It’s gotten pushback from economists and Fed watchers, especially those who question whether policymakers would really sit on their hands in the face of a disinflationary spiral. You can read a lot about it online. I’m going to use this moment instead as an opportunity to start digging into the macroeconomics of AI in a more disciplined way. This is material I’m learning in public here, please feel free to leave critical and technical comments.

My hypothesis is that AI lowering wages will make it harder for the Federal Reserve to stabilize the economy, and will create an ugly situation in terms of labor share and consumption (both central in Citrini’s story). When I embedded that argument in a leading DSGE framework, one of those predictions held up and the other didn’t.

The simplest, representative-agent New Keynesian (RANK) model doesn’t get us very far. We need income distribution to matter for Citrini’s story, where workers have high marginal propensities to consume (MPCs), capital owners have lower MPCs, and shifting income from one to the other affects demand. At the same time, I don’t want us to get lost in the pyrotechnics of a full heterogeneous-agent New Keynesian (HANK) model, where a continuum of agents generates their own wealth distributions.

A two-agent New Keynesian (TANK) model splits the difference. The model I’m going to use is from “Workers, capitalists, and the government: fiscal policy and income (re)distribution” (2021) by Cristiano Cantore and Lukas B. Freund. They are kind enough to include their Dynare .mod files, which makes replication and modification straightforward, a huge plus. Cantore and Freund describe their model as “tractable laborator[y] for understanding various macroeconomic experiments,” which is exactly what we need.

In this model there are two agents, a worker who works to earn labor market income and a capitalist who doesn’t work and receives profits. (I’m already sold.) Both are forward-looking rational agents. Workers can save in bonds but face portfolio adjustment costs. That friction makes them partially constrained and more responsive to current income than a permanent-income consumer, so they have a higher MPC. The rich capitalist class smooths their entire life’s income independently based on their permanent income.1

In order to explore this, I take their medium-scale model and do my best to insert a task-model loosely based on Acemoglu-Restrepo (2019) (especially read through these Jon Steinsson notes). We take a permanent innovation shock to the automation of tasks with three scenarios: one where the shock is fully labor-augmenting, benefitting workers, a second where it is fully labor replacing, and one in-between.

We end up with 43 endogenous equations that we send into MATLAB to be solved.2 (It’s ironic to model the collapse of software while relying on proprietary MATLAB.)

Here’s what we get:

In everyday language, red is where workers get screwed and blue is where workers fully benefit from AI. 0 is the steady-state, or what things were settled at prior to the shock. Don’t worry too much about units, we’re here to see the signs.

What do we see?

Interest rates: All three lines dip below zero, meaning the Fed cuts rates after the AI shock in each scenario. But the cut is dramatically larger under the blue scenario than the red one. This seems backwards. Shouldn’t we expect the scenario where the labor share falls to require deeper cuts?

What I think is happening is mechanical but unintuitive. In the labor-augmenting case, productivity rises faster than wages because of nominal rigidities. Unit labor costs fall sharply, generating deflationary pressure. The Taylor rule responds with aggressive cuts. In the displacement case, human marginal product falls along with wages. Deviations in marginal cost, which is what drives inflation in New Keynesian models, barely moves. So despite the collapse in labor share, there is little disinflation and therefore little pressure for rate cuts.

These kind of New Keynesian results, where inflation is largely about deviations of marginal costs, might feel like its own kind of science fiction. But it’s worth considering. It’s also funny to imagine the Federal Reserve staff who will have to explain to potential new bosses that AI doesn’t magically mean you get to do whatever you want to interest rates.

Workers: It’s not great. In the scenario where workers are augmented by AI they get more consumption and work fewer hours and the labor share fall isn’t that bad. In the displacement case, real wages barely rise, hours initially increase and then drift down, and labor share collapses. The modest rise in worker consumption comes primarily from higher labor supply rather than higher wages. They even, for some periods of the shock, work more hours than the baseline. Their labor share plummets, and their increased consumption is mostly from just working more hours.

I am still not sure if I’m doing justice to the AI shocks. I get a similar result when I instead just use a productivity shock to the model and make it so wages are fully inflexible, they are basically impossible to update. This extreme example gives us the same kinds of movements.3

From a standard business-cycle perspective, this remains within the reach of conventional macro stabilization. In the displacement scenario, inflation barely moves because marginal cost barely moves, so the Taylor rule doesn’t face dramatic tradeoffs.

That said, I’m still uneasy. If AI were sufficiently disinflationary, or if nominal rigidities behaved differently than in this calibration, we could find ourselves closer to the zero lower bound and facing a more prolonged demand shortfall than this baseline suggests.

I don’t like AI art. But one thing I always do to test the picture and video models is prompt it with “the world as if John Maynard Keynes’s ‘Economic Possibilities for Our Grandchildren’ came true.” In the two videos above you can see Sora, OpenAI’s video producer, from its public launch in October 2025 (top) to a video yesterday (second), each with that prompt, and chart the progress.

So I feel a visceral pain seeing a DSGE model that gives us periods where workers are compelled to work more hours as a result of AI-based skyrocketing productivity. My worry now is less that AI causes a recession. It’s that AI can raise output while worsening workers’ position, a distributional transformation without a traditional macro crisis.

That distributional question of who captures the AI dividend and how it feeds back into demand and political economy is where the real work lies. I’ll be digging into that more in the months ahead.

2

Code here. For the sickos, here’s the whole thing. 9-11, 13, 15, 33-34 are the lines where our additional task-based elements show up:

3

Funny enough here the Fed cuts less because output is so high because workers are working a lot more hours with their wages frozen. But as a quick check with a completely different methodology it gives us similar results.

Subscribe now

But much of the discussion has been caught between people building applications as fun personal hobbies on one end, and massive enterprise software on the other. Most of us who use these professionally will live somewhere in the middle. (The discourse is also wedged against whether AI will cause large-scale unemployment or otherwise destabilize society.)

I’ve integrated these terminal tools into my workflow over the past two months. Below are three specific ways I’m using them that are genuinely new, and where I’m not going back to how I worked before. Terminal AI compresses the setup and robustness-checking phase of knowledge work. I’ll also explain what makes them different from the browser-based chat tools, and whether I, Mike Konczal, am about to be automated out of a job. My relevant background is in a footnote here.1 This is not a setup guide, though I’ll try to explain ways these examples might be applicable to you. That will be helpful, as these tools are not going away.

Use 1: Real-Time Analysis, Without the Prep Work

I maintain a set of R files, the statistical programming language, for instant analysis of the economic data releases. Those numbers generally come out at 8:30 a.m. several times a month, and by 8:35 a.m. I need to know the big takeaways.

I’m not trading on this information, so I don’t need to know within milliseconds. But I do make statements that other political actors use, and accuracy is vital. For instance, I was interviewed by the New York Times shortly after the jobs numbers last week, and I want to be able to say something useful, interesting, and, most important, true.

