Preet Bharara:

From CAFE and The Vox Media Podcast Network, welcome to Stay Tuned. I’m Preet Bharara.

Daron Acemoglu:

I think today when people call themselves Democratic Socialists, it’s a bit unclear whether they are in favor of a more radical economic agenda based on redistribution opportunities or whether they are still aspiring for partial nationalization of means of production. And that’s why I think it’s not a particularly good label, especially in the U.S. context where the public’s sympathy for socialist ideas has been always weaker than in Europe.

Preet Bharara:

Welcome to Stay Tuned. I’m Preet Bharara. My guest this week is Daron Acemoglu. He’s an economics professor at MIT and a recipient of the Nobel Prize. He joins me this week to discuss his new book, What Happened to Liberal Democracy?: Remaking a Politics of Shared Prosperity. That’s coming up. Stay tuned. Darren Acemoglu, welcome back to the show.

Daron Acemoglu:

My pleasure. Thanks for having me again.

Preet Bharara:

Since we spoke last, have you won any more Nobel Prizes?

Daron Acemoglu:

Nope. No, I’m done. I’m waiting for the Peace Prize, but it hasn’t arrived yet.

Preet Bharara:

And no Emmys or Grammys?

Daron Acemoglu:

No, no, no. Those are not forthcoming yet, either.

Preet Bharara:

If you’re very persuasive, and I know that you are, if you ask someone else for their Nobel Peace Prize, they might give it to you.

Daron Acemoglu:

Yeah. Well, I’m afraid that somebody may ask for mine.

Preet Bharara:

I know. I don’t think he wants the one in economic sciences. So you’re a very busy guy. I really enjoyed our conversation last time. I’m glad you’re back so soon.

Daron Acemoglu:

So did I.

Preet Bharara:

And you have a new book out.

Daron Acemoglu:

I do. I do.

Preet Bharara:

By the way, you know-

Daron Acemoglu:

That makes me even busier because of the book. Yes.

Preet Bharara:

… it does. Although less busy since you’ve finished the book presumably.

Daron Acemoglu:

That was like 11 months ago.

Preet Bharara:

I’m holding it up. So folks who listen and have been listening for a long time very loyally should know you can also watch these interviews on YouTube. We have a YouTube channel, you should subscribe. What Happened to Liberal Democracy? By you, and it’s not until you get to the title page, interestingly, that you see that there’s a subtitle. And the subtitle is Remaking a Politics of Shared Prosperity. Should we read anything into the fact that the subtitle didn’t make the cover?

Daron Acemoglu:

Yes, it’s long.

Preet Bharara:

You know why it doesn’t fit? I’m going to tell the listeners why it doesn’t fit. It doesn’t fit because there’s a big, gold circle on the cover that said, and I don’t blame your publisher, that says, “Recipient of the Nobel Prize in Economic Sciences.” Right there could have gotten the subtitle, but you needed to brag about the Nobel. So what are you going to do? So let me ask you a question. A lot of people presume that with an open and free liberal democracy comes a particular economic system called capitalism. I’m humbled to be talking to somebody who is as expert in these areas as you, but the originators and the theorists of capitalism who say that you really can’t have thriving democracy without capitalism and vice versa, how much time did they spend in the origins of capitalism in this country? And during the battles of the 1930s and minimum wage and child labor, how much did they think about and care about shared prosperity?

Daron Acemoglu:

So I would say that there’s a difference between recognizing that in the 1950s, early 1950s especially, there was huge racial inequality in the United States, largely undergirded by racial discrimination and lack of opportunity for Black Americans. But it’s also very important to note that the 1950s and 1960s and early 1970s were periods of rapid economic improvement for Black Americans. Employment expanded and in sectors that they were shut out of previously. Real wages of Black men and women started converging towards those of whites. By the way, then in the late 1970s and from 1980 onwards, that process stops and it is reversed, which, again, emphasizes that there’s nothing automatic about any of these things. I think the issue of capitalism and democracy is a particularly vexing one, but a lot of the complication comes from the use of the word “capitalism,” for which I am not a big fan.

The reason for that is because it’s become sort of an ideological label that extends beyond its initial remit. We tend to call Egypt under Mubarak, the United States under Reagan, Sweden under the social democratic government, Argentina. all of those we tend to call capitalists. But when you look at the details, they are tremendously different. All of those are so different that a single label for these models, I think, obscures more than it clarifies. And in that way, I think that what you would view as better forms of capitalism, I think are consistent with democracy and worse forms of it are going to start eroding the support and possibilities for democracy.

Preet Bharara:

To dive into the current political debate, because this is everything that you see on cable news in the last number of weeks.

