Elon Musk Needs to Teach Our Government How to Lose More Money
Josh Zoffer how to really rebuild international trade
The markets have massively gyrated the past few weeks as hourly pronouncements from the White House and Mar-a-Lago seize investors with terror or relieve them of stress. At the heart of the vociferous debate around tariffs and trade is nothing less than the future economy of the United States. What should be built here? What should we outsource? Who pays for that new economic structure?
Joining me and Riskgaming director of programming Laurence Pevsner is Josh Zoffer. Josh was formerly special assistant to President Biden for economic policy, and today, he’s a principal at Clocktower Ventures. He’s penned a number of recent op-eds on trade policy and global interconnectedness (as well as one on the rise of metaverses).
The three of us talk about all of the news this past week around tariffs, what kind of manufacturing we want in America, why globalization remains critical, the challenges of administrability, workforce development and finally, why it is so hard for governments to take on high-risk investing in emerging technologies.
This Q&A has been edited for length and clarity.
For more of the conversation, please subscribe to the Riskgaming podcast.
Danny Crichton:
Josh, you recently had an op-ed in The Economist called “The global trading system needs new rules, not tariffs.” Now, all of a sudden we have the complete opposite of your recommendation, which is lots of tariffs, not a lot of rules. In all of the chaos over the last couple of days, we’ve witnessed the crashing of the markets, huge shifts in trade, massive complaints from China, Japan, Korea, and exporting countries all over the world. What’s your prognosis of the situation as we speak Monday afternoon?
Josh Zoffer:
In short, not good. The market reaction has borne out my and my co-author, Wally Adeyemo’s, take that indiscriminate tariffs are not the right answer.
But to take a step back, part of what we wanted to say is, in theory, there is a set of rules that is supposed to govern when countries use tariffs and other tools to alleviate some of the imbalances that President Trump has rightly pointed out. The fact is that China is not, in many ways, playing by the global trading rules. Those rules are supposed to do something. But, as we say in the op-ed, they haven’t really done much to stop imbalances from forming or to prevent the kind of large-scale subsidies and unfair behavior that we’ve seen not just from China but from a whole bunch of countries.
Various U.S. administrations — Donald Trump’s first administration, the Biden administration, and now on a more extreme level, Trump’s second administration — have used tools like subsidies and tariffs in response. And the WTO, the system that is supposed to regulate all this stuff, has had nothing to say about it, either to prevent problems or address extreme responses.
The breakdown of this rules-based system is what you’re seeing today. The Trump administration is taking very extreme actions. It has totally disrupted the system and has, in many ways, shown why a better alternative would not be to go back to what we were doing before but to say, “All right, we need something new.”
That something is a rethink of the rules so that everybody is playing from the same rule book.
Laurence Pevsner:
During the invasion of Ukraine, U.S. Secretary of State Antony Blinken said, “It’s not about might makes right. It’s right rules make right.” This is the basis of the rules-based international order: the powerful use rules to bind ourselves.
The problem is that the system only works as long as the most powerful countries agree to do that self-binding. So is the WTO the problem? Or is the problem that the countries that have to agree to the WTO’s rules just don’t want to follow them anymore? And if that’s the case, how would a new set of rules address the fundamental political problem?
Josh Zoffer:
I believe that, even if they are not what would be the best thing for you at any given moment, a set of rules that is well thought out and followed is, in the long run, going to be better for everyone. The United States is disregarding rules entirely right now, but the track record of that strategy so far does not appear to be very good.
But you’re right that it is a challenge. We’ve got to reinvigorate the desire among various countries to follow rules. One of the things I feel proudest of from my time in the Biden administration is that President Biden spent a huge amount of time trying to reinvigorate the Transatlantic Alliance and work with our allies and partners.
There is some real work to be done to remind everyone of why rules matter and why they’re helpful. But we’re seeing in real time why everybody doing the “go it alone” approach does not lead to a great outcome.
Danny Crichton:
What’s striking to me is, in the first Trump administration, the ire was fairly focused on China. But the tariffs announced this past week are remarkably broad. Why not just continue down a path for which there’s a lot of bipartisan support? Clearly the U.S. trade deficit with China is the largest. It was $1 trillion last year, in 2024.
Josh Zoffer:
I don’t really have a good justification for the decision to go after the Europeans. They are our best allies, and I suspect we will come to regret it.
When it comes to China, it is hard not to give the first Trump administration a bit of credit for catalyzing everybody to think a little bit differently about the way the global economy had been running. When Trump said, “We don’t want to do this free trade stuff. I want to put tariffs back on the table,” we really started to think about what’s working, what’s not.
I take away two things from that. One is, to some extent, manufacturing and the ability to make things really does matter. There is a certain set of things that it’s actually important, from a national security perspective, to be able to make domestically, or at least to rely on your partners and allies. And ideally you want to be producing the highest technology, most advanced, most valuable stuff. That’s really where you get the most from controlling the manufacturing base.
