Rate futures: the optionality within

With the launch of interest rate futures (IRF) trading at the National Stock Exchange, we begin our second innings with IRFs. The first inning, which was attempted almost a decade ago, was a complete failure with poor product design and an RBI diktat banning banks from taking trading positions leading to eventual death of the product.

In its current avatar, some progress has been made, though not much. Firstly, banks are now allowed to take trading positions in IRFs — this is a huge plus. Secondly, the product design is somewhat better.

This, however, comes after many years of discussion and the end result of two committee reports, both written very poorly. Not only has the committee restricted the introduction to one product, it has also placed some limitations on the product by making settlement by delivery compulsory.

Futures based on the overnight money rates — a market where more than Rs 1 lakh crore changes hands daily — is not being introduced and this decision would continue to rankle many.

Nevertheless, some progress is better than no progress. Gut feel suggests that, in spite of all its relative inefficiencies, the product would be successful. Specifically, only one product — a Treasury bond future contract based on a notional 10-year government bond bearing 7% semi annual coupon — is being introduced.

NSE has provided a list of 11 government securities maturing between 2018 and 2022 that are eligible for delivery for settling the first two contracts that expire in December 2009 and March 2010 respectively.

Since the seller of the contract has the option to deliver any of the eligible securities, a “conversion factor” invoicing system has been devised to reflect the specific pricing characteristics of the security that is tendered. In particular the principal invoice amount paid by the buyer of the future to the seller is given by:

Principal Invoice Price = Futures Settlement Price x Conversion Factor. The conversion factor (CF) is the price of the relevant security at 7% semi annual yield. Further, the CF is supposed to ensure that the seller is indifferent between delivering any security with the eligible basket of securities.

The final consideration that will be exchanged would also include the accrued interest from the last interest payment date to the settlement date:  Total Invoice Price = Principal Invoice Price + Accrued interest.

However, the CF system is imperfect in practice, as a specific security would be “cheapest to deliver” (CTD) after studying the relationship between the price of the security in the cash market and the principal invoice amount. Futures prices would tend to correlate most closely with the prices of the CTD security.

The difference between the price of security in the cash market and the relevant principal invoice amount can be termed as the “basis” for that security. If the basis is positive it would be considered uneconomical to deliver. Conversely if the basis is negative, it would make sense to buy the security in the cash market and deliver it for futures settlement.

This provides opportunities for arbitrageurs or basis traders, who attempt to capitalise on small pricing discrepancies between cash securities and bond futures by buying cheap and selling rich. One can either buy the basis (buy in the cash market and sell the future) or sell the basis (go short in the cash market and buy the future).

Since going “naked short” in government securities for a reasonable time frame is currently not allowed, this may have some limitations. However savvy traders would still attempt to capture this basis spread. Banks and primary dealers can use their existing holdings to mimic short sale of a security and can thus sell the basis. Due to changes in the conditions in the cash market the CTD security would keep changing from time to time.

Intuitively, it is clear that shorter duration securities would become cheapest to deliver if prevailing yields in the cash market are less than the 7% coupon of the notional bond. Conversely, longer duration securities would become cheapest to deliver if yields move higher than the 7% coupon on the notional bond.

Further, the implied repo rate (IRR), which is the rate of return associated with the sale of futures, purchase of security in the cash market and delivery of the same in satisfaction of the maturing futures contract, would be the highest for the CTD security.

In case the CTD security does not change, the basis arbitrageur effectively locks in the IRR as the rate of return from the trade. However, since CTD can change, the basis may go higher or even fail to converge to zero on maturity. In such a case, the return to a basis trader could even be higher than the IRR.

Since the risk is limited on the downside, and theoretically a trader can earn seemingly unbounded returns on the upside, the basis tends to display characteristics of an option. Buying the basis could lead to a payoff profile similar to being long an option, whereas selling the basis could replicate being short the option.

In all probability, the designers of the product probably did not wish to expose participants to vagaries of an option. However, by making physical delivery compulsory, they have done precisely that.

In a way it’s a good thing as this would provide a platform for Indian bond houses to build expertise in handling payoff profiles from trading options. The education and build up of expertise, in itself, is welcome from the perspective of healthy development of our financial markets.

The writer, a former fund manager, runs Finanzlab Advisors, a treasury and risk management consultancy