jump to navigation

The Challenges and Opportunities of Policy Analysis History: Some Notions February 15, 2015

Posted by Will Thomas in Operations Research.

Historian of economics Beatrice Cherrier has asked what a history of policy analysis might look like. She quite reasonably notes that, as a faculty member at the Centre de Researche en Économie et Management at Rennes, she wants her “students to know why and how the theories, tools and practices they will later use on a daily basis were conceived and spread, and a good 80% of them will participate in a policy evaluation in the next 10 years.”

cover_5_2upI suspect we will be best served in answering this call if we admit the poverty of our current historical knowledge. We have a number of useful historical studies of various bits of policy analysis, and many more dislocated fragments of such a history are also to be found in the practitioner literature. However, I do not think we can even, at this stage, outline what a synthetic history would look like.

I arrive at this conclusion out of lessons learned while researching and writing my book, Rational Action, which is due out in a couple of months (and which I will not attempt to dissuade you from pre-ordering). The book uses 300 pages of text to outline the history of a cluster of influential fields—primarily operations research/management science (OR/MS), systems analysis, and decision theory—that developed in the middle of the twentieth century.

One of the key lessons learned is that many, including most historians, have been too quick to assume that they understood the basic outlines of the story as having primarily to do with these sciences’ attempts to apply “scientific” methodology to the realm of policy. Conceived in this way, the history becomes one of various attempts to define what constitutes a properly scientific approach, and of various attempts to command authority through the application of such an approach. As a consequence, the histories of very different fields become blurred together as part of a general mid-twentieth-century movement to make politics and society more scientific.

In fact, Beatrice, having read over my introduction, took me to be making exactly this point: “…Thomas argues that scientists believed that their science’s association with technology and rationality made it de facto policy relevant.”

Reading that was a bit alarming to me, because I actually rewrote my introduction a large number of times to try and convey just the opposite: invocations of “science,” while having some rhetorical (and thus substantial historiographical) importance, had very little bearing on how anyone, scientist or non-scientist, actually judged the legitimacy and utility of these sciences at that time.

Of course, a few proponents of the sciences of policy did suppose that, as people committed to the application of scientific method, they held a skeleton key to problems of policy analysis. Accordingly, they made recommendations such as the establishment of advisory groups to analyze “national political objectives” for the U.S. President, or the problems of economic development for India. These recommendations were rejected not by policymakers, but by these individuals’ colleagues, and primarily because those colleagues were skeptical that they could legitimately claim any special skill or knowledge to address problems in those areas.

One of the more hyperbolic claims about postwar operations research. Physicist Ellis Johnson, pictured, was head of the Army's Operations Research Office, 1948–1961. In 1951 he privately (and vainly) suggested to a National Research Council committee that OR be immediately implemented in the White House.

One of the more hyperbolic claims about postwar operations research. Physicist Ellis Johnson, pictured, was head of the Army’s Operations Research Office, 1948–1961. In 1951 he privately (and vainly) suggested to a National Research Council committee that OR be immediately implemented in the White House.

Legitimacy, I believe, was actually conferred through a more subtle calculus that matched (or failed to match) particular approaches to particular problems in a convincing way. Therefore, the histories of these sciences—the patterns in which they were taken up, rejected, and evolved—only really make sense if we understand the differences between them, and how they developed in response to particular problems of policy.

And here we come to our main challenge: to understand the history of policy analysis or policy evaluation, we really need to get a better grasp on the history of policy itself.

Policy is an extraordinarily broad terrain, encompassing both how policies are made and administered, as well as the various domains (e.g., health, education, military, business, etc.), levels (e.g., Congressional, departmental, local, corporate, managerial, strategic, tactical, etc.), and types (e.g., budgetary, regulatory, program building, etc.) of policy that exist. And this is leaving aside international comparisons!

