The man of system, on the contrary, is apt to be very wise in his own conceit; and is often so enamoured with the supposed beauty of his own ideal plan of government, that he cannot suffer the smallest deviation from any part of it. He goes on to establish it completely and in all its parts, without any regard either to the great interests, or to the strong prejudices which may oppose it. He seems to imagine that he can arrange the different members of a great society with as much ease as the hand arranges the different pieces upon a chess-board. He does not consider that the pieces upon the chess-board have no other principle of motion besides that which the hand impresses upon them; but that, in the great chess-board of human society, every single piece has a principle of motion of its own, altogether different from that which the legislature might chuse to impress upon it.
—Adam Smith, The Theory of Moral Sentiments, 1759
In late 1983, when I was in my second year as a senior economist with President Reagan’s Council of Economic Advisers, I was tasked with writing the chapter on “industrial policy.” Industrial policy was all the rage back then. Various politicians, especially Democratic ones, advocated the idea. Among them was former vice president Walter Mondale, who seemed to have the best odds of winning the Democratic nomination for president and who, in fact, did win the nomination. So we knew it was a hot issue and my two bosses, chairman Martin Feldstein and member William Niskanen, decided that we should devote a whole chapter to the issue.
The essence of industrial policy is that government officials, looking ahead, predict which industries will or should do well, and then use various policy instruments—tax policy, subsidies, subsidized loans, and regulation—to move the economy in what they think is the best direction. They are, in Adam Smith’s words, updated, “men and women of system.”
There are two problems with industrial policy: information and incentives. Government officials don’t have, and can’t have, the information they need to carry out an industrial policy that creates benefits that exceed costs. Also, they don’t have the right incentives. If they spend literally billions of dollars of government revenue on buttressing an industry and the industry fails, they don’t suffer any personal wealth loss and don’t even lose their jobs. The only cost to them as individuals is their prorated share of tax revenues, which will typically be no more than a few hundred dollars. So what ends up happening is that subsidies and preferential treatment are given to the politically powerful, which reduces the amount of capital available for unsubsidized entrepreneurs and innovators.
The MITI myth
In the early 1980s, many observers, looking at Japan’s economic success, touted it as Exhibit A for the success of industrial policy. They attributed it to the Japanese government’s Ministry of Trade and Industry (MITI.) But in “The Myth of MITI,” Fortune, August 8, 1983, I wrote:
Early in the 1950s, a small consumer electronics company in Japan asked the Japanese government for permission to buy transistor-manufacturing rights from Western Electric. Permission was necessary because at the time foreign exchange was controlled by the tax and trade ministries. MITI refused, arguing that the technology wasn’t impressive enough to justify the expenditure. Two years later the company persuaded MITI to reverse its decision and went on to fame and fortune with the transistor radio. Its name: Sony.
Sony, one of the most touted postwar Japanese success stories, succeeded not because of, but in spite of, MITI.
That wasn’t the only such story. I pointed out that in the mid-1950s, MITI had pushed one Japanese industry to develop a prototype “people’s” model of its product. MITI’s idea was that it would choose the winning firm as the sole producer. This was classic government-run industrial policy. In the 1960s, MITI tried to force the many firms in the industry to merge into just a few. In both cases, companies in the industry rebuffed MITI. That’s fortunate for Japan and for US consumers. Why? Because the industry’s product was cars.
What was responsible for Japan’s strong postwar growth, I argued, were the factors that economists often point to: high savings, which were due, in part, to favorable tax treatment of interest income, low and often falling marginal tax rates, and low tariffs, to name three.
Information and incentives
A basic understanding of the economics of information and of incentives shows why industrial policy fails.
Consider information. One of economist Friedrich Hayek’s most important economic insights was that even if incentives were not a problem, central economic planning, aka socialism, would fail because central planners cannot have the information to plan an economy well. That information, he argued, is dispersed, existing in the minds of the millions of actors in an economy. And only a relatively free market can guide us to make good decisions.
