  We may now modify our assumptions and bring them closer to reality by introducing certain factors which may operate to disturb the price level. Suppose there occurs a general decrease in the desire to consume wheat, on account of the introduction of a popular substitute. Or, a more probable occurrence, suppose it is discovered during the year that the size of the crop has been underestimated. Retaining our hypotheses of complete information and of the necessity of getting rid of the entire crop by July 1, it is obvious that the price of wheat must immediately fall to a figure which will stimulate consumption sufficiently to move the entire crop into consumption by June 30. There is no reason to suppose that such a change would affect the premium on futures over spot prices. The triangle ABC in Figure 1 would move toward the line XY, decreasing both XA and XD, but AD would be unchanged. In like manner changes in the proportion of the crop desired for consumption early in the year, changes in costs of production, and many other factors which affect prices would have the same effect on both spots and futures. Even the introduction of a carry over into the scheme would not alter the relationship of cash and future prices, so far as deliveries in the same crop year are concerned (though it would have a marked effect on the relation of cash prices in the spring and futures prices for new crop deliveries).If, however, we assume that, on account of ignorance or economic irrationality or on account of physical difficulties in getting the crop to market, the holders of a large part of the supply fail to sell it during the first half of the year and then dump it on the market for consumption during the last half, the case is entirely different. The only possible result of this situation will be higher cash prices during the first half of the year than during the last half; if the buyers of forward contracts understand the situation futures prices for spring deliveries will correspond to the lower prices which will actually prevail at delivery dates.The result may be illustrated as follows:As before, XA=cash price in July.
YC=cash price in following June.
XD=price of June futures in July.
The prices of intermediate futures will be such as to reflect the anticipated course of the cash grain market. AC will not necessarily be straight, and may be of any degree of irregularity, except that it cannot drop below the level of DC by more than the carrying charges to June 30 from the time represented by the point of intersection.
Under such conditions the market would afford very imperfect facilities for hedging.
The premium DA on cash grain over futures contracts is sure to disappear. Hence the hedger, if he holds grain long, must expect to lose, either by an advance in the price of the futures contract which he is "short," or by a decline in the cash grain which he is "long" or by both. There is no inducement to hold grain longer than is necessary.
In the first place, it must be emphasized that the current assumptions concerning normal spreads do not apply to the relation between spot prices for old wheat near the end of the crop year and contract prices for new crop deliveries. Let us examine the factors which determine the relative value in May or June of wheat for immediate and for future delivery. Consider first the influence of the amount of old wheat on hand. So far as the spot market is concerned, this constitutes the entire potential supply, and its relation to the demand for wheat for immediate consumption is the most important factor in determining prices. So far as the July or September prices are concerned, on the other hand, the supply of old wheat is of importance only as it affects estimates of the total which will be available for comsumption during the coming year, of which total the carry over of old wheat will constitute only a small fraction. More specifically, if in May or June the supply of old wheat is small relatively to the amount needed to keep mills in operation and supply current demands for flour till the new wheat is fit for grinding, the spot prices at terminal markets will have very little relation to the futures prices. On the other hand, if the supply of old wheat is large enough so that a considerable part is likely to be carried over to be sold along with the new wheat, the spot price will come under the control of the factors which determine the price of the futures and is colloquially said to be controlled by the futures. In other words, a shortage of old wheat may send the spot price to a level indefinitely higher than the price for futures, but an abundance of old wheat cannot send it lower than the futures price, except by the "carrying charge," because the surplus of old wheat constitutes a part of the prospective supply which determines the price of the futures.
Spot prices at different points which are not centers of consumption are affected differently by a shortage of old wheat near the end of the season. If there are supplies quite near the terminal markets, and the situation is not complicated by the presence of a local consumptive market, their price will be closely related to the spot prices at the terminals. If there are other supplies so remote that they cannot be gotten to the central market before new wheat is expected, their price fluctuations will correspond to those for the futures. In general, the longer the time required to make delivery at a central market or a point of consumption, the lower the spot price will be, provided the futures are selling below the cash grain.
