Testing the causes of backwardation of stochastic prices in grain markets: A real option valuation approach
Abstract
The existence of backwardation has attracted considerable attention in commodity markets because of the negative carrying cost puzzle. This dissertation estimates the causes of backwardation of stochastic prices in grain markets. The research extends existing research by considering local market responses and uncertainty. The Beta coefficient from the single index model is defined as the measure of risk averse preference for the farmer at a specific location. With the flexible least squares technique, the time-varying beta estimators can reveal the up-to-date price risk change for a site. The uncertainty associated with space and time was examined with the real option pricing evaluation (ROPE) approach, which can calculate the present value for an irreversible store/sell decision option that a farmer faces under uncertainty. To consider the possible stochastic demand and the specific decision behavior for local site, a multi-demand-multi-supply (MDMS) model is developed in this study. Development of MDMS model also contributes to solve the possible aggregation problems by analyzing local data. Price spread between deferred futures price and spot price is defined as the dependent variable. Storage level, transaction costs, risk preference index, option value, demand shock dummy, and a set of four quarterly dummies are considered in the regression equation for corn. Forty observations from the first quarter of 1988 to the fourth quarter of 1997 in Illinois, Indiana, Iowa, Minnesota, Missouri, and Ohio are selected for the corn industry. The wheat market was examined for the same states except Iowa. The empirical outcomes for corn are consistent with the results in previous models. Based on the statistical evidences, transaction costs, uncertainty, seasonality, and external impact are all statistically significant. Very little support for the role of risk aversion was found. Positive coefficients for the first three terms imply that they will expand the size of backwardation if their values increase. This confirms the two major goals for this research. First, the option value can describe the degree of uncertainty well. Second, uncertainty concern significantly influences the size of the price spread and backwardation as well. However, the statistical results are not satisfactory for wheat industry.
Degree
Ph.D.
Advisors
Foster, Purdue University.
Subject Area
Agricultural economics
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.