Industrial Policy Advocates Should Learn From Don Lavoie
By Adam Thierer
Industrial policy is experiencing a major revival in the United States and other countries. In the U.S., several major federal proposals are pending or already set to go into effect:
In June, the Senate passed a massive 2,300-page bipartisan bill, the “U.S. Innovation and Competition Act,” with a $250 billion price tag. Of that, $52 billion goes to the semiconductor industry, and billions more are allocated for government-created “regional innovation hubs” and other new programs. Lawmakers loaded up the measure with favors for their districts and special interests, prompting Sen. John N. Kennedy (R-La.) to describe the effort as an “orgy of spending porn.”
The Biden administration has recently proposed new subsidies for semiconductors and released a 250-page report on how government can help reorganize industrial supply chains. In late July, the Commerce Department also announced plans to allocate $1 billion in pandemic recovery funds to create or expand “regional industry clusters” as part of the administration’s new “Build Back Better Regional Challenge.” The agency’s list of possible winning funding ideas includes an “artificial intelligence corridor” and a “climate-friendly electric vehicle cluster.”
Another new bill was introduced in the Senate in mid-August to create the Industrial Finance Corporation of the United States, which was designed to help America “build the factories of the future” to compete against China.
An assortment of other recent measures have proposed targeted industrial policy steps for specific sectors, including the “Securing America’s Critical Minerals Supply Act,” the “Help Onshore Manufacturing Efficiencies for Drugs and Devices Act” and the “Pharmaceutical Supply Chain Defense and Enhancement Act.” Recent infrastructure legislation has also included a variety of industrial policy provisions.
While industrial policy advocates all propose big plans for economic reorganization, these plans lack consistency in terms of the preferred direction to steer technological innovation and industrial output. Moreover, proponents of grandiose industrial policy schemes cannot even seem to agree on a common definition of the term.
Enthusiasm for industrial policy—and confusion about what it actually means—are nothing new. Economist Donald C. Lavoie documented this same sort of intellectual schizophrenia almost 40 years ago, when the topic was red-hot in the 1980s. With America now embarking on a bold new experiment with many different flavors of industrial planning, it is worth revisiting some of the important lessons that Lavoie’s work teaches us about industrial policy, lessons that still are largely unheeded today.
Don Lavoie taught economics at George Mason University and wrote extensively on comparative economic systems and the role that knowledge and institutions played in them. Sadly, Lavoie passed away 20 years ago on November 6, 2001, at the height of his career. His legacy lives on, however, both in his lasting scholarly contributions to economic theory and also in the work of the many students and colleagues he inspired and who carry on his academic tradition.
Some of Lavoie’s most important research involved his thorough exploration of national economic planning efforts. He wrote comprehensively about all aspects of state planning and identified the many reasons those efforts were bound to fail. Much of this scholarship focused on variants of Marxist planning practiced by socialist nations, but he also explored industrial policy planning efforts, which were all the rage in many countries throughout the 1970s and 1980s, including the U.S.
Industrial policy advocacy fell out of favor in the United States in the 1990s, especially after the “Japan panic” of that era became something of a joke by the end of the century. For a time, the Japanese model of industrial policy planning was deeply feared by American pundits and policymakers, leading to calls for government-led efforts to develop comparable planning efforts. But Japan’s planning efforts imploded and became viewed as such a costly bust that even the Japanese government itself concluded in 2002 that “the Japanese model was not the source of Japanese competitiveness but the cause of our failure.”
Some failed ideas never die, however, and industrial policy has experienced a rebirth in recent years. New proposals seem to pop up almost weekly. Lavoie’s work can help us make sense of these efforts and explain why they will likely come up short of their lofty ambitions.
The Economy Is Not a Machine
Machine metaphors are common in industrial policy discussions. Advocates of state-led planning efforts often imagine government planners are using dials and levers of a computer or device to finely calibrate innovation and growth in certain sectors.