My old workflow was that I’d take one or two hours the day before the major monthly data releases and manually code a few new graphics based on what I expected would be important. Maybe I post them or maybe I don’t, but they help me keep track of how the economy is moving, and they force me to think of new pieces I may have been missing.

Now I just ask Code/Codex to write them. It’s usually 95% of the way there on the first go. I check the math and get it fully there with one or two more exchanges.

What used to be 4–5 hours of work in an average month is now 15 minutes. The reduction in time allows me to ask even more questions and prepare more materials. I have a lot of experience in this, so I’m able to proofread the code and I double-check that it is doing what I want it to do.

Why not just use the Claude or ChatGPT webpage for this? I used to do this. But there is significant copy-and-paste overhead. The old way meant copying code out of the browser, pasting it, running it, copying any error messages or changes that needed to be made back into the chat, and repeating. With the terminal tool, the LLM writes the code and runs it in the same folder. It can see whether what it did worked, investigate failures on its own, and I see the result after every pass without touching a clipboard.

Why not just let the AI find the interesting result? The LLM actually does a poor job of this, as it lacks the context for what people are searching for or what stands out. It can summarize news reporting on those numbers after they circulate, but influencing that coverage is the point of the exercise.

So if your job involves monitoring something that changes on a schedule, such as earnings releases, polling averages, clinical trial updates, or monthly sales figures, and you currently spend the hours before and after each release manually refreshing Excel models or rerunning Tableau dashboards, this is the use case that will feel most immediately different. The tool doesn't just help you go faster. It makes it cheap to add new angles of analysis instead of being locked into what you could cover in limited prep time.

Use 2: Building Out a Report

Last Monday, I published a blog post about the affordability crisis. Check it out! Near the beginning are three graphics I made on the price, spending, and indexing of essential items. To do this I created a folder on my desktop and dropped in the relevant data files. Then I gave this opening prompt:

Note I ask it to create a qmd file (Quarto Markdown), which can render to html, pdf, or slides with minimal changes. You’ll probably be using a similar method in the future.

It got very close on the first pass. I scanned the code to check the methodology, spot-checked one item from each by manually calculating it (knowing the algorithm means that if one is correct, all of them are).

From there the prompts are mostly about cleaning up:

I’m just talking into the microphone at this point, deciding which essentials to include (some categories overlap in ways the LLM wouldn’t know) while simultaneously tweaking colors, legends, titles, and sizing, while it re-renders the graphic.

One thing I worry about is that if I’m not mucking around with the data myself, I’ll miss important side observations. But even in this pass I notice things: cigarettes show a huge price increase (it’s very funny that the LLM thought cigarettes are an essential), and I wonder if it’s driven by excise taxes, the same way tariffs are mechanically increasing the price of goods in 2025 and 2026. Something to investigate.

A benefit here is that stress-testing became trivial. Want to check whether your finding holds under a different index, or a different date range? If you are using Excel this is a nightmare, and even with statistical programming it’s a slog. But I get this done in a minute and can be more confident my results will hold under scrutiny.

Doing this part of my blog post last week would previously have taken 1-2 hours, with an additional 1-2 hours if I wanted to stress-test the indexes to make sure the results are robust. It took me 15 minutes to do this here.

So any future project that involves empirical work will be built out of a folder where the terminal LLM is executing work that I check. The terminal allows it to do much more complicated work, faster.

So if your job requires you to produce regular analytical reports, for clients, for leadership, or for publication, you will use this to cut out difficult first steps. Often the bottleneck is not the writing, it’s getting from raw, unorganized data to a defensible visual that you trust and can defend. The terminal tools compress that step dramatically, and more importantly, they make it cheap to check whether your finding survives different methodological choices. That robustness-checking step is the one most people skip under deadline pressure. Now you don’t have to.

You Bought a New Car and Auto Inflation’s Really Taken Off

I really want to emphasize how much of the work for any report is this basic setup, the data-wrangling and first-pass results to see if the results are worth exploring in full. This can be time-consuming, especially if it doesn’t pan out. Unless you try it, I can’t describe to you how fast the terminal LLM can get through this stage, and keeps the building blocks in place to iterate on.

Two examples I explored recently. There’s an interesting new paper from Louie, Mondragon, and Wieland on whether supply constraints actually explain house price and quantity growth across U.S. cities. I wanted to get into the guts of the paper before I formed an opinion and wrote about it in the future here. But the actual data, merged across multiple datasets of prices, quantities, and restrictions, all from different sources, each with its own formatting and naming conventions for local areas, is a nightmare.

I put the data in a folder with the paper and had the tool merge it all in a single 10-minute session. It closely reproduced their key graphic, giving me a foundation to explore their work more fully. (I was told by one of the authors that the research assistant who led that data merging had a lot of big feelings about this.) Data merging in the browser tool, by contrast, had never worked reliably for me; but being able to iterate in place, with the LLM reacting to exactly what went wrong and adjusting, got it done.

Or take the standard “Phillips curve” style analysis of inflation. From around 2022-2023, a lot of people were running these regressions adding variables that followed the timing of inflation, like global supply chain stress metrics or the vacancies-to-unemployment ratio. I was frustrated because those don’t predict inflation out-of-sample and basically anything with an up-and-down pattern during those years would “predict” inflation.

Back then I had some code that estimated inflation with those variables along with other things that just happened to pop in 2021-2022. First, a dummy variable with a value of 1 in 2021-2022 and 0 every other year. Second YouTube views for Olivia Rodrigo’s “good 4 u” video, which debuted in May 2021, just like the inflation spike. What would be the best predictor of inflation?

I never got around to finishing it all at the time. But then James Stock and Mark Watson, the godfathers of empirical macro, published a paper in late 2025 finding that basically only COVID deaths explain the inflation in a component analysis. It’s a big victory for the cost-push shock side of the inflation debate. We should do a full post on it, as the entire episode traces back to a single supply-side shock. That made me want to dig my idea back out.

Sure enough, Codex one-shotted it.

It got the data and did the analysis with one call.2 From the R-squared above, reflecting a few clean-up calls, Olivia Rodrigo does a better job estimating inflation than vacancies-to-unemployment. (The unit is log(good 4 u views + 1)). If you put them all together, only the dummy variable and the “good 4 u” variable are significant.

You will, no doubt, have more normal and productive projects you want to explore and see if they are worth developing fully. The real work is finding good questions and understanding how to make the results rigorous. Using the terminal gets you to where you can focus on what matters the most.

Use 3: Anticipating Arguments You Might Not Know Ahead of Time

Keeping with that blog post on essentials: I do not know the complicated microeconomic derivations as well as I should. What differentiates Hicksian from Marshallian demand does not come naturally to me. It’s been almost twenty years, and I still occasionally have nightmares about that one chart from Chapter 3 of Mas-Colell.