Daron Acemoglu:

I don’t watch cable news.

Preet Bharara:

You are very smart and you’re going to live a lot longer than I am, although I cut down on X, on Twitter, so I think I’ve extended my life expectancy.

Daron Acemoglu:

Okay, that’s good enough for like 10 years, right?

Preet Bharara:

I want to jump back to the themes of your book, but just while you mention the problem with terminology, we have a new nominee to be a senator in what everyone still says is a purple state, Michigan. The word “socialism” also is wielded by the new mayor of New York City. You talked about other countries. The kind of socialism, whether it’s accepted on the part of the person who is being talked about or it’s thrust upon that person when we talk about it in the United States, is that something radically different from what we have? Or is it incrementally different or is it just semantics? Could you help us? Explain it to us.

Daron Acemoglu:

Look, let’s go back to the main pieces of my book, which I won’t explain in detail, but I think it’s useful as background for answering this question. My argument is that liberal democracy worked very successfully and was able to deliver on its many pledges for quite a long time. And today’s comforts, both economically, politically, as well as socially, could not be understood without liberalism’s aspirations and practice. That being said, it is in crisis today. And a lot of the politics of both left and right cannot be understood but through the lens of that crisis. That crisis creates a strong anti-establishment feeling. And one aspect of the anti-establishment feeling on the left is to find more radical labels. Socialism meant something very specific.

And then, over time it’s changed its meaning. Then, socialists themselves or Marxists invented the word “communist” to describe a different outlook, somewhat different outlook than socialism. Socialist parties, many of whom were originally called Social Democrats, started changing. So for instance, the Swedish Social Democratic Workers Party, which started as a Marxist party at the end of the 19th century, by the beginning of the 20th century had thrown away its Marxist roots and had started moving towards the form of social democracy as we understand it, which does not have as its aim the public ownership of means of production, which was a defining characteristic of socialism in the 19th century.

I think today when people call themselves Democratic Socialists, it’s a bit unclear whether they are in favor of a more radical economic agenda based on redistribution opportunities or whether they are still aspiring for partial nationalization of means of production. And that’s why I think it’s not a particularly good label, especially in the U.S. context where the public’s sympathy for socialist ideas has been always weaker than in Europe. So I think it taps into discontent, but it’s not a particularly good label.

Preet Bharara:

Would they do well to use your subtitle and go on the campaign trail and say-

Daron Acemoglu:

I would think so.

Preet Bharara:

… “I’m talking about a shared prosperity, ladies and gentlemen?”

Daron Acemoglu:

100%. 100%. I think my book’s message is that we do need some radical ideas as well as a new governing philosophy for liberalism and liberal democracy. But it would be unfortunate if that objective conflicted with other core liberal principles like freedom of speech, pluralism, inclusivity. And if it took the form of looking at the world always through oppressor-oppressed dynamics, which I think ultimately is not very inviting for liberalism’s best instinct to develop and it creates backlash.

Preet Bharara:

Shared prosperity, as we said, is a broad term. You spent some time defining it and you did a little bit of that in this conversation. But one way for prosperity to be shared, if you have a particular person or a class of people or an industry that goes bananas, whether it’s the gold rush or it’s oil or it’s the computer, and we’ll talk about that in a moment, or it’s AI, is through… What’s it? Oh, the tax code. What do you think liberal democracy and its promise and its remit for the purpose of ensuring shared prosperity means pragmatically and proportionally for a tax code?

Daron Acemoglu:

First of all, nobody, nobody today, even if they say otherwise, really practices or even imagines practicing what early 19th century economists used to call laissez-faire, where the government has no role in the economy. And as I described in the book, government’s role in regulating, providing the infrastructure, and enabling people’s equal access to opportunities is critical. And a very important part of that is the tax code. And a tax code has two distinct, related, but distinct roles. One is redistribution. You cannot have a liberal society without some public investment in education, healthcare, infrastructure, sanitation, et cetera. That requires tax revenue. So you need to put taxes in order to deal with that and often in the form of progressive taxation whereby the rich pay a somewhat higher proportion. The second is what economists would call corrective taxation or Pigouvian taxation after the economist, Alfred Pigou, would be the role of the tax system to discourage certain activities versus not.

I am hugely in favor of carbon taxes. When you use fossil fuels rather than solar or wind energy, and there’s reason for using sometimes fossil fuels because the wind and solar are still intermittent. I’m not saying that should be banned, but because you create pollution and a negative effect on society, it’s only efficient from an economic point of view and fair, I think, from a sort of societal point of view that you pay a tax. That’s the Pigouvian tax. So both of those are very important parts of it. And we do use the tax code for these kinds of things. But on the other hand, as I also emphasize in the book, actually our tax code is pretty screwed up. One of the most important inefficiencies of the tax code in my mind is that we treat capital and labor income very differently, which creates huge inequality, which creates a lot of room for accounting gimmicks.