Don Lavoie. Image Credit: Mercatus Center at George Mason University
Lavoie believed that such metaphors fail to explain the actual workings of an economy and end up doing more harm than good. “The point is that such relations between the health of different sectors of a modern economy are so intricate and complex that it is the height of pretense to claim that any single agency could take them all into account in its decisions to reallocate credit to certain sectors,” he argued. Moreover, “any argument for offering subsidies in the form of cheap credit to some favored industries, whether old or new, is also an argument for penalizing other (possibly unidentified) industries.”
Lavoie was highlighting how the so-called knowledge problem haunts economic planning efforts, including industrial policy measures. When governments use public resources to favor one sector or technology (i.e., pick “winners”), how might those resources have been better spent? Who does the choosing, and on what grounds? And why should we trust their judgment over that of market actors?
Both then and now, however, industrial policy advocates attempt to sidestep knowledge problem concerns by suggesting that they are not interested in comprehensive planning but merely in more limited or targeted interventions. The problem, Lavoie pointed out, is that they cannot so casually dismiss the knowledge problem, nor can they ignore the longer-term ramifications of failed planning efforts that would necessitate still more economic interventions:
[I]ndustrial policy represents an attempt to accomplish an end—the rational formulation of priorities for a nation’s investment decisions—without supplying the means necessary for that end—the gathering of comprehensive knowledge about the detailed interconnections of the economic system, and the power to alter these details. This means that one can confidently expect the initial policy measures of the investment-guiding agency to fail and to require serious modifications. The nature of these modifications will depend on the policymakers’ diagnosis of the failure. But since the advocates of industrial policy seem to presume that whatever degree of order an economic system attains must be due to deliberate coordinating policies of government rather than to spontaneous competitive forces it seems likely that their diagnosis would be that they lack sufficient central power and control over the economy. So while industrial policy may not be a form of central planning, as such, it could well be a big step in that direction.
His point is that planners cannot completely ignore the grim reality associated with the economic calculation debate: namely, we will always face knowledge constraints. Even if government planners could somehow collect every piece of information about the workings of an economy one day, it does not mean they will become super-forecasters who can accurately predict the behavior of individuals, organizations or other economic variables the next day. “[T]he complex workings of a market economy are driven by the mutually influenced choices of the millions of human minds that comprise it,” Lavoie argued, and societal needs and wants are constantly changing. Markets and rivalry solve these knowledge problems over time—not perfectly, but better and faster than planners ever could, he showed.
When confronted by the failures associated with previous government interventions, the planners’ response is typically of the we-can-do-better variety, with the result being more five-year plans and bureaucracies as one intervention begets another and another. This tendency was (perhaps somewhat surprisingly) best identified by Paul Krugman in a scathing 1983 analysis of industrial policy efforts. Krugman observed that there is “a familiar proposition from the literature on economic development that distortions due to government action may make other offsetting government actions desirable.” To the contrary, he argued, “the appropriate response to government-induced distortions is to try to minimize them, not to target particular industries in which the country underinvests.”
Alas, few industrial policy pundits heed that advice, preferring to assume that this time will be different.
A Tool for Comparing Industrial Policy Proposals
One of the most interesting things about Lavoie’s critique of different planning schemes in the 1980s was the way he contrasted some of the leading frameworks set forth by major industrial policy advocates of the period. He dissected their proposals and then observed the many inconsistencies among them. These differences raised some obvious questions: Whose plan are we supposed to follow when proposed plans conflict? And how much stock should we place in the wisdom of industrial policy when the leading advocates cannot even agree on what sectors and technologies are worth preserving or promoting?
To illustrate the key differences among industrial policy advocates, Lavoie created a helpful tool he called the planning spectrum, illustrated below. On one axis, he plotted “futurist” versus “preservationist” advocates and proposals, with the futurists wanting to invest in new skills and technologies, while the preservationists seek to prop up existing sectors. On the other axis, he contrasted “left-wing or pro-labor” and “right-wing or pro-business” advocates and proposals.
