Deterring Libor Manipulation and Improving Benchmarks

It’s tempting to think that we might be seeing the end of potential manipulation of financial benchmarks and rates, such as Libor.

The story would go like this: the Libor rate was an anomaly – humorous in retrospect – that was structured in a way prone to manipulation because it relied on the subjective (and objectively unverifiable) judgments of bank personnel who were subject to conflicts of interest.  Having weathered that scandal, we have now grown wiser. The FSA’s recent reforms of the rate-setting process have constrained the opportunities for ex ante human discretion in Libor: they treat Libor panel submissions as a regulated activity, and the UK has decided to replace the British Bankers’ Association as the keepers of the rate. And ex post, vigorous government enforcement actions and private lawsuits have made banks realize that misbehavior does not pay, which disincentivizes future malfeasance. So, the story concludes, aberrant rates have been disciplined and domesticated.

Though tempting, this narrative provides false comfort.

Vital as enforcement and plaintiffs’ suits may be, they are unlikely to cause benchmark contributors to adequately internalize the costs that manipulation would create. This is in part because of the practical and conceptual difficulties involved in establishing rate manipulation, some of which I discuss in my recent paper co-authored with Rosa Abrantes-Metz and Gabriel Rauterberg, Revolution in Manipulation Law.

Last week, Judge Naomi Reice Buchwald of the Southern District of New York dismissed all federal and state anti-trust claims by plaintiffs allegedly injured by the purported manipulation of Libor. The core of the court’s argument was that the Libor-setting process was always cooperative and never intended to be competitive; hence, a conspiracy to manipulate it is not a cognizable injury under federal antitrust law. The court will allow only the suit by traders of Eurodollar futures under the Commodity Exchange Act, and there, for only a sliver of the alleged conduct. This is a punishing result for plaintiffs like the Charles Schwab fund family, which alleged that it was sold financial products by the Libor contributor banks, who then manipulated the rate – and their payment obligations – downward.

As for the government enforcement actions in the UK, US, and Japan, billions of dollars have been paid, but criminal guilt shunted off to subsidiaries in order to spare the parent the full implications of their employees’ behavior. For example, RBS’s Japanese subsidiary, but not the parent entity, plead guilty to misconduct.  This may be because, as they have increasingly been candid in admitting, regulators are loath to damage systemically important institutions like the Libor banks.  As Gabriel Rauterberg and I describe in a recently published article, Index Theory, and an upcoming piece for the University of Virginia’s Journal of International Law called Assessing Transnational Private Regulation of the OTC Derivatives Market: ISDA, the BBA, and the Future of Financial Reform, the “too big to fail” problem is exacerbated in the index market as we observe it. When benchmarks are produced in conjunction with numerous other vital economic functions by the same few firms, it can be risky to punish offenders. Furthermore, banks participate in Libor voluntarily: increase the litigation and regulation cost too much, and the panel banks may withdraw from the process (as some have already threatened to do).

The difficulty of establishing credible enforcement makes it hard to believe that ex post enforcement will be successful in aligning incentives for index providers to make trustworthy products.   And because of its wide use, disruptions in the rate can be so widely felt that those banks may simply not have sufficient funds to make victims whole even if prosecutors or private litigants were to demonstrate manipulation and be rewarded full damages.

More importantly, as we describe in Index Theory, Libor is by no means unique. All financial indices implicate some level of human discretion, subjectivity, and opacity, and many vital rates are as manipulable as Libor.  One need only read Tuesday’s Bloomberg article on the ISDAFix to realize that crucial rates are everywhere subject to human influence. ISDAFix is to long-term interest rates much like Libor is to short-term interest rates. Like Libor, it is the base rate for almost $400 trillion of notional value, as banks and corporations use financial derivatives to manage interest rate risk. If it rises or plummets, fortunes are made or lost. And like Libor, it is based on the self-assessments of brokers – a group which may have submitted erroneous or delayed quotes in order to benefit the trading positions of key clients. The structural similarities between the setting of Libor and ISDAfix are many: both are critical components of the financial infrastructure of the over-the-counter derivatives world; depended upon by many; and yet are set based on the subjective – and not objectively verifiable – judgments of individuals who are subject to potential conflicts of interest.

Though discretion and subjectivity carry risks, they do not necessarily lead to self-dealing. Instead, we argue discretion is necessarily part of our financial infrastructure and efforts to purge it from the system cannot be effective as optimal solutions. Judgment is needed to fill in where transactional data has gaps or liquidity is poor, and to otherwise control for anomalies in the market. This conclusion is supported by the fact that the FSA’s final regulations carve out important spaces for discretion and subjective inputs in recognition of the unworkability of a purely market-based or mechanical index. Any financial system must learn to manage, rather than eliminate, these loci of discretion.

Clear and accurate understanding of how benchmarks function is vital if we are to make good policy here. In addition to showing the implausibility of regulating all aspects of the index creation or causing them to be fully “market set,” Index Theory creates a taxonomy of index provider type, motivation and risk. One surprising result of our analysis is that in many cases, indices provided as a byproduct of other commercial ventures (such as Libor’s production in conjunction with banks’ participation in the credit and swap markets) are prone to distinctive manipulation risks. An important corollary is the importance of awarding index providers valuable property rights in their index creations. Major banks provide information to set Libor essentially for free, so they do not gain from investments in governance and quality that might increase its already wide use.

Though my work with Gabriel Rauterberg has been associated with the proposition that index providers need to be more generously compensated for their work than current intellectual property law assures, we acknowledge the need for better enforcement and oversight.  Any policy interventions must address both sides of the ledger, carrots and sticks.