For example, instead of getting income from your labor, which will be taxed at a high rate, you sort of pay yourself in the form of capital gains and then you don’t get taxed. But even more importantly, in the age of AI and automation, it creates an artificial excessive incentive to automate because using robots or AI or algorithms is cheaper than using humans because of the tax code’s distortions.

Preet Bharara:

But isn’t there also an incentive for investment of capital-

Daron Acemoglu:

There is-

Preet Bharara:

… which is also-

Daron Acemoglu:

… and-

Preet Bharara:

… what causes growth.

Daron Acemoglu:

… there is.

Preet Bharara:

Isn’t that how the classical folks justify the lower tax rate on capital?

Daron Acemoglu:

They do, but at the same time, we also want people or companies to invest in humans.

Preet Bharara:

Yeah. So should there be no difference in the tax code as between capital gains and earned labor?

Daron Acemoglu:

That would be my preference, yes. And we can go into the detail for why that is.

Preet Bharara:

What would be the practical consequence of that?

Daron Acemoglu:

The practical consequence of that would be, first of all, that billionaires would pay more taxes, and not in an inefficient way because when you tax, if you want to increase taxation on billionaires right now, keeping the rest of the system as given, you have to jack up their marginal tax rate. That’s very distortionary. If, on the other hand, you broaden the tax base, meaning that a lot of the things like their capital gains that are not being taxed start being taxed, that does not have the same inefficiency consequences. It’s actually efficient in some ways, and it increases tax revenue, it increases the fairness of the tax system.

Second, and very importantly, it would remove this artificial excessive incentives to automate work. So at the moment, if there are two ways of producing that are pre-tax equivalent in terms of costs, one is using workers, the other one is using, say, AI or algorithms, companies would massively prefer the latter because they avoid all the taxes that would come from using labor. That’s unfair and inefficient.

Preet Bharara:

So I have a few questions based on-

Daron Acemoglu:

Of course.

Preet Bharara:

… what you said.

Daron Acemoglu:

I don’t doubt that.

Preet Bharara:

That’s what I do for a living in both jobs. So automation you talk about a lot in the book. And when a new technology arrives, you have a very strong sentence, which is simply, and it has its own paragraph, “Then computers happened.” I think shared prosperity is a good thing. I don’t know that I care that much that it has a lineage going back to Locke or not, or mill or any of those folks or not. It’s a good thing. It’s an important value. But I just want to understand, is there not an argument that over time there’s disruption and there’s displacement, particularly when you have new technology like computers, we’ll come to AI later, that that is ultimately good and creates prosperity and shared prosperity and more opportunities? True or false?

Daron Acemoglu:

Nuanced.

Preet Bharara:

Okay.

Daron Acemoglu:

You know, Nixon said, “Give me a one-armed economist so that they can’t say, ‘Talk about one on the one hand and on the other hand.” But this is a very important question.

Preet Bharara:

They say that about lawyers, too. I heard somebody say once, “Are you one of those lawyers with three hands?” I’m like, “What do you mean three hands? I only have two.” On the one hand, on the other hand, and on the other other hand, people will say-

Daron Acemoglu:

Let me give you two examples-

Preet Bharara:

… yes, please.

Daron Acemoglu:

… of disruption.

Preet Bharara:

Yeah.

Daron Acemoglu:

And then, we can sort of reason about them. The first one is when very large quantity of very cheap Chinese imports started flowing into the United States first in the 1990s and then accelerating in the 2000s after the accession of China into the WTO, that was disruptive. Many companies that produced apparel, toys, furniture, textiles went under. But it was a very concentrated affair. Many communities were thrown into a recession from which they did not recover for three or four decades. Many people could never get back into jobs. And in many ways, they were unprepared for it and the process shook the foundations of shared prosperity in those communities. The second is the move out of agriculture into industry and service that started in the 19th century and went on during the first half of the 20th century in the United States. Today, we would not be the prosperous society that we are if we had remained agricultural.

But in the latter case, the key was that while agricultural jobs were being destroyed by harvesters, combine harvesters, better machinery, other sectors using modern technology, electricity, accounting, service technologies started demanding labor and, in fact, paying much higher wages than agriculture was at the time. And that’s the difference is one is disruption is associated with opportunities.

Preet Bharara:

Right. But was that happenstance and serendipity or was that some kind of industrial policy?

Daron Acemoglu:

Neither. I think-

Preet Bharara:

Nuanced, you’re going to say nuanced again, aren’t you?