One example Hayek used to illustrate his point is the market for tin. In his 1945 classic, “The Use of Knowledge in Society,” Hayek explained how people adjust when the price of tin rises, either because a source of tin has shut down or a new use of tin has appeared. He noted that participants in the market don’t need to know which of those two caused the price to increase in order to know how to adjust. Producers of tin cans, for example, know much more about their production process than a central planner could know. Therefore, they can figure out how best to economize on tin, maybe by using slightly less tin with slightly thinner cans or by raising the price, which, in turn, would lead to fewer sales and less need for tin. Producers of tin would consider, say, opening marginal mines that weren’t quite worth producing from before. With all these adjustments, the market reaches a new equilibrium at the higher price. In doing so, market participants also achieve the results that a benevolent central planner would want to achieve but could never achieve.
Hayek’s information insight makes the case that full-fledged central planning must fail. But does that mean that industrial policy will fail? After all, government officials who carry out industrial policy have the advantage that they can observe prices, for tin and everything else, in doing their planning.
But there’s a still a large information problem. It's true that even the best entrepreneurs make mistakes. But they have a good sense of what to look for. Their ears are to the ground in a way that a government official’s are not. And making the information problem even more lopsided against the government official is the role of incentives. Watch any episode of Shark Tank and you’ll see that the entrepreneurs seeking funding almost always have a huge percentage of their own wealth at stake. And while the sharks’ wealth at risk in any given investment is a tiny percent of their overall wealth, especially for billionaire Mark Cuban, they really do want to, in the words of Kevin O’Leary, “make me money.”
The incentive issue matters in another way also. As the Kenny Rogers song “The Gambler” goes, you need to “know when to hold ’em” and “know when to fold ’em.” When entrepreneurs and venture capitalists have their own wealth at stake, they’re much more likely to fold, and much quicker at folding, a failing venture than government officials with little of their own wealth at stake.
Solyndra, a California company that made a special kind of solar panel and received large government subsidies, is a good example. Robert P. Murphy lays out the issue nicely in “Lessons from Solyndra,” Econlib, February 6, 2012. The Obama administration blew by some of the warnings given by the president’s own employees and proceeded to fund Solyndra. One of the major advocates of Solyndra funding was Energy Secretary Steven Chu, who had won the Nobel Prize in physics in 1997. He was obviously no dummy, but remember that he was sitting in Washington and had none of his own wealth at stake.
One of the rah-rah supporters of Solyndra in the Obama administration, by the way, was Vice President Joe Biden, who, at an event to celebrate the over $500 million government loan to Solyndra, said, “It’s important because these jobs are going to be permanent jobs. These are the jobs of the future, these are the green jobs, these are the jobs that won’t be exported.” Later, of course, the jobs disappeared.
Biden’s industrial policy
President Biden doesn’t seem to have learned anything from the Solyndra failure. He has gone all in on industrial policy with the misnamed Inflation Reduction Act (IRA), the Infrastructure Investment and Jobs Act (IIJA), and the Creating Helpful Incentives to Produce Semiconductors (CHIPS) and Science Act. The latter two were passed with bipartisan support.
As William D. Eggers, John O’Leary, and Kevin Pollari put it in a March 16, 2023, analysis for Deloitte, these “three new laws represent a massive investment in American competitiveness. They seek to rebuild American infrastructure, accelerate the transition to a green economy, and strengthen the domestic semiconductor industry.” They are, in short, a massive industrial policy. How massive? The authors note that the three laws together authorize over $2 trillion in federal funding over ten years. $2 trillion is a large number even for the feds. While Eggers et al. do a good job of laying out the challenges, they seem strangely agnostic about whether the industrial policy will work. If they had thought through the basic economics of information and incentives, their uncertainty would likely disappear.
The political payoffs
In one sense, though, the industrial policy could succeed: by benefiting politically powerful players. Because we economists, whatever our ideology, have a habit of evaluating policies by whether they produce benefits that exceed costs, we sometimes forget about that. But politicians, whatever they say, typically think of policies more narrowly, asking themselves whether those policies will help the special interests they favor. That appeared to be one of the factors motivating the Obama administration to lend money to Solyndra.
Who will benefit? The people and companies that get subsidies, tax credits, and subsidized loans.
One driver of economic growth, which Nobel Prize winner Robert Solow highlighted in his seminal work on growth in the 1950s, is capital investment. Every dollar that governments cause to be put into government-favored projects is a dollar that can’t be invested by non-subsidized investors. One effect of industrial policy, therefore, is slower growth.
The men and women of system are trying to move us around like so many pieces on a chessboard. We “pieces” should rise up and revolt.