The Trouble With Fixing Libor

Peter W. Stevenson for The New York Times

Gary Gensler told the European Parliament this week that Libor deception had not stopped.

Sometimes modern finance has a great need for something, and so bankers invent products that appear to fill that need. When it turns out that the invention was actually something else entirely, people are shocked.

So it was a few years ago with senior tranches of asset-backed securities. Investors perceived a need for risk-free assets with floating rates, and Wall Street banks served up trillions of dollars worth of such paper - or at least they said they did.

So it is now with - the London interbank offered rate - which not coincidentally was an important component of that other folly. That there was fraud based on made-up numbers is clear. That the system can be fixed is not.

But Martin Wheatley, Britain's top financial regulator, has concluded Libor can be saved. “Although the current system is broken, it is not beyond repair,” he said in remarks prepared for delivery on Friday.

He may turn out to be overly optimistic. Libor is, and is likely to remain, a fiction. You can maintain the fiction, or you can embrace a much less palatable reality.

The Libor fiction began in the 1980s, when finance felt a need for a private sector, virtually risk-free interest rate to serve as a benchmark. Banks had learned that there was a big risk to making a long-term fixed-rate loan - the risk that market interest rates would rise and leave them with loans that were paying less than it was costing the bank to pay for the loan. Short-term loans could solve that problem, but at the risk that the borrower might be forced to repay at any time a loan that was taken out for a long-term project.

Enter Libor. A loan could be long term, but with a rate that periodically reset based on the cost of funds to banks. If a loan were priced at, say, three percentage points above the three-month Libor, the bank would be getting a reasonable risk premium, and would face no risk from changing market rates, since the interest rate would be reset every three months. The borrower would get long-term money.

There were two implicit assumptions in Libor. One was that banks were virtually risk-free, or at least that their risk was small and would not vary much over time. The other was that there was a way to actually calculate what the rate was. Both assumptions turned out to be wrong.

Libor rates are calculated each day by the British Bankers' Association, a trade group that makes good money from licensing the use of Libor rates. Each day panels of banks tell the association the rate they will have to pay for unsecured loans at maturities ranging from overnight to 12 months. They do that for each of 10 currencies, including the United States dollar, the euro, the Swedish krona and the New Zealand dollar.

The scandal made clear that those reports were faked before and during the financial crisis by at least some of the banks. But what is not as widely appreciated is that there is substantial evidence that the deception goes on. Banks continue to report figures that strain credulity, both in their level and in their lack of volatility from day to day or week to week. The scandal might never have surfaced, or might have done so in a sanitized fashion, had bank regulators had their way. But the banks had the bad fortune that the investigation of it was spearheaded by the United States Commodity Futures Trading Commission, a market regulator that under Gary S. Gensler, the chairman appointed by President Obama, has changed from lap dog to bulldog. It had no institutional need to protect the banks, and it did not.

This week Mr. Gensler, testifying before a European Parliament committee, laid out the evidence that the deception continues, although he was nice enough not to put it in such stark terms. He noted the wide swings in the cost of on debts issued by major banks, while those same banks were reporting that their costs of unsecured borrowing were varying hardly at all.

“It is critical that markets be able to rely on something that is credible and honest. The data in the market now strains that credibility,” Mr. Gensler said in an interview before Mr. Wheatley's conclusions were announced. “History shows that something that is prone to abuse will be abused, and that even people of good faith can have a difficult time estimating when there are no observable transactions.”