But my recent post on essentials (and a previous one on DoorDash) touches on how groceries and shelter behave differently from restaurants and delivery services as income changes. I wanted to understand how rigorous demand theory would assess what I’d written. So I asked the terminal to take a look:

AIDS here is the Almost Ideal Demand System, a standard model in microeconomics for estimating how consumer spending shifts across goods when prices and incomes change. (Angus Deaton and John Muellbauer published their groundbreaking work in 1980, roughly two years before they would have named it literally anything else.) It generated this analysis:

I have no idea if this analysis was executed correctly. But I do know that on a first pass, an approximation of the analysis doesn’t disprove my instinct that food and shelter behave like my post argues they do. My argument wasn’t conditional on this, and I don’t advertise it. But I now think I’m not going to take obvious incoming criticism on something I missed. And more importantly, this gives me an incentive to learn more and continue digging on this topic.

A lot of analytical work is anticipating good-faith skepticism. You know the argument you're making, but you're less sure what the specialist counterargument looks like. These tools let you run a first-pass stress test against these other frameworks, cheaply enough that you can do it before every major piece rather than only after you've already taken the incoming.

Is It Game Over?

All of this makes me better at my work. My software background helps, as I can proofread the code, double-check what it’s actually doing, and catch errors in the methodology. But I’m also wondering if I should be switching fields ASAP.

Who knows what the future holds? But I keep running a version of this test: I put the Survey of Consumer Finances data, a triennial survey by the Federal Reserve on household incomes, debts, and assets, in a folder, and ask the AI to find the most interesting results. It gives me simple summary statistics. I ask it to find both unique policy insights and business opportunities, and it gave me very generic ideas. I asked it to use linear regressions, figuring it could just p-hack something, and it gave me very little. I keep running a version of this, asking AI to find what’s interesting about a project independent of any direction from me, and it really can’t do it.

So this still looks like an extreme version of labor-saving technology. It makes people faster at a given set of tasks, and it lets you cover more ground with the same resources. Which effect dominates, fewer people or more output per person, is an open question. My experience, at least for now, is that it can complement people who know how to use it, but risks shortcutting those earlier in their careers before they’ve learned the building blocks.

Last summer I was so conflicted about giving up RStudio for Positron, a VS Code fork that integrates Python and Quarto more naturally into R development. Now I keep Positron open to check while I work in the terminal. Whatever happens with AI bubbles and buildouts, the economics of my own work have changed permanently. I can do in twenty minutes what used to take half a day. This means I ask more questions, check more assumptions, and cover more ground. I still believe the tools can’t identify what’s interesting or draw the right conclusions on their own. But they’ve made exploring what’s right a lot cheaper.

useR!: EconTwitter to the White House

I gave a short 15-minute talk last summer at the useR! conference for the R statistical programming language, titled “From #EconTwitter to the White House: Real Time Economic Data with R.” (A sentence I still can't quite believe is autobiographical.) The pitch is four reasons to use R, convincing analysts not to use Excel and economists not to use Stata.

There are a lot of fun tidbits from my experiences in and out of government working on data releases that you might enjoy.

Slides here. Even as more statistical analysis will be done through LLM terminal prompts rather than working directly in IDEs like RStudio, the reasons to base your code in R are still relevant. The ease of the tidyverse, the grammar of graphics framework, and the variety of data libraries available make it a great foundation to iterate with AI tools.

Yale Budget Lab

I was on a panel this past week at the Yale Budget Lab and the Tobin Center for Economic Policy titled “Tariffs, AI, and Our Economic Futures,” looking at the first year of the Trump second term. Moderated by business journalist Alexandra Scaggs, with Michael Faulkender (a Trump first-term Treasury official and Deputy Secretary of the Treasury, the number two, from January to August 2025) and Natasha Sarin (Budget Lab President, and former Biden Treasury official).

Around 32m55s, I pointedly bring up the IRS giving ICE access to ITIN tax records for undocumented workers, which Faulkender signed off on while at Treasury. There’s a back and forth among the panelists that is interesting if you follow the IRS/ICE story, one that ends with a chill coming over the room. (You’ll know when it happens. This was recorded before the Washington Post broke a story about related improper breaches.) There’s also a debate starting around 10m44s about why exactly Trump is doing tariffs the way he is, where I push back on the administration’s rationale.

Columbia / Groundwork Collaborative Webinar

I did a webinar with the Initiative for Policy Dialogue at Columbia and the Groundwork Collaborative examining the impacts of Trump’s economic agenda.

It was chaired by Joseph Stiglitz, with Groundwork’s Alex Jacquez, University of Michigan economist Sarah Miller, and Bharat Ramamurti. It’s a great overview. Even as someone following this closely, I’m still learning and amazed at the damage the Trump administration is doing to healthcare. Sarah Miller’s comments (starting 29m) on healthcare were informative.

Stiglitz & Konczal in the Boston Globe

Stiglitz and I keep our buddy cop routine going with an opinion piece in the Boston Globe covering much of what we discussed there: “The high cost of Trumponomics.” We take on Trump’s recent Wall Street Journal op-ed claiming his tariffs have brought about an economic miracle, and walk through why those claims are unmoored from the economy around us.

I was also quoted in the New York Times about recent data releases and the Washington Post on why affordability has become a campaign buzzword.

If you want me to appear or write for your next event, feel free to reach out!

Current Mood: Burn it down.

Current Music: St. Vincent’s excellent cover of Big Black’s “Kerosene.”

You’ve probably seen some version of Graphic 1 below, which shows real median household income from Census. After falling during the post-pandemic inflation surge, real incomes recovered. By 2024–2025, they had ended up higher than in 2019. This result holds across multiple ways of looking at this data, including hourly wage data. Yet consumer sentiment is stuck near historic lows, about as pessimistic as during the financial crisis and the depths of the Great Recession.

These two facts coexist. And the politics of “affordability” has rushed into the gap between them, as politicians and advocates try to speak to people’s persistent anxiety about prices and living costs.

But there has also been a subtle, yet real, pushback against this focus. That pushback usually starts from the observation that today’s affordability debate is inseparable from the 2021–2024 inflation episode. From there, two critiques follow.

The first is a money-illusion story: people fail to recognize that their incomes rose alongside prices, so their distress reflects confusion rather than material harm. This argument often emphasizes that incomes at the bottom of the distribution rose faster than those at the top, producing a durable wage compression. And yet polling consistently finds that lower-income households report more dissatisfaction with inflation, not less.