The second piece is that the trading system has not done a great job preventing unfair practices. There are cases where a trade imbalance develops because there is meaningful comparative advantage and the system is working the way it should. There are also cases where trade deficits develop because one country has cheaper labor. Fine. We can’t control that. We can’t make everyone pay American wages. But in cases where the lower cost results from lower labor standards, worse conditions, or weaker environmental standards, that’s not really fair.
The international trading system had become a kind of get out of jail free card for a lot of that stuff. This is why you saw the continuation of Trump’s tariff policies against China during the Biden administration. That was a really targeted strategy, though. As someone who worked on the 301 tariffs that the Biden administration announced in May of 2024, we studied that stuff really closely. And there were people in the White House before me who worked on it even before that, going line-by-line through the tariff code, good-by-good, and saying, “Does this really need a tariff? Does this make sense? What’s the right level?” If you get it wrong, it imposes real costs on companies that have their supply chains disrupted and on people.
Danny Crichton:
One of the challenges is that there’s a great lie with trade. On the one hand, we want high environmental standards, high labor standards. We want high quality goods. On the other, the American public likes to go to Temu and get a $2 T-shirt dropshipped from China. We know it’s made in a sweatshop. There’s no other way to produce a $2 T-shirt. You see this with climate and Europe. Many European countries have strong climate and carbon decarbonization policies, but then they’re importing very carbon-intensive goods from China as a way to offshore carbon to the exporter — to push it off the books in Europe.
To what degree do you think that can be undone? How much do you think Americans could actually adapt?
Josh Zoffer:
When it comes to the overall direction of globalization and consumption, the ship has set sail. We’re not going to totally go back. And I don’t think we would want to go back to a world of radically less globalization. The post-war era has, on the whole, been pretty good. Billions of people around the world have been lifted out of poverty. From a human perspective, that’s a good thing. And I don’t think we want to reverse it.
Now, does it have costs? Of course it does. Does it have costs domestically? Yes. And I think we’ve learned those costs were larger than we thought. The effects of de-industrialization are real, and we shouldn’t discount them. But the value of being able to access cheap goods and enhance the purchasing power of Americans is also not something to be discounted.
I don’t think we’ve gotten everything right. But overall, we want trade. We want globalization. We want the trading system to work because, when it is working properly, it does lead to prosperity both in the United States and abroad.
Danny Crichton:
One of the things we try to do with Riskgaming is to complicate things.
This discussion has been focused on the United States and China, but obviously there are companies within those countries. Those companies have multiple divisions. They sell products in different places. An Intel chip is designed here, manufactured in Taiwan using parts from other countries. Those are now shipped to multiple endpoints, for example, a Malaysian cloud infrastructure startup that’s reselling the compute to China in order to get around export controls.
So the layers of complexity keep adding up. When you think about a targeted approach versus the sledgehammers, there’s the challenge of how you handle a network. You may want to prune different relationships, but there’s always a way to rewire and find a way around the export controls, around the tariffs.
The U.S. trade deficit with Mexico has doubled or tripled over the last two, three years because Chinese manufacturers started moving factories into Mexico, particularly in autos, in order to be able to ship to the United States through USMCA.
To what degree do we just have to absorb that complexity?
Josh Zoffer:
Our supply chains are endlessly complex. It is almost impossible to know what is in the final goods we’re buying — where the pieces came from, what those pieces are, how much value to attribute to each of them. There are tools around domestic content requirements or rules of origin, but they are dramatically limited by the data that we have.
There’s this thing you hear about often in government — I’m sure Laurence has heard the term — called administrability. We want to do something but we need to make it “administrable.” It has to be something that is within the boundaries of possibility given the human beings that we have, the number of people we have working on it, the information at their fingertips, and the hours in a day.
Now, if we were redesigning the rules, what do we need? One thing we should pay close attention to as we try to create new rules is putting in place the right systems to gather data so that we can do finer targeting.
But to your question about moving factories. I think this is a really important and underappreciated element of trade policy: What are we actually trying to do? When we put tariffs on just China, companies moved their factories, some to Mexico. But is that what we want? Are we trying to get factories out of China? Or are we trying to get the factory into the United States, for example?
The historical analog I think about is the 1980s. The bogeyman of international trade in those days was Japan, our ally. During the Reagan administration people were really worried about this. President Reagan even put together the Presidential Council on Industrial Competitiveness and convened all the leading executives to think about the problem. Three things ultimately happened.
One, Reagan, the world’s consummate free trader, put tariffs on Japan. Pretty meaningful tariffs — up to 100% in some cases, which complicates the historical narrative around the commitment to free trade at that time. It was not pure free trade or nothing. It was using targeted tools where it matters.
Two, the United States and Japan agreed to what are called voluntary export limits. So Japan basically says, “We agree not to sell you too many cars. We will leave some room for you guys.”
Three, all the Japanese auto manufacturers set up factories in the United States.
I don’t know that anybody in the trade and industrial policy debate has a clear answer to what we want to happen to the Chinese automakers. There is this long-standing narrative that China just cheats. China’s definitely done some unfair things, but Chinese car companies are really good. They make great cars.
