In the wake of multiple criminal and regulatory investigations into alleged manipulation of the London Interbank Offered Rate (“LIBOR”), Andrew Bailey, Chief Executive of the UK Financial Conduct Authority (“FCA”) has announced1 plans to abandon this “unsustainable” benchmark in five years’ time, and transition to an alternative rate. The plan at least is clear, but will this transition assist the FCA in its wider efforts to tackle benchmark manipulation?
A brief history of LIBOR
LIBOR was developed in 1984 by the British Bankers’ Association (“BBA”), in conjunction with the Bank of England, in response to calls from market participants for a universal rate for new financial instruments. Since then, the benchmark has become widely used as a reference rate across an enormous variety of global financial contracts.
The precise definition of LIBOR - and therefore the methodology for its calculation - has developed over time, but key elements have remained fairly consistent. Each day, for each currency, a number of ‘panel’ banks are asked to submit rates at which they consider they could borrow funds across a number of tenors in the interbank market at 11 a.m. Rates submitted by the top and bottom 25% of panel banks are disregarded, and the average of the remaining banks’ rates form the published rate.
Beginning in 2012, multiple international investigations into the operation of LIBOR led to allegations that banks were manipulating LIBOR rates. At the core of these allegations was the suggestion that banks had submitted LIBOR rates that were not a true reflection of their borrowing costs, in order to move the published rate in one direction or another and generate profit.
In light of these investigations, and following recommendations made in the 2012 Wheatley Review of LIBOR2, responsibility for oversight of the benchmark was handed from the BBA to the FCA in April 2013. Since then, both the FCA and submitting panel banks have devoted considerable expense and resource to strengthening systems and controls around LIBOR, with the stated aim of ensuring that when it comes to LIBOR manipulation, history does not repeat itself.
The end of the road
Given the FCA’s significant efforts to strengthen LIBOR in recent years, how do Bailey and the FCA now find themselves at a juncture where, in Bailey’s words, it is “not only unsustainable, but also undesirable” to continue using LIBOR?
The answer lies, at least in part, in the FCA’s method of tackling the risk of LIBOR manipulation. Since taking up the role of caretaker of LIBOR, the FCA’s approach has been two-fold. Firstly, introduce a criminal offence of making false or misleading statements in the course of setting a “relevant benchmark”, and eliminate any ambiguity that might have existed previously about the legality of benchmark manipulation. Secondly, work to remove the element of subjectivity and judgment inherent in the LIBOR definition itself which, according to Bailey, leads to a “greater vulnerability to manipulation”. In lieu of subjective judgments by individual submitters about where a bank could, theoretically, borrow in the interbank market, the message from the FCA to panel banks has been clear: “tie the rate more closely to transactions rather than judgments” (our emphasis).
While this transaction-based approach seems neat enough, in the case of LIBOR it carries with it a fundamental problem. As Bailey recognizes in his speech, “[i]f an active [interbank] market does not exist, how can even the best run benchmark measure it?”. In other words, with no transactions in the interbank market in a particular tenor – which is common for some of the more infrequently-traded tenors for which LIBOR rates are submitted – how can a panel bank make a LIBOR submission at all? The chances of seeing any improvement on this front are gloomy, at least from the FCA’s perspective: “there seems little prospect of these markets becoming substantially more active in the near future”.
Given the apparent absence of an active interbank market, and the FCA’s insistence that the only viable model for LIBOR is transaction-based rather judgment-based, it is easy to see why LIBOR has reached an impasse. However, with hundreds of trillions of dollars’ worth of financial contracts pegged to LIBOR, the rate cannot simply be abandoned overnight: “[w]e could not – and cannot – countenance the market disruption that would be caused by an unexpected and unplanned disappearance of LIBOR”. A suitably low-risk, transaction-based alternative must be found.
The solution, according to the Bank of England’s snappily-titled Working Group on Sterling Risk-Free Reference Rates, is the Sterling Over Night Index Average (“SONIA”). SONIA is administered by the Bank of England, and reflects banks’ and building societies’ overnight funding rates in the sterling unsecured market. Unlike LIBOR, there will ordinarily be high volumes of transaction data underpinning SONIA. The data is gathered by the Bank of England, and a rate is calculated on a quasi-automatic basis, with no real subjective judgment involved.
When one door closes…
Given its controversial history, the FCA’s efforts to move on from LIBOR are understandable. However, in terms of preventing future manipulation, simply transitioning to a different benchmark may not be quite the panacea the FCA would have us believe.
For all its flaws, the subjective nature of the LIBOR-setting process allows submitters to take a holistic view of the market in deciding on an appropriate rate to set. In doing so, a submitter may come to the view that an unexpected movement in the market can be written off as, at best, unsubstantiated hype, or at worst, a more sinister attempt by someone to manipulate or spoof the underlying market. In cases of unexplained market behavior or transactions, submitters can act as a moderating force by making an assessment about how observable transaction data fits into the wider picture.
In transitioning to a rate that is generated automatically from transaction data, such as SONIA, the FCA remove that ability to differentiate between legitimate and questionable market behavior. Unexpected movements in the underlying market will automatically be reflected in the resulting rate without checks or balances, a system which on paper seems inherently vulnerable. As Bailey says, SONIA does not “involve expert judgment”. In the fog of concern about allowing any subjectivity into the rate-setting process, the FCA has failed to acknowledge the benefits that this subjectivity and “expert judgment” can bring.
Of course, it is far easier to take issue with the FCA’s proposed alternative benchmark than it is to suggest a perfect alternative. By replacing LIBOR with SONIA, the FCA would certainly make it extremely difficult to recreate the specific misconduct demonstrated in the LIBOR investigations - in other words, submitting misleading rates in order to benefit a trading position. The question is whether, in the process, the door is left open for manipulation of a new and unexpected kind.