Critiques of national-conservative economics based on the ‘knowledge problem’ usually miss the mark.
Critiques of national-conservative economics based on the ‘knowledge problem’ usually miss the mark.
D riven by coalitional fissures, many conservatives are rethinking long-established principles. One of them is the primacy of free enterprise. Whereas it used to be conservative orthodoxy that allocating resources by markets rather than government was essential for prosperity, national conservatives have begun to question how far this presumption should go.
While certainly not as hostile to free enterprise as their progressive counterparts, national conservatives are not particularly shy about using public power to supersede or supplement market forces. A prominent item on their agenda — often called “common-good conservatism” — is industrial policy. The national interest requires higher employment and output in traditional manufacturing sectors, they argue. If markets won’t provide, government can and should put a thumb on the scale.
Classical liberals are skeptical of this economic mindset in general and of industrial policy in particular. They argue that there are serious incentive problems with top-down economic control, even by those with the best of intentions. Bureaucratic regulation, for example, is susceptible to capture by the very interests that regulators are supposed to check. This is commonplace in health care and finance. There’s no reason to suppose that it would work differently for manufacturing.
Yet while bad incentives are certainly an important part of the industrial-policy critique, they take a back seat to an issue raised by older, fusionist conservatives: the knowledge problem, which is the idea that it’s impossible for the government to acquire and use the necessary information to bolster industry in the ways that national conservatives desire.
I’m an old-school conservative, unashamedly pro-market and distrustful of activist government. Yet despite industrial policy not being my cup of tea, I cannot get behind the knowledge-problem critique for one simple reason: It does not apply to contemporary industrial policy. While I think its goals are undesirable, I also think that they are possible.
To see why, we need to revisit the economic controversy surrounding the knowledge problem. Let’s take a trip back to the early 20th century, when men of incredible intellect debated the foundations of economic systems.
Socialism, Capitalism, and the Knowledge Problem
The greatest economic debate of the past century — perhaps the most important in the history of economics — was capitalism versus socialism. Should productive resources be privately owned and allocated by exchange in the market? Or should those resources be publicly owned and allocated by public authorities?
For some time, many intellectuals saw socialism as the preferable option. Capitalism appeared to have serious defects. Markets permitted wasteful competition via duplication of efforts among firms. The economy periodically swung between irrational booms and crippling busts. Morally, the distribution of income seemed inegalitarian and arbitrary. By directing resources according to a rational, all-encompassing plan, socialists thought they could engineer away these difficulties and achieve an era of abundance. In other words, they thought that socialism could beat capitalism at its own game. Command economies could produce greater abundance than voluntary economies, socialists claimed.
The important thing about this debate was that both sides accepted the same goal. Economic systems were judged according to efficiency: their ability to allocate resources to their highest-valued uses. Other goals, such as equality, played a supporting role but weren’t the main battleground. Tertiary concerns could be accomplished through redistributive policies (taxes and transfers), but these programs were not unique to socialism. Socialists and capitalists understood that the debate hinged on the capacity to produce wealth — not just total goods and services, but the right mix of goods and services.
The socialists were wrong, of course. It’s impossible for centralized, command-and-control schemes to produce more efficiently than decentralized, voluntary schemes. Ludwig von Mises struck the first blow by proving that coordinating production and consumption requires private property, market prices, and profit-and-loss accounting. Without property rights — which socialism does away with — businesses cannot know whether they’re producing the most desired goods in the most affordable way.
Hayek continued the assault by showing how the price system reconciles the decisions of households and businesses. Economic actors possess only a tiny fraction of the knowledge necessary to harmonize consumption and production. Thanks to prices, though, they do not need to know any more. Prices act both as knowledge surrogates and knowledge substitutes. By communicating through market prices, buyers and sellers can say far more than they can comprehend. As a result, the average man can enjoy a standard of living greater than the kings of centuries past.
Unlike his contemporaries, Hayek saw that the true economic problem is about knowledge, not allocation. He both posed the dilemma and demonstrated how capitalism solves it. How can the information scattered among the economy’s millions of participants be harnessed to bring order to the whole system? In other words, how do we get efficient production, despite nobody knowing anywhere near enough to produce efficiently? The answer is prices. Thanks to Hayek, we know that prices aren’t just exchange ratios. They’re irreplaceable pieces of information in a vast communications network, which we ignore at our peril.
The Real Knowledge Problem
However, not every attempt to circumvent the market runs up against the knowledge problem. Hayek was replying to a specific group of economists who insisted that the tools of neoclassical economic theory made central planning feasible. As Hayek proved, socialists begged the question by assuming they had the relevant information to plug into their cost and revenue equations. But their errors do not necessarily mean that other government interventions are impossible. Hayek himself famously defended legislation to protect workers and establish a guaranteed minimum income. He would be the first to deny that his work demonstrates that only complete laissez-faire is feasible.
The whole point of industrial policy is that it is conducted without regard to economic efficiency. Remember: The debate over economic models was productive because both sides agreed on the standard of success. Capitalists and socialists insisted that theirs was the efficient system, which made efficiency the goal. That’s not the case anymore. National conservatives are forthright in their belief that economic efficiency and the national interest diverge. It’s the latter they’re trying to achieve. While their intermediate objectives differ — some want industrial policy for national-security reasons, others for supporting American families, and more still, because they think it might build a winning political coalition — they agree that manufacturing employment and output should be higher than they are now.
There isn’t a Hayekian knowledge problem here. If the government wants to increase the number of factory workers or the output of domestic auto manufacturers, it can. But it won’t be easy. Especially at the industry level, many complex interdependencies exist among production technology, substitutability of labor and capital, and market prices. While subsidies seem like an obvious choice, they could very well backfire. Acquiring good information is costly, even more so for distant bureaucrats than firms and households. For these reasons (and many others), I’m against industrial policy (with the possible exception of tailored security measures). But this isn’t to claim that the knowledge problem makes industrial policy impossible. “It’s hard” and “It cannot be done” are categorically different statements.
Occasionally, industrial-policy advocates push their argument further than it should go. Sometimes they point to South Korea, Japan, or China as examples of successful industrial policy, which show that governments can direct economic growth. This does fail the Hayek test. Economic growth doesn’t just mean rising incomes. It means continually extending social cooperation under the division of labor in a way that uses resources to their highest potential. Believing this kind of industrial policy is feasible means asserting that governments can out-market the market. No chance.
We should call out defenders of industrial policy when they make these arguments. But here’s the key difference: Few proponents of American industrial policy claim that Uncle Sam can do this. Most of them don’t want to make government the arbiter of economic growth. They’re just trying to increase output or employment in certain sectors. Again, there’s a qualitative difference here. Nudging markets to employ or produce more than they otherwise would doesn’t run up against the knowledge problem. Directing markets to employ or produce in a wealth-maximizing way does. But since national conservatives don’t care much about wealth maximization, there’s no reason they should be bothered by this impossibility.
American industrial policy faces monumental challenges. There are all sorts of informational difficulties associated with its design and implementation. Even with an attentive legislature and competent executive it’s a tall order — and we have neither. It’s proper for classically liberal conservatives to be skeptical of industrial policy. But it’s not proper to assert that it’s impossible. Hayek’s insights concerning knowledge and the pricing process are much more subtle than the hackneyed “It’s complicated” objection to government interventions.
If we’re going to debate industrial policy, let’s do so with a clear understanding of the difference between difficult and impossible. Opponents of industrial policy need to focus on why its consequences are undesirable. Hand-waving about the knowledge problem to avoid discussing those consequences won’t cut it.