The second critique follows naturally. If what people really want is their old price level back, that is simply not something policymakers can deliver. Broad-based price declines tend to occur only in deep recessions, and even then only modestly. So it is said to be dangerous politics and bad economics to make affordability central. Doing so risks promising something impossible, or worse, flirting with economic collapse as a policy goal. Matthew Yglesias has made a version of this argument, affordability is “just high nominal wages” and “basically just anger at inflation,” as have others.

This conversation echoes what you often hear about the so-called “vibecession.” The term was originally coined by the writer Kyla Scanlon in 2022 in a nuanced way, to describe self-reinforcing pessimism. Today, it’s more often weaponized to imply that consumer sentiment is untethered from material reality, a reflection of the circulating bad vibes.

One could correctly say that these affordability problems existed in 2019 and are independent of the inflation wave. But we should take the bait. Over the past five years I’ve been haunted and utterly consumed by a keen observer of the debates over inflation. And I think it’s worth being explicit about why a wave of inflation can generate real affordability problems.

These mechanisms point toward specific policy interventions, many of which have already bubbled up in political campaigns. And they also help explain why President Trump’s current policy agenda is depressing sentiment, by placing pressure precisely on these channels. There are several, but the first is what matters the most.

1. The Essentials Squeeze

The simplest story is that essentials have been squeezed: their prices have risen faster than overall inflation, even as households are forced to devote more of their budgets to them.

Consumer prices have increased 26.1 percent over the past six years. But not all price increases are alike. Graphic 2 has some prices that have increased faster than overall prices during that time, ones that are pretty important for people. Food, shelter, transportation, hospitals, and veterinarian services all pop out. They’ve increased faster than both overall prices as well as core services (which rose 27.8 percent during this period).

Last week we discussed DoorDash. From Graphic 2 above, we can see that the price of food away from home (which includes delivery services like DoorDash) increased faster than groceries (food at home), and people shifted their spending to groceries. Given that spending on groceries tends to decline with income, we can understand this shift as a penalty people experience. Even if incomes stay the same, vibes (i.e. utility) decline.

Let’s create two definitions of essentials. Core essentials are groceries and shelter. Essentials are groceries, shelter, healthcare and transportation. Taking the Consumer Expenditure Survey (CEX) data, we can see in Graphic 3 that most households are devoting a larger share of their budgets to these essentials than they did before the pandemic, especially at lower incomes.

Is that because these goods are getting cheaper? No. Using PCE data, we can construct price indices for these bundles, weighted by their consumption shares.1 As Graphic 4 shows, both measures of essentials inflation have run well above overall PCE inflation.

The green line is overall PCE inflation, the blue line is an inflation index of groceries and shelter, and the red is the blue plus healthcare and transportation, all indexed by their consumption weight. As you can see, and as hinted from the CPI data in Graphic 2, the essentials are running much faster than overall inflation. This was true before as well. From end of 2013 to 2019, core essentials rose about 14.3%, essentials rose about 10.5%, and overall PCE rose about 7.9%. But this was supercharged in the recent period.

When prices rise and budget shares still increase, standard demand theory tells us these goods are necessities. If utility comes from consumption above some baseline floor, then a rising share devoted to essentials leaves less room for discretionary consumption and lowers welfare, even if total income keeps pace with total prices.

This also helps explain why lower-income workers are angry about inflation even as their wages rose faster than average. Households traded down to cheaper and generic food brands, a trend coined “cheapflation,” which drove up the prices for the basics that lower-income families depend on most. And as Catherine Rampell noted at the peak of inflation, CEX data show that lower-income households typically consume more than they earn and face more volatile work hours, both of which magnify the pain from price spikes.

The essentials squeeze alone is enough to provide microfoundations for the vibecession and affordability crisis. But it’s not the only channel. Let’s look at other ways inflation creates welfare losses that survive rational expectations. I’ll sketch them briefly.

2. The Housing Tilt Problem

This one is big in the older literature but is oddly absent from today’s debate.

As Modigliani and Lessard (1975) showed, standard fixed-payment mortgages interact badly with inflation even when inflation is perfectly anticipated.2 Higher inflation raises nominal mortgage rates, preserving real interest rates. But because mortgage payments are fixed in nominal terms, higher nominal rates mean much higher initial real payments that then fall rapidly over time. Real payments are front-loaded.

MacGee and Yao (2025) show, using modern life-cycle models, that when borrowing constraints bind at origination, this front-loading tightens credit for first-time buyers. People with steep expected income growth but limited current earnings must qualify against today’s higher nominal payment even though that payment will shrink quickly in real terms as wages rise.

If this is too complicated, it’s just another reason the housing market has been a mess.

3. Planning Under Uncertainty

Binetti, Nuzzi, and Stantcheva (2024) find from surveys that the most commonly cited consequence of inflation is not any specific price increase, but the complexity inflation introduces into everyday decision-making. Eighty-five percent of respondents identify this as a major effect, and more than a third rank it as the single most important one.

When prices are stable, budgeting is boring. When prices are volatile, households must constantly plan when to buy, what to delay, how much to save, and how to interpret nominal changes. That cognitive effort is costly.3

4. The Cost of Money Is Part of the Cost of Living

Another issue is that we largely exclude borrowing costs from how we talk about inflation.

As Bolhuis, Cramer, Schulz, and Summers (2024) show, incorporating borrowing costs into measures of consumer sentiment explains a lot of the U.S. sentiment gap. From a household perspective this is obvious. The interest payment on a new mortgage is several times higher. Interest on new car loans is up sharply. People, especially lower-income ones, experience this as part of the cost of living.

Official inflation measures abstract from interest costs for good monetary and economic reasons. But that creates a gap between measured inflation and lived affordability.

There are others, which we might discuss in the future, but this is a solid grounding.4

Conclusion

When the prices of necessities rise faster than everything else, when housing becomes mechanically harder to access, when planning gets more cognitively costly, and when borrowing is more expensive, welfare can fall even if average real incomes recover.

The good news is that these binding constraints are solvable problems. We have many ideas for tackling housing, healthcare, and food costs. Simple steps like not slapping century-high tariffs on foods like bananas on the fantasy that they create leverage, or not cutting a trillion dollars from Medicaid to cut Amazon’s corporate tax bill in half, would be a great start.

But the first step is to believe that what people have been screaming about their lives for the past several years actually exists. Even a representative agent, forward-looking and fully aware of all the parameters surrounding them, can feel the vibecession.

Nothing gets the internet going quite like a round of hand-wringing about whether too many people, especially young ones, are spending too much money on delivery services, distorting their sense of budgeting and of how the economy actually works.

However, DoorDash Discourse is mostly wrong.

People are not, in aggregate, spending more on eating out and delivery. They are spending more on groceries at home and less on food away from home. This is especially true of young people.

