The Volcker Rule and Goldman Sachs

Andrew Tuch is a Fellow of the Program on Corporate Governance and a John M. Olin Fellow at Harvard Law School, as well as a senior lecturer in the Faculty of Law at the University of Sydney.

In its recently issued report, Wall Street and the Financial Crisis: Anatomy of a Financial Collapse, the Senate Permanent Subcommittee on Investigations considered the conduct of Goldman Sachs in several transactions, including the ABACUS 2007-AC1 collateralized debt obligation. The report “examines Goldman’s conduct in the context of the law prevailing in 2007,” [1] and it asserts that the Volcker Rule provisions of the Dodd-Frank Act, “if well implemented, will protect market participants from the self-dealing that contributed to the financial crisis.” [2] But what justification exists for the conflict of interest restrictions in the Volcker Rule provisions, and how would the Volcker Rule provisions have applied to the ABACUS CDO had the provisions been in force at the time?

In my paper, Conflicted Gatekeepers: The Volcker Rule and Goldman Sachs, I consider the conflict of interest restrictions in the Volcker Rule provisions. These provisions, namely Sections 619 and 621 of the Dodd-Frank Act, purport to impose fiduciary-like standards on banks in their arm’s length relationships with sophisticated counterparties. Section 619 generally prohibits banks from engaging in proprietary trading and affiliating with certain private funds; it permits some activities as exceptions to this general prohibition, but subjects such activities to the requirement that they not give rise to material conflicts of interest, including conflicts between banks and their “counterparties.” Section 621 purports to ban material conflicts of interest between banks (in their capacity as underwriters) and investors in offerings of asset-backed securities.

The limits on conflicts of interest in the Volcker Rule provisions are puzzling since the law generally limits the self-interested conduct of an actor, and thus regulates conflicts of interest, only in fiduciary relationships. But fiduciary obligations rarely arise between sophisticated counterparties in arm’s length relationships. What justification then exists for the restrictions on conflicts in the Volcker Rule provisions? My paper seeks to provide a justification. It then applies the Volcker Rule provisions to the alleged conduct which led to their adoption by Congress – that of Goldman Sachs in the marketing of the ABACUS CDO.

The justification relates to the gatekeeping role banks perform in securities transactions. Building on the extensive literature on gatekeeper liability, this paper explains how the regulation of conflicts of interest – of the type in the Volcker Rule provisions – may promote banks’ gatekeeping role. As underwriters in securities offerings, banks have the power to monitor and control the disclosure decisions of corporate issuers, their clients. Exercising this power, banks may deter disclosure wrongs by issuers, thereby providing some assurance to investors as to the accuracy of issuers’ disclosures. Issuers, in turn, have incentives to engage reputable banks to underwrite their offerings – effectively “renting” underwriters’ reputations for assuring the accuracy of corporate disclosures. The gatekeeping function serves to economize on the information costs investors face in verifying the accuracy of corporate disclosures in securities offerings. This will be so even in a transaction involving sophisticated investors, particularly where the gatekeeper is primarily responsible for the issuer’s disclosures (such where the issuer is a SPV, formed for the purpose of the transaction – as in the ABACUS CDO).

Like gatekeeper liability, limits on a gatekeeper’s conflicts of interest may shape the incentives facing the gatekeeper to assure the accuracy of an issuer’s disclosures. Limiting conflicts may shape a gatekeeper’s incentives not by increasing its expected penalties in the event of a disclosure error, but by limiting the countervailing incentives it faces to less vigilantly police an issuer’s disclosures. As an example of countervailing incentives, consider the gatekeeper that has “bet against” the securities it underwrites and thus stands to benefit financially from their failure; that gatekeeper may be rewarded for inducing investors to participate in the offering. Regulation limiting the countervailing incentives facing gatekeepers (that is, regulation limiting gatekeepers’ conflicts of interest) can serve as a supplement or perhaps serve as an alternative to gatekeeper liability. Unlike gatekeeper liability, regulation limiting conflicts avoids the complex exercise of assessing damages; it also provides regulators with less discretion in enforcement, a relevant consideration if one is concerned about regulatory capture. Accordingly, in addition to the traditional basis for regulating conflicts of interest (to reduce agency costs), this paper develops an additional basis – to economize on information costs, a function performed by gatekeepers in securities transactions.

