Computerization and the Abacus: Reputation, Trust, and Fiduciary Duties in Investment Banking

Editor’s Note: Steven Davidoff is a Professor of Law at the University of Connecticut. This post is based on a paper by Mr. Davidoff, William J. Wilhelm, Jr. of the University of Virginia, and Alan D. Morrison of the University of Oxford.

In our essay Computerization and the Abacus: Reputation, Trust, and Fiduciary Duties in Investment Banking, recently posted to the SSRN, we analyze the 2007 synthetic collateralized debt obligation transaction, ABACUS 2007-AC1 SPV (ABACUS) and the subsequent SEC civil complaint against Goldman Sachs in connection with the ABACUS transaction. We use this analysis as a touchstone to examine the debate over whether to impose fiduciary duties or other heightened regulation upon investment bank/counter-party transactions, the subject of a recently released SEC study (available here).

We situate the SEC’s complaint against Goldman Sachs in the context of recent technological changes within the investment banking market. Investment banking was historically a relationship-based business, sustained by reputationally intermediated tacit contracts. Recent advances in information technology and financial economics have codified many formerly tacit elements of investment banking. As a result, some investment banking deals are now transacted at arm’s length, and rely more upon formal contracts.

The SEC’s complaint highlights the legal and reputational nature of an investment bank’s responsibilities to its counterparties. The SEC’s complaint was premised on a legal violation but was notable because it implicated the trust relationship between an investment bank and its counter-parties. The SEC asserted a formal violation of the antifraud rules, but appeared to be applying the rule to enforce a trust-like relationship among Goldman and its counterparties analogous to a fiduciary duty-type relationship. The SEC’s case was ultimately illustrative of the shifting sands of the investment bank counter-party relationship from trust-based to transactionally-oriented, and the failure of regulators and market-participants to keep up with such developments.

In this light, we argue that the imposition of increased legal regulation is predicated on the type of deal at issue.  If the deal is transactional in nature, at arms-length, there would be a case for bright-line legal rules, but not wide standards like fiduciary duties. Such a case supports the SEC’s application of federal securities laws to the transaction, but not in the fiduciary duty-like manner the SEC purported they applied. And additional formal codified law appears not only unnecessary but economically inefficient and contrary to parties’ norms and expectations.

However, some deals continue to be arbitrated by tacit rules and norms and, for these deals, legal rules are less appropriate, because it is very hard for a third party to ascertain tacit understandings made in the context of a long-lived relationship. An attempt to introduce bright-line legal rules into reputationally intermediated relationships may even impair the counterparties’ ability to arrive at informal arrangements, and so to trade. It is precisely because this sort of determination is quite difficult that trust-based contracting remains important in finance. And it is because of this that the SEC’s application of the anti-fraud rules in a fiduciary-like manner to this transaction was economically wealth-destroying (Goldman Sachs lost $10 billion in market value in the wake of the complaint’s filing) to the extent it purported to regulate a trust-based relationship.

There are circumstances where imposition of a legal gap-filling standard such as fiduciary duties, as opposed to bright-line legal rules, can effectively substitute for a trust-based relationship and produce economically superior outcomes. Such circumstances arise when there are well-established standards and norms of behavior, so that a third party arbiter is able to judge whether such standards have been adhered to. In addition, one or both of two conditions must obtain: first, one of the parties to the trade must lack either the sophistication or the resources needed to identify violations of the received norms of behavior; second, one of the parties to the trade must be unable credibly to threaten to impose reputational sanctions upon the other, either because it lacks market voice or market credibility.

Based on our analysis we preliminarily believe that the issue of fiduciary duties as applied to broker-dealers in their interactions with retail customers requires further study to determine if and how our criteria for imposition of a fiduciary duty standard is met. In particular, the same types of difficulties in identifying transactional versus trust-based inter-actions arise in this context, such that, even if fiduciary duties are appropriate, they may not be economically efficient if applied on a uniform basis.

However, we believe that imposition of a fiduciary duties standard when investment banks are dealing with commercially sophisticated parties, such as in the ABACUS transaction would be inappropriate for a number of reasons, including that in this type of trade it is quite difficult to determine the beliefs and the incentives of the deal parties, and, hence, a failure on an investment banks’ part to abide by the reasonable expectations of its counterparties as set by market norms

The supervision of deals like ABACUS should therefore reflect the extent to which they are transactional or relational; we argue that in neither case is there justification for the application of additional legal rules or the gap-filling standard of fiduciary duties to investment banking transactions with sophisticated counter-parties.

The full paper is available for download here.

Both comments and trackbacks are currently closed.