The last round was kicked off over the weekend with “Freedom With a Side of Guilt: How Food Delivery Is Reshaping Mealtime” from Priya Krishna at the New York Times:

In 2024, almost three of every four restaurant orders were not eaten in a restaurant, according to data from the National Restaurant Association. The number of households using delivery had roughly doubled from 2019, just before the pandemic, the group said. And in a survey last year, about one-third of American adults told the association that they ordered food for delivery at least once a week. […]

That disconnect extends to restaurants, many of which have accepted a trade-off: Delivery helped keep them afloat during the pandemic and expanded their customer base, but they now have fewer in-house diners.

This predictably produced a wave of takes about personal irresponsibility and moral failure. A representative tweet comes from Matt Yglesias: “Part of the affordability crisis is pretty clearly people just refusing to be thrifty — you should not be spending a quarter of your salary on DoorDash.”

I have no doubt that food delivery has picked up and normalized since the pandemic. The relevant economic question, though, is not whether delivery exists. It’s what it is displacing. Is delivery mostly replacing groceries cooked at home? Or is it mostly replacing eating in restaurants?

Positive Analysis

We have a place to look. The Bureau of Labor Statistics Consumer Expenditure Survey (BLS CEX) is a nationally representative annual survey of consumer units that reports annual spending by detailed category. It is the primary federal source for household expenditure patterns and is widely used to understand cost burdens, to inform CPI weights, and to benchmark how budgets shift over time. The CEX makes it possible to compare budget shares across time and across demographic groups such as income quintiles and age.

The latest data for 2024 came out in December 2025, delayed because of the government shutdown. This data can help explain why people’s experience of the economy has been poor even when aggregate income growth keeps pace with inflation. A rising share of budgets devoted to necessities will make households feel worse off.

Here is “food at home” and “food away from home”1 as a percent of total consumption2, for all families and by income quintile.3

The dotted-line and middle number are the average for 2015-2019. Here’s those numbers in a simplified chart:

What do we have here? Start with the post-2019 period. Overall, Americans are spending about the same share of their total budgets on food. For the bottom 40 percent of households, the food share is modestly higher.

But compositionally, there has been a clear shift. Less of the food budget is going to food away from home (which includes delivery and takeout). And much more of the food budget is going to groceries.

In other words, society as a whole is reallocating food spending toward cooking at home, not away from it. DoorDash Discourse gets the direction wrong.

Normative Analysis

Is this shift necessarily bad? And does it support affordability as a real political and economic concern?

I think yes, for two reasons.

First, the shift is bigger for the bottom 40 percent of incomes. Lower-income households are doing much more of the adjustment toward food at home than higher-income households. This is not just a uniform cultural pivot toward staring at your phone rather than going to a restaurant. It is a stratified adjustment likely driven by constraint.

Second, the longer-run trend matters. Historically, as incomes rise, the share of spending devoted to food at home falls. You can see that above, which is the percent of the budget spent versus real total expenditures (the same is true for nominal) for each quintiles of each year from 1980 to 2024. There is a very strong decline in food at home as people get wealthier.4

That doesn’t just mean people dislike cooking. It means that, with more resources, people tend to allocate more toward convenience and variety. A lot has changed in preferences and attitudes these past seven years, but the fact that more of the budget has shifted to a good that declines with income should give us a strong presumption of disutility. The affordability crisis is real.

What About the Kids?

Ok, but can we still blame the kids?

The same exact thing happens, with the pattern being most pronounced among people under 25. They experienced the largest shift toward food at home. The 1.7 percentage point shift toward cooking at home for those under 25 is 2.5 times larger than the 0.7 shift for the general population.

The kids are alright. They're just broke and cooking at home to make up for it.

I want to look at the labor market with fresh eyes as we go into 2026. And the first place I want to do this is the debate over weakening prospects for people in their 20s with college degrees. This was a highly debated topic in 2025, with a focus on whether this was consistent with an overall weakening labor market or whether it was a sign of the impact of AI in the workplace.

Subscribe now

My loosely-held hot take a year ago would have been, echoing Joey Politano making this argument at Big Think, that AI would increase jobs in the short-run. My view is AI makes productive workers far more productive, but its unreliability makes it difficult to automate positions that require certainty over how outcomes are determined. But entry-level workers might be vulnerable, both because of the way the career ladder is structured and because bosses, wanting to cut corners, would try to haphazardly replace workers with AI there first.

And we did see unemployment increase for new college graduates. But, as friend of the blog Guy Berger has noted, unemployment is also higher for young people without a college degree, consistent with a weakening labor market and low hiring. Overall unemployment is up, from 3.5% in the summer of 2023 to around 4.4% now, and we expect that impacts younger people more.

Is there a way to distinguish if it’s increasing more than we’d expect?

Regression Relationship Approach

Let’s build up to how we’re going to approach this. There are many unemployment rates, whether for race, education, gender, or age. But they all tend to track the headline unemployment rate you hear about in the news, called U-3 unemployment, even over decades.

Let’s take two groups: (1) men and (2) workers aged 45-54, and see if their rise in unemployment tracks what we’d expect given the overall increase. (Note that the data sourcing will be important throughout this: here the data is from the BLS itself, seasonally-adjusted and monthly.)

Let’s run a regression on their individual (log) unemployment rates versus the overall (log) unemployment rate for the years 1994 to 2019. We get an equation, which we then use to project forward to 2025. We take the actual increase in U-3 unemployment and, given the historical relationship, we ask: what would we expect the specific unemployment rate for those two subgroups to be?1

Figure 1.a shows the full history back to 1994. The dotted lines are projections based on the log-log relationship between each group’s unemployment rate and overall unemployment, trained on data through 2019. This is a pretty good fit across multiple business cycles.

Figure 1.b zooms in on the post-pandemic period, where we are projecting forward based on the 1994-2019 relationship. Both men and 45-54 year-olds track almost perfectly with their expected values. The labor market slowdown since 2023 has increased unemployment across the board, but these groups are rising exactly as much as their historical relationship with U-3 would predict. Nothing unusual here.

Multiple Groups

Let’s extend this across all demographic groups. I take seasonally adjusted monthly unemployment rates from the BLS by gender, race, age, and education, regress each against the overall unemployment rate 1994 through 2019, and calculate the average difference between actual and predicted values over the most recent 2 months:

A positive value means higher than expected from 4.4% unemployment. These results are consistent across many date ranges. Most groups cluster right around zero, with their unemployment rates exactly where you’d expect given the current level of overall unemployment. But young people (16-19 and 20-24) and BA+ stands out, with unemployment notably higher than its historical relationship would predict.

Many, including myself, have noted the statistically significant increase in black unemployment over the past year. Why doesn’t that flag above? Let’s look at this more closely:

Here we can see that black unemployment has sadly returned to its longer-term trend. The tight labor market and full employment environment of 2021-2024 especially benefited black workers, with the black unemployment rate falling much further than historical patterns would predict, and the black-white unemployment gap collapsed to record lows. It’s a terrible fact that this progress has reversed in this past year. As the labor market has cooled, black unemployment has unfortunately returned to its historical, and far too high, relationship with overall unemployment.