In the cases of the sciences I deal with, their main successes came through their ability to identify and address very particular problems. Military operations research was primarily about the articulation of rationales underlying plans for combat operations, and conducting empirical research to discern whether those rationales could be improved upon. OR, and related activities, also contributed to the problem of developing equipment designs that accorded with military needs, as well as tactics that best made use of available equipment. Naturally, all these problems were also subject to contemplation by military planners and engineers, so the real question was where improvements to existing practices could be made.

In industry, I argue, there were fewer opportunities for generalized study in a business culture crowded with both specialized and generalized expertise. Here their main contribution came through the application of advanced mathematics to optimization problems in logistics, inventory replenishment and production control policies, and other problems of organization that turned on complex quantifiable considerations.

Some figures, eager to have a more fundamental impact on business management and policymaking, later ventured into things like “soft systems” and “wicked problems,” but these were always domains where it was difficult to legitimately claim any special expertise.* The term “policy sciences” (the reason I had to use the construction “sciences of policy”) came into being circa 1970 to connote the study of policymaking from a more sociological standpoint. Such an approach was viewed as essential for confronting the practical and political difficulties of social policy, which at that time was becoming more prevalent.

Here’s a video of soft systems pioneer Peter Checkland talking about the origins of the methodology:

Both more and less formal analytical approaches to policy have had strong academic components. I believe it is important not to assume that ideas moved in anything but the most oblique ways from the academy to policy, despite critics’ and historians’ frequent claims that in the mid-century period the abstract, formal, supposedly rationalist analytical approaches ascendant in the academy carried an unreasonable intellectual and political authority.

For example, although the famous “Measurement without Theory” (pdf) debate of the late 1940s was a clarion call for academic theorization in economics, culminating with the ascendancy of such strict theoretical desiderata as “microfoundations,” practical economic analysis followed its own track (pdf).

Similarly, on the subject of nuclear strategy, I am partial to David Alan Rosenberg’s emphasis on military policymakers’ contributions to U.S. strategic doctrine, which plays down the role of “strategic thinkers,” to whom others, such as Fred Kaplan, have given a prominent role.

Probably the most important players in policymaking and policy evaluation have continued to be people with experience in those areas who are familiar with the broad and detailed questions that go into the craft and implementation of policy. This includes civil servants and contractors, appointed administrators, and then, to a lesser extent, members of external advisory and oversight boards, and even some members of Congress with a dedicated interest in certain subjects. They make policy in consultation with a wide variety of individuals with varying forms of experience and expertise. For the most part, we don’t know who any of these people are or what they are thinking, or why they think they way they do. If we think we do, we’re probably wrong.

For historians, a crucial task would be simply to trace these people’s history, both individually, and, prosopographically (i.e., to trace trends in their training, professional background, socioeconomic status, and so forth). To venture an educated guess, these individuals would not have claimed their measure of influence through any sort of epistemic authority—most were not scientific or engineering figures—but through happenstance, shifting political and social trends, and, no doubt, individuals’ political initiative as well as their ability to formulate and think through detailed problems of administration.

We could go on, for instance to suggest just where technically, economically, and social scientifically (not to mention legally, humanistically, etc.) trained individuals did fit into the larger sphere of policy, and how that shifted through time. I will refrain since our problem is already monstrously complex. But, I argue, therein lies our opportunity.

If we accept that there is no good resource encompassing the history (and, for that matter, the present) of policies, policymaking, policy analysis. And, if we accept that history can bring clarity, by providing information about motivation, context, and reasoning, then, by making available systematic, accessible, well articulated pictures of policy problems and of the thinking of the people who normally address them, it should be possible to lend concreteness, sharpness, and openness to deliberations for those who can benefit from such concreteness, sharpness, and openness.

This sort of thing is not really amenable to pithy conclusions about the mentality of policymaking elites, which has been a temptation that has too often consumed prior history. Nor would it be fit for articles in The Atlantic or for broadcast on BBC4, which are good examples of historians’ preferred vehicles of outreach. But I do think, if done right, it could be a game changer for contemporary history as a relevant discipline.

*I owe my understanding of “soft systems methodology” and its tension with “operational research” in Britain to several works by Maurice Kirby on this topic.


No comments yet — be the first.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: