Trump vs. the U.S. Economy

This episode is supported by Wealthfront.

Markets can be unpredictable, but your cash doesn't have to be.

With WealthFront's cash account, earn 4% annual percentage yield on your cash from program banks with free instant withdrawals to eligible accounts.

Get a $50 bonus when you open your first cash account and deposit $500 at wealthfront.com/slash Ezra.

Bonus terms and conditions apply.

Cash account offered by Wealthfront Brokerage LLC, member Finras IPC, not a bank.

APY and deposits as of December 27, 2024 is representative subject to change, and requires no minimum.

Funds are swept to program banks where they earn the variable APY.

It has been some months since Liberation Day,

and we've seen tariffs come on and off.

We've seen them go up and down.

We've seen all kinds of other economic policies applied and pushed.

And the kind of amazing thing about where the American economy is right now is it's shaky, but pretty stable.

The The place where we're seeing some stress is in the job market.

You've seen the jobs numbers revised downward for recent months.

So is that the beginning of all this really putting pressure on the American economy?

Or is the underlying resilience and power going to push it through?

Then there is this other thing happening.

The administration, President Trump in particular, is not acting like someone.

with a lot of confidence in where the economy is going.

After the jobs numbers were revised down, President Trump in a seeming fury fired the head of the Bureau of Labor Statistics, replacing them with a more ideologically compliant, it seems, person, calling into question the future reliability of government data.

He's also been pressuring the Federal Reserve to lower interest rates.

The administration seems to be acting like they think they need more power.

over the tools of economic policymaking, over the flow of economic data.

So what is that going to mean?

Natasha Surin is president and co-founder of the Elle Budget Lab.

She's an economist and a law professor.

She has experience in academia and in government.

And her lab has been very closely tracking the effect of these policies on the American economy.

We recorded this conversation on August 8th.

As always with President Trump, things are moving fast, so we're not able to talk about his new nominee.

for Bureau of Labor Statistics, but I think the rest of it paints a pretty clear picture of an economy under a fair amount of stress.

As always, my email, ezraklinshow at nytimes.com.

Natasha Sarin, welcome to the show.

Thanks so much for having me.

So we are about seven months into this presidency, four months since the beginning of the trade wars.

How's the U.S.

economy doing?

You know, that's a complicated question because the answer is we don't really quite know yet.

The U.S.

economy before President Trump took office was doing really quite well, especially relative to the rest of the world with respect to our recovery from the pandemic.

So inflation had been very high, but was coming back down to the Fed's 2% target.

They had the last mile to go, but they were directionally there.

The labor market was quite strong.

And then President Trump took office.

And so many commentators at the time, I myself said, kind of the best case scenario for this economy at this moment is literally if the president does nothing.

If he takes credit for the direction the economy is going, it is a strong and robust economy and one that, and I'm sure we'll get to talk about it, is about to get a productivity influx from AI.

And then we didn't do precisely that.

Instead, what we did

was on Liberation Day and since there's been a trade war that's been initiated by the president and the administration with the goal of kind of remaking the global order.

And the consequences of the trade war are that it's the most inflationary policies we've seen in our lifetimes.

And so obviously it's starting to reverberate in the economy.

And the budget lab at Yale that I run estimates that we're going to see household prices increase by something like $2,000 a year.

We're going to see inflation uptick.

And we're going to see a weaker labor market as a result of all that's already been done.

I cover this professionally, and I will say that I am a little confused on where the tariffs are at this exact second,

that they've gone up and down so many times.

They seem to be applied somewhat inconsistently.

Where are we?

What is the effective tariff rate that the U.S.

is placing on the rest of the world and how does it compare to where we were a year ago?

So the effective tariff rate at the moment is around 18%.

Where we were when President Trump took office was around 2.5%.

So that is a very substantial uptick.

And that affects what percentage of the goods that people buy.

That affects essentially everything that people buy because it obviously affects imports, which are around 11% for our economy.

But importantly, it also affects the prices that people are going to see for domestic goods that compete with those imports.

Because if the price of imports are going up, then the price of domestic goods that compete with them are going to go up as well.

And by the way, those domestic goods often, what is even a domestic good?

Because if you take a car that like GM or Ford is selling, something like 60% of those car parts are imported and then exposed to these tariffs.

And so what we're starting to see is we're starting to see these price increases really across the board in essentially everything that consumers are buying.

Not services, though.

Not services, correct.

So importantly, yes, durable goods are the most heavily hit sector by these tariffs.

And durables are things like furniture, apparel, consumer electronics, like goods that people buy as opposed to services that are provided by workers in the economy.

There have been varying levels of tariffs already in place.

And one thing that we were being told was going to happen were huge shortages.

I remember a period of time when the Flexport CEO was everywhere saying that if you look in the shipping data, everything was about to break down.

I think we were expecting to see very sharp price increases on Amazon, at Walmart.

So far, things have seemed somewhat muted compared to some of the more alarming predictions.

Why?

So you haven't seen price increases that are as high as our models predict at this moment.

But to be clear, you have seen price increases.

So durable goods inflation over the last six months was the highest that it has been over any six-month period since the 1980s outside of the pandemic.

So you you are starting to see price increases.

But importantly, knowing that tariffs were a tool that the president was likely to lean on, what you had importers and retailers do in the months leading up to Liberation Day and in the months since, frankly, as these tariffs were paused before taking into effect, is bring in a lot of inventory.

Because the idea is if we can try to front run the tariffs, we can bring in stuff at the pre-tariff rate, and then we won't have to pass on price increases to consumers because we wouldn't have had to pay that tariff when we imported.

Eventually, and you're starting to see the Walmarts of the world and the Procter and Gamble's kind of telegraph this explicitly, your inventories are going to dry up.

There is just not enough space in the margins of those importers and retailers to eat the cost of these tariffs without passing them down to consumers.

Whether they're going to pass down 100% of the tariff or 70% of the tariff or 50% of the tariff, economists in different sectors are going to react differently, and economists have a lot of debates about those particular elasticities.

But the idea that you aren't going to see price increases from this set of policies, it just doesn't work with respect to the way the economy is structured, unless these tariffs are ultimately rolled back in meaningful ways.

So, there was a period of time after Liberation Day.

People were somewhat shocked by the size, scale, and idiosyncrasy, I will say, of the tariff announcements there.

And then it seemed that they reformulated what they were doing into a fairly flat tariff with the rest of the world and then an absolutely astonishingly gigantic tariff on China.

And then they seem to have paused a bunch of those astonishing and gigantic tariffs on China.

and brought back some of the tariffs on the rest of the world.

What is the structure?

It seemed for a while that what we were really doing now is a trade war with China.

Is that still how you would characterize what we're really doing?

Or no, we've flipped back from that.

Deeply not.

And this goes to something that I have struggled with as we have been analyzing these tariffs over the course of the last many months.

And even, frankly, during the election, because during the election, remember, President Trump was proposing a version of what you're describing, a high baseline tariff, but a relatively low baseline tariff, around 10%, and a 60% China tariff.

So the idea was like low rate rest of the world, really high rate on China.

Which seemed like a big deal.

And I was assured by people in the Republican Party, he was absolutely not going to do anything that crazy.

Correct.

And we're higher than that.

We are much higher than that.

Well, China hasn't landed yet.

China's not landed yet.

And neither is Mexico.

So like big open questions.

And all this gets to the question that I have really struggled with and can't quite.

answer for you now, which is what exactly are the tariffs for?

What is the point of the tariffs?

How do we measure success for this new ordering of global trade?

