Consistency of risk premium measures

This research uses the results of a series of within‐sample experiments to elicit risk premium measures from agricultural producers. Results show that there is little consistency between measures in different contexts and using different elicitation methods, suggesting that underlying risk preferences are not consistent. These results highlight some of the difficulty with expected utility theory and risk measurement.     

The (in)stability of farmer risk preferences

We test and quantify the (in)stability of farmer risk preferences, accounting for both the instability across elicitation methods and instability over time. We use repeated measurements (N = 1530) with Swiss fruit and grapevine producers over 3 years, using different risk preference elicitation methods (domain-specific self-assessment and incentivised lotteries). We find that farmers' risk preferences change considerably when measured using different methods. For example, self-reported risk preference and findings from a Holt and Laury lottery correlate only weakly (correlation coefficients range from 0.06 to 0.23). Moreover, we also find that risk preferences vary considerably over time, that is, applying the same elicitation method to the same farmer in a different point in time results in different risk preference estimates. Our results show self-reported risk preferences are moderately correlated (correlation coefficients range from 0.42 to 0.55) from one year to another. Finally, we find experiencing crop damages due to climate extremes and pests is associated with farmers becoming more risk tolerant over time in specific domains.     

Estimating Intertemporal Preferences for Natural Resource Allocation

In this article, we show how the degree of risk aversion, discounting, and preference for intertemporal substitution for a natural resource manager can be structurally estimated within a recursive utility framework. We focus on the management of a reservoir in California, and test the data for consistency with a recursive utility model specification versus standard time-additive separability. The results show that the data are consistent with a risk-averse manager with recursive preferences. The data also reject time-additive separability, with or without risk aversion, such as the standard constant relative risk aversion utility model. The improvement in model fit when recursive preferences are used is notable.     

Heterogeneous Risk Preferences,Discount Rates and Land Contract Choice in Ethiopia

This paper analyses the role of risk and rate of time preference in the choice of land contracts. The analysis builds on the risk‐sharing and imperfect market explanations of contract choice. Unique data from Ethiopia, which contain land contract information and experimental risk and rate of time preference measures on matched landlord–tenant partners, are employed in the empirical analysis. The results show that landlord and tenant time preferences are significant determinants of contract choice. For landlords (but not tenants), risk preference is also significant, indicating the importance of financial constraints and production risk in the determination of contract choice. The results are of particular relevance to land market policy in Ethiopia, where production is risk‐prone, financial markets are imperfect, and where there is a major need for the development of vibrant land rental markets.     

Measuring the risk attitude of decision‐makers: are there differences between groups of methods and persons?

Many studies quantifying individual risk preferences of test persons show that results of different measuring methods may vary. Additional reservations about the reliability of the results regarding the risk attitude measurement arise from the fact that most studies are based on convenience groups, such as students or businessmen in developing countries. With this in mind, we systematically compare different measuring methods to answer the question how the choice of method affects the results. Moreover, we compare the risk preferences of German farmers with those of students and Kazakhstani farmers to investigate whether farmers’ risk preferences can be approximated through those of convenience groups. The methods applied comprise an incentive‐compatible Holt‐and‐Laury‐lottery as well as two psychometric methods. Results show that students respond consistently across all three elicitation methods whereas German and Kazakhstani farmers are more inconsistent. Significant differences exist in the responses of German students and German farmers. The comparison of risk preferences between German and Kazakhstani farmers, however, reveals significant similarities with respect to the psychometric methods.     

Measuring risk attitude of agricultural producers using a mail survey: how consistent are the methods?

A mail survey is used to examine the consistency of alternative risk preference elicitation procedures using five commonly used methods. These elicitation procedures have been used in previous studies to characterise risk preference. Results show little consistency across procedures, supporting strength-of-preference studies. A general recommendation for mail surveys is the development of relatively easy-to-understand risk-preference elicitation procedures that are framed according to the situational construct in question.     

