Spread trading alchemy

Exchange-traded financial futures are well-used, highly liquid instruments, offering regulation, standardization, transparency and removal of counterparty risk. They are usually the first point of liquidity for a hedge fund or institution that is looking to establish or remove an outright interest rate or equity position.

However, it is the spread market that offers the trader another dimension — a relative value marketplace, where it is the relationship between two or more instruments that counts, not the outright direction. Of course, most of these spread markets are familiar to financial futures traders, but the relationships also allow positions with flexible risk profiles that can be adjusted to fit different needs.

Financial futures spreads have two broad categories: intra-contract and inter-contract, both being based on the two popular asset classes of interest rates and equities. Interest rate futures offer the widest range of trading permutations and have the substantial benefit of a mathematical dependency, something that is often forgotten by equity traders, but equity index futures and stock futures also can offer exciting combinations.

Intra-contract spreads are probably the most closely watched and traded. They are also the simplest to understand and follow. Perhaps the most common intra-contract spreads are the bond futures or index futures calendar spread. This is a short lived but highly active trade based around the roll over from an expiring contract into the new front month.

Such spreads are a melting pot of open-interest dynamics and short-term interest rate or repo rate influences. However, a larger, longer-lived intra-contract spread market exists in the Short-Term Interest Rate (Stir) futures markets. These are futures on short-term interest rates and are the largest markets in the world by nominal value. It is by no means unusual for the two largest contracts, the Eurodollar and Euribor, to trade in excess of one trillion dollars (or euros) each day. Also, most financial futures only have one active delivery contract (the front month), but Stir futures can have up to 40, as in the case of the Eurodollar. This means there are an enormous number of spread permutations within the futures complex, which can be traded independently as calendar spreads, or relatively as a butterfly or condor spreads.

The inherent risk in these spreads is curve risk and it follows that the longer the period between the component contracts, the more price movement there will be. A one-year calendar spread such as the Eurodollar June 2007 (M7) and the June 2008 (M8), knows as M7M8, will be more volatile than the three-month June 2007, September 2007 spread (M7U7).

However, despite their different risk profiles, these two spreads are intrinsically linked because the M7U7 spread covers a quarter of the period of the longer M7M8 spread. This relationship can be used to adapt or modify the risk profile of a spread by trading it against another. For example, a position in the one-year M7M8 calendar can be rolled into a three-month M7U7 spread by trading it against a nine-month U7M8 spread or any of its combinations, such as the six-month Z7M8 plus the three-month U7Z7.

Trades like those are popular among intra-contract spreaders who are trying to establish a position in the highly liquid three-month calendar spreads by means other than joining a highly competitive order book. These spreads are always hugely popular because they are simple to follow, transparent and contain only curve risk. Also, because they are instruments within the same complex, there is no basis risk (Eurodollars have a static dollar-point value of $25) and no credit risk.

In contrast, inter-contract spreads (spreads between Stir futures and bond and swap futures) are a fusion of all these kinds of risks, plus some more esoteric ones such as convexity risk, which is dependent on the volatility of short-term rates, but rarely a major influence for a duration of two years or less.

MORE ON RISK

Curve risk and credit risk are, by far, the two biggest influences on price action within inter-contract spreading. Traders sometimes mistakenly believe that basis risk includes curve risk and that their spreads are fully immunized by using duration-neutral hedge ratios. However, basis risk only applies to the effects of parallel movements in interest rates and not changes in the shape of the yield curve.

Where there is a substantial duration mismatch — for example: trading a single deferred month Eurodollar expiry against a Chicago Board of Trade two-year Treasury note future — the position can be akin to a two-year Eurodollar calendar spread but with the added price influence of credit risk. The curve risk in this trade can be largely offset by the use of Eurodollar calendar spreads, but the credit risk remains.

An ideal solution would be to replace the use of bond futures with swap futures, which would nearly remove the credit risk associated with this strategy. Unfortunately, these contracts remain fledging.