If the point of the tariffs were that we need to reassess our relationship with China from a national security perspective, it's incredibly important.

They're our adversary.

We must ally with the rest of the world.

and particularly in certain sectors of the economy, I think you would find like pretty broad-based support for a version of that type of trade policy and strategic decoupling or some such.

But the idea that the right way to effectuate relationships with our allies and our trading partners is by imposing really high tariff rates on them in an attempt, I guess, to move us closer to autarky.

And that's

autarky is a closed economy.

So we just do everything ourselves and sort of isolate from the rest of the world.

I don't quite understand

or have the ability to describe for you what the purpose of that sort of exercise is.

And I can tell you something that has already happened, which is that growth in our economy has slowed.

Over the last six months, our growth rate has been something like 1.2%.

It was supposed to be, as of last November, when we made projections, basically twice that.

So this is having a real effect on the economy.

It is slowing it.

It is shrinking it.

That's exactly what our models predict.

And that's exactly what economists writ large would expect to happen from these types of policies.

Let's hold for a minute in this question of what we're trying to achieve.

So one thing that I think is being attempted here is not properly understood as economic at all, which is a restructuring of

the way global trade works from the American perspective away from

fairly neutral rules governed in multilateral ways in trade deals towards bilateral deals between America and either individual countries or in the case of the EU, collections of countries, in which we fully exert our leverage to get a better deal out of them than we would get from participating in the pre-existing global trading order.

And so I think the big test case here was the EU trade deal.

The administration announced that recently.

They're very excited about it.

What was that deal?

Is it good for the U.S.

economy?

Does it show that Trump is winning the global trade war?

How do you describe it?

The thing that I'm bristling at as you're describing the EU deal and whether we've won or lost is if you look at what practically has happened as a result of, let's take that example in particular, effective tariff rates on imports from the EU were about 1.5% at the the beginning of the Trump administration.

They've gone up as a result of the deal to around 15%.

But just baseline, it's true that some non-tariff barriers went down as a result of this deal.

It's true some tariff rates on U.S.

exports to Europe went down somewhat as a result of this deal.

But practically speaking, the idea that taking a tariff rate of 1.5% and turning it into a tariff rate of 15% plus is somehow a win for Americans.

I'm just like baffled by the concept because no one would say that if you took the sales tax on certain goods and you increased it 15-fold, that was a win for Americans.

But effectively, that's what we've done.

And then it's true, there's these other provisions that are in the deal that the European Commission agreed to do a certain amount of arms purchases or oil purchases or investment.

But if you look under the hood of of all that stuff, and it's true vis-a-vis the Japan deal, also it's true vis-a-vis all these deals, there's not really much there there.

Like they don't even have authority over these are commitments to explore the possibility of different types of companies investing or different types of loans.

It's not really much from an economic vehicle.

Did Trump do this with China in the first term and China just never made the investments?

Never made the investments.

And by the way, if you were to make the investments, that would increase trade deficits, not decrease them.

So the whole thing is.

We're not going to get caught caught up in consistency here.

We're not going to impose a consistency that the Trump administration has not imposed on itself.

But let me try to argue this from their perspective for a minute.

The way I've seen them making the case for their trade deals, including with the EU, is look, we got into this negotiation.

We have now imposed this tariff on them.

They are not imposing, in many of these cases, an equivalently sized reciprocal tariff on us.

So there is a tariff on their goods, which either gets paid, which means more money for the U.S.

Treasury,

or there is an incentive for these things to be made in the United States, which is good for the United States.

And so either we get money or we get more domestic production, but it's win-win.

Yeah.

So a piece that I want to accept and somewhat credit the Trump administration for is if you think about how hard it has been to find ways to raise revenue in this country, and we can talk a lot about why that is.

The

reality is if you look at our estimates and if these tariffs stick at these levels, they are going to bring in something like $3 trillion to the fiscal in a world where I think we are fiscally unsustainable, debt to GDP is rising much too quickly.

And part of the reason for that is the tax bill that was passed last month, but it's rising.

Finding ways to raise $3 trillion

is a feat.

And that is just a fact.

And my problem with tariffs and my problem with this whole line of we've won and they've lost, it's not that tariffs are a tax and taxes are bad.

I actually think quite the opposite.

My problem with the tariffs and why I think you have seen these other countries choose not to retaliate is that tariffs are a bad tax.

And they're a bad tax for two reasons.

One is they're a bad tax because we traditionally think of the tax code as wanting to make it progressive, such that the people at the top with the greatest ability to pay are paying more and the people at the bottom in the middle who have the least ability to pay are paying less.

Tariffs do the opposite of that.

It's been the argument against consumption taxes in this country that actually we don't want to hit low and middle income people who consume most of what they earn.

But the other piece of it that's bad is that tariffs are ultimately a drag on economic growth because they're diverting activity from the most productive sectors in the economy into other sectors of the economy that we're trying to protect in this way and lift up in this way.

And maybe you can make an argument for national security reasons or some such, there are certain sectors of the economy you want to do that for.

But that type of distortion drags on economic growth.

Slow that down.

Give an example of what you're talking about.

So we put this tariff on the EU.

How does that distort U.S.

economic activity?

So we put this tariff on the EU.

We put lots of tariffs on the EU.

Some of the arguments that are articulated is that we want to do more manufacturing in the U.S.

It turns out that other countries have comparative advantages in the ways in which their economies have been structured to be particularly good at those types of activities.

And in fact, they do them in a low-cost way that is better than the way that we do them.

We do other stuff very well, but we don't do that as well.

If you create a structure where it is more costly to rely on the advantages that other countries have,

you are incenting us to try to do more of that domestically.

But the issue is it is not efficient.

It is taking activity that otherwise is efficient in our economy and concentrations in sectors of stuff that we are quite good at and instead encouraging us to do stuff that we aren't as good at.

And that is ultimately not a positive enterprise.

It's a second layer of the challenge for us from an economic perspective is because you are making it more expensive to buy goods.

People are going to buy fewer goods.

Demand is going to go down for stuff.

And you're going to buy fewer TVs.

You're going to buy fewer couches because they're more expensive.

As a result of that, the production and investment in those types of capacities is also going to decrease.

And so you're going to get a drag, not just because you're doing certain types of activities that you're not inherently very good at, but also because there's just less economic activity being done.

And as a result of that, less investment in the future.

This podcast is supported by PMI U.S., U.S.

Businesses of Philip Morris International.

What does it mean to be invested in America?

For the U.S.

businesses of Philip Morris International, it means we're invested in advancing science, giving adults who smoke better options.

We're invested in American manufacturing, helping local economies thrive.

We're invested in community, supporting military veterans and their families, disaster relief, and economic empowerment.

Because we're proud to be invested in America.

See how at uspmi.com.

This podcast is supported by T-Priority from T-Mobile.

At T-Mobile, our critical mission is providing unparalleled 5G coverage for when connectivity is mission critical.

Introducing T-Priority, America's best 5G network experience for first responders, even in times of congestion.

More capacity, faster speeds, better coverage.

T-Priority from T-Mobile.

Built for tomorrow's emergencies, ready today.

Discover more at t-priority.com.

For qualifying organizations on eligible rate plans, coverage is not available in some areas and may be impacted by emergencies.

If you value having choice in your money, your goals, and your future, then you know how frustrating traditional health care can be.

That's where HIMS and HERS comes in.

They offer access to personalized care for weight loss, hair loss, sexual health, and mental health.

Because your goals, your biology, and your lifestyle are anything but average.

Start your free online visit today at HIMS.com/slash Ezra.

That's H-I-M-S.com/slash Ezra to find your personalized treatment options.

