Valuing New Random Genetically Modified (GM) Traits in Corn
Shakya, Sumadhur, Wilson, William W., Dahl, Bruce, Journal of Agricultural and Resource Economics
Numerous genetically modified (GM) traits are currently under development. Those currently being developed for corn include traits for drought tolerance, cold tolerance, and nitrogen use efficiency, among others. The value of these traits is random and sporadic, creating challenges in assessing its ex ante value. This study estimates the ex ante value of a GM trait in corn with random characteristics. A real option model is developed to capture risks and returns associated with the traits, and estimates are derived for drought tolerant corn. Base case results indicate a slight chance that the option value would be out-of-the-money during the discovery phase. In all other phases, the expected value is in-the-money. The results are highly sensitive with respect to trait efficiency and regarding assumptions of randomness in some of the other important variables, particularly trait values.
Key words: real options, risk premium, stochastic efficiency, trait valuation
(ProQuest: ... denotes formulae omitted.)
Genetically modified (GM) crops have had a dramatic impact on agriculture worldwide. Corn has benefited immensely from the introduction of GM traits, and many new traits are under development, including drought and cold tolerance and nitrogen-use efficiency (McMahon, 2011 ; Birger, 2011 a,b). While earlier GM traits had relatively ubiquitous applications, the value of these new traits is less clear because their usefulness depends on sporadic environmental conditions. For agbiotechnology firms, determining the value of future prospective traits is an important managerial decision. There is a high degree of uncertainty regarding many of the factors that determine trait valuation, including efficacy, regulatory approval, commercial acceptance, competing traits, and prices that can be charged for a trait. The high uncertainty of trait valuation means that the ex ante values of traits are also random.
This study estimates the ex ante value of GM corn traits with random valuations, in this case drought-tolerant (DT) corn. Trait development is considered research and development (R&D) and is interpreted as a real option. While there are many types of real options, here R&D is interpreted as a compound-call option. Given the risks and returns across traits, this interpretation is appropriate for determining the ex ante value for these GM traits.
The analysis estimates distributions for farm budgets with and without the trait, the value of which depends on drought probability and trait efficiency. We then determine the grower's value by estimating risk premiums for simulated budgets using stochastic efficiency with respect to a function (SERF). This grower's value is the basis for trait prices in the option model. A real-option model is used to estimate the trait's stochastic value at each stage of development. The study builds on earlier research using real options to evaluate R&D (e.g., Kolbe, Morris, and Teisberg, 1991; Luehrman, 1997; Lee and Paxson, 2001; Jensen and Warren, 2001; Seppä and Laamanen, 2001) and applications of real options to value post-development costs and benefits for GM traits in crops (e.g., Furtan, Gray, and Holzman, 2003; Carter, Berwald, and Loyns, 2005). This study contributes to the literature by using real options to value ex ante, firm-level management decisions, with random variables in a stochastic binomial specification.
Developing and marketing GM traits can be costly and time-intensive. Discovery, proof of concept, early and advanced product development, and regulatory phases can take ten to fifteen years to complete, and revenue from successful development can only be realized after regulatory approval. Estimating development costs is difficult because these are ultimately firm-level activities and information is generally not published. Goodman (2004) estimated that developing a GM trait costs $60 million and that the regulatory approval process can cost $6-15 million (Kalaitzandonakes, Alston, and Bradford, 2006). …