HarvestChoice | Focus on Crop Production Risk

Sugar beet / © Sieto Verver

Complex physical and biological processes make it difficult for farmers to know precisely how their efforts will ultimately affect the quantity and quality of their crop. Weather is an important physical source of risk because over 80 percent of cultivated cropland around the world are rainfed rather than irrigated (Siebert et al. 2006). Risk arises due to untimely, inadequate, or excessive rainfall and varies geographically because varying soils and landscape determine how well farmland can store rainfall for crop utilization or shed excess rainfall. Unexpected or severe frosts, extreme hot spells, hail, and strong winds are additional sources of weather-induced risks faced by farmers. Animal, insect, pathogen, and weed pests are also important biological sources of risk. Animals and insects feed on crops, which can reduce biomass, degrade quality, and create opportunities for pathogens to establish. Weeds compete with crops for precious water, nutrient, and solar resources. Pathogens disrupt the normal physiology of crops resulting in limited growth or death. Furthermore, physical and biological process can interact to further disrupt production. A mild winter can foster the survival of overwintering insects, resulting in increased pest pressure during the growing season. To better understand what types of strategic investments are likely to be most effective in improving poor smallholder livelihoods and economic growth in risky production environments, HarvestChoice is building quantitative and spatially-nuanced insights into different types of production risks.

References

Siebert, S., P. Döll, J. Hoogeveen, J.-M. Faures, and K. Frenken. “The Global Map of Irrigation Areas.” Poster presented at the International Workshop on Global Irrigated Area Mapping, Colombo, Sri Lanka, 25-26 September 2006 and at the Tropentag conference, Bonn, Germany, 2006.

back to top

What is Risk?

Risk is commonly defined as the exposure to the chance of a loss. There are several important and distinct elements to this definition that deserve attention. The first is the idea that the outcome is unknown, but the range of possible outcomes is known—either there is a loss, or there is not. The second is the idea that the some possible outcomes are more desirable than others—not experiencing a loss is better than experiencing a loss where a loss might be defined as a decline in wealth, health, crop yields, etc. The third is the idea that the unknown outcome is a matter of chance, which in statistical terms means probabilities can be assigned to the range of possible outcomes. In the context of this definition, risk is always interpreted as undesirable and reductions in risk can be thought of in terms of either decreasing the chance or severity of a loss.

Namibia / © Bernhard Richter

This notion of risk is not always particularly useful when trying to understand risky behavior because if individuals only care about the chance and severity of loss, they will always choose activities to reduce them. But, individuals often engage in activities that increase rather than decrease the chance or severity of loss: many people use cigarettes even though it increases the chance of developing lung cancer or dying prematurely. What this notion of risk fails to account for is the opportunity cost of reducing the chance or severity of loss: the physical and psychological discomfort of nicotine withdrawal from discontinuing the use of cigarettes.

Two important assumptions play a key role in interpreting risk from an economic perspective: (i) individuals prefer more goods (output) to less, and (ii) individuals do not like variability. In the simplest terms, this interpretation implies individuals will make decisions based on the tradeoff between the expected outcome (what will happen on average) and what is referred to as the risk premium. Chambers and Quiggin (2000) provide a rigorous definition of the risk premium, but for present purposes, it is enough to say that the risk premium reflects the variability of possible outcomes and the degree to which an individual does not like this variability. The risk premium is the fundamental measure of risk from an economic perspective. The difficulty with using the risk premium is that individuals have different tolerances for risk, so it can be difficult and costly to measure. To circumvent this difficulty, the variance of possible outcomes (or some other measure of variability) is often substituted for the risk premium because variability is an important component of the risk premium and in special circumstances, the two are directly related.

It is important to note that these two perspectives of risk can lead to different conclusions regarding the risk consequences of engaging in a particular activity. From the common perspective, activities that decrease the expected loss are interpreted as reducing risk regardless of whether the variability of the expected loss (or risk premium) has increased or decreased. From the economic perspective, activities that decrease the variability of the expected outcome (or risk premium) are interpreted as reducing risk regardless of whether the expected outcome increases or decreases. These differing perspectives are a common source of confusion and miscommunication.

Another source of confusion and miscommunication with economic risk is the distinction often drawn between the risk premium and the marginal risk premium. The marginal risk premium refers to how the risk premium changes as an individual engages in more of a risky activity. The risk premium is important for assessing the welfare effects of risk. When the risk premium is positive, individual welfare is diminished by risk, while when the risk premium is negative, individual welfare is enhanced by risk. Alternatively, the marginal risk premium determines how much of a risky activity an individual will engage in. If the marginal risk premium is positive, engaging in an activity is said to be risk increasing because an individual who does not like variability will engage in less of the activity than an individual who does not care about variability. Alternatively, if the marginal risk premium is negative, engaging in an activity is said to be risk decreasing because an individual who does not like variability will engage in more of the activity than an individual who does not care about variability. There is not always a direct correspondence between the risk premium and marginal risk premium. That is, engaging in an activity can be risk increasing even if the risk premium is negative. For example, a farmer may use less fertilizer when fertilizer increases yield variability because he does not like variability (i.e., the marginal risk premium increases as fertilizer applications increase), yet the risk implications of his fertilizer use can still be welfare enhancing (i.e., the risk premium decreases).

A final caveat worth mentioning when talking about risk is the risk of what: crop yields, crop prices, farmer profits, farmer health, public health, or environmental health. The effect of pesticides on a farmer’s yield risk can differ from the effect of pesticides on profit risk, even though yields are an important determinant of profit. The reason for this is that yield risk does not include the cost of pesticides, the price received for the crop, and other important determinants of profit. If crop yields are all a farmer cares about, then measuring yield risk is sufficient to understand the implications of risk on the farmer, but if the farmer cares about crop yields only to the extent that these yields influence profit, then measuring yield risk may not provide an adequate understanding of the implications of risk on the farmer.

References

Chambers, R. G. and J. Quiggin. Uncertainty, Production, Choice and Agency: The State Contingent Approach. Cambridge: Cambridge University Press, 2000.