Why Industrial Policy Fails, Time and Time Again

As the eminent economist Milton Friedman famously said, “One of the great mistakes is to judge policies and programs by their intentions rather than their results.” Examining recent subsidization of electric vehicle (EV) charging stations, this truism becomes all the more self-evident.

Politico reports that the $7.5 billion Congress allotted in 2021 to subsidize the construction of EV charging stations has produced exactly zero chargers to date. The history of Joe Biden’s domestic agenda will read like a textbook debunking industrial policy’s false assumptions and exposing its practical failures. It seems that practically every one of Biden’s infrastructure projects flounders in one economically edifying way or another.

There is a profound gap between Democrats’ stated industrial-policy objectives and the results their policies have produced. As the EV charger debacle demonstrates, guiding subsidies through the bureaucratic process requires extensive time — often far longer than planners anticipate or advertise to their constituents. Politico notes that “States and the charger industry blame the delays mostly on the labyrinth of new contracting and performance requirements they have to navigate to receive federal funds.” Moreover, the story continues, “While federal officials have authorized more than $2 billion of the funds to be sent to states, fewer than half of states have even started to take bids from contractors to build the chargers — let alone begin construction.”

Politicians’ rosy straight-line projections about their pet proposals rarely resemble the winding and bumpy road that industrialist policies must travel in practice. Politicians and bureaucrats invariably ignore the litany of political, bureaucratic, and economic factors that delay the rollout and inflate the costs of central planning — often rendering it entirely ineffective or worse.

First off, economic forces — rooted in scarcity and tradeoffs; incentives and price signals; and vast, disbursed knowledge — are not as easily subverted as would-be central planners assume. Technocrats can neither know nor understand more than a fraction of the information relevant to any given economic decision. Any attempt to jerry-rigg markets inevitably will miss some crucial datapoint.

Subsidies that incentivize industry and consumers to zig cannot eliminate contravening market signals that incentivize them to zag. As National Review’s Dominic Pino notes, subsidies have persuaded car manufacturers to produce far more EVs than consumers want, leading to many sitting unsold on dealer lots. “The government interventions to promote EV manufacturing are, in a sense, working too well,” Pino writes. “The ones to promote EV purchases aren’t working well enough.” Even with generous tax credits, many Americans prefer to pay half the price of an EV for an equivalent gas-powered vehicle. Registering these market signals, auto brands such as Ford and General Motors have begun to reconsider their mad dash for government handouts and have slowed their announced expansions of EV manufacturing.

The CHIPS Act, which provides $52 billion for domestic semiconductor manufacturing, cannot negate the myriad contributing factors that make America an expensive and inhospitable home for such work. “[T]he cost of producing chips in Arizona could be as much as 100 percent higher than the cost of producing the same chips in Taiwan,” write policy analysts Jordan McGillis and Clay Robinson. They further report that while the industry titan Taiwan Semiconductor Manufacturing Company “will produce many chips” in its U.S. facilities, it will “continue to advance its cutting-edge production elsewhere for cost reasons, leaving Arizona as a legacy producer and leaving our military in the same position of foreign dependency a decade hence.”

For another case study, consider the $42.45 Broadband Equity, Access, and Deployment (BEAD) program, a slush fund intended to universalize access to high-speed internet. Authorized by Congress in 2021, many BEAD grants will likely not arrive until the mid-to-late 2020s. The states, which will receive BEAD moneys and disburse them to individual projects, must first navigate a complex federal compliance regime and build subgrant programs. (And furthermore, NEPA delenda est.) By the time the money is spent, and the networks begin to be built, the broadband landscape will have been altered fundamentally by private-sector investment and rapidly shifting consumer needs.

Industrial-policy makers, moreover, often overlook key considerations when appropriating funds and crafting regulation. “The best-laid schemes of mice and men,” the saying goes; but too often, industrial policy is guided — and doomed — by profound incompetence.

Sen. Joe Manchin (D-W. Va.), a fervent champion of domestic-manufacturing subsidies, admitted in January that he did not know the U.S. has no free-trade agreement with the European Union. Since the Inflation Reduction Act (IRA) limited EV subsidies to cars containing battery components sourced from nations America maintains such agreements with, this oversight has caused havoc due to. To compensate, the Treasury Department simply bent the statute’s plain text.

Consider another example. Biden’s ambition to de-fossilize the power sector by 2035 overlooks the fact that the U.S. lacks necessary infrastructure to effect such a transition. To wit, the U.S. has far too few transmission lines. “Already, a lack of transmission capacity means that thousands of proposed wind and solar projects are facing multiyear delays and rising costs to connect to the grid,” the New York Times reported in June. “In many parts of the country, existing power lines are often so clogged that they can’t deliver electricity from wind and solar projects to where it is needed most and demand is often met by more expensive fossil fuel plants closer to homes and businesses.”

In some cases, ideology, not poor planning, cripples industrial policy. The Department of Commerce’s (DOC) implementation of the CHIPS Act provides a quintessential case study in this phenomenon. In short, it devolved into “social policy with a side of chips,” as put by the Cato Institute’s Scott Lincicome. The DOC conditioned eligibility for CHIPS subsidies on adherence to a slew of progressive orthodoxies. These include pro-union Davis-Bacon standards, “diversity, equity, inclusion, and accessibility” compliance, environmental planning, and much more. Precisely how these will increase the semiconductor output of finite CHIPS funding remains unclear.

In fact, shoehorning in ancillary ideological priorities destroys industry’s ability to use federal funds effectively — a fact many businesspeople understand well. For example, the DOC placed such burdensome constraints on the BEAD program that many small broadband providers plan not to apply for grants. To maintain control over their business practices, they prefer to forgo subsidies.

Further, government spending programs of any sort often cost far more than appropriators expect — particularly given inflation, supply-chain kinks, and rising interest rates. The Congressional Budget Office projected that the IRA’s green-energy subsidies would cost $391 million over a decade. According to further review by Goldman Sachs, however, their true cost will likely exceed $1 trillion. The IRA’s clean-vehicle subsidies alone — initially tagged at $14 billion — will likely cost $390 billion.

Democrats and industrial-policy-curious Republicans are discovering that subsidies don’t function like genies in One Thousand and One Nights; they cannot achieve anything and everything on command. This lesson, however, while apparent to any economically literate observer, will likely fail to reach many politicians, a group that terminally overvalues its own competence and importance.

Milton Friedman was entirely correct: “We all know a famous road that is paved with good intentions.”

David B. McGarry is a policy analyst at the Taxpayers Protection Alliance.