Not available everywhere.

Prescription products require provider consultation.

See website for full details, important safety information, and restrictions.

So one way you might see if part of their plan is working is if we see evidence of a boom in investment in domestic manufacturing.

A lot more money going into factory construction, a lot more training, diversion of workers into manufacturing roles.

Are we seeing something like that?

So the data does not suggest that there has been much of anything that has happened yet i mean if you were making the argument that the data isn't telling us that story yet you would say maybe it takes time and it's not because frankly the terms and the parameters here are far from settled right if this was really about china then investing in vietnam and india as manufacturing centers for your iphones makes tons of sense because they're not china if it's really about the u.s they just put an i think it's a 19 tariff on Vietnam is where we landed.

50% on India right now.

50% on India.

There's also the political dimension of tariffs, right?

We have a 50% tariff on India because it's buying Russian oil.

We didn't put it on China, even though they buy more of that Russian energy.

We also put a 50% tariff on Brazil

because they are prosecuting Bolsonaro.

for attempting to execute a coup.

But we have an international solidarity with right-wing nationalists who tried to do coups.

So you can't have that.

So that is also a way in which the tariffs have become just a tool of geopolitical leverage.

That's definitely true.

And I'm a law professor, in addition to being an economist.

And so I'm kind of interested in some of the legal dimensions of all this.

And a thing that I'm struck by is that a lot of these tariffs are based on this authority called AIPA,

but it's essentially like emergency authority that the president gets in particular national security circumstances.

And the argument that somehow broad-based tariffs on our allies

have something to do with national security is a very hard stretch argument to be made.

And by the way, the argument that the Trump administration has made is that the trade deficits are unusual and exigent and require attention.

How you can then say that countries that have trade surpluses are somehow part part of this same authority.

It's just like an incredibly difficult set of arguments to make.

And I suspect that part of the uncertainty here is going to ultimately play out in the courts because these are far from settled questions, whether they even have the authority to do what they're doing.

Speaking as a law professor, has it been your recent experience that the Supreme Court seems unwilling to countenance arguments that we would have traditionally thought on the level of text are a stretch?

You know, I am an optimist on this dimension for the following reason.

I was struck that the Supreme Court decided in a case about removal that had nothing to do with the Federal Reserve to be pretty explicit that the president doesn't have the authority to fire the Fed chair.

The Supreme Court has the view that actually the president can fire a lot of officials who previously served particular terms that have traditionally not been subject to political whims, but not in the case of the Federal Reserve.

And honestly, it was kind of like a tenuous legal argument how they managed to exempt the Federal Reserve.

You slightly got the feeling from that one that the Supreme Court wanted to give Trump what he wanted, but they also didn't want to see in the long term all their stock investments go to zero.

I think that the Supreme Court has a hard job at the moment, doesn't it?

Because, and you also have to- Yeah, a hard job of doing poorly.

I think Congress isn't doing its job in the way that it should be.

Well, that's also true.

The judiciary, it's slow.

It's measured.

It takes time.

All these cases are taking time.

I think the Supreme Court giving Trump the level of powers and grants of heretofore unknown authority and part of the doing a fair amount of it through the shadow docket is a little shocking.

I hear you.

The sympathy that I have is it must be hard to be the Supreme Court.

And maybe I'm speaking as a law professor and that's why I have this sympathy.

But there's been all this debate over the course of the last many months about whether or not we're in a constitutional crisis, where if the court says something or a court says something to this administration, are they going to agree to abide by it, right?

They don't have any particular enforcement authority.

And I think what you are seeing is this court has a fair deal of respect for the rule of law, but also understands that it is in challenging times and it's kind of trying to pick spots.

And they're picking, they might not be picking the spots to the extent that you would like to see them or other critics of the court would like to see them.

But I am struck by the fact that they have picked some spots.

I think I am more pessimistic about this than you are, but I will bring us back to the tariffs here.

Let's say we have a year of, you know, again, 15 to 20% base tariff, 60-ish% tariff on China.

What does that world look like?

What does that world mean for the economy?

So practically, what it means is it means a smaller economy than you would have had in the world without this set of systems.

And our estimates are we're taking about 0.4 percentage points off of GDP annually forever, as long as these tariffs are in place.

Part of the challenge with talking about fiscal issues and talking about the economic impacts of these policies is it's often hard to make those numbers like tangible to people in a way that is meaningful.

So 0.4 percentage points of GEP.

It's like the way to think about it, I think, and Jason Fuhrman had a good piece about this, is that's about $150 billion a year.

That's about $1,000 out of every American family's pocket as a result of these tariffs.

Probably a little bit more, frankly, depending on where the China number lands.

Is it 50 or is it 80?

The

piece that is then interesting is

how much do people notice that $1,000 out of their pocket each year

if there are other things happening in the economy at the same time?

And so a question that I have wondered the answer to is right now, GDP growth is set to slow this year.

And to the extent there is GDP growth in the economy, it's on the heels of artificial intelligent capital expenditures.

If that ends up going better than we expected and contributes more to growth than we anticipated, is it going to wash away the loss from the tariffs such that it is actually hard to do the precise counterfactual if you're just isolating the impact of the trade policy?

Let's talk about AI for a second.

I was going to do this later, but I think we should do it here.

So one thing that my

economist-y-oriented friends have been debating recently, or I've seen them debating, is,

are we in a recession-net AI?

Yeah, exactly.

And what would it even mean to say we're in a recession-net AI?

So you can look at GDP numbers, we can break things down, and probably if you pull out all the AI investment, we're sort of recessionary, but that's not really, I think, a reasonable way to think about the economy because one, that money would be doing something else in that world.

Exactly.

And two, maybe it's a great investment.

It's all going to pay off.

But the thing about the AI investment is it might not pay off.

It might not pay off immediately.

There might be a huge amount of redundant investment in an industry that's only going to have a couple of winners.

So there's something a little bit frothy in this

in a way that makes it a not totally stable place to be expecting a lot of near-term revenue from.

Totally.

How do you think about it?

I mean, not stable is a bit of an understatement.

People keep asking me, what is going on in the market?

The market was down around Liberation Day.

We're like basically back to like between two-thirds and three-quarters of where the tariffs were at Liberation Day.

And the market doesn't seem to be two-thirds to three quarters of the reaction that it had in April.

Well, can I add one thing to that?

Rejek Karma, my old, who's a senior editor on this show and isn't at the Atlantic, he had a great piece about the stock market recently.

He made this point that there were years when when everybody said, oh, the stock market is just a huge bubble based on low Federal Reserve interest rates.

And then those interest rates went much higher and the stock market is yet higher.

So there's been a lot of people predicting for a long time instability in the stock market.

And it just keeps going up.

But it's very heavily built on these seven tech companies.

Well, that, and by the way, Warren Buffett is one of the people who for a long time has been holding a lot of money in cash precisely because he thinks that the market is overvalued.

And so that piece is true baseline, even without what we're witnessing in AI.

And then within AI, so much of it is about these particular giants that are really driving a ton of investment and frankly are responsible for a lot of the market.

And the reason why I think that is important is your sort of stability piece, which is it very well might turn out to be great.

And I, optimist, hope that it does.

Kind of great the way in the late 90s, you got a ton of unexpected productivity growth that was really about computerization and a bit about the internet in ways that you couldn't have predicted or anticipated a few years before.

But the thing that is nerve-wracking about it is it's kind of like a fundamental restructuring of the way that the economy is going to work and the types of tools that we're going to use.

And there's also, we haven't, we haven't talked about it and might not get to, is it going to displace jobs?

