2012-07-17 07:14:37
LIBOR is badly broken. But for now, a flawed number is better than none
Jul 14th 2012 | from the print edition
THE furore over alleged manipulation of the London Interbank Offered Rate
(LIBOR) and its European cousin, the Euro Interbank Offered Rate (EURIBOR),
continues to rage. In Britain, the deputy governor of the Bank of England and
the chairman of Barclays were hauled over the coals this week by a
parliamentary committee. In America, it emerged that the Federal Reserve Bank
of New York may have been informed of alleged manipulation of LIBOR some time
after 2007; the Senate Banking Committee plans to look into the affair.
Yet all the while the basic mechanism of LIBOR trundles on. Each morning at
11am in London, submitters at panels of some of the world s biggest banks send
their estimates of borrowing costs in various currencies and for various terms.
A few minutes later the benchmark figures flash to life on tens of thousands of
traders machines around the world, and ripple out into the pricing of loans,
derivatives and other financial instruments.
Be generous, and assume that attempts to manipulate LIBOR are in the past. A
deeper problem still besets these numbers: they are almost entirely fanciful
even if the banks that submit them are providing honest estimates. That is
because the unsecured interbank funding market, which is supposed to be where
banks borrow from each other, is frozen solid. In the euro area in particular,
banks are lending almost no money to one another. Most banks that now have cash
prefer to deposit funds at the European Central Bank (ECB), which in turn lends
it on to those that are short of it. At the moment banks have more than 800
billion ($980 billion) parked at the ECB, where it earns no interest. LIBOR and
EURIBOR measure an activity that barely exists.
Even if markets were functioning properly, some of the banks submitting
estimates would struggle to borrow at any interest rate, let alone the one they
have been submitting. This problem is starkest for EURIBOR, where individual
banks have been submitting rates that are likely to be a good deal lower than
the rates they would have to pay in actual transactions. The biggest banks in
Italy and Spain generally estimate the cost of borrowing euros for a year at
about 1.1%. This rate is much lower than the 4% and 5% their governments (and
ultimate guarantors) pay to borrow for the same period.
There is nothing necessarily untoward in this. Unlike LIBOR submitters, EURIBOR
banks are not asked to provide estimates of what they think they would have to
pay to borrow, merely estimates of what the borrowing rate between two prime
banks should be. Yet the definition of prime is essentially now German ,
leading to a widening disconnect between the actual costs of bank borrowing
across most of Europe and the benchmark rates that supposedly reflect them.
The reference to EURIBOR is completely useless for Italian banks, says
Giovanni Sabatini, the managing director of the Italian Banking Association.
EURIBOR is less than 1% and our banks are paying 350-400 basis points above
EURIBOR.
Attempts to improve LIBOR and its equivalents bring fresh problems. Regulators,
politicians and industry groups are now poring over ways to improve the
calculation of such rates (see Free Exchange). These could include using larger
panels of banks and forcing banks to report actual transactions. American
authorities have told Barclays to adopt strict governance and reporting rules
to ensure its submissions are honest. In isolation each of these changes seems
perfectly sensible. Yet in aggregate they pose two big risks.
The first is that banks may simply stop contributing LIBOR estimates to
minimise the risks of being prosecuted or sued. Paul Tucker, the Bank of
England official up before MPs this week, said contingency planning has already
started to deal with this risk. He also raised the possibility that civil
lawsuits against banks might be so large as to undermine financial stability.
A second risk is that changes to the method of calculating LIBOR could lead to
very different numbers being generated. Much bigger panels would include banks
that are smaller and less creditworthy than those currently submitting, leading
to higher LIBOR rates. Since these numbers are hard-wired into tens of
thousands of derivatives contracts and loan agreements, the losers would almost
certainly dispute the changes. If LIBOR 2 is going to give an answer that is
structurally higher than the old LIBOR, then it will be worth litigating over,
says one lawyer. Each basis point might not seem much, but multiplied across
the billions in contracts it soon adds up to a pretty big number.
from the print edition | Finance and economics