July 20, 2012
Libor: Three Scandals in One—A Commentary by Jonathan Macey ’82
The following commentary was posted on ForeignAffairs.com on July 20, 2012.
Libor: Three Scandals in One
By Jonathan Macey ’82
Governments in Europe, Japan, and the United States are now investigating 16 major banks for manipulating interest rates. In the months leading up to the 2008 financial crisis, many of the world's most powerful financial institutions allegedly worked to keep down the London Interbank Offered Rate. Libor, as it is known, is supposed to be the average rate at which the largest and ostensibly safest banks in the world can borrow from one another. Manipulating Libor allowed traders to rig financial markets to their advantage; in the process, they distorted the actual value of key financial instruments such as credit default swaps, derivatives, and home mortgages.
The scandal has sparked calls from politicians, including Mervyn King, the governor of the Bank of England, for stronger regulation of the world's most powerful banks. But such proposals miss a key point: Price fixing and manipulation are illegal. They have been for a long time. So it is unlikely that saddling financial markets with legal constraints that simply double down on what is already on the books will help. A better solution would go to the heart of the problem. Regulators and market participants should set such benchmark interest rates as Libor in a way that makes them reflect movements in the market, making manipulation impossible.
The fundamental principle underlying floating rates is to allow the market to determine borrowing costs. Customers who borrow on a floating-rate basis, if they are sensible, and institutions that loan money on a floating-rate basis, if they are ethical, therefore expect two things from a benchmark interest rate. First, the benchmark should reflect actual conditions in the financial markets. That means no random fluctuations -- money costs what it is worth. Second, the benchmark rate should not be easy to manipulate. No rational, informed borrower would borrow money at a variable rate of interest and then empower the lender to determine when and how the interest rate changed in the future.
So it is startling that Libor, the financial world's most important number, satisfies neither of these requirements. Libor is computed by the British Bankers' Association (BBA), a powerful trade association based in London that represents more than 250 financial institutions. These banks are located in 50 countries and have operations in just about every corner of the globe. But instead of using actual market rates, big banks estimate the interest rate that they think they would have to pay if they borrowed money from other institutions. That is different than reporting the actual interest rate at which they are really borrowing from other banks.
Each day, the BBA sets Libor rates for 15 loan maturities in ten different currencies. In the case of the dollar, 18 banks submit their hypothetical borrowing costs. The BBA discards the four highest and the four lowest submissions, which is supposed to prevent a small number of anomalous results from distorting the calculation, and then averages the remaining ten to come up with the Libor number. Thompson Reuters calculates all of these rates for the BBA, and then publishes the results, usually around 11:45 AM (CET).
Since the BBA launched Libor in 1986, trillions of dollars' worth of credit default swaps, futures, and other securities have been bought and sold at prices linked to Libor. The rate is a key reference point for the International Swaps and Derivatives Association, the trade organization that creates standards for the derivatives market. Today, the prices of $350 trillion in financial contracts created by the ISDA are tied to Libor rates. So even a miniscule manipulation undermines the integrity of financial markets and distorts the allocation of credit and capital to borrowers and companies.
There are plenty of reasons why banks would like to manipulate Libor rates. During the height of the financial crisis, regulators foolishly looked to Libor to determine the market's perception of the health of big banks. A big bank reporting a low Libor rate was thought to be able to borrow money from other banks cheaply and, therefore, was seen as a safe place to invest. Banks worried about attracting the attention of regulators may have submitted low Libor rates in order to try to deflect regulatory scrutiny.
More nefariously, banks also manipulated Libor to make money or avoid losses on their trading portfolios. For example, when U.S. traders at Barclays wanted Libor to rise in order to draw a bigger profit on some of their financial products, they simply asked their colleagues at the rate-setting desk in London to push the numbers up or down to suit their needs. Barclays would then submit artificial bids and persuade their counterparts at other banks to do the same.
In legal papers filed in connection with its recent enforcement proceedings against Barclays, the U.S. Commodities Futures Trading Commission (CFTC) quotes an internal Barclays e-mail sent on February 3, 2006, in which a trader wrote to the rate-setting desk: "Would love to get a low 1 month (LIBOR rate), also, if possible, a high 3 month." Then on March 27, 2006, the trading desk told London, "We need low 1m (month) and 3m LIBOR." In an e-mail that May, a trader asked that Libor be kept at 5.36 because his trading position "would be hurt by a higher rate."
But even in the absence of manipulation, Libor is a poor market metric. Other benchmark interest rates, including the actual rates paid on U.S. government Treasury bills, notes, and bonds, make more sense. Consider the findings by the research department of the Federal Reserve Bank of Cleveland, which reported that in 2007 Libor rates and U.S. Treasury rates began to diverge sharply. While U.S. Treasury rates declined during the financial crisis, as investors sought the safety of government bonds, Libor rose, as investors considered big banks highly risky. Lucky borrowers whose variable interest rates were tied to Treasuries saw their monthly mortgage payments go down, but borrowers whose variable rates were tied to Libor saw their monthly payments soar. As it turns out, beginning in 2005, many banks were steering borrowers into Libor-based loan agreements and away from those based on U.S. Treasury securities. By 2007, according to the Cleveland Federal Reserve, more than 90 percent of subprime adjustable-rate mortgages were linked to Libor. Until everything fell apart, this shift was great for the banks, of course, but ultimately crushed many borrowers and helped sink the global economy.
So the Libor scandal is actually three scandals in one. The first is the way banks reported artificially low Libor rates to regulators. The second is the manipulation of Libor for trading profits. And the third is the use of Libor to determine home-mortgage interest rates when a better metric, U.S. Treasuries, was readily available.
The tragedy here is that all of these problems could have been avoided by choosing a more transparent and objective measurement. Such a system would not have to be managed or administered by any body. For starters, rather than using banks' hypothetical, imaginary prognostications about interest rates, the market could use actual interest rates on observable financial instruments such as U.S. Treasuries and other sovereign debt that trades in highly liquid markets. In other words, do not use the rate at which bankers imagine they can borrow money to determine Libor. Use the rate at which a big borrower, such as the U.S. government, borrows money.
For example, instead of making a floating-rate loan that is three percentage points above Libor, a bank could make a loan that is three percentage points above the interest rate on the one-year U.S. Treasury bill. Such a change would be extremely simple to implement. Until the mid-2000s, almost half of floating-rate home mortgages were priced this way. And to virtually eliminate the possibility of manipulation, one could go a step further. Rather than use a single, isolated rate, one could set the floating rate at the rolling average of the interest rates on a government security over, say, a one-month or one-year period.
In the world of financial contracts, rolling averages are generally considered the best reference point for adjustable rates. From the banks' point of view, Libor's advantage was the fact that it could be manipulated. But unless bankers want to hand over even more control to government regulators, they will need to restore integrity to and trust in the process. They need to drop Libor as the benchmark and replace it with an actual market rate.
If the bankers do not get their act together, one of two things will happen: The market will fragment, as individual banks develop their own benchmarks, and consumers will suffer because it will become virtually impossible to compare and shop for loans from competing banks. Or government regulators will swoop in. Then the benchmark rate will become bureaucratized and ossified -- and consumers will suffer even more. Either way, the banks, without customers, would lose. Which is all the more reason to take care of the problem themselves, and now.