Is it a complement?

Is it a substitute?

Like we don't know the answers to any of these like big structural questions, but we're kind of like taking a bet.

So one way of thinking about it is that if AI pays off,

the way it pays off is a massive forward leap in per-worker productivity.

A massive forward leap in per-worker productivity, where AI does something we've not really been able to do before, and it functionally simulates human workers.

That is in many ways the pitch being made.

Other things have simulated human tools

or beasts of burden, right?

The car replaces a horse.

This is trying to simulate humans.

It talks to you like a human.

It does things that a human might do on a computer.

When you listen to Dario Amade from Anthropic,

a lot of these people talk about data centers full of geniuses or

drop-in remote workers.

It is very hard to see

how that vision of AI works

in a way that actually creates revenue without displacing workers.

What you are functionally doing is increasing by a conceptually limitless amount the pool of labor.

Now, we're not there yet, but there's some, it's very hard to imagine the bet paying off without it happening.

So I agree with you, I think, to a certain extent.

If you look under the

hood of the unemployment numbers, you don't actually see unemployment rising among youths or among anyone in the parts of the economy that are most AI exposed.

So in some sense, like it's just not there in the data.

And by the way, you also don't see any inflection associated with things like the introduction of chat GPT or some of the moment inflection points in the AI boom that would draw you to the conclusion that there's somehow this relationship between AI and the broader unemployment trends that we're observing.

What I have experienced is AI has fundamentally changed the nature of what it means to train for a job like mine.

Because it used to be that economics PhDs, like they would spend a lot of time learning how to code and a lot of time learning how to debug their code.

And they would be in these like state of help centers, like looking things up.

And none of that is happening anymore.

And in fact, that feels like dinosaur times almost, even though I don't feel like I'm that far removed from actually training to be an economist.

And so I have a lot of sympathy with your view that it kind of must be the case that progress or success here has to look like a kind of different workforce doing different things than it was before.

So the version where you do not have job displacement, to even just go to the example you just gave, is

it's great if all the young economists don't have to spend hours debugging their code.

It doesn't displace them to give them a tool that debugs codes any more than it displaced everyone to create calculators or, you know,

laptops or Adobe Photoshop or all the different tools we use.

Although the ATM is an interesting example, I always think about the ATM as an interesting example because Obama always used to talk about the ATM as an example.

And then it turned out actually we had more bank tellers than ever.

It's my favorite example.

I was going to give it to you.

But then that collapsed.

Yes, it did.

And now we have fewer bank tellers.

So there was a lag.

And there's a world where AI just makes everybody a bit more productive.

I can get my work done faster.

And so you can get more out of me.

And maybe that means either you want more people like me because we're more valuable or

something.

It's more that the level of investment I think to pay off requires something that looks more like replacement.

Yeah, that's interesting.

So the claim I'm making is not that you can't imagine a world where AI just makes everybody a little bit more productive.

The claim I'm making is that the level of investment doesn't look like people preparing for that world.

A version where we do get waves of displacement, I have wondered, does it move us in the direction of like the Keynesian, we're going to have 15-hour work weeks because there's going to be all of this work that can be done ultimately in an automated way by agents that are free.

And what does that mean for society is like a question that we just haven't really grappled with.

The transition between here and there is nightmarish.

That might be a great place to end, but between here and there is a nightmare.

I feel like a lot of people draw the lessons of the China shock literature as telling us something about trade.

And it does, but I think it really tells us something about the job market and labor and how difficult actually like the words we use, upskilling, retraining, apprenticeship, how hard all of that stuff is in practice.

Yeah, we are very bad at doing that.

A A future I have considered

is

something happens that pops the quote-unquote AI bubble that makes people pull back in the way that the dot-com boom made people pull back.

But that doesn't make AI a non-useful technology.

And in fact, the recession it causes is the time in which people, in which companies begin trying to build AI into their firms at the ground level, replacing people, which is something we often see in recessions.

Those are often big periods where firms retool themselves around new technologies.

I don't want to be too doom and gloom about this.

It could go many different ways.

It just

seems like one of the ways it could go.

That's like very interesting.

I have, we talk about it in fiscal too, that

the nature of our unsustainable debt, how do you deal with it ultimately in a world and a climate that doesn't feel like it cares that much about deficits?

And a thing that people say sometimes is what you want is you want a small fiscal crisis because a small fiscal crisis, whatever that means, is going to focus the mind and bring all these policy compromises that you couldn't imagine elsewhere.

You're going to raise a lot of revenue.

You're going to find ways to cut spending.

And a little bit what you're saying in AI almost is you want like a small crisis because the small crisis is going to give you the scope to actually weed out what is froth and bubble from what is real and tenable and actually figure out how to deploy a lot of this technology in ways that are ultimately going to be the future of the landscape.

It's just right now, there's like a lot of different, very large figures chasing a bit the same future.

What would it mean for the level of investment we're seeing in AI to pay off, and how quickly would it have to happen?

Like, how patient are the investors here?

You know, it goes back a little bit to what I was saying about the late 90s and the productivity boom.

In that the late 90s, you got like productivity growth of around 3% in an economy that for the prior decade had been closer to 1%.

So big productivity enhancement.

And that productivity enhancement was really about computerization.

It was about learning how to take the technologies that had been developed really 15, 20 years prior and figuring out how to deploy them in the way that that businesses do their work.

I suspect

that even transformational technologies take time to fully realize their potential.

And

I do actually think, having used it, as I'm sure you have, we're talking about like big transformational technology that's going to make my work better.

I don't know whether you should anticipate, but again, famous last words, whether you should anticipate next year, this time, us talking about a ton of displacement in the labor market that's attributable to AI.

But right now, and I think this is an important point because people have been hearing these conversations about what's happening with young workers.

The labor market people often describe as a little bit frozen.

There's not that much hiring, but there's not that much firing either.

In the data for college graduates, in the data for young workers,

we are not seeing AI displacement.

We are not seeing evidence of AI displacement right now.

We are seeing a lot of vibes.

So that's a bit what you're describing, but it's not in the data right now.

It has not happened in any way.

There's another dimension of AI versus jobs.

There's also AI versus wages.

So if you take the economic analysis that the Trump administration applies most often to the economy, which is that more labor means lower wages, That is their fundamental view of immigration,

that more people here is bad for you, even when they're fairly different than most American workers, right?

It's very different to have a worker who can speak English and a worker who cannot.

They are bad for your wages, right?

That is what J.D.

Vance thinks.

That is what Donald Trump thinks.

That is animating a fair amount of administration policy.

I do think that if you apply that theory to AI, and I'm not sure that I do because I also don't apply it to immigration, but I think if you apply that theory to AI, you should be quite concerned.

And I think it's interesting they don't seem to see it that way.

So it's, so first of all, the reason why you don't apply it to immigration is because it is not true.

Right.

There is just the empirical evidence on this question, and it is vast, is that there is not a decrease in home wages that is associated with immigrants or immigration into this country.

And actually, it's a, you're saying one version, which is, shouldn't they be worried about AI on these dimensions?

I feel like the flip side, shouldn't they be less worried about immigration on these dimensions?

Well, they should definitely be less worried about immigration.

And by the way.

And births, right?

Births.

I mean, so.

As I said, consistency is not the strong suit of this administration, but if you have their view of immigrants and wages, you weirdly should probably not have their view of fertility rates.

Because you don't want more labor.

No, you don't want more labor.

And there is much more competition between native-born labor than between native and foreign-born labor.

