Bankers' trust

Economics focus

Bankers' trust

Apr 24th 2008
From The Economist print edition

Why have banks been paying so much in the interbank market?

SINCE last August, some of the world's most powerful central bankers have battled with growing resourcefulness to restore the law of gravity to the market that banks use for short-term borrowing. As the credit crisis has deepened, the central banks have made more money available against a broader range of collateral for longer periods to a wider group of financial firms. By throwing money at the situation, they have aimed to lower the London Interbank-Offered Rate (LIBOR) that banks charge each other for anything up to three-month loans (it helps determine borrowing rates for firms and households, too). But the rate remains stubbornly high; it has jumped again recently on reports that the British Bankers Association, which sources quotes each day from a panel of banks, is investigating whether banks have been reporting lower rates than they are actually paying in order to appear healthier than they are.

Before the crisis broke, typically banks would provide unsecured three-month loans to each other at rates that were barely higher than their cost of borrowing from the central bank. But last August, the gap between LIBOR and the overnight indexed swap (OIS) rate (a gauge of expected central-bank rates over the same period) widened sharply in Europe and America (see left-hand chart). At times, the gap was almost as wide as during the Y2K fears at the turn of the century. According to a recent paper* published by the Bank for International Settlements, the higher risk premium reflected in the rise of LIBOR over the OIS rate responds to several factors, particularly credit and liquidity risk. The first points to the bigger chance that a bank will go bust over a three-month period than overnight. As for liquidity, a three-month loan can less readily be exchanged for cash than an overnight one and lenders require a reward for that risk.

Disentangling the two is tricky, but the paper notes that the Y2K gap was driven more by liquidity concerns than by credit ones. Using the spread between secured and unsecured interbank rates as a gauge of credit risk, it believes that in the second half of 2007, credit concerns may have played a significantly larger role than in late 1999. The Bank of England, which has also sought to disaggregate the interbank spreads, said in February that most of the more recent pick-up appeared to be due to worries about credit risk.

In another recent paper† John Taylor of Stanford University and John Williams of the Federal Reserve Bank of San Francisco find support for this thesis—that the higher cost of interbank borrowing is due to banks' fears that their rivals will go bust, leaving any unsecured debt unpaid. They argue that the Federal Reserve's Term Auction Facility, introduced in December to increase the supply of term (28-day) loans, has not helped much, because spreads are driven primarily by such fears.

Another way of looking at credit risk is by comparing the LIBOR spread with the premiums charged on banks' credit-default swaps (CDSs), which measure the risk of default. The latter have fallen recently (see right-hand chart), which appears to suggest that credit fears are easing in the banking industry. Why, then, do interbank rates remain so high?

Perhaps, argues Tim Bond, a market strategist at Barclays Capital, the falling cost of CDSs is a harbinger of things to come. He notes that interbank illiquidity reflects not so much banks' mistrust of one another as a loss of confidence in the banking market as a whole by those, such as money-market funds, which normally supply it with cash.

He finds that these funds have shunned banks' short-term offerings, fearing that their money might not be paid back. That has forced banks to scramble for extra cash in the interbank market. The banks' funding shortfall has been made more acute by the seizing-up of trading in asset-backed securities, closing off another vital cash-raising route.

Mr Bond reckons that in America, the combination of a rapid increase in commercial banks' assets and slow growth in customer deposits left a funding gap of more than $660 billion in the year to February. Although money-market funds had enough fresh investment to fill this gap, they chose to put most of their new money into safer government bills. Their risk aversion is not without foundation, as the funds suffered badly when they lent to bank-sponsored entities that invested in ropy mortgage securities. High LIBOR rates will persist, says Mr Bond, until money-market investors are lured back into lending to banks.

Disaster averted

That has not happened yet, but Mr Bond sees hopeful signs of improving confidence. The decline in CDS premia, he says, may reflect the commitments made by central banks, led by the Federal Reserve, to ensure that a solvent bank will not fail for the want of enough cash to meet its immediate needs.

That effort was bolstered on April 21st when the Bank of England unveiled its own ambitious plan to tackle the problem: the Special Liquidity Scheme. Under this measure banks can—for a market-determined fee—swap high-quality asset-backed securities for a smaller, but far more liquid, supply of nine-month treasury bills. The bank's governor, Mervyn King, said the scheme was designed to end once and for all the worry that a solvent bank might go under because it could not gain access to emergency lending.

At the very least, that should help reduce the illiquidity risk in the interbank spread and go some way to soothing solvency fears. But central banks can only do so much. Banks could also help solve the problem by revealing the full extent of their exposure to distressed mortgages, and by raising enough capital to offset those losses. Perhaps the CDS market is saying that they are already heading in the right direction.

* “What Drives Interbank Rates? Evidence from the LIBOR Panel”, BIS Quarterly Review, March 2008.