Degree Type


Date of Award


Degree Name

Doctor of Philosophy



First Advisor

Bruce A. Babcock

Second Advisor

Dermot J. Hayes


The general theme of this dissertation is risk and uncertainty in agriculture, with each chapter addressing a specific topic related to agricultural risk and uncertainty. Chapter 2 examines the effects of production uncertainty on the types of contract structures used in specialty grain markets. A theoretical model of a contractual relationship between a monopsonistic processor and risk-neutral producers is presented. Two common contract structures, and their resulting effects on the sharing of production risk between buyer and seller, are compared. The spatial structure of yields and farm-level yield volatility are shown to have significant impacts on the processor's preferred choice of contract structure and expected profits of both the processor and farmers in the resulting equilibrium. Chapter 3 provides a critical look at a classic definition regarding the relationship between input use and risk, and attempts to reconcile an apparent paradox in the production literature. Experimental corn yield response data is used to estimate a stochastic production relationship between applied fertilizer, soil nutrient availability, and corn output. Optimal fertilizer application rates for risk-averse and risk-neutral producers are found using numerical methods. In addition to the empirical analysis, primary data collected through a farmer survey instrument, designed to elicit information from farmers regarding their risk attitudes and subjective beliefs regarding the relationship between risk and fertilizer use, is presented and compared with the results of the empirical analysis. Chapter 4 turns to the opportunities for managing weather risk using weather derivative markets. Developing regions are areas in which weather based risk management tools show significant potential. However, the success and long-term viability of insurance programs depends heavily on the availability of accurate and reliable historical data. The lack of this type of historical data for developing regions is one of the largest obstacles to insurance program development in these regions. A framework which utilizes statistical methods to estimate unbiased rainfall histories from sparse data is developed. To validate the methodology's usefulness, a drought insurance example is presented using a rich data set of historical rainfall at weather stations across the state of Iowa.



Digital Repository @ Iowa State University,

Copyright Owner

Nicholas David Paulson



Proquest ID


OCLC Number




File Format


File Size

131 pages