Editor’s Note: The author delivered these remarks, as prepared, as his opening statement in a debate with Richard Reinsch at the National Conservatism Conference in Washington, D.C. on July 14, 2019.
I’d like to begin with a quote attributed to Michael Boskin when he was Chair of George H. W. Bush’s Council of Economic Advisers: “Computer chips, potato chips, what’s the difference?” My argument, in its simplest form, is that there’s a very large difference, and policymakers should take it into account.
My argument rests on three claims, moving from economics to policy and then politics:
First, that market economies do not automatically allocate resources well across sectors.
Second, that policymakers have tools that can support vital sectors that might otherwise suffer from underinvestment—I will call those tools “industrial policy.”
Third, that while the policies produced by our political system will be far from ideal, efforts at sensible industrial policy can improve upon our status quo, which is itself far from ideal.
Why Manufacturing Matters
Economics first: Why do we care about the economy’s composition? For one thing, it has serious distributional consequences. If we only care about consumption, we might believe we can remedy those with redistribution—let the wealth get created by people working in tech or finance in just a few cities, then collect taxes from them and mail checks to everyone else. But if we value, as we should, the ability of individuals, their families, and their communities to participate as productive contributors to society, then our economy needs to generate good opportunities for workers of different aptitudes in different places.
We also care because some industries matter more for the economy’s health and long-run trajectory. Some achieve greater productivity gains. Some rely on, and foster, broader ecosystems of researchers and specialists, suppliers and customers. Some generate larger “multiplier” effects for other employment.
Manufacturing, from these perspectives, is particularly important. Our popular obsession with manufacturing isn’t some nostalgic anachronism. (Here I use “manufacturing” to encapsulate the sector of our economy that makes physical things—traditional manufacturing, resource extraction, energy production, agriculture, some construction, and so forth.) Manufacturing provides particularly well-paying, stable employment—especially for men with less formal education. Manufacturing also tends to deliver faster productivity growth because its processes are susceptible to technological advances that complement labor and increase output.
Echoing Boskin on potato chips, President Obama’s CEA Chair, Christina Romer, once wrote: “consumers value haircuts as much as hair dryers.” Fair enough, for consumers. But not for workers. A barber today is barely more productive than one of past generations, while someone making hair dryers might help churn out ten times the product of his predecessors (assuming, of course, we still made hair dryers here). If that barber wants his wage to rise, he’d better hope his customers are productivity-gaining hair-dryer makers, not just personal-services providers themselves.
Note also that, while our economy can be predominantly services-based, not everyone can cut each other’s hair. If the local hair-dryer factory moves overseas and the laid-off workers all try to shift into local services, the community will face a rather existential crisis: what will it send the rest of the world, in return for all the things that it wants the world to send it?
When communities lose manufacturing—which is not the only form of tradeable production, but certainly the primary one—they begin to “export need.” You see this across America, in the dilapidated shopping centers that still have sparkling occupational therapy offices. They are literally the exporters for those towns, exporting to the nation’s taxpayers the care of local residents on disability. That’s how the community attracts resources. This might look fine in the aggregate consumption data, but we should not consider such outcomes equal, or acceptable.
Finally, manufacturing is unique for the complexity of its supply chains and the interaction between innovation and production. One of the most infuriating face-palms of modern economics is the two-step that goes like this: First, wave away concern as other countries with aggressive industrial policies that attract our critical supply chains overseas, explaining that it doesn’t matter where things get made. Second, wait for people to ask “why can’t we make this or that here,” and explain that of course we can’t because all of the supply chains and expertise are entrenched elsewhere. It’s enough to make one slam one’s head into the podium.
We are also discovering that innovation and production are not so easily disaggregated. Where manufacturing goes, research and design follow. Here I’ll quote Andy Grove, long-time CEO of Intel:
Our pursuit of our individual businesses, which often involves transferring manufacturing and a great deal of engineering out of the country, has hindered our ability to bring innovations to scale at home. Without scaling, we don’t just lose jobs — we lose our hold on new technologies. Losing the ability to scale will ultimately damage our capacity to innovate.
So, computer chips, potato chips, hair dryers, haircuts… the differences are enormous. Still, the case for industrial policy requires recognition not only of certain sectors’ value, but also that the market will overlook the value in theory and that we are underinvesting in practice.
That the free market will not solve this should be fairly self-evident, and is entirely consistent with economic theory. The long-run value, both social and economic, of a robust manufacturing sector is of no concern to a given individual allocating his own resources. It has nothing to do with the most efficient allocation of resources at any point in time. It does not offer a higher return on capital.
And sure enough, we are suffering from a failure to invest. Output growth in the manufacturing sector itself has slowed dramatically and employment has plummeted, even as compared to other developed economies. Manufacturing output is only 12% of GDP in America, compared to 19% in Japan and 23% in Germany. Economy-wide, private-sector investment has declined, so much so that the private sector is now a net lender in the economy and only the government borrows. Productivity growth has slowed nationwide, even flatlining in recent years. Wages have stagnated. Our trade deficit has skyrocketed, even in advanced technology products like life sciences and electronics.
That is the economic case for industrial policy: manufacturing is important, markets ignore this, and we are paying the price.
A Path Forward
It’s one thing to identify a problem, quite another to suggest we can do anything about it. The good news here is that we can—indeed, not by coincidence have other developed, market economies like Germany and Japan chosen to adopt industrial policies and also reaped the rewards we might expect.
I will describe briefly the types of policies we should consider, from least to most aggressive; Richard can draw the line where he begins to object and we can debate specifics from there:
- Fund basic research across the sciences
- Fund applied research, focusing in fields like advanced materials, robotics, and logistics
- Support private-sector R&D and commercialization with subsidies and specialized institutes
- Emphasize vocational education and target higher-education support at relevant disciplines
- Give engineering majors greater assistance than comparative literature majors
- Increase infrastructure investment and reduce regulatory burdens on it
- Fast-track approvals of projects that expand industrial capacity
- Bias the tax code in favor of profits generated from the productive use of labor
- Retaliate aggressively against mercantilist countries that undermine market competition
- Reduce Chinese student visas toward zero until Chinese policies change
- Tax foreign acquisition of U.S. assets, making U.S. goods relatively more attractive
- Impose local content requirements in key supply chains like communications
Certainly, the policies that emerge from our political process will be imperfect, opportunities for regulatory capture will abound, market distortions will emerge. But here we arrive at the third and final point: for all the limitations of our politics, adoption of an industrial policy will improve upon the status quo.
One reason to believe this is to observe that other market democracies like Germany and Japan have pursued the approach successfully. Also probative, we should admit, is China. Of course, China’s circumstances are radically different from America’s. But it would be hard to reconcile an assertion that policy support for a manufacturing sector makes it inevitably weaker with the actual experience of the Sino-American trading relationship over the past two decades.
Within the American context, we should work toward the policies that we believe can make a positive difference, even recognizing that the result will be imperfect. Libertarians often posit an ideal world of policy non-intervention as superior to the messy reality of policy action. But that ideal does not exist—messy reality is the only reality and we should not give preference to our existing mess, built on an incorrect understanding of our economic challenges, over one that at least aims closer to the right direction.
That’s especially the case here, because you can have free trade, or you can have free markets, but you can’t have both. The market fundamentalists who insisted on eliminating the barriers between our market and China’s have, by their policy choices, introduced massive distortions into our market. Insisting on allowing the distortions, and then announcing that a dislike of distortions precludes any response, is irrational. Insisting that our workers, alone, fight with their hands tied behind their backs is frankly immoral.
In the real world as we find it, America has no choice but to adopt an industrial policy, and we will be better for it.