Safety-first? A context-rich experimental study of the role of safety-first risk preferences in a post-harvest grain marketing setting

Purpose Roy (1952) proposed the conceptually simple concept of Safety-First (SF) as a behavioral basis for decisions made under risk and uncertainty. SF is based on the simple and appealing principle that decision-makers will make choices to minimize the probability of a disastrous outcome occurring. Agricultural economics literature has mostly ignored SF preferences as a factor that may influence these decisions. In this paper we demonstrate in practice the importance of the SF rule in decision-making under risk in a grain marketing environment. Design/methodology/approach Using technology in the form of a context-rich grain marketing game, we track individual grain marketing decisions within multiple marketing years. Experimental economics methods are incorporated into the grain marketing simulation game, enabling us to elicit individual risk preferences in a controlled environment and track how those preferences affect grain marketing decisions under diverse market conditions represented by different grain price paths. Our approach incorporates the dynamic characteristic of realistic price variation at each decision point without introducing considerable cognitive load on the experimental subject. Findings Our results within and across marketing years indicate that SF risk preferences significantly impact grain marketing decisions. Specifically, individuals displaying SF risk preferences sell an additional 8.45% of harvest in the first month of the marketing year over those not displaying these preferences. Such upfront high grain sales are observed across all market conditions. This outcome represents the fact that selling early in the marketing cycle guarantees a profit in each pricing scenario and reduces the probability of financial disaster if prices fall during the marketing season. Research limitations/implications Producers would be ideal subjects for this study; however, recruiting them to participate in experiments is often extremely challenging due to high demands on their time, concerns about confidentiality of responses, complexity of the decision-making process and other privacy issues. Student subjects were used to study human behavior in the current context involving dynamic decision-making and learning, which is hard to isolate with primary data of any other kind (Rosch et al., 2021). We are hopeful this work will motivate future research to tackle the issues presented by studying producers’ grain marketing decisions. Originality/value We investigate a conceptually simple, yet not well understood, concept of Safety-First (SF) framework for decisions under risk and uncertainty. Our approach also merges technology, experimental economics methods and a risk management tool of forward contracting, allowing us to gather primary data and match individuals' risk preferences with a type of decision that is otherwise particularly challenging to be tracked over time.