Daron Acemoglu:

… exactly. So there is an element of serendipity and nobody planned it. So there is an element of serendipity, but several things were critical. First of all, the U.S. had already a good education system, which meant that children who were coming to working age in 1900s had the skills to go into the service sector and the manufacturing sector. That’s very, very important. Second, the U.S. was very open to new technologies. That’s why Ford and many other farsighted entrepreneurs jumped on the bandwagon of using electricity, reorganizing the factory floor, attempting to get into mass production and increase efficiency. Third, U.S. was already a democratic society by many means and had a relatively strong, by today’s standards, labor movement, which meant that none of those employers could shortchange, cheat, or manipulate workers that easily.

So the way that they actually tried to transition to that new production system was by paying higher wages and trying to attract, compete for workers. So all of these are very important elements of creating opportunities. So there is no guarantee that those opportunities are going to be created by some magic if we do nothing, if we don’t do the infrastructural investments, if we don’t protect the rights of workers, if we don’t adopt technology the right way and try to increase productivity. So that’s why today this agenda morphs into what I argue at the end of the book that we need a pro-worker AI agenda.

Preet Bharara:

I’ll be right back with Daron Acemoglu after this. This discussion is reminding me of an anecdote from Lloyd Blankfein’s memoir. We had him on the podcast, and I’ll get some of the facts wrong and listeners will correct me, but I’m just curious your reaction. His father, I think, worked in the postal service, and among other jobs that his father had was, may have been his grandfather, but I think it was his father, was sorting mail. And at some point during his career, towards the end of his career, the post office developed or purchased automated sorting machines. And I think Lloyd tells the story of either hearing about or seeing these machines in central post offices still wrapped or still in their boxes. And the plan was not to take them out and not to use them until there was enough attrition on the part of people like his father who would otherwise immediately be out of a job.

And so on the one hand, you have a delay in the use of something that, I don’t know, arguably could bring the price of stamps down or slow the rate of increase of the price of stamps, which is a consumer good. And on the other hand, doing something, some would say paternalistic, some would say humanitarian. Some would say just good labor policy. Do you have a reaction to that anecdote?

Daron Acemoglu:

Yeah, I think a better strategy would be, if we can do it, is what German and Japanese companies, especially car companies from whom these anecdotes come from, did with the introduction of robots. By the way, they were much, much ahead of American companies in terms of robot adoption, partly because they were already facing labor shortages. But rather than mass layoffs was they introduced robots for painting, assembly, welding jobs. What they did is that they took the blue collar workers and retrained them to do more technical jobs, more supervisory jobs, and other maintenance related jobs in a more robotized workplace. And it worked quite well. So the alternative of, “Let’s slow down everything,” versus, “Let’s just go as fast as possible and lay off everybody,” are not the only two options. And the reason why I mentioned that is because that’s more similar to the way that I described the move out of agriculture.

We go fast, but we at the same time create jobs. Now, if it is impossible to create jobs, then we have to ask why not? I don’t think there’s anything in the nature of technology then or even today that says you cannot create new opportunities. But in some other places you might actually favor a slowing down strategy. For example, would I be in favor of if I could go back now with knowing what I know about the inflow of Chinese imports to say, “No, the United States should remain a closed economy?” No, I think that wouldn’t be good either because we would not benefit from the International Division of Labor. But I think the speed with which that took place without any guardrails was not optimal. So perhaps a slower process may have been favored. So there are some circumstances in which slowing down things might be justifiable, but we have to decide that on a case-by-case basis.

Preet Bharara:

As you know better than anyone else, lots of behavior can be explained by incentives.

Daron Acemoglu:

Mm-hmm.

Preet Bharara:

True in the law, true in criminal conduct, true in government, true in business. What you’re describing that happened in Japan with robots, was that altruistic or was that good business or was it a combination? Or-

Daron Acemoglu:

No, it was good business ultimately because, A, they had labor shortages, so labor was valuable. B, they needed technicians. Robots are so much more efficient when you have a lot of engineer, a lot of technicians integrating them better. And C, they had employees that had years of training for that sector, so their knowledge was valuable. And then, D, and this is the nonprofits but relevant aspect, but very relevant, all of this was being done, especially in Germany, in the context of strong labor organizations. So unions also favored that. But employers didn’t at the end lose out.

Preet Bharara:

How about that? How about that? That’s a coalition you don’t see very often.

Daron Acemoglu:

Well, we used to see it much more in the U.S., too. It is the heart of the social-democratic compact in many parts of Europe that went on for eight decades.

Preet Bharara:

So how did the elites, and you have a lot to say about the elites, MIT professor talking to senior partner at white-shoe law firm, how do these goddamn elites screw this up, Daron?

