Libor (London Interbank Offered Rate), one of the most famous acronyms in global finance, is a casualty of the current credit crisis. The credibility and reliability of this universal benchmark of short-term interest rates has come into question, with a potentially incalculable impact on the smooth functioning of markets and financial contracts.

Entrenched for over 20 years as the money-market measuring tool, it came under suspicion earlier this year, mainly for not being measured correctly. It now turns out that the problem goes much deeper, to the very heart of the credit crunch, involving the banks’ shortage of capital and their general unwillingness to lend to each other.

Background

The Libor system was developed in the 1980s and rapidly progressed to its current status as the basis for calculating payments on trillions of dollars of loans.

Libor, as the name suggests, is set daily in London. It covers 10 currencies, and ranges over 15 maturities, from overnight to 1 year. The most frequently used maturity is the 3-month rate as the benchmark for a vast number of short-term interest rate deals and contracts.

Every day, panels of leading banks submit their estimates of interest rates to Reuters. The latter calculate Libor on behalf of the responsible body, the British Bankers' Association (BBA). To avoid the distortion arising from manipulation by individual banks or from unusual circumstances, exceptionally high and low estimates provided by the banks are excluded. Reuters then calculate the Libor rates to be published.

Suspicion grows

Earlier this year, widespread suspicion arose that a number of banks had been understating the interest rates they were paying, particularly for 3-month money, and that official Libor rates had therefore been falling short of actual rates in the market place. At times the gap was believed to be as high as 0.3%, a figure that might seem tiny but is considerable as a proportion of short-term interest rates in the low single digits. While millions of borrowers have clearly been better off as a result of the understatement, this small difference translates into billions of dollars collectively for those on the wrong side of the contracts.

Though collusion between the banks was suggested, a less blameworthy reason is more likely. Individual banks were apparently scared of being seen to be paying too much for their loans, fearing that this might spark off questions about their creditworthiness and lead to bigger problems, such as those suffered by Northern Rock and Bear Stearns.

As the suspicions spread, the BBA, working closely with the Bank of England, started investigating and, lo and behold!, the published Libor rate rose, which tended to confirm the suspicion that previous underreporting was being brought to an end for fear of getting caught.

Procedures under fire

The above problem brought to light several deficiencies in the procedures used to calculate Libor, in particular the dollar rates.

A problem was that Libor was set at 11.30am London time, well before New York opened for business. Moreover, of the 16 banks contributing to Libor’s calculation with their estimates, only 3 were based in the US.

To counter this, the large inter-dealer broker ICAP established the New York Funding Rate (NYFR) in early June, for 1-month and 3-month dollar interest rates. Interestingly, NYFR has tracked Libor quite closely since its introduction. In late June, 3-month NYFR was 2.799 whereas Libor was 2.804, indicating that the Libor measurement problem had at least temporarily subsided.

This confluence of rates has taken the heat away from the calculation of Libor, a development welcome to policy-makers worried about Libor’s credibility undermining confidence in money markets generally.

However, the procedural deficiencies in calculating Libor persist and the BBA is undertaking a review of how to improve the situation.

Credit crisis barometer

A much bigger problem turns out to be that Libor, even when measured correctly, was much too high and no longer representative of money market rates as it used to be. The gap between overnight interest rates set by central banks and 3-month Libor, as measured by the co-called ‘Overnight Index Swap’ (OIS), is now in the region of 60-70 basis points in the UK, US and Eurozone, whereas prior to the summer of 2007 the figure used to be about 15bp and averaged 11bp in the ten years prior to August 2007.

That the banks are scared to lend to each other is a widely held belief of this crisis. Apparently it is wrong. Interbank lending rates have deviated from overnight rates not because banks were unwilling to lend to each other, but because money-market funds, with $4trillion of assets, have not provided banks with cash as they used to before the summer of 2007.

Rather than banks not trusting each other, it is a question of money market funds not trusting the banks. Banks do not often lend to each other, except for the very short term, but have relied instead for 3-month-plus funding on the money markets. Given that they are currently deprived of the latter, they are themselves short of 3-month money, let alone having enough to lend to other banks.

The elevated level of Libor reflects the fact that the banks are short of capital to lend, particularly for the 3-month stretch, and Libor therefore represents a front-line symptom of the banks’ capital shortage. The gap between Libor and overnight rates is expected to subside to more normal levels only when the banks’ capital stock is restored to adequate levels. Thus, Libor remains a barometer of the financial system’s health.