Lavoie’s planning spectrum remains a useful tool that helps explain what drives today’s industrial policy discussions. For example, in the wake of the COVID-19 lockdowns, the health of the American semiconductor industry became a major concern as supply chains broke down, creating shortages. As a result, some policymakers started a major push to boost that sector using subsidies and other industrial policy levers. Others have been suggesting that high-speed 5G wireless communications networks deserve more government attention and support. Some officials in the Trump administration even wanted to create a nationalized 5G network to compete with China, and they took regulatory actions to block China’s leading 5G supplier from doing business in America. These policies indicate a preservationist attitude at work.
At the same time, others have been promoting grandiose government planning and subsidies for artificial intelligence and quantum computing. The Obama and Trump administrations released similar reports outlining a National Artificial Intelligence Research and Development Strategic Plan. The Trump administration also pushed a National Quantum Initiative and proposed doubling federal R&D spending on AI and quantum computing. Meanwhile, in early March 2021, the congressionally authorized National Security Commission on Artificial Intelligence released a 756-page report detailing ways to counter China on the AI front and advocating $40 billion in new government AI investment immediately and then “hundreds of billions in federal spending in coming years.” This is the “futurist” instinct at work, with planning advocates hoping that government will wisely calibrate investments in cutting-edge sectors and technologies and gain an advantage in the next global technology race.
The other axis of Lavoie’s chart helps us identify how industrial policy advocacy is shaped by economic or political preferences. For example, Robert D. Atkinson, president of the Information Technology and Innovation Foundation (ITIF), is probably the most prolific scholar on industrial policy issues today. In his writings and recent books such as “Big Is Beautiful: Debunking the Myth of Small Business,” he unapologetically advocates for larger enterprises as the primary drivers of advanced R&D and global competitive advantage.
By contrast, President Joe Biden and many other Democrats advocate for an industrial policy vision in which labor unions and small businesses play a greater role. They are pushing the most radical antitrust agenda seen in decades, which would intentionally hobble some of America’s largest technology companies. But Atkinson, who is also a Democrat, rails against the revival of such trust-busting, explaining in a recent Wall Street Journal editorial how “Antitrust Can Hurt U.S. Competitiveness.”
Thus, while these Democrats share a technocratic desire to intervene in the economy to impose new industrial policy visions, there exists a tension among them about whether to promote innovation outcomes through larger enterprises or smaller entities. Again, to channel Lavoie, we can ask: Whose plan should we follow?
Lavoie’s planning spectrum helps us identify the tensions between the values underlying other new industrial policy proposals, such as environmental sustainability versus national security. For example, both the Democrat-supported “Green New Deal” and Sen. Marco Rubio’s “national development through 21st-century American industrial policy” represent industrial policy planning blueprints, but the values and policy priorities are radically different. Like most Republicans, Rubio thinks the Green New Deal is an effort “to repackage socialism” and fears that it “would bankrupt our nation.” But that does not stop him from endorsing grand industrial policy schemes on national security grounds, primarily as a way of countering China. In the end, it is all central planning, but the Democrats and Republicans are motivated by very different policy rationales.
The Importance of Humility
The most fundamental lesson today’s policymakers can learn from Lavoie is that humility is a high virtue. We must recognize the limits of our knowledge, especially as it pertains to predicting (or promoting) future technological and industrial outcomes. Economies are not like machines to be programmed, but rather like complex organisms or highly varied and ever-changing ecosystems.
The limitations of industrial policy exist regardless of the policymaker’s intentions. There are no “good guys” versus “bad guys” when it comes to industrial policy efforts; there are just many people with many different technocratic plans, all of which are constrained by limited knowledge and resources.
Finally, Lavoie reminds us that if you find yourself in a hole, it is wise to stop digging. Constantly doubling down on planning efforts is not going to help governments escape the problems created by their earlier interventions. Unfortunately, this is exactly what many industrial policy advocates do: They insist that America already has an industrial policy, but that it lacks the sort of conscious design or coherent form or direction they desire.
This is the ultimate folly. “In light of the inherent deficiencies of central planning,” Lavoie said, “it might be argued that the U.S. should instead try to reduce current government interference with the competitive process to the absolute minimum consistent with other political goals.” It remains wise advice for today’s policymakers.