Herb Simon was quite the intellectual hero. He earned his PhD at the University of Chicago in social science, what they called decision science, but made significant contributions in economics, experimental psychology, and artificial intelligence while a professor at Carnegie Mellon University. He won the Nobel Prize in Economics in 1978 and learned Swedish to give his acceptance speech, which I think is just showing off. Your correspondent never met him but saw him once sitting next to my adviser on a panel. I did hear a story of a student who was tasked with driving Prof. Simon to the airport, and not knowing what to talk about with Simon, he asked him about his Nobel Prize, and Simon’s response was surprising because he was mad. Simon felt the prize was something of a lifetime achievement award and was angry because the ideas and topics he had developed and written about had not been adopted by academia.
Simon wrote two books that particularly influenced your correspondent, Reason in Human Affairs and Sciences of the Artificial. Together they develop a concept Simon calls bounded rationality, the idea that people tend to make decisions based on few and certain data, and when decision makers are stressed, they make decision based on even fewer and more certain information. Bounded rationality squares with observation and experiment, but it doesn’t square with economic theory that tends to be more flattering of human cognition.
The consequence of bounded rationality is that high-status decision makers can make claims of competency and prediction that just aren’t true, that just don’t stand up to observation and experiment. I’ve always thought that public policy should be crafted with the same intelligence and care that engineers use to design their systems, especially electronic systems, given the stakes of such policy decisions, but that is not usually the case. However, the changing international system and the seriousness of policy consequences provide the incentive for revisiting Simon’s work that was doubtless ahead of its time.
Bounded rationality manifests itself politically when understanding why centralized planned economies can’t solve the distributed market information problem. That is, markets process more information than humans can process, which doesn’t sit well with ego-driven political leaders who have risen to positions of leadership. Additionally, it doesn’t do well for one’s career prospects that political leaders are cognitively incapable of controlling markets and comparable social systems even if it is true. Hayek wrote intelligently and presciently about complexity in the 1960s and the social sciences still haven’t incorporated their insights into the discipline.
Part of the reason Simon’s and Hayek’s good and accurate work hasn’t been picked is that their positive observations are sometimes in conflict with normative preferences. For instance, Simon considers the voting associated with democracy—the sine qua non of American social science—as just another decision-making mechanism. Social scientists may feel that a democratic vote confers legitimacy, but Simon argues that there are other systemic characteristics that are more important and indicative of policy success than voting.