Husbands’ and wives’ risk preferences and improved maize adoption in Tanzania

Research on technology adoption typically assumes that the preferences of a single individual—the household head—determine household-level decisions. This study uses experimentally derived prospect theory-based risk preferences from couples in Tanzania to test whether men's and women's risk preferences, which often diverge, influence the adoption of improved maize varieties. We find that women's risk aversion and men's loss aversion are negatively correlated with improved variety use. The tendency of men to overweight small probabilities is negatively associated with adoption, whereas the tendency of women to do so is positively associated. These findings suggest that technology adoption research should look beyond the preferences of the household head, and that technology promoters should consider targeting both men and women in their education and dissemination efforts.     

Production risk,risk aversion and the determination of risk attitudes among Spanish rice producers

Agricultural production is subject to risk and the attitudes of producers toward risk will influence input choices insofar as these affect production risk. Risk attitudes in turn may be affected by certain socioeconomic characteristics of producers. Using 2004 survey data from a cross‐section of 130 Spanish rice farms, we estimate risk‐aversion coefficients of farmers and investigate the influence of a series of socioeconomic variables on their risk attitudes. Our results show that farmers exhibit risk‐averse behavior and that risk attitudes are related to a series of socioeconomic characteristics. In particular, the belief that the farm will continue after the producer retires is found to increase the degree of risk aversion, while age is found to have nonlinear effects on risk aversion. Off‐farm income, especially from nonagricultural activities, is found to reduce risk aversion. Neither the educational level of the producer nor the presence of dependents on the household is found to have an effect on risk preferences. Regarding the production technology, we find that land, labor, and fitosanitary products are risk‐reducing inputs, whereas capital, seeds, and fertilizer all increase risk.     

Stochastic efficiency analysis with risk aversion bounds: a correction

A recent paper by Hardaker et al. (The Australian Journal of Agricultural and Resource Economics, 48, 2004a, 253) and book by Hardaker et al. (Coping with Risk in Agriculture, 2004b) describe a procedure for determining an efficient set from among a set of random alternatives. This procedure, called stochastic efficiency with respect to a function (SERF), is claimed to make the same assumption concerning the risk aversion measures as does stochastic dominance with respect to a function (SDRF). This is claim is incorrect. SERF imposes an additional requirement on the risk aversion measures of the decision makers. Both procedures assume a lower and an upper bound on risk aversion, but SERF also assumes that all risk aversion measures are of the same functional form as these lower and upper bound functions. This additional strong requirement on risk preferences implies that the efficient set identified under SERF is usually smaller than that identified using SDRF.     

Risk Preferences of Dairy and Hog Producer in Quebec

The objective of this study is to investigate the nature of risk preferences of Quebec dairy and hog producers. The direct elicitation of utility method is employed to determine producers'degree of risk aversion. The Delphi process is used to obtain more refined and realistic responses. The results reveal that the risk preferences of the randomly selected Quebec farmers are highly diverse. The percentage of risk-taking farmers ranges from 8% to 23% depending upon the level of investment and the nature of the enterprise. On average, the majority of farmers in both groups are found to be risk averse. Based on the differences between the means as well as distributions, although not significant in all cases, hog producers are found to be consistently more risk averse than dairy producers. Moreover, the gap between the two groups widens as the level of investment increases. The implications of this result are that the stability of farm income due to supply management in dairy sector may facilitate investments of a given risk (for example, adoption of a new technology) more so than it would in the hog sector.     

Risk, uncertainty, and learning in adoption of a crop innovation

Risk and uncertainty have often been suggested as causes of poor adoption of rural innovations, but empirical evidence has been scarce. This study focuses on a new crop‐type, chickpeas, in Western Australia to gather such evidence. The empirical models developed are based on a theoretical framework that conceptualizes adoption as a dynamic decision process involving information acquisition and learning‐by‐doing by growers who vary in their managerial abilities, risk preferences, and their perceptions of the profitability and riskiness of the innovation. Learning encompasses improvements in skill as well as reductions in uncertainty. An annual face‐to‐face survey of over 100 farmers was conducted over 3 years, eliciting the farmers' risk attitudes and their subjective distributions of yields and prices. Two limited dependent variable models, Tobit and Probit, are used to estimate the empirical model. There is a high degree of goodness‐of‐fit for both models. The study provides strong empirical support for the primarily economic character of the adoption decision, and highlight the importance of economic risk in the process. The two risk‐related factors with greatest impact on the adoption decision were risk aversion and relative riskiness of the innovation. Risk aversion tended to reduce adoption, and to do so to a greater extent as relative riskiness and scale increased. Results also reveal the key role that trialing of the innovation plays in adoption.     

