An IBM 650 in the basement of the business school — a history of computing and economics at Carnegie GSIA

In 1956, the fledgling Graduate School of Industrial Administration (GSIA) at Carnegie Institute of Technology received an IBM 650 computer to use jointly with the engineering and the mathematics departments. In the coming years, this acquisition led to a number of research activities which not only produced multiple economics Nobel Prizes and Turing Awards, but also two very distinct, somewhat incompatible economic frameworks: bounded rationality and rational expectations.

Pittsburgh, the Iron City. Photo by Library of Congress.

Centered on the multifaceted interests of Herbert Simon, the research agenda of the GSIA faculty has been frequently described as “interdisciplinary”, but the focus of most of these efforts were on opening the black box of the firm and describing both its technical processes as a production function and its administrative processes as an organizational decision-making structure in far more detail than most economists of the time deemed useful.

One key building block of this revolution was the Carnegie Tech Management Game, the first computer-based simulation of corporate decision making, developed on the IBM 650 in 1957. Even if its original purpose was to be used in graduate instruction, the game brought together academics from the three main branches of the first wave of research, later to be dubbed “Carnegie School”.

Merton Miller participated on developing the finance module at the same time he was tasked, along with his colleague Franco Modigliani, to teach a class in corporate finance. Finding the existing literature lacking, they developed what came to be known as the Modigliani-Miller theorem (1960), demonstrating that under certain circumstances capital structure (debt or equity) didn’t matter.

At the same time, Franco Modigliani also participated in a research project on modeling dynamic control of warehouse inventories, joint with Herb Simon, Charles Holt, John Muth, and two graduate students: Charles Bonini and Peter Winters. Winters also participated in developing the management game.

The best-known outcome of the U.S. Navy-funded research project was a book titled Planning Production, Inventory, and Work Force (1960), usually shortened to HMMS after its four main authors.

This now largely forgotten book not only presaged the enterprise resource planning software industry, but also the internal rift between the two camps in the Carnegie School. The trigger was an article by Jack Muth (1961), where he took aim at Simon’s concept of bounded rationality by positing that forecasters were, under certain mathematical conditions, still pretty good at predicting future price movements in the aggregate — a concept he dubbed “rational expectations”.

Other than creating internal friction, this insight did not receive a lot of attention until ten years later, new faculty member Robert Lucas made rational expectations the centerpiece of his dynamic model of the macro economy (1972). In the wake of Lucas’s model a second generation of economists affiliated with the GSIA, among them Ed Prescott, Finn Kydland, and Tom Sargent, developed the new classical macroeconomic (or “Sweetwater”) framework. Unlike Jack Muth himself, their efforts were rewarded with multiple Nobel prizes.

Another member of the management game team was economist Richard Cyert, who along with Herb Simon and James March shaped the organizational behavior research within GSIA. Two major outcomes of this productive partnership were Simon & March’s book Organizations (1959) and Cyert & March’s classic A Behavioral Theory of the Firm (1963), building on computer simulations created with the help of Ed Feigenbaum (1959).

Herb Simon himself, disaffected by the increasingly rationalist bent of his colleagues at GSIA, focused more and more time on his research in computer science, psychology, cognition, and the emerging field of artificial intelligence, along with research partners Alan Perlis and Allen Newell. All three would eventually receive the Turing Award for their efforts.

As Simon commented in a later interview, at the time Carnegie acquired the IBM 650, computers were considered tools to do accounting arithmetic. The computer science department did not even exist yet. To even conceive them as tools to augment human decision making, to model and simulate organizational behavior, and to forecast complex nonlinear dynamic processes, both in the micro and the macro economy, was likely alien to most practitioners outside that group at Carnegie GSIA and their allies, bent on implementing Simon’s goal of hardening the social sciences.

Herbert Simon, Administrative Behavior, 1947.

James March & Herbert Simon, Organizations, 1959.

Charles Holt, Franco Modigliani, John Muth, Herbert Simon, Planning Production, Inventory, and Work Force, 1960.

Richard Cyert & James March, A Behavioral Theory of the Firm, 1963.

Herbert Simon, “Dynamic programming under uncertainty with a quadratic cost function”, Econometrica 1956.

Franco Modigliani & Merton Miller, “The Cost of Capital, Corporation Finance and the Theory of Investment”, AER 1958.

Richard Cyert, Edward Feigenbaum & James March, “Models in a behavioral theory of the firm”, Behavioral Science, 1959.

K. J. Cohen, R. M. Cyert, W. R. Dill, A. A. Kuehn, M. H. Miller, T. A. Van Wormer and P. R. Winters, “The Carnegie Tech Management Game”, Journal of Business, 1960.

John Muth, “Rational expectations and the theory of price movements”, Econometrica 1961.

Kalman Cohen and Richard Cyert, “Computer Models in Dynamic Economics”, QJE 1963.

Kalman Cohen & Merton Miller, “Management Games, Information Processing, and Control”, Management International 1963.

Herbert Simon, “Rational decision-making in business organizations”, Nobel Memorial Lecture 1978.

Franco Modigliani, “Life Cycle, Individual Thrift and the Wealth of Nations”, Nobel Memorial Lecture, 1985.

Merton Miller, “Leverage”, Nobel Memorial Lecture, 1990.

Robert Lucas, “Money Neutrality”, Nobel Memorial Lecture, 1995.