L-5306 RM2-5.0
5-99

Seasonality and Its Effects on Crop Markets

William I. Tierney, Jr., Mark L. Waller and Stephen H. Amosson*

Seasonality and Its Causes 

Seasonality is the phenomenon that causes crop prices (including cash, futures, basis, option volatility, intramarket, intermarket, and inter-commodity spreads) to behave in a relatively predictable manner, year in and year out. Generally speaking, there are two major components to crop seasonality: 1) the "harvest lows," followed by 2) the "post-harvest rally." Sometimes seasonality is a strong element of the pattern of crop consumption (domestic usage as well as exports).

The dominant (but not the only) factor driving seasonality is the "on-off" nature of crop harvest. Most of the principal field crops grown in the U. S. have a single harvest season. Consequently, the total supply of the crop becomes available to the marketplace in a relatively short period of time. It is this sudden increase in supply that provides the most dramatic evidence of seasonality --the "harvest lows.

Following "harvest lows," the supply of the commodity is reduced by inevitable (but not always steady) domestic consumption and export demand. In order for the market to ensure that some portion of the year's crop will be available for use later in the marketing year, forward price bids at harvest generally are higher than harvest prices. In most years, prices follow an upward trend, staying "on-track" with the pattern of forward price bids initially laid down at harvest. Therefore, a corollary to the "harvest lows" is the "post-harvest rally."

Seasonality vs. Cycles

In most cases, seasonality is restricted to one production cycle (the period of time that passes between one production event and the next). For most of the principal field crops produced in the U. S., seasonality occurs over a 12-month period (stretching over all four seasons -- hence the name "seasonality"). 

Seasonality should be distinguished from other cycles. Seasonality is related to the calendar, such as months or seasons, and is usually based on changes in supply and demand. Cycles can last any length of time (from minutes to decades). While  there is ample statistical evidence (and a sound theoretical explanation) for seasonality in crop markets, there is only limited evidence that other types of "cycles" affect the markets for most of the principal U. S. field crops. 

Unlike price cycles, which may have a "technical analysis" explanation, the few fundamental crop "cycles" that have been identified are widely believed to be triggered by external events that have an unusual impact on the market. These market shocks (often associated with droughts), in turn, trigger production, demand, and even policy reactions. The effects of such market shocks gradually dampen over time and do not continue indefinitely.

Some economists argue that because these crop cycles lack continuity (are not self-perpetuating), they are not true cycles. There is far more support for the concept of cyclical influences in livestock markets (particularly cattle and hogs) than in crop markets.

Seasonality vs. Trends

Many other factors besides seasonal fluctuations in supply and demand affect crop prices. Price trends are the result of gradual one-directional changes in supply and demand that occur over a period of time. These trends can have a powerful influence on market prices and can significantly alter seasonal patterns (Fig. 1). Consequently, trends and other inconsistencies can cause prices to deviate substantially from those that would be expected based on the crop's seasonal pattern.

"Normal" vs. "Conditioned" Seasonals

The "normal" seasonal pattern that prevails can be estimated as the average of all years or the average of the majority of years deemed to be free of unusual market shocks. Or, a "conditioned" seasonal could be constructed using data from years in which a specific condition is applied. Sometimes referred to as analog modeling, it is a technique commonly used in forecasting other things besides commodity prices. For example, meteorologists often look for distinctive and anomalous weather phenomena. If a particular unusual weather event is present in the current period (such as the occurrence of an El Nino or a major volcanic eruption), then meteorologists look at past years when these events occurred to see if certain weather patterns necessarily followed the event in question. For example, do strong El Niño events strongly correlate with drought in North America?

In commodity analysis, it is common to separate corn and soybean seasonals into two groups: 1) "short crop" years-- years in which yields fell significantly below the trend because of drought, freezes, floods, lack of growing degrees, blight, etc.; and 2) "normal" years -- all years other than short crop years. A recent study of optimal corn marketing strategies (Wisner, Baldwin and O'Brien) found that the "best" marketing strategy in years following short crops was a futures hedge on 100 percent of the expected production in the fourth week of February, covered by a $. 20 out-of-the-money call on new crop futures that was offset in the first week of July. Conversely, in years that did not follow a short crop, the "best" marketing strategy was to purchase $. 20 out-of-the-money puts on new crop futures on 80 percent of expected production in the third week of May, hedge with futures the remaining 20 percent in the first week of July, and offset the puts in the second week of September.