Young People

Now to focus on younger people. Let’s take data from the New York Fed’s The Labor Market for Recent College Graduates website. This longer-standing measure was given a lot of attention last year in an article from Derek Thompson. These are 22-27 year olds. Recent grads have a completed BA or higher; young workers are everyone else. The researchers make their own seasonal adjustment. This is their creation from the microdata. Let’s run the same regression from 1990 to 2019, their starting years, and project:

(EDIT: as requested by readers, added a new figure that shows the relationship back to 1990, and extends the recent figure back to 2017. Still same analysis.)

As you can see, people without college degrees who are 22-27 more or less track exactly what we’d expect given the labor market slowdown. But young college is much higher than our trendline we’d expect from the slowdown.

To be clear: college-educated unemployment is still lower than non-college unemployment, and everyone's unemployment is up. Let’s not get tripped up on this on social media. What's unusual is the gap between where college unemployment should be historically and where it actually is today.

Now, as John Burn-Murdoch and Guy Berger have each noted in different ways with excellent analysis, college-educated workers are likely to be new entrants to the labor force in their 20s compared to those entering the labor force earlier.2 So in a slowing labor market it might not be correct to compare these two age groups.

To dig into this, I take the BLS microdata from IPUMS and create my own categories. I use 12-month rolling values instead of seasonally adjusting the data. I take 3-year age bands (so, e.g., 22-24 for age 23) for college-plus and high-school-plus (HS+) without a BA. The College+ series starts at ages 21-23, and the HS+ series starts at ages 18-20. I then run a log-log regression on each from 1989 to 2019 and take the average difference between the projected and the actual over the past 3 months (ending in September, to avoid the month of missing data). I then chart it across the age spectrum:

To read this graphic, for someone who is 23 (the age-range 22-24) with a BA+, they have unemployment that is over 2 percentage points more than we’d expect based on their historical relationship with overall unemployment. That excess fades by their late 20s. This gap is also high for non-college workers in their early 20s, but their peak is lower than the college one and it collapses faster.3

So my current read is that young people have higher unemployment than we’d expect at 4.4% overall unemployment. It’s especially higher at its peak and throughout their 20s for people with a college degree. Their recent unemployment rate is historically a surprise. The bad kind of surprise.

This was organized before Senator Elizabeth Warren called out the Abundance movement in a speech about the party’s future, which kicked off fighting on X, as well as posts by Ben Winsor of the Open Markets-affiliated Liberty and Power Substack and Jonathan Chait of The Atlantic.

Subscribe now

Given the interest in the topic, I thought I’d share the notes I prepared, written up after reflecting on the discussion.

On tensions between Abundance, as popularized by Ezra Klein and Derek Thompson’s book, and more populist approaches, I think this is mostly a solved problem as we enter 2026.

Democratic campaign veterans like James Carville and David Plouffe emphasize populist messaging focused on affordability and economic pain. Virginia Governor Abigail Spanberger and New York City Mayor Zohran Mamdani each spent their first days announcing serious efforts to identify ways to streamline housing production. Spanberger launched her Commission on Unlocking Housing Production, and Mamdani created his LIFT and SPEED task forces. We are arriving at a synthesis that emphasizes both challenging high prices and bad actors while also prioritizing longer-term supply-side issues.

This isn’t surprising. Whether creating an effective new agency in the CFPB or rebooting an older one in the FTC, progressives care about how well the government works. There is also a long-standing streak of YIMBYism across parts of the left.1

But when it comes to rank-and-file progressives, there’s still a genuine confusion and discomfort with Abundance. This isn’t just a positioning fight within the Democratic Party. To keep things grounded, here are three specific questions that point to bigger issues.

1. “Does Abundance require the federal government to overrule state-level AI regulation?”

The Abundance Institute is one of the leading policy advocates calling for the federal government to block state-level AI regulations, a live issue during the tax bill debate and again recently with a potential executive order.

This led to funny moments where Abundance people were mad that someone might assume the Abundance Institute speaks for Abundance. The Institute had to clarify that they actually had the name first, years before the book. This created some genuine confusion: who exactly speaks to the boundaries of this movement?

If Abundance means YIMBYism, building more housing and preempting exclusionary local zoning, I am 100 percent a supporter and encourage people to contribute. If Abundance means e/acc, letting tech and AI rip across every domain regardless of consequences, I am not in favor and encourage people to be critical. When I encounter Abundance in Washington, D.C., I’m never sure which of these two I’m going to get. It takes time and cognitive work to figure out which is which, time most people don’t have.

A movement driven by magazine writers and academics will find it difficult to police the boundaries of what’s included and excluded. When the Center for American Progress writes convincingly against state AI preemption, we know how that stands institutionally. If an Abundance-affiliated writer argues against it, does that count for the movement? There are big stakes here: $150 million in AI lobbying is being deployed against elected officials, including on this basic preemption question. How should progressives understand what gets included and excluded?

2. “Does Abundance require endorsing charter schools?”

You can sometimes see Abundance thinkers say that the theory calls for charter schools. Setting aside your own views on charters, they are clearly a different matter than the YIMBYism policies Abundance has been associated with. But they do fit if Abundance is meant to be a centrist omnicause.

The omnicause is a term for the gravity well that pulls politics into ideological alignment. If you are progressive or conservative on one topic, you tend to end up the same on all topics. There are interpersonal, institutional, and technological (e.g. social media) reasons this happens. Right now you can read many Abundance authors and thinkers argue that the term should function as an omnicause for centrists.

This is how I understood Abundance in early 2024, well before the book came out. Abundance was pitched then as a self-conscious moderating “faction” within the Democratic Party. The surprise bestselling success of the book and the increased prominence of YIMBYism in Democratic electoral politics were in some ways a temporary detour away from this idea.

To avoid sounding conspiratorial, consider two prominent YIMBYs who are uncomfortable with this move. Editor-in-chief of The Argument Jerusalem Demsas writes: ‘Don’t make abundance the moderate omnicause.’ She worries that Abundance will get “watered down into a sort of umbrella term used by vaguely pro-business centrists who want a new slogan.” If this happens, Abundance will lose the more radical implications of YIMBYism (and the practical political anchor it provides) and instead become a buzzword for people who want to moderate the party.

And former policy director for California YIMBY, Ned Resnikoff, had a Roosevelt Institute paper, “Lessons from YIMBYism: Taking ‘Abundance’ Back to Its Fundamentals,” on how to apply YIMBY principles across economic topics. The piece starts with an important intervention about how Abundance has become so “ideologically capacious,” often “pursuing entirely different and mutually exclusive objectives” where it isn’t “possible to build a coherent synthesis that accommodates” them all. As a result, he’ll just stick with describing YIMBYism instead:

I use the term YIMBYism instead of abundance throughout to emphasize that this essay is about a particular policymaking approach, and not about the larger ideological debates that have become part of the abundance discourse.

