Implementation Rationality : The Nexus of Psychology and Economics at the RAND Logistics Systems Laboratory, 1956-1966

In October 1956, the RAND Corporation established the Logistics Systems Laboratory (LSL) with the goal of using simulation to translate the broad findings of normative microeconomics into detailed, implementable procedures for US Air Force operations. The laboratory was housed in the training simulation facilities that had been recently vacated by psychologists working at the RAND Systems Research Laboratory. Economists at the LSL, interwove their marginal cost-benefit analysis with the psychologists' focus on process, adaptation, and group behavior. Over the course of a decade, economists and psychologists at the RAND Logistics Systems Laboratory conducted game simulations structured by the four separate laboratory problems. Economists went from using simulation to demonstrate the superiority of optimal policies derived from deductive economics to using the experiment as an inductive tool. One of the concerns in this historical case study is with how economics leveraged psychology to grow a regulatory system when individual units pursuing their own interests did not promote effectually the interests of society. This dilemma was one of a few stimuli generating a new focal point for rationality, that of efficient implementation. More recently, economists on the BIS Basel Committee on Banking Supervision were engaging in implementation rationality through simulation in the form of the Regulatory Consistency Assessment Programme (RCAP). The examination of iterative modeling and solving for rules of action at the LSL and in the RCAP suggest that the explicit narrowing of modeling choices that bind the rationality of the individual units would be best iterated through a process that takes into account the human factor. Interactions with experimental psychologists opened a door for economists to non-standard modeling and an iterative, heuristic specification of economizing rules of action that had a greater chance of implementation