Mechanics of Futures Markets. Chapter 2. Futures Contracts. Available on a wide range of underlyings Exchange traded Specifications need to be defined: What can be delivered, Where it can be delivered, & When it can be delivered Settled daily. Futures.
Exp Spot Price
Exp Spot Price
Price of Underlying
at MaturityProfit from a Long Forward or Futures Position
Price of Underlying
at MaturityProfit from a Short Forward or Futures Position
Case Study: GNMA CDR Futures
The contract was introduced in 1975 on the CBoT.
First interest-rate futures contract traded on an exchange.
Grew explosively over first five years,
Declined slightly over 1980-82.
Collapsed back to near-zero levels by 1985.
The initial success and subsequent failure of the contract can be traced to the poor specification of delivery options.
GNMA CDR Futures: Trading Volumes - Death of the future...
An obvious but important point:
Hedging is an offsetting of risks.
For a futures contract to provide a good hedge for a spot risk, futures price must bear close relationship to spot price of asset being hedged.
That is, futures price changes and spot price changes must be highlycorrelated.
In many cases, the underlying risk is well defined (oil, lumber, corn, exchange rates, etc.).
For interest-rate securities, one must be careful in identifying which risk it is that investors are seeking to hedge.
GNMA CDR futures contracts are futures on mortgage-backed securities.
But which mortgage-backed securities are investors seeking to hedge?
Hedging demand is concentrated in current-coupon mortgages.
Primary source of hedging demand: mortgage bankers.
Mortgage bankers are exposed to interest rate risk on mortgages written at current coupon rates, between the time the loans are made and the time they are sold on the secondary market.
For the futures to be a successful hedge vehicle, futures price must bear a close relationship to current coupon mortgages.
Futures contracts also have delivery options that provide for alternative deliverable grades and the price adjustments for each grade.
The actual delivered grade will be not the standard grade, but the cheapest-to-deliver grade.
This means the futures price will always bear a close relationship to the price of the cheapest-to-deliver grade.
However, we want the futures price to be closely related to the price of current-coupon mortgages.
The delivery options must be specified such that the current coupon mortgages are typically the cheapest-to-deliver.
Did the GNMA CDR futures contract meet this criterion?
Specification of the contract:
Underlying asset: GNMA mortgage backed securities.
Contract size: $100,000 in face value of mortgages.
Maturity: 29–30 years.
Any coupon in place of the standard 8%.
Cash flows from delivered grade discounted at 8% for 12 years.
Implicit assumption: all mortgages are prepaid in 12 years.
High-coupon mortgages are more likely to be prepaid early than low-coupon mortgages.
Holder of mortgage has an option to "call" (repay) the mortgage at any time.
Right is more valuable to holder of high-coupon mortgage.
The 12-years-for-all-mortgages assumption undervalues the call in high-coupon mortgages relative to low-coupon mortgages.
Equivalently, it overstates the adjustment factor for high-coupon mortgages relative to that for low-coupon mortgages.
In general, high-coupon mortgages will be the cheapest-to-deliver grade.
Thus, as long as high coupons are also current coupons (i.e., interest rates are constant or rising), the contract will be a good hedge.
This was actually the case between 1975 and 1982.
Coupon rates were around 8% in 1975, rose to 17% in late 1981, and remained at 16–17% until early 1982.
As a consequence, trading in the contract grew rapidly.
However, in late 1982, interest rates declined steeply.
The contract was no longer a useful hedge vehicle.
Trading in the contract dropped to near-zero levels by 1987.
Interest in the contract never revived as Treasury futures and other contracts supplanted it as the hedge vehicles of choice.
Case Study: Metallgesellschaft
Protagonist: Metallgesellschaft Refining & Marketing (MGRM), a subsidiary of Metallgesellschaft AG.
Beginning in 1992, MGRM began selling 5–year and 10–year fixed-price oil supply contracts to a number of customers.
The customers were also provided with an option that allowed them to void the contract (at a profit) if oil spot prices rose rapidly.
The contracts were marketed aggressively and very successfully.
By Nov '93, MGRM had build up long-term supply commitments of over 150 million barrels.
This was 8 times the commitment of Oct '92, and more than twice the commitments of May '93.
These contracts left MGRM exposed to increases in the price of oil.