I think Resnikoff has ended up at the correct place. YIMBYism has important insights across domains. The rest of Abundance, though, can feel like a confusing grab-bag of priorities.

3. “Are the Affordable Care Act and lowering prescription drug prices part of Abundance?”

A recurring move among some Abundance advocates (it’s in the book) is to characterize the Affordable Care Act as a pure demand subsidy, throwing money at coverage without any effort to manage supply or delivery.2 This is confusing. From creating the Center for Medicare and Medicaid Innovation, to implementing bundled payment initiatives, or launching the Hospital Readmission Reduction Program and the Medicare Shared Savings Program, the ACA spent enormous energy on healthcare delivery and “bending the cost curve.”

It’s the right of every writer to say their idea has never been tried before. But if more direct measures to bending the cost curve are written out of Abundance, where does this leave us? Take a voter worried about their kid seeing a doctor. Imagine a candidate responding that their main effort will be removing certificate of need requirements for new hospitals and taking on the AMA to expand residencies. And that’s it. How does that help this parent now? And how can they be certain it helps them later? Will new hospitals actually get built? Will insurers pass along that supply as actual access? Will it translate into care?

These are good ideas, but supply-side reforms work on long timelines, and the benefits and their distribution are uncertain. The causal chains are long, and voters are right to be skeptical of politicians promising that they will automatically do things and help. And as Bharat Ramamurti notes, without some measure to directly address people where they are, it is difficult to pass these supply-side reforms.

The Affordable Care Act, whatever its limitations, got over 20 million people coverage while reducing projected healthcare spending by hundreds of billions of dollars relative to baseline. This looks like Abundance to me: better outcomes at lower prices by reducing unnecessary spending and harnessing the government’s capacity to operate at scale. If the ACA gets read out of the Abundance framework, what should progressives, who want to expand Medicare and create public options, make of the agenda?

The book itself seems conflicted on drug prices. Klein and Thompson note that European countries achieve lower costs because their governments negotiate prices, versus our “hodgepodge of private and public insurers who do not coordinate and do not effectively negotiate,” which they characterize as weak “state capacity.” But they don’t follow through on this logic. And the broader Abundance conversation is even more muddled. I’ve spoken with people adjacent to this world who suggest that real Abundance means abandoning the prescription drug price controls in the Inflation Reduction Act, with the Democrats deprioritizing lowering drug prices through negotiation because of the risk to pharmaceutical innovation.

If Abundance is a supply-side complement to traditional liberal priorities like healthcare, then that’s valuable. But it can come across as if it’s meant to displace or subordinate those priorities instead.

It’s not surprising that progressives are confused by what Abundance is offering. If it’s YIMBYism plus supply-side complements to traditional liberal goals, that’s a coalition worth building. If it’s a centrist omnicause that wants to replace liberal priorities with smaller goals unlikely to win politically or deliver substantively, and whose only checks against business capture are writers otherwise busy getting people to click, like, and subscribe, progressives are right to be skeptical it’s a good direction to go.

I enjoy writing empirical pieces that surface new arguments, work that’s too specific for traditional outlets but too substantial for social media. Twitter/X no longer supports that kind of work, and Substack does. And 2026 is going to be a consequential year, so I want a place where I can be part of the public conversation as it happens.

My initial goal is to write one post a week. Sometimes I’ll do more. I want to grow my readers. I don’t know what a realistic target is, but my stretch goal is to triple to 9,000 subscribers by year’s end. So please consider subscribing and adding me to your recommendations.

I plan to follow the secret of blogging as described by Max Read:

But the key lesson, the thing I would impart to any aspiring bloggers, content creators, or newsletter proprietors, is that the cornerstone of internet success is not intelligence or novelty or outrageousness or even speed, but regularity. There are all kinds of things you can do to develop and retain an audience [...] but the single most important thing you can do is post regularly and never stop.

Which in practice means I’ll stress less before hitting publish, posting without fear of perfectionism, knowing I will write again soon.

This Substack uses a sans-serif body text font because empirical and ideological work should feel clean, modern, and legible, and because I’m building toward a futurism of endless economic possibilities.

I’m turning on paid subscriptions, though I will keep everything free. Consider subscribing if you want to support this endeavor.

Disclosures

If I’m asking you to read, engage, promote, and support this Substack, I should make some disclosures. My full-time job is with the Economic Security Project, where I’ve led their policy and research team since March 2025. I expect to make more than 90 percent of my income from that job. This Substack is on my own time and independent of that work.

I recently served in the White House in 2024. That means I can state that I “favorably adjudicated Tier 5 investigation; Top Secret/Sensitive Compartmented Information (TS/SCI) clearance granted.” So this Substack is not compromised and displays “unquestionable loyalty to the U.S.” (My favorite part was an F.B.I. agent asking me about being on the editorial board of Dissent Magazine, and me getting to clarify that “Dissent was founded as the non-Stalinist magazine of the mid-century Left.”) It also means my personal finances are public. They are very boring, and I’m flagging them here so you know there’s no financial shenanigans in what I write.

I use AI daily for research work, particularly with data analysis. I’ve used R for 20+ years and spent time as a software engineer, so it’s easy for me to sanity-check AI code in real time.

For posts, I write a first draft myself, and then do three AI passes. The first is as an editor, asking what needs more work, what can be tightened, and what phrasing is unclear. I’ve been writing as a contributor for over 15 years, so I have a sense of what great editing looks like, and approach AI in that way. The second pass is fact-checking and anticipating objections. The third is proofreading.

I think AI images are fascistic and will try not to use them.

Hope you join me on this trip. As I am learning, please leave any suggestions for what makes a Substack work well in the comments.

And if you read this far, you can subscribe below.

Subscribe now

With my colleague Becky Chao, I wrote an Affordability Framework (pdf) for my new job at the Economic Security Project. Even as we began drafting, it was clear that affordability was becoming a central political focus, but also one in need of a fresh economic understanding of the problem.

It was well-received, with write-ups in Vox and Politico. I discussed it on The Realignment podcast, and Becky discussed it on the Liberal Currents podcast. Mark Schmitt wrote about the political dynamics of affordability and got what we were trying to do. I plan on expanding the framework further; expect more in the coming months. If you haven’t yet, check it out.

Another Thing I Enjoyed Writing

For Democracy Journal, I wrote The Abundance Doctrine, a joint book review of Abundance and Why Nothing Works. Rereading it now, I think it gets both the strengths and limits of the argument right. Law professor Noah Kazis gave it first place in his ‘review of the initial reviews’ competition.

