Is the gold/silver ratio a tradable spread? Was it ever a good spread trade? Probably not. The relationship is mainly dependent on public perception of both items as all enduring store of value. The same is true for the platinum gold ratio (Figure 1 3.4) or the value of gems. Prices remain strong as long as the public confidence is unshaken. There is no fundamental reason for these prices to remain in the same ratio as there is for feedgrain prices to adjust relative to their protein content.

The public was hurt during the silver crisis. Many people made their purdiases during the latter part of 1979, then lost heavily in 1980. At the same time that the U.S. dollar m-as gaining incredible strength and interest rates were still high- gold and silver prices were dropping rapidlj

Naturally, the value of gold did not decline as much as some other inveshnents. Many countries still hold gold as a store of value. But times change and there has been a notice

FIGURE 13-4 The platinumigold spread, similar to goldisilver, shows periods where it might be successfully hded; however, markets such as 1980 alwajs remain a possibility

1980 1981 1982

urce NewYi,.ikltecantileE::cL.iuee

able shift to the U.S. dollar as an intemational standard. Investors feel safe with therr money in U.S. banks and m U.S dollars; they are willing to suffer short-term currency swings in exchange for security and diversification. For exanple, the price of oil. the largest traded commodity in the worid, is quoted in U.S. dollars rather than gold.

There is no doubt the U.S. dollar will lose glamour if U.S. leadership or its economy weakens; if worid peace becomes a reality, the security of the dollar will also be less important. If there is significant economic instability, there may be a return to gold and diamonds as an implied standard. However, these relationships are too uncertain to be hdable. except at great risk

Intramarket and Intermarfcet Financial Spreads

Spread relationships in financial marfcets represent anticipation of economic policy. The spread values themselves are well defined by the interrelationship of all interest rate vehicles, but they will varj on the interpretation of the impact of govemment policy on money supply in reaction to the balance of hde, unemployment, and an expanding economy, and the time period in which the action will occur.

An intramarfcet spread, or delivery month spread, in a single financial inshument such as T-bills, is a conservative speculation in changing policies as discussed in the previous section. A bull spread, long the nearby and short the deferred, shows a belief in temporarily declining interest rates. An intermarfcet financial spread, involving different maturities, is a speculation in the changing yield curve. A long bonds/short bills position favors a negative carrj, while long billsshort bonds expects a retum-to-normal carrj (with respect to ftitures prices).

Intercrop S

A special case involving carrjing charges is the intercrop spread, which can be highly volatile, even though there is an old crop carrjover that ties the two seasonal marfcets together. Soybeans, for example, are harvested mainly in September and October. The August delivery is clearly the old crop, and November is the first new crop month; the September contract often reflects the shift from old to new.

Normally, the old crop trades at a premium to the new crop. Carrjing charges, accumulating since the previous wmter, are part of the August price; export demand may cause shortages in the old crop, which moves prices further above the cost of cam. Figure 13-5a shows the anticipated price pattem resulting fran carrjing charges during a normal year. The minimum storage commitment for the 3 months immediately following harvest is shown as a larger increase in price. Normal carrj adds equal amounts to the price until the following September, where old and new crop mix. Finally, any carrjover must assume the price of the new crop, which is in greater supply

The theory behind an intercrop spread is that prices must come together when the old crop merges with the new crop. But can it be a profitaUe hade? Examine the two possible events that affect this trade:

1. Problems involving development of the new crop making supply uncertain. This causes prices in the new crop to rise more quickly than the old crop (Figure 13-5b).

2. Rxport demand in the old crop results in old crop prices rising faster than new crop prices (Figure 13-5c).

In case I, the spread between crops narrows and can only be haded as a spread that is short the old crop, long the new crop. This spread has limited potential since the November delivery cannot exceed the old crop by more than the normal carrjing charge. At that

FIGURE 13-5 Intercrop spreads, (aj Normal carrjing charge relationship, lb) New crop supply problems result in narrowing spreads, (c) Rxport affects the old crop more than the new, resulting in a widening spread.

I-I-1-I-1-I-1-I-1-1-1-I I I I i

NDJFMAMJJA S N D J F

I-1-1-1-1-1-1-1 I I 1 I L-

point, processors will buy, store, and redeliver the old crop against the new crop. \ien prices in the new crop are nearing the old crop value, it makes sense to liquidate or even to reverse the spread. Once reversed, there is litUe risk that the spread can move adversely. A revised crop estimate that showed better yield than originally e; ected would cause the new crop to decline sharply and the spread to widen.

in case 2, export demand in the old crop causes the intercrop spread to widen at first. A spread trader will tjpically wait until export commitments are complete, then sell the old crop and buy the new. \ien old crop prices rise to a large premium over the new crop, many processors will reduce purchasing based on-

I. Low or negative profit margins at the current price levels

2. Use of reserves or inventory to carrj processing through until the lower new crop prices are available

It is remarkable how demand is inversely related to price, even when it is considered inelastic. \ien a delay in purdiasing or processing will result in greater profits for the commercial, it is somehow achieved

Butterfly Spreads

A butterfly spread, normally referred to as just a butterfly, is a low-rid; technique for capturing shortterm interdelivery distortions. It is most appropriate in hily volatile markets, where the concentration of trading in one oi two deliverj months causes those contract prices to move away fran the normal delivery month relationship.

For example, in April a combination of poor planting, declines in the U.S. dollar, and new export agreements changes the soybean prices as shown in Figure 13-6. Increased demand results in sharply higher July futures prices with a tapering-ofiF of the effect in the more deferred conhcts. The normal carrjing charge relationship is shown as a shight line in the old crop, beginning again in the new crop. A butterfly entered in the old crop would mean selling two contracts of July soybeans and bujing one conhct each of May and August This is the same as executing two spreads: long May, diort July and short July, long August.

Eadi spread in the butterfly has agood chance of being profitable; the combination of the two is exceptionally good. The July conhd cannot remain out-of-line with the deliveries on both sides, because a hder could take delivery of the May conhd and redeliver it in July at a profit exceeding the cost of carrj Under normal circumstances, commercial users of soybeans will defer their purdiases to later months, depleting their reserves, to avoid pajing a short-term premium and causing July prices to drop.

FIGURE 13-6 Delivery month distortions in the old crop making a butterfly spread possible

Oldcrtv Sell 2

Buyl

The butterfly spread guarantees that any adjustment in the three conhds back to a normal relationship will prove profitable. If ttie prices of both May and August rise to be inline with July or if May rises to form an inverted relationship, the spread will be profitable (Figure 13-7).

The problem with such an ideal spread is the short life and difficult execution. Because profits are nearly riskless, opportunity is small. The beneficiaries of these hdes are usually the floor ttaders who can act quickly. Once the position is entered, liquidation can be easily accomplished.

Pseudoarbihge of S&P Day and Nit Markets

intended to take advantage of distortions caused by S&P Market-on-Close orders and GLOBEX session, Connors and Hajward have created a GLOBEXIS&P Fair Value Program.

1 . Determine ttie daily fair value difference between ttie S&P futures and cadi maitets, W = S&Pfutures - S&P cadi. Use the price as soon after the close of the NYSE as possible, for example, 4:03 EST, leaving 12 minutes to the close of futures hiding.

2. Calculate your price ertrj target, T, as T = S&P cadi + W + .50, where ttie value .50 represents 1/2 an S&P point.