So maybe their argument to you would be, again, I don't agree with any of this, but I'm just trying to, maybe the argument would be that we have such deep demographic challenges, which we do, right?

We're an aging population, that you actually want a bunch of new kids to enter the labor force in 20 years, but you don't want them now.

But again, that doesn't make any sense.

I don't think that would make it.

They would make that argument.

And I don't think it makes any sense, right?

In fact, by the way, a thing that I was just talking to someone at the Congressional Budget Office about this this week.

A bunch of the challenge that we face right now as a country has to do with labor supply.

Obviously, you know this.

And the productivity estimates that exist, which show productivity growth in the next decade, but relatively limited productivity growth, about 1.8% in the next decade.

They are on the backs of models that believe that we are going to have a lot of immigration into this country over the course of the next decade, like we have in the last decade.

And so that productivity growth kind of falls away in a world in which you're not getting this labor supply.

Yeah, just in a very simple simple way of describing the future that the administration is trying to build, we have an aging population.

We have a falling birth rate.

And they want to squeeze immigration down to a trickle or near zero while deporting large numbers of workers.

And if you held that policy for an extended period of time, that would look just structurally very bad for the economy.

You would have no productivity.

They really need AI.

They really need.

That's a little bit what it's like you're thinking about is, are we in an absent AI recession?

I think we're banking a lot on AI.

And to that, we are also adding this like man-made uncertainty in trade that doesn't actually have the potential upside that you can get from AI.

It's just risk.

It's just bad.

We've sort of been circling jobs.

Let's talk about the job market.

There was a jobs report that came out very recently.

It led to a sort of secondary story we'll talk about about firing the head of the Bureau of Labor Statistics.

But before we get into that, just what did that job report say and what should it make us think about the labor market?

So over the course, first half of this year, there was a bit of a puzzle because the economy was slowing down.

And that is because of, again, our estimates at the budget lab suggest that is what you would expect in a world in which you have tariff rates that have gone up seven or eight times relative to what they were in January.

But at the same time, the labor market was looking very resilient.

And so you'd expect as the economy is slowing, that the labor market would be slowing.

Fewer firms would be hiring people, or even they would be firing people because they're anticipating or they're watching the economy shrink.

And what the jobs number told us is that in fact, there is no puzzle because that is happening.

Over the course of the last few months, hiring has stalled very significantly in essentially all sectors of the economy outside of education and health services and AI a little bit.

And I think that that fact shouldn't be that surprising to us.

Again, we still have a pretty strong labor market.

Unemployment is still just a tick above 4%,

but it is starting to show the signs of a labor market that is under some pressure from this trade policy.

One of the things I thought was interesting in the jobs report is if you break the labor markets down by sectors.

And as you mentioned, the growth in labor was coming from really primarily healthcare.

So there's a sign from Bloomberg.

Without healthcare, the last three months of payroll gains look like this.

Minus 53,000 jobs in May, minus 45,000 jobs in June, and minus 300 jobs in July.

Now, there's nothing wrong with healthcare jobs.

Many of them are very good jobs, but it's not a high productivity sector of the economy.

It's primarily, in a way, caring for older people.

It doesn't look very dynamic where we're headed here.

Well, and another thing about healthcare, by the way, is it's one of the sectors of the economy that's least impacted by the trade policy, right?

So, in part, I think what it's telling you is that the rest of the economy is like really bearing the brunt, like the tariffs are really starting to bite.

And it is going to continue to show those signs in the months ahead, I predict.

This podcast is supported by T-Priority from T-Mobile.

At T-Mobile, our critical mission is providing unparalleled 5G coverage for when connectivity is mission critical.

Introducing T-Priority, America's best 5G network experience for first responders, even in times of congestion.

More capacity, faster speeds, better coverage.

T-Priority from T-Mobile, built for tomorrow's emergencies, ready today.

Discover more at t-priority.com.

For qualifying organizations on eligible rate plans, coverage is not available in some areas and may be impacted by emergencies.

If you value having choice in your money, your goals, and your future, then you know how frustrating traditional health care can be.

That's where HIMS and HERS comes in.

They offer access to personalized care for weight loss, hair loss, sexual health, and mental health.

Because your goals, your biology, and your lifestyle are anything but average.

Start your free online visit today at HIMS.com slash Ezra.

That's hims.com slash Ezra to find your personalized treatment options.

Not available everywhere.

Prescription products require provider consultation.

See website for full details, important safety information, and restrictions.

This podcast is supported by GiveWell.

This year's cuts to U.S.

foreign aid have made headlines, but it's hard to know what's actually happening or how to help.

Over the past 18 years, GiveWell, the independent non-profit research organization, has helped guide more than 135,000 donors and $2.5 billion to highly cost-effective aid.

Its researchers are analyzing the impact of USAID funding cuts in real time and sharing what they're learning for free.

For trusted, evidence-backed insights into this rapidly evolving situation, follow along at givewell.org/slash USAID.

So, one of the things that got a lot of attention in this jobs report was the pretty big revisions to the previous couple of months.

Just walk me through what revisions are, why they happen.

Totally.

The way way that the bureau of labor statistics collects data in real time about the labor market is it surveys about a third of non-farm employers and it gives them a relatively short period of time to respond to that survey because again it's trying to be in live time these are monthly numbers and they do a bunch of sample size adjustment and you know sort of trying to make it more representative of a sample but that's basically the exercise and then there are always revisions twice.

They revise the numbers as they get more information, more people answer the survey.

I know the numbers sound big to people when they hear them, and the president says 258,000 jobs or some such were revised.

You really need to put the numbers into a ton of context.

So, first of all, 258,000 jobs revised downward for May and June.

That is just a tick above point, it's about 0.16%

of the labor force.

It is a relatively small revision.

And in fact, over the last two decades, the Bureau of Labor Statistics has gotten much more accurate with respect to its reporting in real time, which again is a hard thing to do, about the state of hiring and firing in the American economy.

Thing that I find really concerning

about all of this and this moment is that the rest of the world sits with envy about how good government data collection is in the United States.

It is like the gold star of data collecting.

There is a ton of trust by markets, by regular people, by other countries in what these numbers mean and what they represent.

And a ton of transparency with respect to the methodology, like how the data is collected and what it's telling us.

I just worry that the politicization of our data collection is like a much bigger and more important and more troubling story than I think it's gotten.

It's gotten a ton of attention as it should, but it deserves even more.

And it's not an isolated example in some sense of what we've seen in the last few months.

One thing that people who understand this data better than I do tell me is that the commissioner of the Bureau of Labor Statistics or the head of it doesn't actually have that much

in the data.

They only see it very shortly before it goes out, anyway.

It's survey response, it's coming in.

You can actually see the raw numbers.

So, on the one hand, the opportunity for monkeying around with this

is not so dramatically increased by this move.

And on the other hand, the

intention

of the move

and the effort to sort of bring things under a kind of control

is very worrying.

How much do you take this as just like Donald Trump got mad at somebody,

but it doesn't really matter versus in a year we may not be able to trust the jobs numbers?

I hope that you are right.

I've always made the point that BLS has about 2,100 employees and one political appointee.

And the political appointee is essentially the person who like delivers the envelope with the numbers.

Like they're not involved in the construction of or in the presentation of these estimates.

And there were no signs that there was any reason to be concerned, obviously, because if there were any politicization or monkeying around in the data, you would have legions of dedicated civil servants who would be out the door and telling you that this is like not to be trusted.

But I think the idea

of

politicizing economic statistics

is so deeply disturbing and dystopian and authoritarian.

You know, in Argentina, there was a ton of pressure put to report friendlier inflation and poverty estimates.

