The revival began under the Trump administration, as part of its anti-China offensive, but accelerated under President Joe Biden, who supplied his economic agencies with figures willing to intervene in the market. “Part of our effort is to create space again for very serious people to really fight for the idea that government has a legitimate role to play” in industrial development, a senior administration official said.

But big industrial policy raises a big question with billions of dollars at stake: what works and what doesn’t?

Advocates point to a long track record of achievement, beginning with Alexander Hamilton whose “Manufacturers Report” argued for tariffs and subsidies to help the fledgling industry. Government research and funding has helped create commercial jet aircraft, the Internet, communications satellites, digital mapping, and has repeatedly kept major automotive companies alive during economic downturns. More recently, the Trump administration’s “Operation Warp Speed” bet $10 billion on a number of Covid vaccine candidates and came up with winners.

Critics count a long list of failures, including decades-long efforts to create “clean coal”, nuclear reactors that use recycled plutonium, nuclear fusion, synthetic fuels and supersonic commercial jets. The most recent poster is Solyndra Corp. who convinced the Obama administration to co-sign $535 million in loans and went bankrupt about two years later.

Even proponents say industrial policy has its limits. While government aid can be effective in launching new industries and strengthening those facing competitive challenges from abroad, it cannot go back and revive industries that the United States has lost.

This emerging consensus means heartbreak for many companies seeking government assistance. U.S.-owned solar power makers were counting on the White House to help them revive a domestic industry by blocking foreign competitors. But China and other Asian countries dominate the market so completely that the administration has turned its back on their pleas and paved the way for imported solar panels to continue flowing into the United States.

“It’s not that industrial policy doesn’t work,” says Robert Atkinson, who has championed government aid to industry since the Clinton administration. “But you have to do things at the right time. Once you’ve lost important abilities and your competitors have gained them, there’s not much you can do.

These days, Atkinson is the chairman of the Information Technology and Innovation Foundation, a think tank supported by the semiconductor and other technology industries. He championed the funding of computer chips as the bill passed late last week.

Ultimately, once an industry is lost to foreign competition, the chances of its recovery are extremely low. Then it’s not primarily about inventing new technologies – it’s about finding a way to attract manufacturers and their supporting industries to the United States. In the case of solar panels, that would mean creating a large domestic supply of the steps needed to manufacture panels, including refining polysilicon and producing silicon ingots, wafers and cells. In some of these segments, Chinese companies control more than 90% of the global market.

Industrial policy is most effective in helping new industries gain momentum, where they can capitalize on America’s longstanding strength in science and technology. But even then, difficult political decisions remain.

As manufacturing becomes increasingly international, it’s not clear which companies are “American” enough to warrant government assistance. Are American companies those that do business in the United States or are they those that have facilities in the United States and, if so, do they have to be factories? Or are American companies headquartered in the United States, even though many of their shareholders and employees are overseas?