Daron Acemoglu:

I don’t think you can blame just elites. I think we-

Preet Bharara:

Well, let’s start there. How about we start with the elites?

Daron Acemoglu:

Okay. Who are the elites?

Preet Bharara:

I don’t know.

Daron Acemoglu:

Perhaps the most controversial part of my book is that we cannot understand the crisis of liberalism without bringing two very important things together. One is our transition to a post-industrial economy. That’s why computers and digital technology as well as globalization plays such an important role. They really changed the pathways for shared prosperity in a critical way. But also, second, in a post-industrial economy, the nature of coalitions supporting the center-left changed everywhere, not just in the United States, not just in Britain, in Germany, in France, in Spain, in Italy, in Canada. In particular, workers’ parties, as their name suggests, got a lot of votes and gave a lot of representation to workers. Social democratic parties in Europe received more votes from workers where we people without a college degree, and those with college degrees used to vote for the center-right. Exactly the same thing in the United States.

Sometime starting in the late 1970s, that started changing. Today, everywhere in the developed world, center-left parties get their votes from the college-educated, and the lower education individuals are increasingly voting for the center-right or sometimes for the hard right. That is a big change. In fact, the numbers don’t tell the full story because at the same time that the center-left became more of the party of the college-educated, the college-educated in terms of their values, in terms of how they live, in terms of their priorities, started diverging away from the rest of society. But I wouldn’t say the college-educated are the elites no more than I would say every business person is an elite, but they are obviously in a more privileged position economically. And through the last half century, they have become more and more socially influential. And that’s why the solution moving forward that I suggest is to form a broader coalition in support of liberal democracy and liberal ideals.

Preet Bharara:

So on the one hand, we are told that our college campuses are full of woke young men and women. And part of that is caring about the oppressed and caring about inclusion and caring about income inequality and caring about the climate and caring about all these liberal causes, one of which presumably would be shared prosperity. And on the other hand, as you point out and others do, there’s an increasing gulf between those people when they graduate and the working class. How do you square those two observations?

Daron Acemoglu:

I think that is a critical question, and we need to go back to two important observations to make sense of it. The first one is that supporters of the left are still broadly committed to creating a fairer society. I think any account from the right that would say, “Oh, democratic parties, elites don’t care about workers,” is not fully true. That being said, both the priorities and the form of achieving those priorities shifted within the Democratic Party and to some extent diverged from the preferences of lower education working classes, those people without a college in the United States, for example. You see that, for instance, in the policies that the Democratic Party advocates for dealing with inequality. In the 1960s and ’70s, that included minimum wage, unions, trade controls, job programs.

Today, especially with a transition in the 1970s, which some people identify with the Watergate Babies or the Atari Democrats, that changed. Minimum wage still plays some role, but unions became much less important. Trade controls became taboo among the democratic elites. Job programs became downgraded, and the emphasis was more on taxation redistribution and transfer programs. The remarkable thing is that if you look at during the same period, the preferences of the working class hasn’t changed. They still wanted unions, jobs, job programs, trade restrictions, minimum wages. So that’s one aspect of the divergence. But even more important, cultural topics became much more important.

Now, I do not think that is completely frivolous. U.S. has always been a complex, diverse society and it was becoming more so during that process. So cultural issues have to be part of the debate. But the way that cultural politics became central to progressive activism created a bigger wedge between it and the working class. And I think to a large extent is the main reason why the working classes started leaving the Democratic Party in the same way that the reason why they left the Labor Party in the UK, et cetera. And I think all both of those are still consistent with many people on the left wanting to create a fairer society. But the methods sometimes diverge from liberalism and the objectives diverge from what part of the coalition that was so essential would have liked.

Preet Bharara:

The agenda has changed. It’s very interesting. Martin Luther King, Jr., died very young. And to most people, he’s known for what? Advocacy of civil rights. But people don’t appreciate, and I think people also lament his early death because he began speaking more and more about economic opportunity and economic equality. Those speeches don’t get as much attention in the history books. But a lot of what he was talking about is what you’re talking about.

Daron Acemoglu:

100%. And, you know, one thing he understood is actually that it would be both more durable and more effective if that economic agenda wasn’t just for the betterment of the Black people, who were being discriminated against, but he said, “No, let’s provide opportunities for all low wage people because that’s going to be a better way of pulling people up.”

Preet Bharara:

Right. And I think if he had lived and had a choice of what to talk about more and more, he would have been talking about these issues. And the reason I raise it is to the extent you talk about these issues and these agendas sort of splitting and diverging, they were unified in King, I think.

