The LIBOR Mess: How Did It Happen -- and What Lies Ahead?Published: July 18, 2012 in Knowledge@Wharton
When regulators in the United Kingdom and United States announced a settlement with Barclays bank over its manipulation of LIBOR, the benchmark interest rate used around the world, there were plenty of reasons for jaws to drop. First and foremost was the whopping fine of $450 million, reflecting the seriousness of the case, along with analysts' predictions that LIBOR rates could influence interest rates on between $350 trillion and $800 trillion in loans and investments. That's not a mistake -- trillion, with a "T".
Now it looks as if many major banks may have been involved, with about a dozen of the biggest names in the world under investigation for rate fixing intended either to pad profits or to make themselves look financially healthier than they were. Regulators may have glanced the other way. Civil suits from investors, pension funds government entities and others dependent on LIBOR may eventually cost big banks billions in damages.
"What interests me the most is that it shows once again how unexpected failure in the system can pop out of almost anywhere and become so major," says Wharton finance professor Itay Goldstein.
How did this mess happen, who was hurt and what's to be done about it?
'Shocked' at the News
Among the surprising aspects of the story: Many thought this issue had been addressed years ago. The first public hints of trouble came in 2007 and 2008, after which the British Bankers Association (BBA), which calculates LIBOR rates every day based on submissions from participating banks, took steps to stop the fudging of these numbers. It turned out those measures were not at all ironclad.
"It's remarkable that people are surprised LIBOR could be manipulated," notes Wharton finance professor Richard J. Herring, comparing the recent reaction to the police official who was "shocked!" to learn about gambling at Rick's place in the film Casablanca. "It's been more or less an open suspicion since at least 2007," Herring says, adding that "the rate after the crisis was determined by the British Bankers Association according to a slightly modified approach which was intended to be more transparent, more broadly representative and harder to manipulate."
Twenty of the world's largest banks submit their interest rate data to the BBA at 11 a.m. each day. While the shorthand term LIBOR, for London Interbank Offered Rate, makes it sound like one rate, there actually are calculations in 10 currencies for 15 loan terms ranging from overnight to 12 months.
The procedure for setting each benchmark is the same. "The BBA then throws out the top 25% of quotes and the bottom 25%, and takes the average of the remainder," Herring says. "On the surface, this makes it seem highly unlikely that any one bank could manipulate the rate, but on closer inspection, it is possible."
If a bank's true rate was actually in the top quarter of all those submitted, but the bank reported a figure in the bottom quarter, another bank's rate might be lifted from the bottom into one of the middle quarters, according to Herring. Thus, a low rate that should have been thrown out would end up being used in the calculation. "The impact on the quoted LIBOR could be rather substantial." If numerous banks understate their rates, the average is sure to be lower than it should be.
"The fundamental problem," Herring adds, "is that LIBOR is a hypothetical rate -- the rate at which each of the 20 banks on the panel believe they could borrow funds at 11:00 a.m. It is not a transaction rate, and although it is possible to see what each of the banks has quoted, it is not possible to verify the quoted LIBOR rate contributed by each bank against an actual transaction."
At its heart, it is an honor system.
Why did the BBA changes not discourage underreporting? Because the banks felt they were in a corner, says Wharton finance professor Krista Schwarz. "Especially in 2008, the biggest problem was that all the banks wanted to claim they were able to borrow more cheaply than was in fact the case, so as not to heighten concerns about their creditworthiness."
But if LIBOR's rates can be so easily manipulated, why use them? After all, the financial world offers a cornucopia of rates that are set by the marketplace and are therefore virtually impossible to rig. Rates on government and corporate bonds, for example, are governed by supply and demand and can be tracked through trading data, while LIBOR rates cannot.
"It's surprising that LIBOR has become so important, because of the imprecision and the lack of verification" by regulators of the rates that banks report, states Wharton finance professor Jeremy Siegel. "To some degree, it's just convention," adds Mark Zandi, chief economist of Moody's Analytics, describing LIBOR as "an historical artifact." Banks continue to use it because they have done so in the past.