And ultimately, investors realized that they were being duped and decided that they, for the longest time, would not actually make new international loans to Argentina because they couldn't trust the data.

Or in Greece, you literally triggered a sovereign debt crisis on the heels of underreporting deficits and having sort of public arguments that resulted in criminal prosecution for statisticians who tried to present truth.

And obviously, though we are not Argentina and we are not Greece, but I think going in the direction of starting to draw into question some of these like fundamental truths that we like believe in numbers, we report the numbers, we can trust the numbers.

I think it's just like a really slippery slope.

And frankly, I don't know how you undo the new nervousness that we all seem to be feeling.

There's also another way that data can degrade, which is the Bureau of Labor statistics has seen fairly large attrition.

20%.

20%.

This has been amidst Donald Trump and Doge sort of declaring war on the administrative state and plenty of people not really wanting to work for them.

And then you just imagine that Trump puts in charge,

even if they can't do all that much damage, just some right-wing member of Congress who is primarily known for being a lick spittle to Donald Trump.

And you might just see a lot of good people leave.

And the people who'd be coming in would be less good because the reputation of the Bureau of Labor Statistics would have degraded to the people who it needs to attract, which is

very, very,

very

literal-minded statisticians who believe very heavily that the integrity of data collection and data inference is like the highest good in an advanced democracy or an advanced society.

And what you do is you just begin to break the talent there

and you demoralize it and you dispirit it.

And if you do that to any organization that does anything whatsoever,

you will get a worse output.

You will get a worse product.

And in this case, the product is economic data.

So I am so worried about that.

And you're already starting to see it happen across government.

And particularly with respect to the BLS, like a sort of statement that they make, which tells you a little bit about the temperament of the people who work there, is they won't tell you whether a glass is half empty or half full.

What they'll say is that eight ounce glass has four ounces of liquid in it.

So that is the degree to which they do not spin in any direction.

And they take that responsibility so immensely seriously.

And it's one of the things that makes government like a remarkable place to get to spend some time is to be around these civil servants who take so much pride in the work that they do and the contributions that they make to our ecosystem.

BLS has already lost a very significant share of the labor force and its data collection efforts have already degraded.

So it's already the case that certain aspects of price indices that they used to collect in certain parts of the country, they don't have the capacity to do that work anymore because they've already suffered.

It's also the case that response rates for the survey that I was describing, they've actually gone down post-COVID.

And that's a problem.

Yeah, from about 70 to 40s, right?

And particularly for small businesses.

So response rates have declined.

And so I do think there's like some deep irony here.

And the survey responses are in a situation where we would like to see improved response rates.

And it's something we would like to see the BLS invest in.

And yet precisely those aspects of government that we want to improve, we are simultaneously taking away their resources and capacity to improve.

We are making them places where civil servants don't want to work because like Russ Fought promised, every day that they wake up is a nightmare.

And so I think these things, they have longer term repercussions beyond this administration.

We are so lucky to have people who could get paid much more if they go to the private sector, feel like they're dedicated to government statistics or to improving tax collection in this country.

And we're sort of chasing a lot of talent out the door in ways that are really going to redound and make some of the problems that have rightly been identified by this administration much worse.

Among things that depressed me about this episode, one of the ones that was high up there was Kevin Hassett, the head of Donald Trump's Economic Council, endorsing the decision.

And like a lot of reporters who have covered economic policy in Washington for a long time, I've known Kevin a long time.

He's very much part of the sort of Washington economic policy community.

And the Kevin I knew for many, many years, and I believe the Kevin who existed even during the Biden administration, like if Joe Biden had done this, Kevin Hassett and others, frankly, in the Trump administration, would have lost their minds correctly.

Correctly.

And so, on the one hand, I don't really think that the replacement of the BLS lead is going to change economic statistics that much, at least in the near term.

But it was the bottomlessness of complicity

that disturbed me most?

Because there are things that you can politically influence.

And some of them are very big and well-known and happen in public.

Many of them don't happen in public.

And they're smaller.

And that there's just no line for people in this administration, that the nature of being in the administration is that you do not have a line,

is very worrying.

And I know, I can imagine how you would rationalize it to yourself.

Say, Donald Trump is a president.

If he doesn't like the Bureau of Labor Statistics Commissioner, like we get rid of them, that's his prerogative, right?

And I'm sure he'll replace him with someone good.

But this world where you fire people because data came out that you don't like,

that is a terrible regime.

We've seen it in many other countries.

It does not work well.

And lots of the people in Congress, in the administration,

know it

and won't say it because of the rules of being a Republican Good Standing or whatever Donald Trump does,

you stand up and you clap.

I am of the view that it is impossible to defend these types of decisions.

And I hope and feel confident, frankly, that if I were ever around them,

that this would be the type of line that would mean that you no longer are going to be serving in an administration in this capacity.

I understand why if you are so short-termist that you are thinking tomorrow, maybe you tell yourself, like, this chart looks cool or these numbers are good.

But if no one believes it, and if in a year or two years you've denigrated the U.S.

economy, a little bit you have to ask yourself, like, for what?

Well, I think in the case we're talking about, I think the for what is

There are a lot of people who would like to be Fed chair.

I think Kevin Asset is one of those people.

You're not not going to be Fed chair if you oppose something.

Donald Trump does.

And maybe that's a good bridge to the Fed, which is another place Trump has been attempting to exert some pressure.

There was a burst of news that maybe a tried to fire Jerome Powell, which, as you know, Didi doesn't really have the authority to do.

Then there was maybe he will accuse him of a form of fraud related to the rehabilitation and

refurbishing of the Federal Federal Reserve's headquarters, which, as you can imagine, I'm sure Jerome Powell spends a lot of time thinking about the specific contracting decisions being made in that.

And it sort of seems to have blown over.

But Powell's term is limited.

His term is up in 2026.

And behind this is Donald Trump and the administration really want interest rates to go lower.

I guess a good place to start here is

first just on the merits, are they right?

We are seeing the economy slow down some.

Maybe the Fed should lower interest rates.

On substance, I think the Fed actually has a bit of a hard task ahead of it because they're in a situation where, like where we started in this conversation, the tariffs are the most inflationary policies of our lifetimes.

So every model, our models, everyone else's external models are predicting that inflation is going to rise as a result of tariffs at these levels.

And remember that the Fed just had a very significant bout of inflation that it hadn't even fully managed back down before this next round of inflationary policies kicked in.

And until this jobs report, by the way, the Fed was operating with a fair bit of let's wait and see with respect to the economy because the labor market looked quite strong.

And so if the labor market is strong and you're anticipating inflation, there are real reasons to think like this is not the environment in which to cut interest rates.

And you see that there is actually like room for debate on this particular topic.

The Fed made its last interest rate decision ahead of the most recent jobs estimates.

But you saw two governors dissented and kind of explained their dissent as exactly about this.

They were looking at indicators that were suggesting that the economy was slowing down.

And they were of the view that it was time for the Federal Reserve to cut rates.

I will note their view was to cut rates by something like 0.25%.

And the president has been calling for a decrease of around 3%.

Which would be a lot.

Which would be a lot.

What would happen if they did that, right?

Let's imagine next meeting.

Yeah, yeah.

Chair Powell just comes down and says 2% cut, which would that be the biggest cut in history?

I'm sure.

It would be huge.

Yes.

What would happen if he did it?

So there is a difference between the interest rate that the Federal Reserve sets, which is the interest rate that impacts how banks lend to one another and borrow from the Fed.

And then there's like interest rates in the economy that you and I and others care about, like your mortgage rate or your student loan rate or your small business loan rate.