Daron Acemoglu:

They might well have been. That’s a great observation. But this is why I call these things sins of omission of liberal establishment. The inequality that was forming was already pretty clear in the 1980. And it was also clear that globalization and especially automation using new digital tools were deepening that inequality. But not many people spoke up against it and not many people try to find a way of still using new technology in order to improve productivity in our lives, but in the process create opportunities for the working people.

Preet Bharara:

Okay. So we have technological advances, huge disruptions. I mentioned we had the Industrial Revolution. We had the assembly line. You say, “And then computers happened.” In 30 years when you’re writing a book and you write the sentence, “And then AI happened,” what will be the immediate paragraphs after that sentence?

Daron Acemoglu:

And then all went to hell. No, I don’t know.

Preet Bharara:

We can end right there. Okay, thanks, folks. Thank you for being a guest again on the show. But is each wave of this totally different qualitatively or just in degree?

Daron Acemoglu:

In degree.

Preet Bharara:

Yeah.

Daron Acemoglu:

AI is in many ways continuation of digital trends. It’s very slow. So it’s not going to be like next year the world is different. I mean, not very slow. It’s much faster than previous technological waves, but it’s still slow because it’s still a developing technology and it’s spreading only slowly into the use by companies and workers. But there are scenarios in which AI could be completely destroying whatever is left of liberal democracy. And most importantly, if AI massively reduces job opportunities, we are starting from an already very unequal society. It would make us get much closer to a two-tier society in which there are some very well-positioned people who earn a lot, who control AI, who do things that are important for the AI economy, and then the rest who are dispensable either do menial jobs or no jobs. That kind of society is not consistent with liberal democracy.

Jefferson and American Republicans understood that very well. They thought that democratic governance is only possible in a society in which people are big stakeholders and have some equal standing or quasi-equal standing in society. And that’s an idea that goes back to the 17th century in Britain and in France as well.

Preet Bharara:

So there are some tech folks who think you’re very pessimistic-

Daron Acemoglu:

Absolutely. They do.

Preet Bharara:

… not by name-

Daron Acemoglu:

I’m crazy. Yes.

Preet Bharara:

… and who say what the paragraphs that they think would follow, “And then AI happened,” is that lots of people won’t have to work, but they won’t be deprived of food and shelter and entertainment and everything else because there’ll be so much prosperity. I don’t know if they mean there’ll be universal basic income, but they think these people who are displaced from the… I mean, seriously have said this, right?

Daron Acemoglu:

No, of course.

Preet Bharara:

You’ve seen this.

Daron Acemoglu:

Yes, they have. They have. Absolutely.

Preet Bharara:

And that instead of catastrophe, we’re talking about utopia. What do you think of those statements?

Daron Acemoglu:

Well, this is why it’s important to recognize, as you did, Preet, that this is a continuation of previous trends. Did we do that with digital technologies? We created much higher levels of productivity thanks to the internet. But look at what’s happened to inequality. I don’t think looking back at our track record, you can say, “Oh, of course we’re going to find the right way of steering this technology so that everybody benefits from it.” And moreover, while I cannot rule out a very different future in which people don’t work and are happy, our existing evidence suggests that people also need a sense of contributing to society, doing something useful, having a dignified existence, which isn’t just being a consumer that gets transfers from the government, whether it’s called a UBI or something else. And that’s why it’s so important that we provide the means for people to earn a decent living and share in the prosperity.

So no, I’m not optimistic that if we just do nothing everything’s going to work out. And I’m certainly not optimistic that ideas that sometimes even come from the left, one of them is called fully automated luxury communism, that somehow we’re going to create Marxist communist ideal-

Preet Bharara:

Boy, fully automated luxury, FALC.

Daron Acemoglu:

… I mean, it just seems so farfetched to me.

Preet Bharara:

Yeah, but AI seemed farfetched.

Daron Acemoglu:

Yeah, perhaps. Perhaps.

Preet Bharara:

It’s interesting how fast we get used to things. I was at a conference two or three years ago, maybe it’s three years ago, where someone said, “Throw out a topic and we will have this computer instantaneously write in the style of the economist a paragraph about the topic.” And everyone was bowled over, I mean, and it was a sophisticated crowd.

Daron Acemoglu:

AI is a very promising technology and it’s going to get more promising. But that promise doesn’t mean that every direction in which AI is developed is going to be beneficial for society. And that’s why I spend more than half a chapter on what I call pro-worker AI. Can we develop AI in a way that actually creates jobs and shared prosperity opportunities? And I think the answer is yes, but unfortunately that’s not the direction we’re going. And that’s why we still need more important debates on this issue.

Preet Bharara:

I know you have a lot to do. Thank you for your time.

Daron Acemoglu:

And so do you.