The spread between spot and future prices is the most obvious instrument which society uses to control the carry over of grain from one crop year to another. If large offerings of old wheat or a prospective shortage of new results in a premium on future contracts large enough to cover the full cost of carrying, storage is encouraged; if a shortage of old wheat or the prospect of a large new crop results in cash prices higher than futures, or even above the figure where the premium on the futures will cover carrying costs, holders are encouraged to market their holdings of old and replace them by purchases of futures. The social utility of an adjustment of prices which will effect a distribution of the old grain between the two crop years in accordance with the relations of present and prospective demand and supply conditions is obvious, and in general the system, though imperfect, seems fairly effective. The usual situation is a premium on cash over the futures. 1 If the market always behaved in a perfectly rational way a very small premium on spots over futures would suffice to insure that no grain would be carried forward beyond the amounts actually needed to supply demands for old grain before the new became fully equivalent in quality; for no one could afford to carry forward for sale in a future month a commodity which he could more economically buy for delivery in that month. Some grain is carried over, however, especially by farmers, in excess of what is needed, and in the face of a practical certainty of lower prices in the fall.
The situation during the fall months is apparently simpler than that which exists before the harvest, and the expression "normal spread" has more meaning, but even here there is no uniformity in the spreads between cash and futures markets in different years. In the assumed setting given earlier we have seen how "a normal spread" might be made possible. But the marketing of grain is so irregular that there occurs surprisingly often a relative shortage of cash grain and surplus of prospective supply. Given a certain state of demand, the price of futures tends at this time of the year to be controlled by the reported size of the harvest, while the spot price depends upon the supply of grain actually available. Whenever, on account of bad roads, car shortage, railway strikes, farmers' holding movements, or for other reasons, the supply of grain actually coming into the central markets is smaller than current reports concerning the size of the harvest led the trade to expect, the spot price rises above so-called "normal " relation to the futures. A premium on cash over futures is a very common phenomenon. For example, on the first trading day in November from 1901 to 1915 the lowest price of spot contract grain in Chicago was higher than that of May futures, six times, and in only three years out of the fifteen was the May future as much as seven cents higher than the cash.
Even during the spring months, when we should expect the closest correspondence between the spot and the futures markets, there is no discernible tendency for the futures prices to equal the cash price plus a carrying charge. From 1901 to 1915 there were only nine years in which at their respective low points on the first trading day of March the May future was higher than the cash contract wheat. A similar situation prevailed in other markets. Comparison of the lowest prices for cash contract oats and for May futures on the first trading day in April, for the years 1901 to 1915, gives the following results: May price higher, four times; lower, nine times; the same, once; data insufficient, once. Examination of figures for corn prices has also failed to show any tendency for the futures to run uniformly above the cash.
It is therefore clear that any attempt on the part of a miller or other hedger to make money by taking off hedges when the spread is abnormal is purely a speculation. No one can ever predict from the quotations themselves what either the cash markets or the futures markets will do. All that can be said is that the two markets will come together by the last day of the month of delivery, but whether the spread will disappear by a rise of the lower price or a fall of the higher, or a rise or fall of both with one moving more than the other, is a question for the speculator.
The only element in the entire relationship of the spot and futures which may be called a "control" is the fact that at any given time the futures prices cannot get above the spot prices by materially more than the carrying charge from the date in question to the date of delivery. This is true because of the possibility previously noted of buying grain in the spot market, selling futures to the same amount, carrying until the delivery date, and delivering the actual grain. The profit of any trader doing this would consist of the difference between the carrying charge (plus commissions and other charges) and the premium on futures over spot prices at the time he made the two contracts. Such operations would at once narrow the spread by forcing spot prices up and future prices down.
In this same connection, however, it is worthy of notice that it is not to be expected, even if grain is marketed freely, that the futures contract prices for all deliveries or for any two deliveries will fully discount the carrying charge between them. If the spread between the December and the May futures at any time is fully equal to the cost of carrying from December to May, any speculator may buy December grain and sell the same amount of May, knowing that the spread will not become wider, to his detriment, and may become narrower, to his profit. A speculation offering a chance of profit and no chance of loss, even if the chance of profit is not great, is an ideal speculation, and the competition of speculators to buy the nearer and sell the more distant future is bound to be sufficient to make such a situation quite abnormal and only of momentary duration. Spreads between futures contracts vary widely.
The carrying charge bears exactly the same relation to the movement of grain into storage and out into the channels of consumption which the cost of shipping gold bore to the movement of specie in international trade in the days before the war, with this important exception, that whereas gold could be shipped in either direction, grain can be carried forward for future consumption but cannot be transported back into the past. Hence, as just noted, the premium on futures over spot cannot exceed the carrying charge, but the premium on spot over futures has no limit, while in the gold trade a shift in the premiums could cause a movement in either direction, so that the premium and the discount on foreign exchange both had definite limits.
|