MGRM hedged its exposure using exchange-traded futures contracts via a "stack-and-roll" hedging strategy.
Such a strategy involves the following steps:
The firm takes long positions in futures contracts to cover its entire exposure.
All positions are in the nearby futures contract, i.e., for delivery at the end of the current month. (This is the "stack" part.)
At the end of each month, the company closes out its position, and opens new long positions to cover its remaining exposure. (This is the "roll" part.)
Theoretically, a stack-and-roll strategy should provide a good hedge for an exposure like MGRM's.
If oil prices rise, there would be a loss on the forward contracts, but a gain on the long futures positions.
If oil prices fall, there would be losses on the long futures positions, but these would be offset by the increased economic value of the forward commitments.
There were three specific risks that MGRM's strategy entailed:
A steep fall in oil prices.
A change in the oil market from backwardation to contango.
Basis risk from the futures/forward mismatch.
Position limits made it impossible to completely hedge MGRM's total commitments of 160 million barrels using only futures contracts.
MGRM had long futures positions of 55 million barrels on NYMEX.
It also entered into OTC swaps arrangements to hedge the remaining exposure.
Every $1 fall in oil prices would lead to a $55 million cash outflow on the futures margin accounts alone.
A steep oil price fall would thus create an immediate and large cash requirement to meet margin calls.
The corresponding gains on the short forward positions would not translate into cash inflows until some date in the future.
Thus, although the economic value of the position is unaffected (it remains hedged), a severe short-term cash flow requirement is created.
Unfortunately for MGRM, this scenario came true: oil prices plummeted in late 1993.
This led to a cash requirement of around $900 million to meet margin calls (on the futures positions) and extra collateral (on the OTC positions).
Backwardation to Contango
A futures market is said to be in backwardation if futures prices are below spot (expected future spot price).
It is said to be in contango if futures prices are above spot.
In a typical commodity market with a positive cost-of-carry, the futures will be above spot, i.e., the market will be in contango.
However, in some commodity markets (especially oil) futures prices have remained below spot for long periods of time.
This phenomenon is commonly attributed to the presence of a large "convenience yield“ (the ability to profit from temporary shortages, and the ability to keep a production process running)from holding the spot commodity.
The Problem with Contango
Recall MGRM employed a stack-and-roll strategy.
Rolling over futures positions at the end of each month involves
Closing out the existing long futures position by taking a short futures position in the expiring contract.
Taking a long futures position in the new nearby contract.
This is effectively selling at the current spot price and buying at the current futures price.
In backwardation, rollover creates cash inflows.
However, in contango, rollover creates cash outflows.
Nightmare Scenario Continued...
Through much of the mid and late 1980's, the oil futures market was in backwardation.
If this situation had continued, MGRM could have expected to make large profits on rollover.
Unfortunately for MGRM, in late 1993, the oil market went into contango.
As a consequence, by end-1993, MGRM was incurring a cash outflow of up $30 million each month on rollover costs alone.
A final technical issue that may have hurt MGRM is basis risk.
MGRM was hedging long-termforwards with short-term futures.
Since these two prices may not move in lockstep, there is basis risk
In the presence of basis risk, a well-developed theory shows that it is not, in general, optimal to use a hedge ratio of unity (i.e., to hedge exposure one-for-one).
However, MGRM does appear to have used a hedge ratio of unity which may have further degraded the quality of the hedge, adding to losses.
When MGRM's cash requirements became public, the problems compounded.
NYMEX doubled margin requirements.
Later, NYMEX also removed MGRM's hedger's exception, effectively halving their position limits.
Counterparties on their OTC contracts also demanded increased collateral for rolling over contracts.
In response, MG's senior management then decided to close out their positions and terminate the hedging strategy in place.
Cash requirements had become excessive.
Rollover costs were around $30 million a month.
The long-term forward contracts were not "watertight," i.e., significant credit-risk existed.
Basis risk also existed from mismatch in assets underlying forward and futures contracts.
Termination of hedge converted paper losses into real ones.
If market went back into backwardation (which had been the market's normal" state for several years), rollover profits would arise.
Removal of hedge left MGRM vulnerable to price increases.
As it happens, MGRM's positions were unwound near the bottom of the market: oil prices rebounded during 1994.
By waiting a few months to unwind, the company could have recouped a substantial portion of its losses.