This paper argues that the limits on conflicts in the Volcker Rule provisions may be justified on the ground that they promote the gatekeeping role of banks. To the extent that the provisions limit conflicts between banks and their clients or customers, the regular rationale for regulating conflicts – the reduction of agency costs – may apply, of course. But the provisions also limit conflicts between banks and their counterparties; in doing so, they may be justified on the basis developed in this paper. However, the provisions should be narrowly interpreted to apply to relations between banks and counterparties only where banks can be considered to be performing a gatekeeping role, such as in underwriting a securities offering. The provisions should not be interpreted to limit conflicts of interest in other relations between banks and counterparties.

The paper also considers the 2010 enforcement action by the Securities and Exchange Commission against Goldman Sachs alleging securities fraud in the marketing of the ABACUS CDO. The SEC’s focus was on Goldman’s role as an underwriter in marketing asset-backed securities. The paper considers how the Volcker Rule provisions would have applied to the bank’s alleged misconduct.

The application of the Volcker Rule provisions to the ABACUS CDO turns on whether Goldman faced a material conflict of interest. Many commentators have argued that Goldman faced no conflict because, like investors in the CDO, the bank ultimately lost money in the deal. This paper explains, however, that the ABACUS deal comprised a series of transactions, which – contrary to popular impression – were not executed simultaneously, but over the course of several weeks. The paper shows why, at the time the ABACUS CDO transaction closed in April 2007, Goldman faced a position of conflict with the interests of investors. At this time, the bank had not neutralized the conflicting interests it faced as a counterparty to a credit default swap entered into with the CDO entity. The paper also explains how, although Goldman – like investors – lost money in the CDO, Goldman had apparently intended to offload that residual interest in the CDO. In other words, the bank’s loss should not be taken to suggest that the bank’s incentives were aligned with those of investors.

In a world where sophisticated parties can shift and transform their financial positions rapidly through the use of myriad financial instruments, the upshot for regulators is that banks’ incentives may be based not on their immediate or ultimate financial positions but on their intended or anticipated positions. Determining whether a material conflict of interest exists is a complex task in this context. Regulators should take a liberal approach to identifying material conflicts under the Volcker Rule provisions. The ABACUS transaction illustrates the importance of this point: that Goldman eventually lost money on the deal does not mean it did not face a conflict of interest with investors earlier in the course of the transactions, nor does it mean that the bank’s incentives were aligned with those of investors. Rather, it seems clear that Goldman would have violated the Volcker Rule provisions had they been in force at the time.

In examining the application of the Volcker Rule provisions to the ABACUS CDO, the paper offers some other preliminary comments on interpreting the provisions. The paper argues that the concept of materiality in the expression “material conflict of interest” should be based on the definition of materiality under Section 10(b) of the Securities and Exchange Act. Since the regulation of gatekeepers is primarily intended to ensure the accuracy of an issuer’s disclosures, a material conflict will arise where there is a substantial likelihood that a reasonable investor, in the position of the person with whom the bank’s interests conflict, would consider the conflict as having significantly altered the “total mix” of information about the investment. The paper also discusses the intended role of information barriers, or so-called Chinese walls, under the legislation, and raises doubts about their intended utility.