Money market funds, on the other hand, are still playing safe. If their risk-aversion decreases, and they resume lending to the banks, then the latter’s liquidity problems can be alleviated, hastening the end of the credit crunch. So, perhaps money funds hold the key to the banks becoming more active lenders again.

While there are signs that money market funds’ risk aversion is decreasing somewhat, it is still nowhere near enough. The currently excessive deviation of Libor from overnight rates is symptomatic of the banks’ problems and can end if the money-market funds once again lend to banks.

Alan Greenspan, the former Fed chairman, said in June that this spread should serve as a measure for identifying when markets return to normal. He reckoned that a narrowing to 25bp would be viewed as positive, but that the forward markets were signalling that this would not happen until after June 2010.

Alternatives to Libor

Understandably, the continuing use of Libor is now being challenged and alternatives are being put forward. Changes are proposed on two fronts. Firstly, whether Libor itself should be used as the indicator of short-term interest rates and secondly whether the 3-month rate should be regarded as the bellwether for calculating the general level of short-term interest rates.

David Rule, the chief executive of the International Securities Lending Association (ISLA), has argued, both in the Financial World (February 2008) and in the Financial Times (10 April 2008), that 3-month Libor should give way to a measure based on overnight interest rate derivatives.

Reflecting this changed situation, LIFFE, the futures exchange, has got into the act of trying to supplant Libor. They have introduced two new futures contracts. SONIA (Sterling Overnight Interbank Average) is meant to represent the average for sterling overnight transactions between selected counterparties. The second contract, EONIA (European Overnight Interbank Average), is the corresponding tool for the Eurozone.

However, these two contracts are far from having widespread acceptance, as whether they are sufficiently liquid is still to be demonstrated. There are also copyright arguments stemming from the Wholesale Market Bankers’ Association (WMBA), who claim copyright for the idea.

Aftermath

Libor is something all financial market participants take for granted and its solidity as a benchmark has not been doubted on this scale since its inception. If one owns a brand new Mercedes Benz, it is taken for granted that it will run smoothly and well unless an extremely unlikely event occurs. One doesn’t worry about what’s under the bonnet. The same applies to the hitherto universally accepted Libor benchmark. Very few people knew how it was put together and calculated, and hardly anybody was bothered about it.

It is now, however, a different story. The way it is put together and calculated has been dissected in the financial media and things are unlikely to ever be the same again. Shortcomings in the calculation of Libor are now widely known and it will be impossible to put the genie back in the bottle in terms of it recovering its previous credibility and acceptance.

Furthermore, when it was first introduced in 1985, it did give an accurate reflection of interbank lending rates, more so than the overnight rate. The tables have now been turned and the latter has become the better yardstick.

According to the BBA, over $350trillion worth of financial contracts are linked to Libor. This is a huge problem because Libor’s suitability as a measure of money market rates can no longer be taken for granted.

Conclusion

The alternatives to Libor suggested above are unlikely to make much headway in the near future, as the widespread consensus is that too many existing arrangements are based on Libor.

While it might be true to say that it would be nigh impossible to switch from Libor because of the huge mass of existing contracts, at least those entering new contracts might take more care to define exactly what they mean by short-term interest rates, rather than automatically assuming that it should be 3-month Libor.

A greater use of overnight rates could be on the cards, but a challenge to the use of Libor itself would be more difficult. Who knows whether the alternatives could be constructed in any better way. It would seem more sensible to stick to the Libor system, but with much greater supervisory vigilance than before.

In late August, the Libor-OIS dollar spread was 77 basis points and clearly some way away from normal levels. At some point, the gap between Libor and overnight rates will revert to more normal levels, but it is now clearly established that Libor cannot ever again be regarded as a risk-free rate, an appellation that was never justified anyway. The idea of a risk-free rate in the money market is perhaps just an abstraction, and all investors are faced with the problem, firstly, of choosing a benchmark and, then, measuring the deviation from this benchmark of their money market return.

How does the Libor saga affect fund management? Many performance-based fees are linked to exceeding Libor, based on the natural view that something close to Libor was actually obtainable. This might come under greater scrutiny, and those marketing absolute return funds, as well as hedge funds and their clients and customers, will need to pay close attention to this issue. Performance attribution analysis will also need to take into account divergence between the benchmark used and the rates actually obtainable.