Shaping science policy
Forged Consensus: Science, Technology, and Economic Policy in the United States, 1921-1953, by David M. Hart. Princeton, N.J.: Princeton University Press, 1998, 267 pp.
This is a path-breaking book, destined to influence subsequent academic discussion and historical interpretation of its important topic. David Hart, for the very first time, analyzes the history of public policy debates over the role of government in U.S. science and technology policy in the first half of the 20th century.
Hart sorts the major historical figures involved in U.S. science and technology policymaking over the period 1921 to 1953 into five bins. There are the “conservatives,” who hold that the state has no role in science and technology other than guaranteeing intellectual property rights and getting out of the way. Then there are the “associationalists,” who believe that because markets sometimes fail as a result of poor information, we need to create hybrid publicly supported but privately controlled institutions that develop this information and transmit it to the private sector. There are “reform liberals,” who see dangers in the private concentration of economic power and call for the government to step in to regulate or even directly participate in the market.
“Keynesians,” in Hart’s view, believe that the economy suffers from occasional mechanical malfunctions and that the role of government is to shove it back on track when required. He further distinguishes between “commercial Keynesians,” who believed that appropriate technological innovation would naturally emerge from firms if the government got the macro stuff right, and “social Keynesians,” who believed that “market failures in provision of science and technology for economic purposes appeared far more widespread.” Hart argues that “commercial” and “social” Keynesianism merged in the late 1940s and that the resulting merged Keynesian prescription for science and technology policy was for federal funding for academic science and the provision of subsidized funding to small business.
The last box belongs to partisans of the “national security state.” In this view, national security comes before all other goals of government. The development of military technologies to support security trumps all other arguments, and any effects in the economic sphere are the incidental (if not entirely unwelcome) consequence of the security imperative. Hart argues that the security statists and the converged Keynesians found common ground and formed the coalition that put into place the policies of the early 1950s, the so-called “postwar consensus.”
Back to the beginning
Hart’s history begins with the little-known history of U.S. science and technology policy before World War II. His account revolves around, of all people, Herbert Hoover, the “Great Associationalist,” taking the first pass at an activist technology policy in the 1920s while serving as Secretary of Commerce. Beefing up the National Bureau of Standards (NBS) (predecessor of today’s National Institute of Standards and Technology), Hoover beat back opposition from conservatives in the Treasury and elements of the private sector to make the NBS the centerpiece of a major technology partnership program with U.S. industry.
The Great Depression and the New Deal in the 1930s, Hart argues, saw the associationalist program evolve into crude protectionism. With the death of the National Recovery Administration at the hands of the Supreme Court in 1935, the associative state created by Hoover and continued by Roosevelt ground to a halt. With conservatives on the run in Washington, says Hart, liberal reformers seized the opportunity to develop new government programs to stimulate the economy through technological progress. The brightest star in this firmament was the Tennessee Valley Authority and its efforts to build a technologically advanced fertilizer industry in the South.
By the late 1930s, however, the center of gravity for technology policy, which Hart sees moving into the hands of the “reform liberals,” had shifted to the Department of Justice. There, reform liberal and assistant attorney general for antitrust Thurman Arnold embarked on a crusade to “break the patent bottleneck” and prevent entrenched corporations with a few key patents from blocking the emergence of new technologies and industries and thus suppressing the expansion and modernization of markets. Antitrust enforcement was combined with new financing and R&D programs in the construction industry, which Hart describes as a key example of New Deal reform liberal policies in action.
With the rise of the Keynesians in the early 1940s, however, with their focus on macroeconomic forces and demand management, what at first had been a loose collaboration evolved into a policy competition. “War mobilization briefly delayed the estrangement of the reform liberals from the Keynesians,” writes Hart, because “national security was a potent weapon in the battle for both public investment and antitrust enforcement.” With the war on, however, the military and corporate America joined and gained operational control over a vastly expanded government investment apparatus that “shoved ‘Thurman Arnoldism’ aside,” while embracing the Keynesian macroeconomic agenda. Hart says that in the early 1940s, social Keynesians, particularly Alvin Hansen, argued that “secular stagnation”–a characteristic of an economy dependent on technological innovation but incapable of generating a sufficient quantity of it–was a serious problem and supported the reform liberal agenda of antitrust, R&D funding, and public investment. By 1948, notes Hart, Paul Samuelson had turned this idea on its head, writing in his introductory economics text that “secular exhilaration,” not stagnation, was the problem of the postwar years. With the focus now squarely on aggregate demand management, social Keynesianism and commercial Keynesianism had fused into a new synthesis with little interest in technology policy, or for that matter, any policy based on microeconomic analysis of particular industries or problems.