Preet Bharara:

Thank you for sharing the, not financial prosperity, but your insights. The book, once again, What Happened to Liberal Democracy? I hope a lot of people read this. It’s very important and very good work. Thank you, sir.

Daron Acemoglu:

Thank you for having me on the program. This was a wonderful conversation. Until next time.

Preet Bharara:

Until next time. My conversation with Daron Acemoglu continues from members of The Insider community.

Daron Acemoglu:

And liberalism is a big tent. And I am more concerned with the left-leaning part of liberalism because workers’ parties, Social Democrats, the Democratic Party in the United States have often been at the forefront of reforms.

Preet Bharara:

To try out the membership, head to staytuned.substack.com. Again, that’s staytuned.substack.com. After the break, I’ll answer your questions about FEMA emergency funding and whether the Attorney General should be an elected position. Now, let’s get to your questions. This question came as a post in the Stay Tuned Substack chat from Carolyn. “Can the federal government deny entitlement, Congress-approved or FEMA emergency funding to blue states like California?” Carolyn, that is a great question and one that’s on the minds of a lot of folks. The second Trump administration has repeatedly tried to condition FEMA funding on states’ compliance with policy priorities that are totally unrelated to the purpose of those grants. Many people, of course, would view policy priorities as a proxy for pure politics, red states versus blue states. In any event, those efforts have not always held up in court. Most people know FEMA from its responses to hurricanes, wildfires, and other natural disasters, but the agency does much more than respond after catastrophe strikes.

As part of the Department of Homeland Security, FEMA also helps states prepare for emergencies by funding measures such as fire prevention programs, dam safety initiatives, and cybersecurity improvements. To support that work, it administers, as you might imagine, billions of dollars in grants for emergency preparedness, Homeland Security, and other public safety programs. In 2025, the Trump administration tried to condition a broad range of FEMA grants on states’ cooperation with a totally different thing, federal immigration enforcement. President Trump issued an executive order directing DHS to withhold federal funding from so-called sanctuary jurisdictions. DHS then imposed new conditions on its 2025 grants requiring recipients to cooperate with federal immigration authorities, and states that refused actually risked losing DHS funding.

So what happened? Multiple states went to court. Where did they go initially? Rhode Island. They argued that DHS had imposed conditions Congress never authorized and that the policy violated both the Administrative Procedure Act and the Constitution’s Spending Clause. And they won. The judge ruled that the immigration enforcement conditions were arbitrary and capricious under the APA and, therefore, unlawful. But the administration has now tried a variation of that approach again. In July, DHS and FEMA imposed new conditions on the Homeland Security Grant Program, one of FEMA’s largest preparedness grant programs. Under the new rules, states can lose at least 20% of their funding if they refuse to comply with a series of election-related requirements. Once again, something completely unrelated to emergency preparedness and something that reeks of politics.

So now, we’ve got 25 states and the District of Columbia going back to federal court in Rhode Island. Their core argument is that Congress created these grants to help states prepare for terrorism, disasters, and other emergencies, not to give DHS authority over how states administer their elections. As the complaint put it, “Congress never gave DHS or FEMA authority to rewrite state election law.” So we’ll have to see how this case plays out. But if the earlier lawsuit is any guide, the administration looks to face an uphill battle, which I think is correct and good.

We can leave for another day an extended conversation about how, in particular, pitting red states against blue states is galling. But just a reminder that blue states like California are major contributors to the Federal Treasury. Governor Gavin Newsome has repeatedly noted that states like California send more to Washington in federal taxes than the state receives back in federal spending, which is not irrelevant to this discussion. When a natural disaster strikes, the people who live in those states, red, blue, or purple, should not be punished because of political disagreements between their state government and the White House. Period.

This question came as an email from Greg, who has an interesting thought. Greg writes, “What do you think of the idea of making the U.S. Attorney General an elected position? Would that create greater independence from the Executive and Legislative branches? And more broadly, would that be good for our democracy?” Well, Greg, thanks for the question. It’s a really interesting thought experiment. First, a little background, as usual. When Congress created the Office of the Attorney General in the Judiciary Act of 1789, there was actually a lot of uncertainty and debate about where the office should sit within the new government. An early draft of the bill gave the Supreme Court the power to appoint the Attorney General, but Congress ultimately decided that the Attorney General should be appointed by the President. And there’s a variety of ways that the chief federal law enforcement officer of a country is appointed and selected and vetted depending on the country.

