MSU Agricultural, Food, and Resource Economics
In creating this framework, the research presents a review of relevant literature in the fields of Agricultural Economics, Economics, Management, and Strategy. Building on this broad body of published work, a sequential decision making framework regarding the choice of vertical coordination strategy is proposed and tested: (1) at the level of an industry for the Michigan seed potato, Wisconsin seed potato, and Michigan celery industries, and (2) at the individual decision maker level for producers within the Michigan seed potato and Michigan celery industries.
The validity of the four decision nodes and the resulting four research propositions were tested empirically through twenty five case studies, based on in-depth face-to-face interviews, the investigator's work experience and contact with the three industries, and phone calls made to professionals knowledgeable of the forces shaping these industries. Qualitative and quantitative analyses used to examine the data included a comparative analysis of the cases and discriminant analysis of the interview responses coded primarily as categorical variables.
Principle findings of the empirical research support the theoretic assertations of the overall framework. The reduction of a coordination error and the acceptability risk/return tradeoff were found to provide high explanatory power regarding the willingness or unwillingness of decision makers to change from their current vertical coordination strategy to an alternative strategy they had considered. The implementability of an alternative was found to provide additional explanatory power, but not to the same extent as costliness of fit or acceptability of the risk/return tradeoff. All the alternatives considered by decision makers in this research were found to be programmable, which did not allow for testing via discriminant analysis. This result suggests a change in research design or a reconstruction of the programmability variable to improve measurement ability. Further refinement and testing of each decision variables is warranted. A parallel decision making framework is offered as the next logical step in this line of inquiry.