OR, Management, and Economics: Historiographical Gains, Context, and Questions June 2, 2012Posted by Will Thomas in Commentary Track, History of Economic Thought, Operations Research.
Tags: Abraham Charnes, Alan Waterman, Alfred Sloan, Allen Newell, Beatrice Cherrier, Charles Holt, Edward L. Bowles, Eli Shapiro, Franco Modigliani, Henry Stimson, Herbert Simon, Horace Levinson, Hunter Heyck, Jay Forrester, Joel Isaac, John Muth, Judy Klein, Karl Compton, Marion Foucarde, Paul Samuelson, Philip Morse, Rakesh Khurana, Robert Solow, William Larimer Mellon Sr., William W. Cooper
This post continues my provision of supplementary commentary for my Business History Review article, “Operations Research vis-à-vis Management at Arthur D. Little and the Massachusetts Institute of Technology in the 1950s” (Thomas 2012). In it, I look at a history split between this article and my 2009 article with Lambert Williams, “Epistemologies of Non-Forecasting Simulations, Part I: Industrial Dynamics and Management Pedagogy at MIT” (Thomas and Williams 2009).
When MIT established its new School of Industrial Management (SIM) in the early 1950s, the institute’s administrators sought a signature approach to the subject reflecting its strengths in science and engineering. This search moved from operations research (OR) to Jay Forrester’s “industrial dynamics”. In the end, neither approach became the distinguished approach to management that MIT sought, though SIM and OR would both become individually successful within the Institute.
The last part of this post puts this story in the context of the more successful effort of the Carnegie Institute of Technology to develop a high-profile program for its Graduate School of Industrial Administration, which was established around the same time. Carnegie Tech’s approach to management had strong intellectual connections with academic economics — an intellectual model that soon attracted the field of OR into its orbit. The equivalent intellectual and institutional movement at MIT was to be found in the ascendancy of its economics department.
For some general institutional background to the foundation of SIM at MIT, see the MIT website.
In 1950 GM magnate (and MIT alumnus) Alfred Sloan met with physicist, former MIT president, and chairman of the MIT Corporation, Karl Compton to discuss a large Sloan Foundation donation for the creation of a new management school at the Institute. The plan went forward immediately, and discussion quickly turned to what sort of spin MIT could lend the subject.
OR quickly rose to the top of the heap as a possible approach. MIT already had strong links to the subject. MIT physicist Philip Morse had directed the United States Navy’s wartime Operations Research Group (ORG), and had also been first research director of the Department of Defense’s Weapons Systems Evaluation Group from 1949 to 1950 before returning to the institute full time. The institute also administered the Navy ORG’s postwar successor, the Operations Evaluation Group (which still exists, but not under MIT’s administration).
But, what put OR on the agenda for SIM was a visit paid to MIT just after the Sloan-Compton meeting by Horace Levinson and Alan Waterman. Levinson was a former physicist and retired research director for Bamberger’s and Macy’s department stores, while Waterman was chief scientist of the Office of Naval Research (and soon to be first director of the National Science Foundation). They represented a National Research Council committee looking to foment interest in a civilianized version of operations research, which regarded MIT as a likely place to establish an educational program in the subject.
Yet, as Thomas 2012 relates in detail, enthusiasm for OR fell flat on SIM’s dean, and leadership in OR at MIT passed back to Morse, who went on to become the first director of the OR Center at MIT (which still exists and remains highly respected). SIM (since 1964, the Sloan School of Management) has always been heavily involved in the OR Center, but since the failure to place OR within SIM, interdisciplinarity has been a central feature of the Center’s identity.
With OR out of the picture, SIM still lacked a unique approach to management, and this is where the story picks up in Thomas and Williams 2009. Following SIM’s foundation, MIT’s “consulting professor” of electrical engineering, Edward Bowles, was brought in to advise on SIM’s development. For someone most historians have never heard of, Bowles had an extraordinarily high position during the war as the head “expert consultant” to Secretary of War Henry Stimson. Bowles had supported wartime OR, but in the 1950s he was unconvinced of its viability as a new profession.
