Over at the Washington Examiner, Alexander Salter has a piece about how we should reject calls by the national conservatives to implement industrial policy. His case doesn’t rest on any claim that industrial policy is not doable but, rather, precisely on the claim that it is doable, but not for the reasons, we, free market people often claim.

Here’s Salter:

Classical liberal errors about industrial policy stem from a misreading of Ludwig von Mises’s and Friedrich Hayek’s seminal contributions to economics. Supposedly the same knowledge problem that plagues comprehensive economic planning applies to industrial policy, too. This is false. Mises and Hayek proved governments cannot plan economic efficiency because private property and markets are the sources of the knowledge necessary to “solve” the planning problem. But as we’ve seen, national conservatives have much more modest goals. They want more factory workers and more of what factory workers produce. Direct subsidies, tax credits, and similar policies are fully capable of achieving this.

I’ll not address Salter’s claim that the knowledge problem presents no obstacles to industrial policy here. Others have already addressed this claim.

But it’s still worth noting that the whole point of Don Lavoie’s 1985 book National Economic Planning: What is Left? is precisely to correct misunderstanding of the sort that appears in Salter’s op-ed. The knowledge problem isn’t born from questions about the size or scope of central planning. Whether it’s partial, with “modest goals” as Salter notes, or comprehensive, any attempt by government to override the pattern of resource allocation determined by markets will suffer from the knowledge problem. Moving on.

My bigger issue with the piece is that Salter seems to be fighting a strawman. Indeed, no one is saying that industrial policy is impossible to implement. All we are saying is that such industrial policy, much like comprehensive central planning, fails over time because resources aren’t allocated through the price system but instead are allocated through the political process, and hence the plans suffer from public choice problems and knowledge problem (See pages 96 to 100 of Sam Gregg’s excellent forthcoming book, The Next Economy, for a comprehensive discussion on this issue). That all but guarantees all the disastrous outcomes that many scholars have warned about and that we have experienced in the past when these policies have been implemented. For instance, a closer look at industrial policy projects, even the ones labelled “a success” are often found to have economic and political costs that almost always far exceed their intended benefits. (See this excellent piece by Scott Lincicome on the reasons why he opposes using industrial policy to create manufacturing jobs).

Now, I get that Cass and company think that these costs are worth incurring for the sake of getting the allocation that they prefer. Salter notes this. But, in practice, they may be surprised at how hard these costs are to ignore (especially, politically) after a while.

Finally, I caution Salter about an even more basic point – namely, about his claim that direct subsidies, tax credits, and loan guarantees will indeed surely result in more factory workers and more output from these workers. He makes it sounds as if this outcome is a given. Spend the money and you get the jobs.

After spending over twenty years in the trenches of cronyism research, I can tell you that, first, in practice industrial policy will be riddled with cronyism, and second, that this cronyism obstructs the ability of the policies even to deliver on their “modest” goals, such as that of creating and sustaining more manufacturing jobs. In other words, it’s not impossible, but it is harder than it seems on paper. There are many reasons for this, but the main reason is that special favors from government either go 1)  to companies that don’t need them and are already doing what they are now subsidized to do (such as Intel getting massive subsidies from the CHIPs Act for building a plant it would have built even without the subsidies), or 2) to companies that will sooner or later fail (Solynda comes to mind).

Think about the 1705 energy loan program. The loan-guarantee program was supposed to encourage new entrants into the green-energy business and create many jobs. In practice, most of the money went to companies that were already in the green-energy business, already employing people for these jobs. These firms were paid to do things that they were already doing. Sure, the government would claim that the money ‘created’ or ‘supported’ X number of jobs, but this claim is bogus because these jobs existed independently of the handouts. As for the rest the money, it was showered on companies that afterwards went under, hence creating no jobs.

I could tell a similar story for each of the other many corporate-welfare programs I have studied, whether it be the Export-Import Bank, farm subsidies, or the Small Business Administration. And don’t get me started on subsidies to FoxxCon ($3.6 billion in state subsidies and a total failure to create the 13,000 workers promised or build a new plant). Other state governments’ economic-development subsidies are no better and I assume national conservative subsidies to create manufacturing jobs could very well go the same way.

Salter also seems to assume that there are plenty of manufacturing jobs out there and that anyone can become a manufacturing worker. Remember when President Obama realized that there weren’t enough shovel-ready projects to make his stimulus plan work as promised? In practice, research showed that stimulus recipients weren’t hiring people from the unemployment lines (and hence, not putting unemployed contractors in infrastructure jobs) but, instead, poaching trained employees from other companies. Hence, the impact of the “targeted” government spending wasn’t more people employed. I suspect the same may be true while trying to create manufacturing jobs with subsidies.

The good news is that while industrial policy isn’t the way to go, there is a lot government officials can do to connect more workers to the workforce and connect them with better opportunities, especially in the most affected areas in the United States. Indeed, before policymakers rush to implement industrial policy programs, even modest ones, they should acknowledge that some of the challenges workers face today are often created by existing government programs. The list of potential reforms that would better the lives for those who have been frozen out of the gains enjoyed by most workers is too long for this post. But here are a few nonetheless.

Many of these barriers exist at the state level (occupational licensing, zoning and land use regulations, and such). Yet, there are many things the federal government could also do, like removing the Trump tariffs and other trade remedy regulations include tariffs, import taxes, antidumping, countervailing duties, and safeguards. The federal government could also reform the disability insurance program and other programs that at the margin create disincentives to work. It could deregulate the permitting process to allow Americans to build more infrastructure, produce more energy and more drugs. And so much more…

Veronique de Rugy is a Senior research fellow at the Mercatus Center and syndicated columnist at Creators.