17 August 2017
Better the devil you know?…
Libor may be a dead man walking – but the pace will be very slow.
by Frank O’Nomics
There are many people in financial services who must wish that Libor had never been invented. Obviously those traders who are currently detained in Her Majesty’s prisons will be in that group, but also those bank employees whose bonuses were cut due to the huge fines that were paid for rigging the rate ($9bn and counting); also the regulators and central bankers who were widely discredited for their inability to fulfill their supervisory roles. There should then be widespread celebration following the announcement from the Financial Conduct Authority that Libor as a benchmark will be phased out by the end of 2021. However, this may be easier said than done, when an estimated $350 trillion worth of interest rate swaps, forward rate agreements, exchange traded futures and options, syndicated loans and floating rate notes have been priced using it. We are then faced with two key questions: What will replace Libor?, and, how quickly can it be fully replaced? Neither has a simple answer.
First it is important to understand the background to the Libor fiasco. The London Interbank Offered Rate was created as a benchmark in 1986 to help price the growing number of loans and other financial contracts that depended on a floating rate. There are 35 different types of Libor, across 5 currencies and 7 maturities ranging from overnight to 12 months. The rates are calculated from submissions by banks who estimate what it would cost to borrow money from each other. The problem came when, due to the financial crisis, banks weren’t prepared to lend to each other, meaning that the levels were artificial and subject to manipulation. Even now, 10 years after the crisis, it is estimated that only one third of the 3 month Libor submissions (the most commonly used measure) are based on real transactions, and one less used measure was found to have had only 15 relevant transactions over a twelve month period.
It seems obvious that a replacement is needed, but what should it be? There is a difference of view depending on which side of the Atlantic you sit. In the UK a reformed version of Sonia (Sterling Overnight Interest Average), which has been in use to some extent since 1997, is favoured. Similarly in Europe there has been a distinct shift from Euribor to Eonia as a benchmark to price transactions. The US on the other hand favours using a repurchase or “repo” rate as a benchmark. Banks often borrow money in return for lending US Treasury Bonds as security, and the repo rate (which will be termed the BTFR, or Broad Treasury Financing Rate) determines the difference in the price of the bonds between their sale and repurchase. Both benchmarks have their problems. The US measure is only measuring the cost of secured borrowing (and over relatively short periods), whereas Libor is an unsecured rate and provides a good indication of the creditworthiness of banks. Sonia on the other hand, while unsecured, is not widely used for longer maturities and, as with Libor, the problem of there not being sufficient bank-to-bank transactions could remain an issue.
These differences become very important when one considers the vast amount of outstanding transactions that exist. For example, many pension funds have attempted to match their long-term liabilities by entering into long-term interest rate swaps, many of which run way beyond 30 years, with almost all using Libor as a benchmark. All of these trades will need to be re-agreed between the two parties, one of whom is currently paying the rate set by reference to Libor and the other receiving it. This process will take a great deal of time. Of the $100 trillion of $ swaps outstanding, 65% are tied to Libor, with many having maturities beyond 30 years. If agreement can’t be reached over the repricing of these swaps using a different benchmark there may be some need for a provision to perpetuate the generation of Libor rates way beyond 2022. This may be easier said than done, with many banks wary of the reputational and financial risks of being accused of misrepresentation in making their submissions. Without a quorum of banks Libor will cease to exist.
None of this negates the arguments for disposing of an obsolete benchmark, and there are a number of positives about starting the process. The very act of doing so will focus banks on helping to make the new benchmark work and, while it may be difficult to reprice existing deals, a by-product may be a widespread termination of interest swap deals, helping to tidy up many bank balance sheets. Nevertheless, the FCA and other regulators should be careful what they wish for. Instead of having one, albeit highly flawed (especially in times of financial crisis) universal standard benchmark, we face a new system of benchmarks that vary according to currency and which either have no greater prospect of reliability in terms of gauging the temperature of the banking system, or struggle to be adapted for longer maturities. The sheer size and length of maturity of the products that use Libor means that it will probably be a feature of financial markets for much longer than the next five years and maybe that is not a bad thing. As the other great Sonia (the 1990’s Liverpudlian singing sensation) once trilled in the Eurovision Song Contest “better the devil you know”.
If you enjoyed this article please share it using the buttons above.
Please click here if you would like a weekly email on publication of the ShawSheet