And there's not a direct translation between those two objects.

But they are connected to each other.

They're connected to each other because of a belief that when interest rates go down, the Fed, they are telling us a story about the direction the economy is going.

But the thing that's perverse about the idea of tomorrow we wake up and the Fed funds rate is 1%

is if it was 1%,

what the world would understand is that interest rates over some relatively medium-term, long-term horizon are gonna have to go way back up because we're gonna get a ton of inflation.

And the result of that is gonna be that you are actually not going to to see translated into your mortgage rate or into your small business loan rate anything like a 1% interest rate, even if magically the Fed turned on and said that's what the interest rate should be.

So there's like a perversity in all of this that even if the Federal Reserve did in fact what the president seems to be asking them to do, it wouldn't actually deliver the type of economic benefit in any real horizon that would be meaningful to households.

And in fact, the reason I know that is because we have played this game before in the U.S.

So Nixon had a Fed chair named Arthur Burns, and he, ahead of a presidential election, put a ton of pressure on Arthur Burns to get the rest of the FOMC to lower interest rates dramatically.

And in fact, they did.

And what happened as a result of those types of politicized decision-making by the Federal Reserve is that we had inflation inflation in this country go up.

So the idea that this is somehow going to accomplish anything positive for the economy is like silly.

So the Trump administration is doing something really dangerous, which is they've got a bunch of policies that are slowing down the economy and a bunch of policies that are pushing prices up in the economy.

And if you have a slow economy that's not really adding jobs, not growing that much, and you have inflation going up, well, you have a stagflation.

And then the Fed doesn't really have an obvious move because if it lowers interest rates to speed up the economy, it pushes up prices.

If it raises interest rates to bring down inflation, it slows down the economy.

In the past, it has broken stagflation by raising interest rates so high, it pushed the economy into a recession, which I also think the Trump administration would prefer to have not happen.

But I've heard a lot of people saying that the way to think about the economy right now is we are in a mild stagflation.

And that if that continues and gets worse,

right, if the things we are talking about here happen,

the tariffs begin to pass through even more.

The big, beautiful bill is highly inflationary.

It's putting a huge amount of money on the national credit card, a bunch of tax cuts.

That

you sort of look around and the conditions are there for us to get into something that is pretty tricky to break when we get into it.

I think that is true.

And I've been saying that we're starting to see,

I'm nervous to say it's like the West Wing episode where they won't say recession.

Like, I don't want to say the word stagflation, but stagflationary signals or stagflationary adjacent type of information from the

time when economists don't like something when they treat its name like Voldemort.

Yeah, exactly.

I can't say it.

I don't even want to invoke.

But I will say that this isn't a crisis that has dropped from the sky.

We're not in COVID or even a financial crisis.

We have gotten ourselves from economy that was riproaring, just like let it go

and inflation coming down

to strongest economic recovery out of the pandemic to the S-word adjacent

by a set of policy choices that have been made that didn't have to be made this way.

And ultimately, like we couldn't even articulate exactly what they are meant to accomplish.

And so, I guess maybe that should make you a little hopeful, because in some sense, when markets have gotten shaky or the bond market with its disciplining device has gotten nervous, you've seen some pullback from those types of policies.

But it also makes you kind of dismal about the ways in which mistakes are being made that seem so obviously avoidable.

Aaron Ross Powell, I think the question it raises is

what happens if you have an economy that has this much uncertainty in it, this much weird policy in it,

and then you had some kind of external crisis that is actually hard to manage, right?

In the way that the inflationary shock was hard to manage in the sort of post-pandemic period, in the ways that many of the things that have caused recessions, like the dot-com bust was hard to manage.

I mean, maybe you can say the Trump administration has given itself a lot of tools to manage it because it could turn down the tariffs and that would be stimulus.

But it has added a lot of stress in a pretty good scenario that has gotten us to a point where people have begun at least talking about, to use your terminology, recessionary signals, if not a recession.

And

it seems to me they're hoping for a lot of luck here.

Well, you know, I don't want to be overly pessimistic in that I hope we are.

I don't think we are in a recession.

I hope we are not in a recession.

The probability of a recession has declined actually since Liberation Day as measured by market indicators.

But we have done a lot of self-inflicted harm here, and there's no doubt about that.

I guess it could be worse because part of what you're worried about is this fiscal situation becoming even more untenable.

And eventually you hit a fiscal crisis and everything was deficit financed in the big, beautiful bill.

And so that's really concerning.

But you should feel a little bit less bad about that than you would have baseline, because it is true that the tariffs are bringing in some revenue.

And I guess it could be worse because we talked a lot about AI and the possibility of AI and absent that sort of hope in the economy and whether or not it actually is realized, things could be worse.

But it's kind of a weird place to be from an economic policy perspective to continually be saying, I guess it could be worse as your barometer of progress.

It is interesting to think about the tariffs as a pay for for the BBB's tax cuts.

I have been thinking a lot.

about the tariff revenue because again, it's like $3 trillion over the decade.

It's roughly the size of the bill that was just passed.

And, you know, I wouldn't have used the deficit reduction from the tariffs to do a bunch of tax cuts that disproportionately benefit the top and social safety net cuts, but like just from a fiscal perspective.

And you're also in a situation where let's play out the scenario where it's 2029 and these tariffs have been in place for a while.

And for whatever set of reasons that have come, maybe the economy has absorbed them.

Maybe there's been a slight downturn, but then an uptick, maybe AI, something, but they're in place.

In that universe, I think the revenue is somewhat sticky, right?

I think the tariffs are terrible.

I think they're a very bad way to raise revenue and a very inefficient way to raise revenue.

But they are a $3 trillion tax increase that has been essentially not even legislated.

It's been sort of sign of a pen put into law.

And so where do you go go from there?

Because can you turn the tariffs into something like

a destination-based cash flow tax?

Like nobody knows what that is.

I'm going to help.

I'm going to, I'm here to do.

Can you turn the tariffs into a progressive consumption tax or just any kind of consumption tax, frankly?

Walk me through this.

So I am a weirdo about this.

I love progressive consumption taxes.

I'm like an old devotee of Robert Frank books, The Economist, who's like a progressive consumption tax obsessive and has been for decades.

So I've always thought that's a kind of interesting tax structure.

So the theory of a consumption tax is by taxing consumption more, particularly among rich people, you get more kinds of saving and investment, which is good for the long-term growth rate of a country.

So one, do you think we actually need that?

It doesn't really seem like we don't have enough investment, right?

The stock market is booming.

The other thing is, are there other things we should want to tax, both because they raise revenue, but also because they discourage the thing?

Pollution in the form of carbon, online gambling.

Externalities.

Yeah, externalities, as economists like to put them.

Are there parts of a code like this that you think could raise substantial money, but also

nudge us in a better direction?

Auchin Klaus likes the idea of taxing digital advertising

that sort of monetizes our attention, right?

You could come up with a lot of things like that.

But the reality is, if you ask any economist what the right tool is to try to deal with the fact that there are too many carbon emissions, they are going to tell you that it is a carbon tax.

And one of the sort of benefits of some progressive consumption tax version is you can imagine layering on a very small, by the way, carbon fee to that and rebating it and doing all the stuff.

The

idea of trying to think seriously about what the types of activities are that you might want to, you know, the soda taxes, right?

These types of activities where we're trying to get people to have less sugar.

Yeah, exactly.

I think there's like a lot to that.

So there is a real revenue challenge that this country faces, which is that we basically don't have capacity.

Republicans have said no tax increases on anyone ever.

