Finance & Investing

Darrell Duffie: Why LIBOR Manipulation Poses an Ongoing Risk

A finance professor explains what it will take to fix the problem.

September 10, 2014

| by Edmund L. Andrews
Men walking outside a bank

The London Interbank Offered Rate has become an international benchmark for trillions of dollars of debt and hundreds of trillions of derivatives. | Reuters/Toby Melville

Two years ago, many of the world’s biggest banks were caught systematically manipulating the so-called LIBOR interest rate. A normally placid fixture in the financial system, LIBOR is a benchmark that is used to determine the rates on trillions of dollars in adjustable-rate debt. Darrell Duffie of Stanford Graduate School of Business and Jeremy Stein of Harvard Business School helped craft recommendations for the Financial Stability Board, an organization of international banking regulators. Here, Duffie explains why LIBOR is important, why manipulation poses an ongoing risk, and what it will take to fix the problem.

What is LIBOR, and why is it so important?

LIBOR is short for the “London Interbank Offered Rate,” an estimate of the interest rate at which large banks can borrow money from each other.

It’s almost the result of an accident of history, but LIBOR has become an international benchmark for trillions of dollars in adjustable-rate mortgages and other floating-rate debt, as well as hundreds of trillions of derivatives. If you have an adjustable-rate mortgage, your rate is probably pegged to LIBOR. Borrowers, lenders, and investors all need an agreed-upon settlement index – a “reference rate” – for determining their interest rate payment. Without that benchmark, it would be hard to know if the bank was charging a competitive or fair rate at any particular time.

I often compare LIBOR to a public utility that provides reliable tap water or electricity. People may not think about it, but it’s a service that is always running in the background and allows many other things to function smoothly. Clearly we have a problem if LIBOR doesn’t work.

Why were banks manipulating LIBOR?

LIBOR is measured by asking banks to report the rates at which they can borrow. To some extent, we have taken them at their word. Making matters more difficult, there are surprisingly few actual interbank loans of the kind that are used to set LIBOR each day. Even if a bank wants to tell the truth, on many days it must estimate its daily rate because it hasn’t done any LIBOR-type transactions that day.

There were two basic reasons why banks manipulated their reported rates. The first was “low-balling,” which became very significant during the financial crisis. When banks had problems with their creditworthiness, they didn’t want to advertise that by admitting that they could only borrow at higher rates than their competitors. So all the banks began claiming that they could borrow at virtually the same low interest rates. It was like a financial Lake Wobegon: Every bank’s creditworthiness was above average.

The second motivation for manipulation came from bank trading desks, which had‚ and still have huge positions in derivatives that were greatly affected by tiny changes in LIBOR. Pushing LIBOR up or down by just one one-hundredth of a percentage point may not seem like much, but it could mean significant extra profit to a derivatives trader. So traders would call whoever at their bank was responsible for reporting to the LIBOR poll and plea for a slightly higher or lower report, depending on which direction made their derivatives positions more valuable. In emails and phone texts that later surfaced, traders would promise things like a bottle of champagne for the help.

Why does any of this pose a larger risk?

Despite all the manipulation scandals, the market still seems quite happy with LIBOR. But leaving things as they are creates potentially major new risks. The first is that there is nothing to prevent these scandals from coming back, and the next round of scandals would be worse. The regulators would be more embarrassed, and they would punish banks more harshly than before. The banks would suffer from even more litigation from investors, and they might decide they’ve had enough and drop out of the LIBOR polling. That could reduce trust in LIBOR to the point of disrupting markets, even perhaps causing a significant drop in lending. Market participants could be less willing to do transactions and much less able to hedge risk. That would make loans more expensive and difficult to obtain — exactly what you don’t want, especially if it happens during a financial crisis.

During the last crisis, credit markets froze up significantly. At times, even blue-chip corporations couldn’t borrow for more than 24 hours at a time.

Ironically, the LIBOR manipulation actually eased that panic a bit, because the banks claimed they could still borrow at normal rates – and people believed them. That would not be the case if we have another crisis. In fact, the opposite might happen. Market participants could panic even more than otherwise, because they might assume the banks were manipulating LIBOR and that actual conditions were worse than the banks were admitting.

What are your recommendations to the Financial Stability Board?

The first broad recommendation of the Market Participants Group, which I chaired, is to base LIBOR on a wider set of lending transactions. As I mentioned earlier, there are surprisingly few actual interbank transactions that underlie the LIBOR rate. We suggest using other kinds of transactions, such as the rates that banks pay on certificates of deposit or on bank borrowings known as commercial paper. The goal is to base LIBOR on real transactions, rather than on a bank’s word about the rate it would have paid on a hypothetical loan. Widening the base of types of transactions makes that more feasible.

We don’t think that would be enough. Even with that wider base of transactions, LIBOR would still depend on a relatively small number of transactions, yet would affect an enormous quantity of derivatives transactions, literally hundreds of trillions of dollars. LIBOR is like a tiny rowboat pulling a fleet of ships.

It was like a financial Lake Wobegon: Every bank’s creditworthiness was above average.

Our first recommendation would amount to improving the rowboat’s navigation system. But a tiny boat can easily be knocked temporarily off course by huge waves or high winds. The financial system is huge, and powerful waves in the form of big transactions are normal. If there are only 10 transactions, it’s not hard to rock the market with a single shrewd and possibly corrupt trade that is designed to cause LIBOR to move up or down from its true fundamental level.

At the risk of straining the boat metaphor, we would like to see multiple boats pulling those big ships. In a storm, one or two of them might be momentarily off course but the group as a whole would stay on target.

We’ve identified a number of alternative reference rates that could do the job alongside LIBOR – Treasury rates, the federal funds rate, and several more technical ones that might be even better competitors.

That sounds like an easy solution. Is it?

That’s the 64-trillion-dollar question. The problem is that the trading has clustered around LIBOR, and it will be difficult to change that. It’s a chicken-and-egg problem. People like to trade in markets that have the most activity, the most buyers and sellers, because that’s where they are likely to get the best prices. If almost everybody is using an alternative reference rate, then the others will flock to it too. But if only a few are using that alternative, most people will be reluctant to switch.

As Jeremy Stein and I wrote for Bloomberg View, this is where the national policy makers come in. By speaking publicly about the advantages of reform, and perhaps by using their regulatory powers, they can guide the markets in the right direction. Left to themselves, the markets may not get there.

For media inquiries, visit the Newsroom.

Explore More