As the secretary of commerce in the Biden administration, Gina Raimondo spearheaded the CHIPS and Science Act to boost US investment in cutting-edge technology, especially semiconductors. But she sees some obstacles in the United States’ path to leading in the global economic and technological competition. In this conversation with McKinsey Partner Matt Watters, Raimondo shares her views on the use of tariffs and other geopolitical levers, the risks in governments investing in companies, and the pros and cons of the AI revolution. The following transcript has been edited for clarity and length.
‘America First’ cannot become ‘America Alone’
Matt Watters: My colleagues and I have argued that since 2020, we’ve entered a new world order—one defined by multipolarity where economic and technological competition is on the rise. Do you agree with that view?
Gina Raimondo: I think that’s right. I worry about China, Russia, and North Korea seeming to be working together. The question for the United States is whether we are doing enough to expand our list of allies and deepen our relationships with them. That is essential for the country and our businesses, because if your analysis is right and these groupings and nodes are forming, “America First” cannot become “America Alone.”
Matt Watters: One reason we see a new world order is the significant increase in governments’ use of geopolitical levers, whether trade agreements, tariffs, export controls, investment restrictions, or industrial policies. Do you think this heightened intervention will be the norm and, if so, how should business leaders navigate this landscape?
Gina Raimondo: It’s hard to know how long the current chapter will last, but tariffs, once put in place, are hard to remove politically. Tariffs tend to protect certain groups of workers, certain industries, or certain groups within an industry. If a president later removes a tariff, they are exposed to the criticism that the move hurts the protected group. I tried unsuccessfully on several occasions to convince President Biden to lift or reduce certain tariffs that the previous Trump administration had instituted. Congress opposed it, and the president didn’t want to do it. That being said, I don’t think President Trump or any other president will maintain triple-digit tariffs. Tariffs are inflationary, but I think they are here to stay for some time.
The pros and cons of industrial interventions
Matt Watters: Do you expect to see continued use of industrial policies such as tax incentives or even governments taking equity stakes in companies?
Gina Raimondo: I think so, because the United States has woken up to the very real risks of being overreliant on China for certain critical items, including biotechnology, active pharmaceutical ingredients, rare earth and other critical minerals, and semiconductor chips. We’re too vulnerable, and that view is bipartisan. There are different approaches—you can stockpile supplies or use the Department of Defense procurement power, tariffs, tax incentives, grants, or equity stakes—but I do see a collective recognition that it’s too dangerous to do nothing.
Matt Watters: The US government took equity stakes in companies during the global financial crisis. Do you support this approach outside of a crisis situation? Are there other steps the government could take to address the economic security concerns you mentioned?
Gina Raimondo: Taking equity stakes worries me. I was the architect of the CHIPS Act. I had to build the team and provide $40 billion of grant money to dozens of companies, so I can speak with some authority about how hard this is to get right. State-owned or heavily state-subsidized companies the world over are not the most efficient, productive, or innovative. If the US government takes a large equity stake in a company, and that company thinks the government, as a shareholder, won’t let it fail, the company may lose its edge over time.
That being said, it’s more expensive to build a chip factory or do clinical trials for a new drug in the United States than it is in China, so providing some subsidy is appropriate. I favor one-time subsidies, tax incentives, and other measures that require fewer investment decisions by politicians or policymakers and that are one-time infusions of capital rather than long-term interests in companies.
Matt Watters: The CHIPS Act has successfully helped start construction projects and deliver funding, but it has also been a learning exercise in the delays that US infrastructure initiatives face. What have we learned about doing infrastructure in this country?
Gina Raimondo: We need to get better. We have to ask why it is so much more expensive to manufacture or do clinical trials or build a memory [chip] fab in the United States. Some of it is due to labor costs, and that’s a good thing—we don’t want to pay our construction workers suboptimal wages, and we want construction sites to be safe. Some of it is due to long permitting timelines and government processes for zoning and environmental regulations. Some of it is due to our political system. It’s good to have due process; it’s not good when that process takes three years, and one lawsuit can tie things up for five to seven years.
Some of the extra cost is due to the lack of supply chains in the United States for some industries. It’s approximately six times as expensive to build a ship in the United States as it is in Asia—not because of significantly higher wages here but because we don’t have extensive ship-building capacity in the United States. We need secure supply chains and we have to make our government work faster and more efficiently if we’re going to be competitive globally.
The need for collective action on AI
Matt Watters: You’ve spent a lot of time thinking about how to deal with AI—its great potential for drug discovery or personalized, affordable medicine but also the danger of mass unemployment and social or political upheaval. What do you see as the biggest risk and the biggest opportunity in AI, and who is responsible for finding the balance between those two?
Gina Raimondo: We all are. It’s in no one’s interest for this country to have a double-digit unemployment rate. It’s all in how we manage the transition. I’m an AI enthusiast—I want the United States to win, and I want our companies to be the most innovative. To compete, we need to lean into AI and use it to innovate. But if we let the market play out and let AI advance at the current rate without providing for a labor market transition, we will get 10 to 15 percent unemployment, social unrest, political violence, and a degrading of our democracy. Shame on us if we allow that to happen. It’s a problem for collective action. No company will unilaterally slow down [AI development], and that’s where government regulations, industrial bodies, trade associations, and business leaders have to step in.
Matt Watters: Let’s talk about export controls. You’ve said that they need to be “precision strikes.” In certain industries, such as telecommunications, solar, batteries, or high-speed rail, China has become a market leader. Are we at risk of creating an affordability gap by placing export controls on advanced chips?
Gina Raimondo: Yes. This was the hardest part of my job as secretary of commerce, because, on the one hand, if we have cutting-edge technology that China doesn’t have and could use in a military capacity to hurt us, we’d be foolish to let China have it. On the other hand, diffusion is the way to win in tech. I always said to President Biden: We want the world to run on American AI rails, our data centers, our models, our chips. That’s why export controls have to be precision strikes, covering only the most sophisticated chips—or the equipment used to make those chips—where we are truly ahead of China. To keep America safe, we want to always be ahead of our competitor.