Risk preferences of commodity crop producers and specialty crop producers: An application of prospect theory

Producers’ decisions, such as crop insurance, contract agreement, and technology adoption, involve considerable risk and uncertainty. Particularly, specialty crop production is more vulnerable to risk and requires more intensive management than commodity crop production, while risk mitigation tools for specialty crop production are comparatively limited. We apply Prospect Theory (PT) to analyze risk preferences of U.S. producers, and further compare the preference differences between commodity crop and specialty crop producers. Reference dependent, diminishing sensitivity, loss aversion, and probability weighting, as well as certain farm characteristics and producer demographics, are found to have a significant impact on grower risk attitudes. In addition, we do not observe significant differences in the base PT estimates between commodity crop and specialty crop producers. However, the relationships between risk behavior and individual characteristics vary between the two types of producers, which shed lights on the development of agricultural policies and provide implications for the design of contract and insurance.     

Measuring the effect of risk attitude on marketing behavior

Despite extensive study, researchers continue to search for consistent and reliable measures of risk preferences to explain market behavior. We find that a measure, combining experiments rooted in expected utility theory and measures derived from surveys, explains spot and contractual sales, but does not exhibit substantially greater explanatory power than its underlying components. Survey‐based measures are generally more significant indicators of marketing choices, but experimental measures reveal how risk attitudes vary over a range of probable outcomes, which is important in light of increased commodity price volatility. Given recently identified limitations on the applicability of expected utility theory, we suggest that researchers include survey methods to obtain low‐cost supplemental measures.     

Social capital,risk preference and adoption of improved farm land management practices in Ethiopia

Many developing countries grapple with high rates of farmland degradation and low agricultural productivity amidst increasing climate variability. Considerable efforts have been exerted to promote the diffusion of improved farmland management to address these challenges. Despite these efforts, adoption rates, especially of soil conservation and water harvesting technologies, are still low, which has been the subject of investigation in several studies in Ethiopia and elsewhere. Most studies on the adoption of these technologies, however, tend to focus on economic incentives only, paying little attention to the role of social capital. This article provides evidence of the effects of different dimensions of social capital on innovation adoption across households holding different levels of risk aversion. We address this issue by using cross section and panel data from Ethiopia. Results show that social capital plays a significant role in enhancing the adoption of improved farmland management practices. We also find evidence that the effect of social capital across households with heterogeneous risk taking behavior is different.     

Is Risk Aversion Really Correlated with Wealth? How Estimated Probabilities Introduce Spurious Correlation

Economists frequently focus on correlations between wealth and risk preferences but rarely observe the probabilities needed to test this relationship empirically. These unobserved probabilities are typically estimated via profit or production functions conditioned on wealth correlates, which may leave statistical fingerprints on subsequently-estimated risk aversion coefficients and confound correlations between wealth and risk preferences. Using data from an experiment with observable probabilities, we compare risk aversion coefficients based on true probabilities with those based on probabilities estimated using standard approaches and show how estimated probabilities can change risk aversion coefficients substantially and introduce spurious correlation between risk aversion and wealth.     

How Sure Can You Be? A Framework for Considering Delivery Uncertainty in Benefit Assessments Based on Stated Preference Methods

The economic valuation of benefits resulting from environmental policies and interventions often assumes that environmental outcomes are certain. In fact, these outcomes are typically uncertain. This article proposes a methodological approach to incorporate delivery uncertainty into benefit estimation based on stated preference methods. In the study design of a choice experiment survey on land‐based climate change mitigation, we explicitly include delivery uncertainty as the risk that a proposed mitigation project fails to deliver emission savings. We find that respondents’ preferences do not change significantly after being confronted with choices that included risk of failure. However, failure risk itself does have an important impact on the preferences for delivering emission reductions. We show that delivery uncertainty can have a large impact on stated preference estimation of benefits of public programmes. This result should condition conclusions drawn from ex‐ante environmental cost‐benefit analyses that make use of such benefit estimates.     