By the mid-1950s, Bowles and a new associate dean, Eli Shapiro (a professor of finance recruited from the University of Chicago), had begun pressing the need for a unique approach for SIM. Long story short, their advocacy led to the recruiting of electrical engineer, servo-systems expert, and Project Whirlwind mastermind Jay Forrester. By the late 1950s, Forrester had formulated a digital simulation technique called “industrial dynamics” (later “system dynamics”), which he continues to promote to this day.
Forrester certainly obtained notoriety for his simulation project, particularly after it became embroiled in the “limits to growth” controversy of the early 1970s. However, system dynamics never found a major place in the mainstream of simulation and other forms of technical policy analysis. Oddly, though, Forrester has remained steadfastly convinced of the particular value of system dynamics among other simulation techniques. In Thomas and Williams 2009, I argue that Forrester’s commitment to system dynamics is to be found in his understanding of it as a highly flexible pedagogical and learning tool appropriate for the mission of SIM. Unlike OR or many computer models, it was not a tool to be used by expert advisers to policymakers — it was a tool to be used by policymakers themselves to correct and expand on their personal understanding of how the systems they govern operate. Unsurprisingly, given this background, some of Forrester’s most recent advocacy for system dynamics casts it as a useful tool for K-12 education.
Ultimately, while both the OR Center and the Sloan School of Management have been great successes for MIT, MIT’s effort to establish its own special take on management was never very successful. This failure can be contrasted with the Graduate School of Industrial Administration (GSIA) founded at almost the same time at Carnegie Tech in Pittsburgh, with an endowment from Gulf Oil founder William Larimer Mellon. (It would be interesting to look at these developments from the vantage-point of educational philanthropists.)
Hunter Heyck, Marion Foucarde and Rakesh Khurana*, and Judy Klein have given us a good starting portrait of the GSIA and its signature intellectual program. At Carnegie Tech in this era, figures such as Herbert Simon, Allen Newell, William W. Cooper, Charles Holt, John Muth, Franco Modigliani, and Abraham Charnes developed an intensely academic approach to management and organization studies, economics, and cognitive science, which successfully distinguished their institution. In fact, Simon and Modigliani would win Nobel Memorial Prizes in economics, followed by several other second-generation economists at the GSIA.
I briefly discuss some of Simon’s lesser-known work on inventory and production control theory in both Thomas and Williams 2009 and Thomas 2012. But, it is Klein’s (unpublished) work, which argues in the most depth that the specific thing that distinguished the work at Carnegie Tech was a commitment to developing a theoretical understanding of practical decision-making processes. Seeking out complicated decision-making problems (in the military, in business, in games of chess, or wherever), theorists could then study the general logical principles governing “best” or common solutions to them, whether at the level of the macroeconomy, the firm, or the individual mind. Since the objective was to articulate a body of general concepts (usually expressed symbolically), research results could accumulate within the framework established by a well-organized literature where the state-of-the-art in concept articulation was recorded.
The Carnegie Tech work certainly set one important model for the growth of academic OR — including at MIT — but the clearest point of comparison at MIT is not with the OR Center or SIM, but with its economics department, where Paul Samuelson and Robert Solow became extraordinarily influential figures. Happily, historians of economics — see, particularly, Beatrice Cherrier’s preliminary history (pdf) — are undertaking a concerted effort to understand better the history of MIT economics. Hopefully this effort will plug nicely into Klein’s.
What we have yet to obtain is a detailed picture of how theoretical projects in the engineering and social sciences (OR-economics-cognitive psychology, or Keynesian-Carnegie-MIT-Chicago-Cowles) related to each other. I am also yet to see an adequate picture of how new theoretical tools related to the practical toolkit that was taught in management and engineering education, and used selectively by engineers, consultants, and managers. Joel Isaac has written recently about the “tool shock” in the social sciences of this era, but I reckon we can all agree that we still need to get a better handle on the full array of tools that were used, how they compared with each other, and what they were used for, before we can develop more compelling historical conclusions. But the potential for genuinely insightful historical work here is presently very strong.
By the way, also recent and relevant: Q&A with Ivan Moscati at the History of Economics Playground
*Foucarde and Khurana offer the most detailed institutional history we have of the foundation of the GSIA in “From Social Control to Financial Economics: The Linked Ecologies of Economics and Business in Twentieth Century America,” Harvard Business School Working Paper, 11-071 (pdf).