The Relationship Between Risk Attitude and Strength of Preference: A Test of Intrinsic Risk Attitude

• Published in: Management science Vol. 43; no. 3; pp. 357 - 370
• Main Author: Smidts, Ale
• Format: Journal Article
• Language: English
• Published: Hanover, MD., etc INFORMS 01-03-1997; Institute for Operations Research and the Management Sciences
• Series: Management Science
• Subjects: Agricultural management; Attitudes; Crop economics; Decision; Decision analysis; Decision making; Expected utility; Exponential functions; Family farms; Farmers; Hypotheses; Lotteries; Management; Management science; Mathematical independent variables; Measurement techniques; Operations research; Power functions; relative risk aversion; Risk; Risk aversion; Risk aversion preference; Studies; utility assessment; Utility functions; utility theory; value assessment
• ISSN: 0025-1909; 1526-5501
• DOI: 10.1287/mnsc.43.3.357

In a field study, the concept of intrinsic risk attitude is investigated. Intrinsic risk attitude concerns the relationship between risk attitude, measured by the utility function u ( x ), and strength of preference, measured by the value function v ( x ). We study farmers' decision-making vis-à-vis price risk and obtain assessments of risk attitude and strength of preference in two consecutive years in a sample of 253 respondents. This design enables us to investigate the temporal stability of intrinsic risk attitude. Our findings show that risk attitude and strength of preference are two distinctive constructs. More specifically, the hypothesis of a linear relationship between u ( x ) and v ( x ) is clearly rejected in favor of an exponential relationship. This exponential relationship implies that our respondents exhibit a constant absolute intrinsic risk attitude for different price levels, during a time period. Differences between respondents in direction and degree of intrinsic risk attitude are substantial and the majority of the respondents are intrinsically risk seeking. No statistically significant change in intrinsic risk attitude could be detected between the two years of measurement. However, the correlation between measures across time appears to be low. In the discussion, issues for further research are identified.