Farmers’ risk preferences and their climate change adaptation strategies in the Yongqiao District,China

Farmers’ risk preferences play an important role in agricultural production decisions. This study characterizes risk preferences among farmers in Yongqiao and determines how these risk preferences are related to their choices regarding climate change adaptation strategies. We find that most farmers in the study area were aware of climate change. They were taking measures to protect their livelihoods against perceived changes to the local climate. The risk experiment result shows that the representative subject was risk averse, and women were more risk averse than men. The relationships between farmers’ risk preferences and different climate change adaptation choices were different. Farmers’ risk aversion was negatively and significantly related with adaptation strategies on planting new crop varieties and adopting new technology, but it had a significantly positive effect on purchasing weather index crop insurance. The results also indicate that the level of education, farming experience, farm size, household income and perception of climate change impacts influence farmers’ adaptation decisions.     

Risk,learning, and technology adoption

This article explores how decision makers learn and use information, with an application to the adoption of biotechnology in agriculture. The empirical analysis relies on experimental and survey data measuring risk preferences, learning processes, and the adoption of genetically modified (GM) seeds among U.S. grain farmers. While controlling for risk aversion, we link individual learning rules with the cognitive abilities of each decision maker and their actual GM adoption decisions. We find evidence that very few individuals are Bayesian learners, and that the population of farmers is quite heterogeneous in terms of learning rules. This suggests that Bayesian learning (as commonly assumed in the analysis of agricultural technology adoption) is not an appropriate characterization. In addition, we do not find a strong relationship between observed learning styles and the timing of GM seed adoption. To the extent that learning is a key part of the process of technology adoption, this suggests the presence of much unobserved heterogeneity in learning among farmers.     

Modeling preference heterogeneity in stated choice data: an analysis for public goods generated by agriculture

Stated choice models based on the random utility framework are becoming increasingly popular in the applied economics literature. The need to account for respondents' preference heterogeneity in such models has motivated researchers in agricultural, environmental, health, and transport economics to apply random parameter logit and latent class models. In most of the published literature these models incorporate heterogeneity in preferences through the systematic component of utility. An alternative approach is to investigate heterogeneity through the random component of utility, and covariance heterogeneity models are one means of doing this. In this article we compare these alternative ways of incorporating preference heterogeneity in stated choice models and evaluate the sensitivity of estimated welfare measures to which approach is selected. We find that a latent class approach fits our data best, but all the models perform well in terms of out-of-sample predictions. Finally, we discuss what criteria a researcher can use to decide which approach is most appropriate for a given data set.     

Partial equilibrium analysis of vaccination as an avian influenza control tool in the U.S. poultry sector

The study addresses the economic effectiveness of vaccination in the case of highly pathogenic avian influenza (HPAI) outbreaks. The effectiveness is evaluated in a U.S. case study comparing the outbreak costs under the currently recommended HPAI control practice versus one which also includes vaccination. An integrated economic‐epidemic partial equilibrium model is used to empirically simulate HPAI outbreaks in three different poultry regions within the State of Texas. The simulation results show that the effectiveness of vaccination depends on flock density in the region and risk aversion preferences of the decision maker. In regions with high flock densities, if the decision maker is less risk averse the currently recommended strategy—which includes depopulation, movement restrictions, and flock testing—is more cost‐effective than the same strategy supplemented with vaccination. Addition of vaccination improves cost effectiveness only if the decision maker is highly risk averse or if surveillance, carcass disposal, and cleaning costs are high. In regions with medium and low flock densities, adding vaccination to the currently recommended strategy is not cost‐effective except in the case of high surveillance, carcass disposal, and cleaning costs.