So one answer might be to do what we did in the 1980s and say, “The right outcome is they set up factories here and they employ Americans.” Another is to say, “No. Actually what we really want to happen is for American car companies to employ Americans and to sell more cars.” And those would require very different policies.
Laurence Pevsner:
If one solution is to try to block out other countries from putting their goods into our markets, another solution is on the supply side. Our housing is going to get cheaper, for example, because we build more houses rather than cutting the cost of everything else.
Josh Zoffer:
One of the things that was most exciting to me during the 2024 presidential campaign on Vice President Kamala Harris’s side was the commitment to build 3 million new homes. And the Biden administration had the CHIPS Act, the idea of which was to build factories. The Inflation Reduction Act, too, was about trying to put in place tax incentives, loan guarantees, and financial support for production in the United States.
But we are not going to secure America’s economic place in the world purely by building houses for ourselves.
Think about where abundance really comes from — where national wealth and prosperity come from. It comes from making things that everybody in the world wants. It comes from being on the leading edge of technology and being able to build businesses that employ people and create economic opportunities that allow them to buy houses and to consume energy and to have prosperous lives.
U.S. tech companies are the envy of the world. But how do we have more of them? Where is the future of innovation? And how do we ensure that American companies, engineers, and scientists are the ones not just inventing new technologies but also commercializing them?
This is another lesson we have learned over the last 50 years of international trade. A lot of the things China is really good at making today were invented in the United States. The LFP Battery is a great example. It was invented in the United States. One of the first companies trying to commercialize it, I think it’s called A123, is an American company that gets funding from the Department of Energy’s Loan Programs Office. But the company ends up not making it. The technology is bought by a Chinese company, and then they take it all the way to the bank.
Danny Crichton:
This is one of the largest, most difficult questions, and I don’t think we have a great policy answer. The United States can be on the cutting edge. We can develop great new technologies. But ultimately, if we don’t have the factories or workers to manufacture them, we are losing the market anyway.
And there, internal migration is complicating the story as well. Let’s look at, say, the CHIPS Act. It is trying to bring the fabs back. But the question is, which states should have these industries?
In Korea, you have the shipbuilding industry concentrated in Ulsan. You see this in China across a whole chunk of industries. You have the Christmas ornament town, for example, and the whole business of that town is Christmas ornaments.
We don’t do that here, for a variety of reasons. We don’t do that because Congress, when it comes to government programs, wants to split them across congressional districts. No one wants to commit to one place and say, Birmingham, Alabama is a hub for aerospace.
Josh Zoffer:
Our system is hard. Would I trade it for anything else? No. Is it easier for China to do long-term technology planning? For sure. But in the long run, our economy has all of the advantages to be successful. It has the best innovation economy in the world, the best research institutions, and the most sophisticated capital markets.
But we have to learn how to get out of our own way. The most important thing is to have the factory. It doesn’t matter where it is if you don’t have it at all. We have to build the muscle to do the hard things, and not everybody’s going to like it.
We used to be fantastic at this. Look at mid-century America and the hard choices that were made under Presidents Roosevelt and Eisenhower. We have rightfully gotten more attuned to the trade-offs. But to the extent that we have let that become a barrier to doing anything at all, that’s a problem.
Laurence Pevsner:
This leads very naturally into another piece you had in the New York Times not too long ago. The headline was, “Elon Musk Needs to Teach Our Government How to Lose More Money.” Could you say a little bit more about it?
Josh Zoffer:
I will say the original version of the piece that I pitched did not mention Elon but, yeah, I think the point stands. Historically, the United States has been able to catalyze innovation with government support. There is this amazing book by Mariana Mazzucato called The Entrepreneurial State, in which she points out that pretty much all of the most important parts of the iPhone benefited at some point or another from early stage government investment in R&D.
But our risk appetite should be higher when we do really early stage stuff. We have R&D and grant programs. For the most part they’re not huge. ARPA-E, which is the early stage grant making part of the Department of Energy, has an annual budget of, give or take, $400, $500 million. It’s like the size of a midsize venture capital firm. DARPA, which is our flagship innovation program, has a program budget of, call it $4 billion. That’s great, and DARPA is an amazing program. But compared to where most of our industrial policy support and dollars go, it’s pretty small. Most of that money goes towards supporting more mature, later stage enterprises.
In my view, if we really want to do the thing we were talking about earlier, which is to figure out what’s going to be at the core of the global economy in 10 or 20 years, then the U.S. government should have more willingness to put dollars to work. Not all of it is going to pan out. This is the “lose more money” piece.
A venture capital firm knows that if you take risks and back early stage technologies, not all of them are going to work. But if you find the right ones — the ones that end up being important and the ones that do work — the gains are enormous. And in this case, those gains accrue not to the LPs of a fund but to a country. If the companies in that country are the ones employing the people who are getting paid higher wages because that technology is commercialized domestically, that’s an incredibly important mechanism for creating prosperity.