How to take the lies out of Libor

August 9, 2012

By George Hay and Agnes T Crane

The authors are Reuters Breakingviews columnists. The opinions expressed are their own.

Libor has become a dirty word. Attempts to manipulate the London Interbank Offered Rate benchmark have already claimed the scalps of three Barclays executives. With further fallout likely, regulators need to fast-track reforms.

Unfortunately, binning Libor isn’t an option. The rate is so embedded in the world’s financial architecture – the benchmark underpins more than $350 trillion of derivatives contracts and loans – that any attempt to rip it out risks creating mayhem. That means regulators have to work with what they’ve got.

Libor is set every day by a group of banks that publish rates at which they can borrow from their peers. The panels – administered by the British Bankers Association – submit a total of 150 rates in 10 currencies, with maturities ranging from overnight to one year. Thomson Reuters tabulates the submissions, lops off the top and the bottom quartiles, and averages them out to produce official Libor rates.

This approach has two problems, both of which were exploited by Barclays. The first is that banks submissions are a best guess, raising the risk that traders attempt to manipulate the rate. The second is that, during periods of market stress, banks have an incentive to submit artificially low rates in order to avoid signalling that they are in trouble.

Reform should start by ensuring that Libor submissions are based on actual transactions, rather than banks’ best guesses. That would make it harder for traders or executives to try and fiddle submissions.

But using actual data raises the question of what happens during a market panic. If banks aren’t able to borrow from each other – as happened during the credit crunch of 2008, and as is currently the case in parts of the euro zone – very few will be able to submit rates. This could be a problem even when markets are calm: there may not always be sufficient volume in, say, longer-dated Swedish kronor for rates to be truly representative. Besides, in recent years, a large chunk of the interbank market has shifted from unsecured to secured borrowing.

Yet this problem isn’t insuperable. Other indices, like Platts’ oil price benchmark, deal with temporarily weak volumes in less heavily-traded types of oil by restricting real-time data to the more heavily-traded ones. That’s similar to what the Federal Reserve Bank of New York suggested in 2008, when it recommended that the number of Libors be restricted to the most liquid maturities. According to market sources, many derivatives contracts are priced off three month Libor anyway.

Libor’s overseers would still face a problem if trading volumes collapsed even in these most liquid maturities. In that scenario, there might be a case for basing submissions on average borrowing rates from the previous week or two. Judging at which point Libor should be based on historic rather than real-time data would be a tricky and controversial call. But such a system would help to smooth out any hiccups from volatile market conditions, giving banks less scope to submit artificially low rates.

Another way of reducing the stigma involved in Libor would be to only publish the submissions used in calculating the average rate, but keeping the highest and lowest rates secret. That way, markets would not know if a bank that had been left out had submitted a high or a low rate.

One other key reform is oversight. The BBA’s self-regulation has been shown to be seriously flawed. An obvious remedy would be to involve the UK’s new anti-market abuse regulator, the Financial Conduct Authority. It could either take a seat on oversight committee charged with policing Libor submissions. Or it could elbow the industry aside and directly hold banks accountable for their submissions.

If a bank’s rates looked dodgy, the FCA could press for an explanation. It could also perform random audits by asking banks for data to back up their submitted rates. A repository for trades could also help to restore credibility by making it easier to double-check submissions.

Over time, the Libor headache may cure itself. Investors may migrate to other benchmark rates to underpin loans and derivatives. But it will take years before any alternative to gain traction. That makes fixing Libor a priority. FCA chief executive-designate Martin Wheatley will present his initial thoughts in a discussion paper due on Aug 10, while global bank regulators are due to debate the issue in September. They should not pull their punches.

Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/