And Democrats are actually, I don't like doing the on the one hand, on the other hand, because Democrats have identified $3 trillion-ish of tax increases over a decade that they would like to see levied at the top.

But if you're trying to have a social safety net.

But they said no tax increases on people making under $450,000, which is not a good way to think about the tax code.

No, pledges are not a good way to think about the tax code.

And frankly, if you're trying to have a social safety net that looks like other countries, social safety nets, they have higher taxes on a much broader swath of the population than people making $400,000 or more.

In that environment, if you think about the tariffs, they're kind of like a workaround for this stuff, right?

Because like it was the case that the Republicans said no tax increases on anyone whatsoever, but here's $3 trillion of tax increases, essentially.

And they're in the system and in the world in which we're talking about 2029 and a new world order where we've had the tariffs in place for years,

they have become a revenue source and a meaningful one.

And by the way, tariffs are kind of like consumption taxes because they are a tax on a thing that is imported and then consumed.

They're not as

good.

as an actual consumption tax because

they are going to raise prices for consumers more than they raise revenue for the United States, because they are going to also increase the price of domestic goods that aren't hit with the tariffs.

The way that you get from a tariff, though, to something that looks like, in my dream, progressive consumption tax, is you kind of start to follow the playbook of Speaker Paul Ryan in 2017 when he he proposed what, and I gave you my acronym, the destination-based cash flow tax, what it essentially was, was that rather than having to figure out what corporate income is and it's a concept and where is it sit in a worldwide system, he said, let's just do cash flow taxes, okay, which is kind of like what other countries do with a VAT.

And the concept

was that

when stuff is sold in the U.S.,

it is then taxed in the US.

And when stuff is sold outside of the US, well, those cash flows accrue outside of the US.

And so you pay taxes on it outside of the US.

And that kind of structure sounds a lot like a tariff.

And the complaint at the time, there was a fair bit of support for it at the time among Republicans.

The complaint was really the retailers in the US who said, wait, the Walmarts and the Home Depots, you're going to weigh increase the cost of goods to consumers by this new type of tax.

But the tariffs are already doing that.

And so I wonder whether you don't manage to shift the tax base in the direction of a better designed consumption tax and better than the version I hope in 2017 because we managed to find a way to make it progressive.

Is it a rebate to people for their consumption tax adjacent thing?

I don't know.

I kind of think that there might be more to that idea than I had been grappling with previously.

So we are in a world where the tariffs are raising a lot of money.

So the big budget bill, big beautiful bill, is not quite the fiscal disaster we'd have seen if it were just alone.

Yeah, it's true.

So we're just kind of

in a new world around all that.

So I am eternally looking for reasons to be optimistic.

And

the reality is that finding $3 trillion of tax revenue that we didn't previously have any way of raising is, from a deficit perspective, an accomplishment.

Let me ask you something.

Let's say, you know, it's 2029.

If a Democratic president came to you and said, Natasha, I just want to propose root and branch tax reform in a way that would be really good for the economy.

I want to, you know, raise the amount of money we need to raise,

but I want a healthier economy.

The tax code is a huge lever.

I want it to be simple.

I want it to be explainable to people.

I want it to be progressive.

And I want it to be good for the long-term future of the American economy.

What in like a, I know tax codes are complicated, but in a stylized way, would you tell them to try to do?

It's such a, I'm so happy with this question because it would be nice to dream a world in which a lot of this was possible.

The way I guess you would start is you would start with the principles that we need to raise enough revenue.

We want a tax code that is simple.

And we want a tax system that is competitive.

And we want to minimize distortions in the economy.

And right now we have a tax code that does none of those things.

And particularly what you would, I think, be concerned with is the idea that over time, what's happened is that the economy has grown more complicated just to take advantage of opportunities that that complexity poses in the tax code.

So, for example, if you are a corporation, you pay a corporate income tax.

But if you decide to structure yourself like a partnership or a pass-through,

you get a totally different tax structure because your income is taxed at the individual level and often frankly, not taxed that meaningfully at all because of the ways in which you're able to characterize it.

The other thing I would do after trying to streamline the code is I think we're too rich of a country to have so many children living in poverty.

And the interesting thing that I realized when I was in government, and I knew it before, but kind of got to see it on the ground in a meaningful way, is the IRS is a really important administrator of federal benefits.

A lot of them run through the tax code.

And a super important one is the child tax credit.

But because it runs through the tax code, in part because the IRS is quite good despite having very few resources at administering things,

you only really get the full value of the credit if you're rich enough to have $2,000 or $3,600 at times when I was in government to deduct from your taxes.

So like the poorest people are being helped the least by our benefit system.

That seems nuts, right?

So obviously it is the case that you should design a system that lets us do more for those who need it most.

And the third thing I would say in this sort of magic world of being able to think seriously about tax reform from scratch is that so much of the tax code has been distorted because there happen to be particular interests that are able to get an exemption here or a carried interest loophole there for their particular pet type of income or type of benefit.

And I think we really need to find a way, and this is true writ large and is why this exercise is more of a dream than it is in practice.

You really have to find a way to push against the impact of those constituencies because there isn't a counter constituency to say, no, that's really bad.

And I think that's a pretty fundamental problem.

I'm going to leave it there.

I'll use our final question.

What are three books you'd recommend to the audience?

It's good because we were just talking about tax reform.

I'm teaching federal income tax this fall, and I was prepping around the 1986 Tax Reform Act, which was a time when we thought seriously about tax reform.

And Showdown at Gucci Gulch is one of my like all-time favorites and a great read and will leave you hopeful for the possibility of this type of reform effort.

I also

just recently, I had twins last year, I was telling you before, and so have been slow to be able to do that much reading for fun of recent.

So I just recently picked up Remarkably Bright Creatures, which is this like lovely story about a woman who finds companionship with an octopus.

And it sounds kind of wild, but it's just a true, in a world that's feeling intensely more isolated, it like brought me a lot of joy.

And I also really love, we talked a little bit about economic models and the ways in which we try to measure and derive truth about the world.

There's a great book by Michael Lewis called The Undoing Project that's about the relationship between Danny Kahneman and Amost Versky, but also the ways in which the field evolved that I think is like pretty profound and a great story.

Natasha Sarin, thank you very much.

Thanks so much for having me.

This episode of the Azure Clown Show is produced by Roland Hoo.

Fact-checking by Michelle Harris.

Our senior audio engineers Jeff Geld with additional mixing by Isaac Jones and Amin Saota.

Our executive producer is Claire Gordon.

The show's production team also includes Annie Galvin, Marie Cassione, Elias Isquith, Marina King, Kristen Lin, Jack McCordick, and Jan Koble.

Original music by Amin Sahuta, Carol Sabaro, and Pat McCusker.

Audience Strategy by Christiania Samuluski and Shannon Busta.

The director of New York Times Opinion Audio is Annie Rosstrasser.

Special thanks to Catherine Abraham, Skanda Armanoff, Kim Klausing, Catherine Ann Edwards, Matthew Klein, and Claudia Somme.

You just realized your business needed to hire someone yesterday.

How can you find amazing candidates fast?

Easy.

Just use Indeed.

Join the 3.5 million employers worldwide that use Indeed to hire great talent fast.

There's no need to wait any longer.

Speed up your hiring right now with Indeed.

And listeners of this show will get a $75 sponsored job credit to get your jobs more visibility at indeed.com/slash nyt.

Just go to indeed.com/slash nyt right now and support our show by saying you heard about indeed on this podcast.

Indeed.com/slash nyt terms and conditions apply.

Hiring?

Indeed, is all you need.