Foreword

by Galen Burghardt

February 2022

Writing this foreword is a little like writing an obituary for a dear friend. LIBOR, and the Eurodollar futures and options contracts that have been tied to it for nearly 40 years, will be put to bed soon. And so ends a remarkable era of financial innovation that transformed the world of interest rate risk management and academic research.

Still, if you're reading this Foreword, chances are that SOFR, and its related futures and options contracts, have made the competitive cut and are serving as replacements for my old friends. So let's spend the next few paragraphs reflecting on what we've learned.

I think I can be most helpful by recounting some of the reasons the Eurodollar futures contract helped to revolutionize the world of banking and finance. And by finance, I mean both applied and academic.

A LITTLE BIT OF HISTORY

First, it's worth remembering that at the time Eurodollar futures were first listed in the 1980s, there had never been a futures contract that cash settled to an abstract concept. In talking with Rick Kilcollin, who was largely responsible for the contract's design, I learned that the LIBOR market in the early 1980s was thin, and that the development of an index that could capture a relevant financing rate and resist attempts at manipulation was still unfinished. With that in mind, what the Chicago Mercantile Exchange (CME) devised was an ingenious survey in which banks of whatever credit rating were not asked what rate they were paying for interbank funds in London. Instead, they were asked to provide the rate at which they perceived funds were offered to prime quality banks. This, combined with the practice of throwing out the high and low responses, produced a survey outcome with an astonishing degree of agreement.

Second, it's worth remembering that when the contracts were first listed, they were the runty cousins of the certificate of deposit contract. A special, and less expensive, membership was created by the CME for trading the contract, which took place in a small corner of the CD pit. I may have made up this story, but I recall someone saying that Fred Arditti, who was the CME's head of research at the time, would visit the pit each day and come back saying, "I die a little each day when I see how little is going on there."

Then all hell broke loose. Continental Illinois, whose CDs were deliverable into the CD futures contract (and whose motto was "We will find a way") suffered some substantial loan losses and took a hit to its credit rating. It didn't take long for the market to start worrying about credit risk in the deliverable instrument and to look elsewhere.

At the same time, the interest rate swaps market was beginning to take hold and grow, and the Eurodollar futures market was poised perfectly to go along for the ride.

A REVOLUTION IN FINANCE

Eurodollar futures proved to be a financial engineer's dream tool. In the 1980s, the idea of zero-coupon bonds was largely found in textbooks. As was the idea that one could break up the yield curve into three-month (3M) segments and use those segments to study yield curve behavior and the sensitivity of one's financial position to each of those segments.

Now these ideas seem commonplace, but at the time, the world of bonds was almost unbelievably primitive - at least in the world of actual bonds. And the market for forward rates was nearly nonexistent. Try to imagine, for example, what it was like to extract a continuously compounded forward rate curve from the traded bond market. Even if one used data from the Treasury market - possibly the deepest and most liquid bond market in the world - the results could be almost hilariously erratic. In contrast, with Eurodollar futures, one had the next best thing - a quarterly compounded forward rate curve - served on a platter.

Well, almost. It's one thing to know that convexity matters, and another to know just how much. In the late 1980s, Terry Belton and I published a piece for our clients at Discount Corporation of New York Futures called The Financing Bias in Eurodollar Futures. The idea was a simple one based on the daily settlement of gains and losses on futures. That is, if one were short Eurodollar futures, one would be able to invest cash coming in at higher rates (i.e., when rates were going up and you were making money on your short position) and borrow the cash you paid out at lower rates when rates were falling. This was an obvious advantage to the shorts, and if you could do it long enough and over big enough swings, the advantage could add up to real money. At the time we published the note, though, the Eurodollar futures curve only went out a year or two, and the advantage proved not to be worth much for such short-dated contracts. So that research note sank without a trace.

In time, though, the CME extended the Eurodollar futures curve out to 5 years and then to 10 years. And when it did, the interest rate swaps market used these newly available futures rates to price their swaps. The problem, though, as Bill Hoskins and I discovered when we published The Convexity Bias in Eurodollar Futures - perhaps one of the most important research notes of our working lives - was that the market had failed to take the value of convexity into account. Swaps were priced as if futures rates were forward rates so that it was possible to receive fixed on a swap and hedge the position by shorting Eurodollar futures and make completely riskless money as rates rose and fell. Not long after we published that note, the market became aware of the mispricing and completely readjusted.

Another lesson that Bill Hoskins taught me, although it took him a while, was that forward rates (or prices) are breakeven values. That is, if you finance a position to any given forward date, you know just how much the price of what you have can rise or fall before you make or lose money as of that forward date. This is a hugely valuable tool.

One example of just how valuable a tool it is came when Gavin Gilbert, a wonderfully voluble friend of mine, rang me one day to announce, more or less at the top of his lungs, "Galen! You won't believe it! I just bought the forward 2-year TED for zero!" For this to make sense, you need to know that we had just published a good piece of work called Measuring and Trading Term TED Spreads. This was the basis for much of what you could find on Bloomberg if you visited that particular page. We had not, however, considered the buying and selling of term TEDs forward. But Gavin had. He found that if he bought a two-year note two months forward and sold the appropriate strip of Eurodollar futures, he basically owned the spread at 0. Since the two-year TED spread at the time was trading at roughly 20 basis points, he expected to make 20 basis points on the trade. And he also knew that the spread would have to go negative for him to lose money.

I, of course, checked into it and found that by the time I got there, the spread could be bought forward for 10 basis points. So we published a note (as Gavin knew we would) telling our clients about the trade. What made the trade remarkable, though, was that even with highly sophisticated and integrated markets, the term repo market was not yet in sync with the term LIBOR market. Hence the glaring mispricing.

One of the things you learn in any class on derivatives is that the gains and losses on the derivative look just like the gains and losses you would experience on a trade that you could construct in the cash market. So, for example, a long Eurodollar futures position has the same payoff as a cash position in which you borrow money for a term equal to the contract's expiration date and lend for a term that is three months longer. As a result, a long Eurodollar futures position is the equivalent of a simple borrow short/lend long yield curve trade.

Once, during one of our classes on Eurodollar futures, a young man from Panagora asked me what the Sharpe ratio of a Eurodollar contract would look like. It was the first time I'd ever heard the question, so I had to beg off. But when we got back to the office, we tackled the question and found that we could analyze the gains and losses combined with their standard deviations and calculate very straightforward Sharpe ratios. When we did this, we learned that the most profitable part of the yield curve carry trade was in the first two or three years of the yield curve. If you're interested, you can find these early results on page 64 of The Eurodollar Futures and Options Handbook, at least until it disappears from the face of the earth. Or you can look for one of our yield curve carry notes such as Yield Curve Carry Rides Again.

It was neat, too, that these results conformed to what Antti Ilmanen had written in one of his extraordinary monographs at Salomon Brothers. The note was called Does Duration Extension Enhance Long-Term Expected Returns? (Ilmanen 1995). He concluded that once you got past the two-year mark, you had more or less exhausted any useful excess returns and that no, you didn't get paid for taking extension risk.

I should add that one of the greatest contributions of Eurodollar futures in the banking industry can be attributed to one of its most prosaic features. That is, they were futures contracts, which meant that one could buy them in the morning and sell them in the afternoon and have the positions offset. For asset/liability managers, this feature was a godsend....