Interviewed by Tal Cohen

Feb 24, 2025

Having served in both academic and government roles, how do these different perspectives inform your understanding on the gap between theoretical frameworks and practical governance challenges?

That’s a good question. In academia, one has time to pick, explore and consider alternatives for debate and discussion. My advice to young scholars – and the advice that I was given – was to make sure that when you publish your article or book, you’ve reached essentially the flat of the curve of your ability to improve it. If you could buy another month or six months to make it clearer, you should do it, because the timeline is more flexible. In government, however, the crisis is this morning and the memo that you need to deliver to whomever you’re trying to be helpful to might be due in an hour or in half a day. And if you come tomorrow with a memo that’s ten times better than the one you did today, the train may already have left, so that’s one big issue.

The second key difference is that, in academia, you are framing an issue and helping conceptually to clarify what’s going on. For example, I’ve written about a Thucydidean dynamic to understand what’s happening in the dynamics between the US and China. But when you’re in government, the question becomes: among the feasible options, what are the pros and cons? In an academic setting, you can often say, ‘Well, if I had this and this and this [the ideal conditions], then we would achieve this outcome’ – but of course, you don’t have this and this and this [those conditions don’t exist in reality].

There’s a famous old joke about the economist who finds himself on a desert island with a can of beans. The question was ‘How are we going to open the can of beans’? And he says, ‘Let’s first imagine we have a can opener’. That’s a good idea, but you don’t have a can opener. And frequently, if you look at policy commentary, whether in op-eds or academic discussions, people offer their assessments as if the parties had capabilities that they simply don’t have.

For example, look at what is currently going on: the Trump team has told the Europeans they must step up and provide more military capability for protecting and sustaining a ceasefire or a peace agreement in Ukraine. That’s a good idea in theory. They had a meeting yesterday [11/02/25] in Paris where everyone agreed they should do this. But then they begin to have a discussion and Starmer says, ‘Okay we’re prepared to do this’, but with a footnote that says ‘as long as the Americans guarantee to be there’. Then you come to Schulz in Germany, facing an election, who says ‘We shouldn’t be talking about this topic’. Why is he saying that? Because he’s living in a political environment in which Germans won’t support their citizens fighting for Ukraine. Even the Poles, who consistently lead in defense expenditures, say they’re not talking about sending troops.

So while from an American point of view, the proposition of telling Europeans to ‘get your act together and send up’ sounds straightforward, the reality is there’s no unified Europe. I mean, yes, there’s a geographical entity and various countries. They have the common market and the Euro as a common currency – that’s great. But in the security area, there is not so much consensus. I could give you 99 more reasons, but that’s a start.

During your time as Assistant Secretary of Defense, there was a presumption that senior policy positions required deep subject matter expertise – something reflected in your own appointment. Recent administrations, particularly under President Trump, have marked a departure from this tradition, often appointing individuals with limited domain knowledge to critical positions. What does this apparent devaluation of expertise tell us about the changing relationship between knowledge, power, and legitimacy in contemporary governance?

That’s a great question – one that would make for a good essay for somebody to look at over time and in different settings. Speaking about the American government, which I know the most about, as American politics has become more splintered, divided, and fractious, the relative weight of loyalty and competence has shifted increasingly towards loyalty.

This shift has been evident across the last ten administrations. And there’s a certain logic to why a president might prioritise trust – just look at Trump’s administration, where many staffers were simultaneously working on their memoirs and exposés. I’m trying to have you give me advice, and you’re thinking about how you can write something that shows that I was a jerk and you were brilliant?

Even John Kennedy, when selecting an attorney general, didn’t seek out the best lawyer in the world – he chose his brother. He needed somebody loyal that he could trust, particularly for cases that would impact him personally. The fact that Trump is trying to do his version of that, I think, is consistent with the way this has happened in other governments.

This emphasis on loyalty might actually make preliminary analytical work more important, so that when you have someone who’s loyal but also intelligent, they can effectively engage with the material. Let’s take Pete Hegseth as an example, he is the Secretary of Defense. He was a Kennedy School student. He’s smart enough, so is able to learn and evaluate information, whether he ultimately agrees with it or not. Or consider Stephen Miran, who is the head of the Council of Economic Advisors. He has produced academically sound and thoughtful analysis of economic policy, even though it it is very contrary to the prevailing consensus. Or look at Robert Lighthizer, who was the trade representative in the first Trump administration, and whose chief of staff is now the trade representative. Lighthizer wrote a book that you can read that details his perspective.

I think the virtue of having expertise in an analytical community is that you can actually have debates about issues on which people can agree and disagree that improve their views. I would wish that the loyalty requirement was less powerful than it has become in the US. Singapore, to take another example, prioritises competence. They assume that the people in government are going to be loyal, and it’s viewed almost like an ethic, which is what it used to be in the American context. So we’ll see. The British are another story entirely.

As you mentioned earlier, you’ve written extensively about US-China relations, coining the term ‘Thucydides’s Trap’. In your recent article published in the Washington Post, you noted Trump’s surprisingly conciliatory campaign rhetoric towards China, which stands in contrast to the hawkish positions of his own national security team. How do you see this tension between presidential messaging and institutional resistance playing out in US-China relations, particularly given that key figures like Secretary Rubio have built their careers on confronting Beijing?