The Cold War and Korea saw the rise of the national security state–in some sense, the institutionalization of an alliance between associationalists and committed believers in technological investment as the key to military advantage that was forged in the 1940s. Converged Keynesians played a supporting but largely passive role as managers of aggregate demand. This latter period, which included the struggles over control (in Hart’s analysis, largely between Keynesians, associationalists, and reform liberals) that delayed the creation of the National Science Foundation, takes us to more familiar ground.
But Hart is surely right–in fact, probably does not make the point strongly enough–that the history of the alleged postwar consensus in subsequent decades is far from the stable, smooth execution of an agreed-upon consensus model. He sees constant tugs and pushes, zigs and zags in U.S. science and technology policy, propelled by policy entrepreneurs promoting agendas that are the lineal descendents of the same forces he documents in the first half of the 20th century.
One problem with Hart’s analysis is that he squeezes a very fuzzy and amorphous reality into five very sharply bounded boxes in classifying policy and policy entrepreneurs. Although the five viewpoints he enumerates are, on reflection, useful in defining the sometimes antagonistic forces shaping U.S. science and technology policy, they also confuse and confound historical understanding at times. For example, Vannevar Bush, a gigantic figure in the policy debates, is portrayed as an associationalist in the late 1930s, a social Keynesian in the mid-1940s, and a converged Keynesian in the early 1950s. I question whether Bush would have identified himself as any one of these things, or whether his views changed as sharply as these reclassifications suggest. When Bush made common cause with Frank Jewett of AT&T, Hart’s archetypal “conservative,” were they simply cutting a deal, or is it not possible that both men actually had relatively complicated and nuanced views of the world, with convergence on some concrete questions reflecting genuine agreement outside the sharply drawn bounds of Hart’s simple taxonomy?
Another problem is that the study does not target a wide audience. The book originated as Hart’s Ph.D. thesis and bears the resulting baggage. The author drags the reader down occasional meandering byways as he labors mightily to show the novelty of his theoretical framework, which boils down to the observation that policy is made by a political process rather than being the determinate outcome of a rational and optimal choice. This may be hot stuff in political science departments, but most policy-savvy readers will be distinctly unimpressed when he writes that “I theorize science and technology policymaking as a political process.” Nonetheless, the book works well as an excellent and path-breaking political history and deserves a wide readership.
There is one glaring omission in this important book. Hart completely overlooks a major development in the economic theory of science and technology investment that occurred in the late 1950s and early 1960s and that profoundly altered the nature of the policy debate. Over this period, economists (in particular, Richard Nelson, Charles Hitch, and Kenneth Arrow) for the first time precisely articulated a reason for market failure and a theoretical rationale for active government policy: the inability of those investing in science or technology to completely capture the economic fruits of this investment. This was a precise and persuasive diagnosis of market failure, far removed from vague hand waving about secular stagnation. Other economists (notably Zvi Griliches and Edwin Mansfield) then produced pioneering empirical work demonstrating that this theoretical argument appeared to be significant in real life, reflected in large gaps between social and private returns on technology investments. This research directly motivated a new interest in a more active role for government in funding research in the 1960s and many more such proposals over the years, as the tides of policy in an increasingly high-tech economy ebbed and flowed.
Indeed, so great was the impact of this new scholarship that the meaning of “conservative” changed. In the 1920s, it was indeed possible for a reasonable conservative to argue that the government should simply get out of the way and do as little as possible in the realm of science other than to strengthen and enforce intellectual property rights. Today, one is hard pressed to find a conservative, or at least a conservative economist, who does not concede that the government has some appropriate role in supporting basic scientific research, and particularly academic basic research, as the least easily appropriated investment in technology.
Kenneth Flamm ([email protected]) is Dean Rusk Chair in International Affairs in the Lyndon B. Johnson School of Public Affairs at the University of Texas at Austin.