The probability of nonconvergence and its spatial effects in the hard red winter wheat market

Abstract

Nonconvergence in commodity markets has caused some market participants to question the effectiveness of using futures contracts to effectively set prices. This failure of the price discovery function of the futures market increases a farmer’s basis risk exposure when hedging their grain. A variable storage rate (VSR) mechanism was adopted in 2018 for the hard red winter (HRW) wheat market to prevent nonconvergence. The VSR adjusts the storage rate on delivery instruments dependent upon the amount of financial full carry present in the market. This thesis will examine the efficacy of the VSR and show the spatial effects of nonconvergence in the HRW wheat market throughout the state of Kansas. A dataset consisting of daily spot prices at 91 grain handling facilities across Kansas, daily closing futures prices, and daily three-month London Interbank Offered Rate from 2004 to 2019 was used in the analysis. To determine the effectiveness of the VSR on preventing nonconvergence, each contract month was categorized into three convergence outcomes based on the average basis during the delivery period for four facilities in three delivery locations in Kansas. A multinomial model was used to determine the probability of each convergence outcome given an average percent financial full carry in the observation period. The occurrence of nonconvergence was found to be significantly more probable in all locations when the average percent financial full carry in the observation period is greater than 80%, supporting the underlying theory behind the adoption of the VSR. Next, basis at 90 grain elevators across Kansas was predicted using a naïve pricing model with respect to basis in Kansas City, a delivery location, to determine the effect of nonconvergence on basis at outlying locations. Results indicate the presence of nonconvergence at delivery location has a significant effect on basis at some, but not all, locations. Locations with a higher grain storage capacity are less likely to be affected by nonconvergence, lending support to the theory that nonconvergence can be caused by low storage availability.