Good question. If I try to think about the relations between the U.S. and China, this presents a great case for students to analyse what is playing out in real time. I, as a student of international affairs, start with structure, which I believe determines 75-80% of what happens. So that’s the place to go to begin with. And in the structural relationship between the U.S. and China, this is and will be the fiercest lucidity and rivalry of all time, which is why the relationship is likely to become increasingly contentious.

Next, I look at the views of citizens via public opinion, since we have elections. American views of China have grown increasingly negative over time. They used to be positive, then they became neutral, and now they’re negative. This is typical in a Thucydidean rivalry: when someone who used to be ‘the little guy’ is suddenly in your face everywhere, tensions naturally rise.

Then we consider key administration members, whose relative importance varies between administrations. During Trump’s first term, he was a novice. He didn’t know exactly what he was getting in for. He appointed people that look like they should be appropriate. Mattis, who looks like a Secretary of Defense or McMaster who looks like a National Security Advisor. They came with their own views, and when Trump would propose something they disagreed with, they would push back and say ‘That makes no sense, we think you should do this instead’. And then he would fire them. Now, having experienced this during his first term and having four years to reflect and plan, he’s made it clear to everyone in key positions: ‘You work for me. You execute my decisions. If I want your opinion, I’ll ask you’.

So now how does that work? As a student of government and bureaucratic politics, I recognise that the emperor cannot know everything that everybody’s doing at every minute. There’s simply too much going on. The predilections of people in key positions will still end up being important. But here’s where we have the wild card in this case – Trump himself, who has a very different view, as I argue in that Washington Post Piece, about China than almost everyone else in the US. His view aligns more with business leaders like Tim Cook at Apple or Elon Musk. But I don’t think there’s anybody in the foreign policy community that’s been representing a version of a view like that. So that’s the stew, and now we wait to see how it will work.

In fact, at Davos, I made a big public bet with Ian Bremmer that by January 2026, we’ll be surprised by positive developments in U.S.-Chinese relations. I understand that that’s a counterview. If you haven’t read the piece yet, I would recommend it. It just looks at Trump’s campaign statements. He consistently says ‘I respect China, I especially respect Xi Jinping’. He says ‘my people tell me I shouldn’t say this, but he’s brilliant. He and I can do deals on anything’. This is one of the quotes that he’s used three or four times, including on the phone with Xi.

I’m watching for China to become involved in bringing a rapid end to the war in Ukraine. That’ll be interesting. They could both influence Putin and potentially serve as guarantors of any peace agreement – something even Zelensky has shown interest in. If they were to do that for the next six months, that would be a big achievement.

Trump has also announced plans to visit China within his first 100 days, so that would be the end of April. If he and Xi establish a framework for the relationship, and if officials like Rubio understand their role, things could change significantly. This administration also has a very serious, able chief of staff, Susie Wiles, who reminds people they probably ‘don’t want to be the first person to be fired’. Take Rubio’s stance on Ukraine as an example, he strongly supported American aid until last summer when, seeing Trump’s likely victory and potential cabinet position, he reversed his position in line with Trump’s messaging. I expect that, at least until a few people are fired, many will be looking over their shoulders and falling in line, but we’ll see.

Finally, what gives you hope about our ability to overcome the political polarisation in the US and what most concerns you about its trajectory?

I think that’s a question everybody’s wrestling with, and I haven’t seen anyone with a truly convincing answer. Looking at many items on Trump’s agenda that contrast with Democratic positions, roughly two-thirds of Americans actually align with Trump’s view. Take border security, for instance. Open borders, essentially, are insecure borders with millions of what Trump calls ‘criminals’ entering the US, using vivid rhetoric about them ‘sometimes eating our cats’. How many Americans think that [unrestricted immigration] is a good idea? 10%, maybe 15%.

However, when it comes to enforcement, there’s a disconnect. Americans want stricter border control, but they don’t want to see the consequences. When people are deported and there are cases of children being separated from their families, public sentiment might shift if the implementation seems cruel. It will be interesting to see how this plays out.

Similarly with DEI – about 75% of Americans think that’s nonsense, myself included. But I suspect the Trump administration will likely overreach in dismantling these programs in significant ways, which could trigger some pushback and adjustment.

Regarding regulation, almost all the business community loves the idea of eliminating regulations, and there’s no question that many of the regulations are strangling or limiting. But on the other hand, these regulations didn’t come from nowhere. They mainly emerged because letting financial institutions operate purely in their own interests, without any regulatory framework, led to events like the 2008 crash. It will take a little while to adjust, like a pendulum swinging between extremes.

Finally, the biggest factor will be whether the Democratic Party can reconstitute itself with a platform that makes sense to enough of the population. I’m particularly concerned about how technology, including social media, has fed polarisation in ways we don’t fully understand and continues to feed it in ways that will accentuate the polarisation, which is good for Meta or X but not good for the country.

So I don’t know. I don’t have a definitive answer.

---