In theory, LIBOR reflects what banks expect to pay to borrow every day in deals with one another. "They look at the rate they expect to borrow at rather than the actual rates" they do borrow at, according to Goldstein. This system is intended to reveal the banks' real cost of money, incorporating all the market's up-to-the-minute assessments of the risk of lending to the participating banks. In contrast, a U.S. Treasury bond rate, while set more transparently, does not include default risk. Investors believe the U.S. government will pay its debts, while they cannot be so sure a bank will, so the two rates reflect a different set of concerns.
There have not been many problems with LIBOR prior to the financial crisis. Therefore, there was little reason to question its use, which has broadened because of London's dominance in the world financial markets, a certain pack mentality and tradition: Since many institutions use it, many others follow suit.
"The LIBOR rate affects a large number of end borrowers whose loan terms are variable, which includes both corporations and municipalities with floating rate debt and, especially, households with adjustable rate mortgages, student loans and some credit cards," says Wharton finance professor Nikolai Roussanov. In fact, about half of adjustable-rate mortgages in the U.S. track LIBOR, he adds. In a typical case, the mortgage is recalculated every 12 months by adding a fixed number of percentage points to the LIBOR rate that day. The higher the rate, the larger the borrower's monthly payment, and vice versa.
While an artificially low rate benefits borrowers by reducing monthly payments, that boon was probably small, he notes, since instances of understatements apparently did not last very long.
Minuscule Change, Major Gain
More important than consumer loans is the use of LIBOR for fixed-income derivative contracts, Roussanov says, adding that the most common are interest rate swaps, in which the party on one side makes regular fixed payments to the other party, while the second party makes a floating payment to the first based on LIBOR. Swaps have various uses, such as managing the investor's exposure to changes in interest rates. With swaps, each side therefore makes a bet based on the relationship of fixed and floating rates. If LIBOR rates are not what they ought to be, some traders will make more than they should, others less. At the end of 2011, the value of outstanding swaps contracts was about $18 trillion, according to Roussanov.
Government investigators are also looking into evidence that some banks tinkered with LIBOR to boost gains on specific contracts, as a minuscule change in LIBOR could be worth millions of dollars on a large position.
Most of the time, LIBOR rates move in tandem with other rates that are easier to verify, indicating LIBOR does indeed accurately reflect the rise and fall of prevailing rates. But occasionally these relationships change, causing raised eyebrows. Beginning in 2007, the gap between LIBOR rates and market-set rates like the Fed Funds rate began to widen, an alarming development that worsened in 2008, Siegel says. "I was appalled by how much that rate went up, hurting borrowers."
The LIBOR rate spiked after the Lehman Brothers bankruptcy in 2007, reflecting worries about the banks' financial health. They were wary about lending to one another, and the volume of lending plummeted, Siegel notes. Then the process reversed, and LIBOR fell to a full percentage point below that on a comparable U.S. Treasury bill, reversing the usual relationship.
"Analysts found this highly suspicious," Herring recalls. While the comparatively low LIBOR rate indicated lenders thought the banks were relatively safe bets, other rates governed by the marketplace, such as the Overnight Index Swap Rate, showed just the opposite. Amid the worst financial crisis since the Depression, with the credit markets all but frozen, it was impossible to believe the banks were lending to one another at such reasonable rates. "This may be another good example of Goodhart's Law: Whenever you focus on a rate for policy purposes, you will set up incentives to manipulate the rate," Herring notes.
After Lehman, investigators began digging into the LIBOR-setting system. "The suspicion was that several banks that were having trouble funding themselves in the market tried to disguise their distress by quoting much lower rates than they would be obliged to pay -- if they could borrow at all," Herring says.
Aside from this image polishing, investigators began to suspect other motives, he adds. "While [looking into] allegations that banks were understating their rates, officials uncovered emails suggesting that at least one bank attempted to manipulate LIBOR to increase its profits at rollover dates, in settlement of Eurodollar futures transactions and swaps based on LIBOR." Goldstein points out that "banks have all sorts of contracts that depend on LIBOR, and decreasing the rate can increase their profit." Manipulating the rate to boost earnings is fraud, he says.
