counterargument is that elective shareholder liability is politically infeasible. That may well be true. But political infeasibility is something of a non sequitur to the fundamental merits of the arguments presented here. Elective shareholder liability will rise or fall as a concept of academic interest on its own merits--its political success or failure is a separate issue entirely. Even so, it is interesting to note the tension between the political-feasibility and unenforceability arguments. That is, the Grundfestian arguments posit that capital market participants will simply be able to evade the regulations through the mechanisms described above. If so, one should expect that bankers would be in favor of elective shareholder liability, and indeed would elect it. If evasion is "trivially easy," to use Grundfest's phrase, see Grundfest, supra note 20, at 411, 414 n.105, 424 n.149, then the bankers have very little to lose by electing shareholder liability, and indeed may gain the appearance of supporting meaningful banking reform.

(195.) See id. at 389.

(196.) See Susanne Craig & Eric Dash, Study Points to Windfall for Goldman Partners, N.Y. TIMES DEALBOOK (Jan. 18, 2011, 9:40 PM), partners/?hp.

(197.) See id.

(198.) See Lynn A. Stout, Why the Law Hates Speculators: Regulation and Private Ordering in the Market for OTC Derivatives, 48 DUKE L.J. 701, 714, 722-24 (1999); see also JOHN STEELE GORDON, AN EMPIRE OF WEALTH: THE EPIC HISTORY OF AMERICAN ECONOMIC POWER 226-27 (2004) (describing the continuing existence, despite their legal unenforceability, of contracts requiring payment in gold specie, during a financial crisis in the late nineteenth century).

(199.) Grundfest, supra note 20, at 391.

(200.) See Hansmann & Kraakman, supra note 20, at 429-31.

(201.) See, e.g., Ronald D. Orol, Geithner Urges Global Capital Rules for Swaps, MARKETWATCH (June 6, 2011, 4:29 PM EDT), -urges-global-capital-rules-for-swaps-2011-06-06-127390.

(202.) See Janet Cooper Alexander, Unlimited Shareholder Liability Through a Procedural Lens, 106 HARV. L. REV. 387 (1992).

(203.) For Hansmann and Kraakman's response, see Henry Hansmann & Reinier Kraakman, A Procedural Focus on Unlimited Shareholder Liability, 106 HARV. L. REV. 446 (1992).

(204.) See 15 U.S.C. [section][section] 78/(i), 78m(d)-(e), 78n(d)-(f) (2006).

(205.) See Guhan Subramanian, Bargaining in the Shadow of Takeover Defenses, 113 YALE L.J. 621, 635 (2003). For a broader discussion of the problem of structural coercion, see Lucian Ayre Bebehuk, Toward Undistorted Choice and Equal Treatment in Corporate Takeovers, 98 HARV. L. REV. 1693, 1723-25 (1985).

(206.) See, e.g., Catarina Saraiva & Mariko Ishikawa, Australian, N.Z. Dollars Rise After Obama's Jobs Address, China Inflation, BLOOMBERG (Sept. 9, 2011, 10:37 AM), obama-s-iobs-addresschina-inflation.html; Catarina Saraiva, Franc Falls as Economy Minister Says Currency Is "Massively Overvalued," BLOOMBERG (Sept. 9, 2011, 5:58 AM), says-currency-is-massively-overvalued-.html.

(207.) See, e.g., David Zaring, Administration by Treasury, 95 MINN. L. REV. 187, 188-91 (2010).

Kirsty A. Macdonald, Academic Fellow, Rock Center for Corporate Governance, Stanford Law School and Stanford Graduate School of Business. I thank the Stanford Fellows Workshop, the Stanford Law and Economics Seminar, the faculty workshops at BYU Law School, Stanford Law School, and St. John's University School of Law for helpful discussions; ! thank Christopher Cole, the editorial staff of the Stanford Law Review, and the anonymous reviewers for their excellent feedback; and l thank individually Anat Admati, Marco Becht, Sam Bray, Brian Cheffins, Nikki Conti-Brown, Dick Craswell, John Crawford, Rob Daines, Brigham Daniels, Darrell Duffle, Ron Gilson, Jeff Gordon, Joe Grundfest, Mark Kelman, Mike Klausner, Adam Levitin, Michael McConnell, Larry Mitchell, Britton Olson, Paul Pfleiderer, Mitch Polinsky, Elizabeth Pollman, Jordan Segall, David Skeel, Richard Squire, Lynn Stout, Colin Sturt, John Turner, Art Wilmarth, Andrew Yaphe, and Jonathan Zittrain for substantive feedback. I am also grateful to my extraordinarily able colleagues at the Robert Crown Law Library at Stanford for their patient assistance.
COPYRIGHT 2012 Stanford Law School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2012 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Conti-Brown, Peter
Publication:Stanford Law Review
Date:Feb 1, 2012
Previous Article:Incriminating thoughts.
Next Article:Harrington's wake: unanswered questions on AEDPA's application to summary dispositions.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters