Content area
Full text
Literature Review
Forecasting in Agriculture
The nature of agricultural production sets it apart from most economic activities (Allen, 1994). As Allen puts it (p.84), "relative to most manufacturing activity, agriculture is greatly influenced by unpredictable random events such as droughts, floods and attacks by pests." In addition, supply is dependent on biological processes in two ways. First, in the short term, farmers have little ability to alter the rate of development of a crop (Allen, 1994) because of long biological lags in the production process (Freebairn, 1994). Second, for most agricultural crops, the production cycle is measured in months or years (Allen, 1994), and quantity and quality produced between years varies greatly due to fluctuating seasonal conditions (Freebairn, 1994).
Figure I illustrates the impact of long lag phases and supply fluctuations in grape production. The first lag period, of approximately six to eighteen months, lies between ordering plant material in the form of cuttings or grafts and planting (Franks, personal communication, December 1996). A second lag period, of approximately three years, exists between planting and full production (Jackson & Schuster, 1994). Following harvest, there is a further lag of six months to five years before the finished wine is released onto the market.
These lag phases and long production cycles require a long-term commitment by growers to a particular area of land. In addition, growers have limited opportunity to adjust production should demand change. In fact, it is perhaps more difficult to reverse land use and grape variety decisions by grape growers compared with similar decisions made in general farming and other forms of horticulture (The Wine Institute of New Zealand, 1979). While some top grafting might be possible in order to change the varietal mix, the lag phase between changing the variety and harvest is still long and costly.
Given the long term consequences of any production or marketing decision, one might expect that forecasting industry trends would be seen as reasonably important. However, in his review of agricultural forecasting, Allen (1994) stated that farmers practically never produce formal forecasts, although most of them doubtlessly form some kind of judgement about the future outcomes oftheir business choices. Allen stated that while farmers may rarely make forecasts, they form the largest group of forecast users.