Finally, in examining the ABACUS deal, the paper separates the issue of whether Goldman faced a material conflict of interest (for purposes of the Volcker Rule provisions) from the issue of whether the bank committed securities fraud. The latter issue, which was the crux of the SEC case, depends on the factual materiality of the bank’s alleged misstatements, a matter on which the available evidence is inconclusive. This paper focuses instead on whether Goldman faced a conflict of interest. That is the question raised by the Volcker Rule provisions. The provisions tackle disclosure misstatements in securities transactions by shaping the incentives facing banks acting as gatekeepers. By preventing banks from having conflicts with the interests of investors, including sophisticated investors, the Volcker Rule provisions attempt to reduce countervailing incentives banks may face as gatekeepers to less than adequately monitor and control the accuracy of disclosures. This will not prevent banks from committing securities fraud, but it should mitigate the incentives they may face to do so.

The paper is available here.


[1] United States Senate Permanent Subcommittee on Investigations, Committee on Homeland Security and Government Affairs, Wall Street and the Financial Crisis: Anatomy of a Financial Collapse, Majority and Minority Staff Report, April 13, 2011, at 624 (“This analysis examines Goldman’s conduct in the context of the law prevailing in 2007. Since then, the Dodd-Frank Wall Street Reform and Consumer Protection Act of 2010 has established new conflict of interest prohibitions that would apply to this type of conduct…”).
(go back)

[2] Id. at 638 (“Together, these two prohibitions [Dodd Frank Act, sections 619 and 621], if well implemented, will protect market participants from the self-dealing that contributed to the financial crisis.”).
(go back)

Both comments and trackbacks are currently closed.

One Comment

  1. Ami de Chapeaurouge
    Posted Sunday, May 15, 2011 at 9:44 am | Permalink

    Couldda and Shouldda
    This is a sophisticated take, thanks. Any reservations about an analysis pondering the retroactive application of a recently adopted statute notwithstanding, the question is whether such gate-keeping function derived from Goldman being the Cayman-Island SPV originator and underwriter be a reliable standard for the future. You find rightly fault with Professor Duffie’s asseveration that there could be no conflict of interest in that Goldman failed to off-load a remaining slice of risk (worth 50 million if I remember correctly), which struck me at the time when I listened to his presentation (with Grundfest’s) at the Rock Center (generously made available to the rest of us via Internet-video) as counter-intuitive. As far as Grundfest’s own axiomatic assumption is concerned that there could be no conflict of interest, you counter with the details and sequence of the time-line in which the transaction evolved. But this doesn’t amount to much of an argument, as, you better believe me, Goldman would not have sought out and marketed the notes to IKB were it not for the original agreement with Paulson as a matter of intent.
    Steve Davidoff’s dichotomy between trust-based business and computer-generated trading didn’t convince me in the article, nor in his recent summary of it in the New York Times either. People all over the world still look with some awe and respect at Goldman for its reputation and they would be the first to tell you how much such reputational capital matters for them, as you rightly point out in your reference to their self-searching and soul-searching earlier in the year. They may even have inflicted evil from a moral point of view and earn our disdain for certain practices, if you follow the thread of analysis from the Magnetar analysis in ProPublica to the most recent polemic in Rolling Stone. But was there a reliable legal standard that would have rendered their ABACUS-related comportment illegal? Or in some other transactions to an effect that would require the Justice Department to step in? Maybe for lying to Congress, but I can’t see a securities laws violation in what they did or left undone, as nobody distinguishes in the Big Boy defense between sophisticated and semi-sophisticted investors.
    I sat down with the head of the legal department at KfW (the ultimate parent of IKB at the time of the ABACUS transaction, IKB now belongs to Lonestar) after the SEC charges were lodged and before the SEC settlement became known and obviously such conversation is privileged and they did not need to retain outside counsel as the US$ 300 million recovery inured to their (KfW’s) ultimate benefit and not Lonestar’s. But against the so-called ‘Big Boy’ defense that Goldman and others have oftentimes raised, please consider that a German Landesbank (or its functional equivalent such as IKB) never was an intellectual equal to Goldman in terms of sophistication. Rather, they inhabited a space best demarcated as ‘Semi-sophistication’ Square, the most dangerous place on earth. ACA is out of business and so is IKB as an independent entity. Say no more.