Looking the Other Way
The recent spate of news suggests that a number of the world's largest banks were doctoring the LIBOR rates, and that regulators had some knowledge of what was going on but did little, perhaps because they, too, wanted to shore up confidence in the banks amid the financial crisis. "There's a kind of weird irony in all this," says Zandi, noting that it truly was important to boost confidence in the banks during the crisis. "This is really scandalous, but, on the other hand, [the rate trimming] may have helped forestall a much worse crisis."
"I can see no reason [why] they would have condoned manipulation of the rate for the profit of particular banks in particular transactions," Herring says of regulators. "Their concern about overall confidence in the banking system in the depth of the crisis, however, makes it plausible that regulators would have been pleased to see LIBOR decline and perhaps [were] not tempted to ask very probing questions."
Schwarz adds that regulators had a lot of other things going on at the time and "probably did not think that this was the most pressing problem. Also, regulators had no easy tools to address this. It is, in particular, hard to see what U.S. regulators could have done in real time to deal with a rate produced in London."
This is not, however, a case of no harm, no foul. Borrowers such as homeowners would have benefitted if an artificially low LIBOR kept their payments below what they should have been. But that would have occurred at the expense of lenders and their shareholders. "If they charge the wrong rate, it's just not fair to one party or the other," says Siegel.
Because changes in LIBOR also govern returns on various types of investments, it is clear that pension funds, mutual funds, municipalities and other investors are likely to have been hurt as well. A string of lawsuits has been filed by investors and, though the cases may take years to resolve, some analysts predict the major banks will pay out billions in damages, which will not please their shareholders.
Another casualty, says Herring, is public confidence in large banks. The industry has been trying to stave off heavier regulation following the financial crisis, and public distrust could make that harder. In addition to the LIBOR scandal, the public has been bombarded with news about the multi-billion dollar trading losses at JPMorgan Chase, as well as allegations that the bank's financial advisors improperly steered investors to house-brand mutual funds that performed worse than competitors' offerings. To bank critics, the LIBOR case is just more evidence the banks cannot be trusted and need tighter oversight and restriction.
LIBOR could even become a factor in the U.S. presidential campaign, as Republican Mitt Romney wants to repeal the Dodd-Frank reform laws passed, with President Obama's backing, after the crisis.
In fact, there is little to nothing in Dodd-Frank that applies to the LIBOR case, says Schwarz, noting that she expects the LIBOR scandal to strengthen the hand of those seeking tougher regulations. "It seems to be having a stronger effect along those lines in the U.K. than in the U.S.," she says. "Since LIBOR is produced in London, it is U.K. regulation that matters the most. For now, U.K. regulators have no direct supervisory authority over LIBOR, but I expect that will change."
Regulating financial markets is often a tightrope walk, adds Goldstein. "When bankers don't face constraints or regulation, then they do whatever serves their interests. If you want them not to do [something] and not to put the system at risk, then you need to put in more regulations and constraints.... But you don't want to go to the other extreme, where government controls everything."
Choosing Another Benchmark
Going forward, a key question is whether LIBOR should be replaced with another benchmark less susceptible to manipulation. But LIBOR is so embedded in the world's financial system it would be impossible to eliminate its use overnight. In addition, says Herring, LIBOR is unique in providing a very wide variety of terms, from overnight to one year.
Herring notes that rate setters could ask banks what rates they would be willing to lend at, rather than what they think they could borrow at. "This may reduce the incentives for understating rates." Another alternative, he adds, would be to use actual transaction rates, such as those on the Overnight Index Swap Rate, the US Treasury bill rate, or something else. These would be harder to manipulate, but currently do not come in as wide a variety as do LIBOR rates.
"Given this [scandal], I think we should be rethinking how all these debt instruments are priced," says Zandi. "Why not price off something like the Federal Funds rate, or the interest rate on reserves, or something we know for sure is accurate?"
Schwarz, who once worked on the Federal Reserve's money market desk, believes there are options if players conclude LIBOR is too tainted to redeem. "The obvious thing is to use a rate that is based on actual transactions," she says. "That's already done for the effective Federal Funds rate published by the Federal Reserve Bank of New York. It's even done for eurodollar transactions based in New York, called NYFR. I can't see any reason why this could not be set up fairly quickly."