An alternative is to look to Europe where the credit spread is significantly less volatile compared to the United States spread. A combination of lower interest rates and a preference for European traders to concentrate on the inter-country bond spreads among Germany, France and Italy results in a relatively static credit spread, thus allowing the trader to tailor an inter-contract spread with little curve or credit risk.

Here, a similar position of a Euronext.Liffe Euribor futures can be traded against a Eurex two-year schatz futures in a variety of ways, each modifying the trade to order. First, however, it is necessary to present the trade as a price spread that incorporates the prices of the relevant instruments, their quantities as determined by the hedge ratio and the respective tick values per basis point.

This is expressed as:

(Pbf x T x Qbf) - (Ps x T x Qs)

Where:

Pbf is the price of the bond future

T is the tick value per basis point

Qbf is the number of contracts used

Ps is the price of the STIR futures.

Qs is the number of contracts used

The chart “Mapping stability” (below) illustrates the visual representation of the price spread. The top graph depicts a Euribor/schatz spread (schatz H7 vs. euribor Z8 in a 1.5:1 ratio). Although the value of the spread is large, it behaves like an intra-contract spread: An increase in the value of the spread will lead to a profit for a long spread position — that is, a long position in the bond and a short position in the Euribor and vice versa. Furthermore, the difference between the purchase and sale price of the spread will indicate the overall profit of loss on the trade. The price spread (black line) is volatile, reflecting the curve risk, and small credit effects as shown by the credit spread in the lower pane, but by introducing a liquid one-year calendar spread (red), a near identical mapping is achieved.

Although a one year-calendar spread does not match the duration of the two-year schatz, it effectively captures the majority of curve action in the bond futures. The trader can then use one as a leading indicator for the other or use them both as a relative value trade. The ratio of price spread to calendar spread can be determined by regression analysis.

ANOTHER WAY

Another way to trade the Euribor futures against the two-year schatz futures is to stagger the ratio and expiration dates used (see “Playing with time,” below). The top pane shows the standard price spread of a two-year schatz vs. a single Euribor contract.

At present, the variance of the spread stems from the curve (and credit) movement due to the fact that a two-year cash flow is being matched with a three-month forward period. If the schatz were spread against a Euribor bundle (eight sequential delivery months), then much of this variance would disappear.

However, bundles are not quite the solution they might appear to be. Not only are they a little clumsy to manage, they are not an entirely accurate price match due to differences in duration matching and because a bundle entails an equal number of contracts per delivery month.

Financial theory dictates that futures strips, when spread against bonds, need to be weighted to compensate for the nearer dated bond cash flows being worth more because they benefit from the effects of reinvestment. This means that the nearer-dated coupon payments of bonds are more important than further-dated ones and this is compensated for by a higher weighting assigned to nearer-dated Euribor contracts.

This is a relatively easy mathematical procedure. However, this still leaves an even more unmanageable spread consisting of schatz vs. eight weighted Euribor contracts, but this can be reduced to as little as two weighted contracts by a process of optimization. Because most Euribor delivery contracts are highly correlated to each other, one can easy substitute for another.

There are several ways to optimize the strip. Some, such as principal components analysis, are highly mathematical and others, such as tweaking trial Euribor weightings on any charting package that incorporates composite strategies, are simple. Given that it is known that the nearer-dated futures will be weighted higher than the further-dated ones, reduces time spent in this trial-and-error approach significantly.

The red spread in the chart shows an optimized schatz/Euribor spread. Instead of being a duration neutral spread between schatz Z6 and Euribor U8, it is now shown as being a spread between schatz Z6 and a weighted combination of Euribor H7 and U8 in a 10:8 weighting (vs. 25 Schatz).

It can be clearly seen how much less volatile this spread is compared to the black version and yet no more contracts were used; they were just redistributed.

Both these examples show how just one kind of inter-contract spread can be metamorphosed into different but more agreeable risk profiles.

Other types of spreads, such as duration positive versions, permit the use of more esoteric products such as packs and bundles to offset risk. Hopefully, this article might initially stimulate studies beyond the confines of exchange-traded products to how their variables of duration, curve and credit risk can be modified to provide a greater range of trading strategies.