Another criterion commonly used to discriminate between years is to examine years when another major fundamental supply/ demand factor changed. For example, crop prices may have a distinctive pattern in years in which two events occurred-- both total beginning supplies and ending stocks increased. The logic of this type of seasonal is this: "Did prices behave differently in years in which the market had to struggle with a persistent, year-long over supply of the crop (over supply relative to the final quantity consumed)?"

Figure 1. Relationship between seasonal, cyclical and trend effects on prices for a hypothetical crop. 

It is sometimes useful to construct a conditioned seasonal, picking the appropriate years based not on a particular supply/ demand fundamental but on an "unusual" price phenomenon. For example, if the December corn futures contract set a new life-of-contract (LOC) low in July (a month in which the contract normally is setting its high), is that a reliable predictor that the contract will trade lower in the succeeding months?

One example of a seasonal of this type is provided in Figure 2. It examines the pattern of wheat prices in years when the January price was close to or below the July price (something which occurred in 1998 and eight other years since 1973). Clearly, this phenomenon is associated with even weaker prices in the February-May period.

Figure 2. Wheat price seasonal (1973-1996). 

Another variation of a conditioned seasonal is shown in Table 1, the frequency at which KCBT July futures traded higher (or lower) than the previous month's high (or low). Here is a detailed description of the table:

Col. (A) the month examined;
Col. (B) the average amount that the month's high exceeded the previous month's high;
Col. (C) the maximum high and the year that it occurred;
Col. (D) the percent of time that this month's high exceeds last month's high;
Col. (E) the current year's pattern of monthly high prices (notice very few months set new highs);
Col. (F) the current year's pattern of monthly low prices (notice most months set new lows);
Cols. (G, H and I) the same data as columns B, C and D, except that it shows the history of the monthly lows.

Table 1. KCBT July wheat futures seasonal.

(Price Moves From One Month To The Next)

Timing and Magnitude of Price Changes

There are two purposes of seasonal analysis: 1) to correctly identify the timing of a season's high and low; and 2) to estimate the magnitude of the difference between the high and low price. Sometimes market analysts rely on timing to identify the seasonal lows (which may be more consistent than the highs) and then rely on magnitude to predict the high. For example, a particular crop's seasonal low may have occurred in October-November 80 percent of the time. The seasonal high was 12 to 15 percent above the seasonal low 75 percent of the time. Based on this analysis, one would expect the seasonal low to come at harvest (in October or November) and the high to be 12 to 15 percent above the low.

Of the two, timing is the more important for speculative purposes, whereas magnitude is often more important for hedging purposes. Farmers (or other hedgers) may make or lose money in their commodity futures/ options
accounts, but the ultimate profitability of the agricultural enterprise depends on the net profit of the crops produced (net any futures/ options gains or losses). It follows, therefore, that a farmer should be more interested in selling a crop at a profitable price than selling it at the seasonal high.

References

Wisner, Robert, Baldwin, Dean and O'Brien, Dan. "An Evaluation of Pre-Harvest Corn Futures and
Option Marketing Strategies in Selected Iowa and Nebraska Locations." Department of Agricultural
Economics, Iowa State University, 1995.

*Extension Agricultural Economist, Kansas State University Agricultural Experiment Station and Cooperative
Extension Service; Associate Professor and Extension Economist, The Texas A& M University System; Professor and Extension Economist, The Texas A& M University System.

Produced by Agricultural Communications, The Texas A& M University System

Extension publications can be found on the Web at: http://agpublications.tamu.edu

Educational programs of the Texas Agricultural Extension Service are open to all citizens without regard to race, color, sex, disability, religion, age or national origin.

Issued in furtherance of Cooperative Extension Work in Agriculture and Home Economics, Acts of Congress of May 8, 1914, as amended, and June 30, 1914, in cooperation with the United States Department of Agriculture. Chester P. Fehlis, Deputy Director, Texas Agricultural Extension Service, The Texas A& M University System. 1.5M, New ECO