But here in the USA, the attorney general did not become the head of a formal Executive Branch Department until 1870. That’s when, interestingly, Congress created the Department of Justice and placed the Attorney General at the helm. By the way, there’s a long succession of SDNY U.S. attorneys who like to point out that our district was created a hundred years before the Department of Justice, thereby presumably trying to justify a bit of our sovereignty. Anyway, so the idea that the attorney general must necessarily be structured exactly as the office is today is not something that was fixed from the very beginning. So your question raises a great point. Now, what would it look like if the President didn’t appoint the attorney general? So instead of looking to other countries, let’s look at some of the states within our Union. Only five states have Attorneys General appointed by the governor.

So only a 10th of the states use the federal model. Those states are Alaska, Hawaii, New Hampshire, Wyoming, and the state I grew up in, New Jersey. Maine takes a different approach, unsurprisingly. Its Attorney General is chosen by the state legislature. And in Tennessee, the Attorney General is selected by the State Supreme Court. In the remaining 43 states, the Attorney General is elected by popular vote. Now, I’ve been a vocal advocate for the need for the U.S. Attorney General to be independent from the President, but the Attorney General is part of the Cabinet, is part of the Executive Branch under the current system. But I would also worry that if you made the U.S. Attorney General an elected position, that could create a whole different set of problems. An elected Attorney General would not necessarily depoliticize the Justice Department.

First of all, by definition, the person would be a politician having been elected. And, of course, if the Attorney General were from the opposing party, the office could become a political rival to the President. And depending on the nature, character, and intentions of an elected Attorney General, he or she could wreak a lot of havoc. So imagine, and we’ve seen this in the states, but on a national level, imagine a Presidential election followed two years later by an Attorney General’s race in which a candidate campaigns on a promise like, “Elect me and I’ll prosecute the current President.” That wouldn’t be great I don’t think either.

We can see a version of this tension in the States. Pennsylvania currently has a Democratic Governor, Josh Shapiro, but a Republican Attorney General, Dave Sunday. They don’t always agree on a variety of things, but in particular on when the state should challenge the federal government. In the FEMA funding lawsuits I just discussed, Attorney General Sunday declined to join challenges against the President of his own party. So Shapiro instead brought suit through lawyers in his own Governor’s Office. And sometimes the conflict between a governor and an independently elected attorney general can become even more direct. State Attorneys General have on occasion sued their own governors over alleged abuses of power. That can be a valuable check on executive power, but it could also be a vehicle for partisan conflict depending on the motivations behind the lawsuit, and as I said, the integrity of the person.

So an elected Attorney General can create genuine independence from the Chief Executive, true. But independence on paper doesn’t necessarily mean de-politicization. And by the way, it’s usually the case just because of the kinds of news that breaks and that we talk about on the show that I and Joyce Vance and others are critical about the lack of independence of an Attorney General from the President. When we do that, we’re almost always talking about a particular enforcement action in the criminal realm. In other words, whether an Attorney General is taking a direct order from a sitting President to prosecute enemies or to protect allies. That is bad for the country. That is bad for justice. That is bad for ethics. That is bad for the bar, and that is bad for the country.

However, on the other side of the coin, it is important for the President and the Justice Department to be aligned on many policy priorities, not on questions like I just said, like which of the President’s enemies should be prosecuted, but on legitimate matters of law enforcement policy. Things like how to prevent crime in cities, how to set immigration enforcement priorities and allocate federal prosecutorial resources. On those things, there should be general policy alignment. So at the end of the day, Greg, it’s a complicated issue. There are complicated solutions. None of them are happening anytime soon, but there’s a structural kind of paradox or dilemma here. And as I’ve always said, and as I wrote in my book, and as has been a theme of my work in the show, a system depends on the people who are running it and the people who are interpreting the rules. And if they have good character and integrity, then you don’t see these problems. And if they don’t, you will.

Well, that’s it for this episode of Stay Tuned. Thanks again to my guest, Daron Acemoglu. If you like what we do, rate and review the show on Apple Podcasts or wherever you listen. Every positive review helps new listeners find the show. Send me your questions about news, politics, and justice. You can reach me on Twitter or Bluesky, @PreetBharara, with the hashtag #AskPreet. You can also call and leave me a message at 833-997-7338. That’s 833-99-Preet. Or you can send an email to letters@cafe.com. Stay Tuned is now on Substack. Head to staytuned.substack.com to watch live streams, get updates about new podcast episodes, and more. That’s staytuned.substack.com.

Stay Tuned is presented by CAFE and The Vox Media Podcast Network. The executive producer is Tamara Sepper. The supervising producer is Jake Kaplan. The associate producer is Claudia Hernández. The audio and video producer is Nat Weiner. The senior audio producer is Matthew Billy, and the marketing manager is Liana Greenway. Our music is by Andrew Dost. I’m your host, Preet Bharara. As always, Stay Tuned.