Best Help to Stop Doomscrolling

I took the social media apps off my phone, which is a big help. But the biggest improvement was reviving a very Web 1.0 technology, the RSS feed. I started using Inoreader, though any reader works. More sites than you’d think have RSS feeds. Substacks all have them on the main page with /feed, such as here, for this one.

Take an hour or two to move your favorite reads over and customize it. You regain control over what you see and how you see it. Nothing is trying to hijack your attention or push you to engage beyond simply reading.

The late Aaron Swartz helped develop RSS at the age of 14, a technology closely aligned with his lifelong commitment to an open, decentralized internet. Aaron was a friend, not a close one, but one I did some work with and still miss. He is still remembered fondly by those who knew him, as recent pieces from Ryan Grim and Henry Farrell show. It makes me angry that the state prosecuted him for downloading files, something AI companies now do openly, at scale, in the service of turning the internet into something more enclosed and less human. Rebuilding an RSS feed is a small act to remind you of an earlier, better version of the internet.

Best Low-Stakes Personal Crisis

I was born in 1979, and turned 46 this year. I used to just think of myself as a very young Gen-Xer until a decade ago, when Anna Garvey coined the term “Oregon Trail Generation” to describe those of us born between roughly 1977 and 1983. (I will not be using the term Xennial.) Others have written about this cohort before: old enough to remember a pre-internet childhood, young enough to come of age alongside the web, but before social media colonized daily life. As Anne Helen Petersen describes that experience, “In 1999, the internet was not yet real life.” This in-between state fit.

But I think this year all of us have to pick a side. Gen-X, as a political and cultural bloc, is now widely coded as Trumpy and MAHA, and in 2024 they were among his strongest-leaning cohorts. That politics feels alien to me. Obviously this isn’t true of everyone, but it has become a dominant Gen-X posture, one I increasingly don’t recognize as my own.

Looking at it now, it makes sense that Gen-X’s vague distrust of authority, an ironic refusal of commitments, and a lowering of expectations have turned into measles outbreaks and deportation detention camps. It’s telling that many of the people who went Trumpy from liberal, left, or moderate spaces are now in their mid-50s, the core Gen-X years. For at least some, what once passed as ironic distance hardened into a politics comfortable with cruelty so long as it carried the aesthetic of anti-authority. (And the best book on this happening, Doppelganger by Naomi Klein, came from a Gen-X leftist who could see it happening in real time.)

Meanwhile, the millennial reconsideration has infiltrated my social media. As a recent ICYMI podcast described it, “millennial cringe” became “millennial optimism.” There are all these posts about living in Brooklyn and writing online in the early 2010s. And it’s like, yes, my blog did move me across the country to Greenpoint in 2010 to write online for a policy startup, and I even wrote a cringe, earnest blog post announcing it.

So the award for the best low-stakes personal crisis is whether or not to mentally rebrand as an elder millennial, perhaps the eldest millennial.

Best Fight I Failed to Break Up Between My Kids

<Scene> My two daughters, then 5 and 3, were in the backseat of the car while I was driving. 5 had a new interest in the Ghostbusters, just the theme song, not the actual movie. 3 loves to play a game where she says “no” to everything, even factual descriptions of the world. I, on request from 5, started playing the theme from Ghostbusters.

5, to 3: This song is about the Ghostbusters.

3: No it’s not.

5: It is! It’s the theme song for the Ghostbusters.

3: No, it is not.

5, getting upset: But it is!

3: No.

Me, to 3: Just tell your sister this song is about the Ghostbusters.

3: No, it is not.

Me: I think it’s a commercial in the movie? He’s saying to call the Ghostbusters.

3: No.

<5 has started to cry>

Me: The Ray Parker Jr. song Ghostbusters is on the movie soundtrack for the movie Ghostbusters. He wrote it for the Ghostbusters.

3: No.

<5 is wailing now, just absolutely lost it, tears down her face>

Me: Listen to the words. He’s not afraid of ghosts, busting ghosts makes him feel good. This is describing what it is like to be a Ghostbuster.

3: No. It isn’t.

5 stopped crying, after a while.

Watch I Wore Most This Year

I kept it basic. I bought a Casio G-Shock GW-M5610U-1 at the beginning of the year. This was a year of survival, of watching some pretty bad stuff happen. I wanted something stripped down for it. It’s a classic for a reason. I recommend the upgrade for solar charging and auto-time.

Last year was an early-1960s Omega Seamaster De Ville. I was (and still am, but less) daycare broke so I needed a dress watch that conveyed seriousness, but was also not crazy expensive. I originally found this in a small alleyway shop in Kensington. It was perfect, and I’ll be bringing it back out next year.

Meme of the Year: I’m a Mommy/Mamacita

@celebsnapzHuda telling nicolas that she is a singlemom#Lovelsland#nicloveislandusa #loveislandusa #huda #chelley #loveislandusaseason7 #loveisland #foryoupage#fyp #viral#foryou #fyp #loveislandusaseason7

Tiktok failed to load.

Enable 3rd party cookies or use another browser

This short exchange from Love Island between contestants Huda and Nic, where Huda explains she’s a mom and Nic doesn’t get it, causing Huda to give an Oscar-worthy laugh to his questions, is my meme of the year. At a baseline, it required no understanding of the show, which I’ve never seen.

But it’s a perfect meme for two reasons. First, you can play it several different ways depending on your mood. You can do an ironic-yet-poignant address to yourself on becoming a parent:

@itscaseyjoeI just know 25 year old me is gaggged! How did this happen 😂 #loveislandusa

Tiktok failed to load.

Enable 3rd party cookies or use another browser

You can use the remix for a crazy baby shower theme:

@nadiarassoulhad to use this sound at the baby shower 🤭 #mamacita #babyshower #loveisland

Tiktok failed to load.

Enable 3rd party cookies or use another browser

But you can play it up wholesome, Peppa Pig can join.

But the other, more important, reason it is a perfect meme is that you can play it equally from both sides. You can enjoy it from the perspective of being confused that someone is a parent, while also being confused that someone would be confused to learn someone is a parent.

So you can plug in when your friends with no kids come to visit:

@courtneyyykingMommy? Mamacita?? #loveisland #loveislandusa #mamacita #imamommy @Love Island USA #huda #nic #peacock

Tiktok failed to load.

Enable 3rd party cookies or use another browser

But also when someone without kids realizes that their coworker is raising a child on the same paycheck they get:

@elc_illawarraMOMMY…MAMACITAAAAA💃💃💃#workhumor #trending #loveislandusa #fyp

Tiktok failed to load.

Enable 3rd party cookies or use another browser

Plus the more general trend of people having kids later, as noted by someone worried their 36-year-old sister has become a teen mom:

@cherie.brookeWe’re both teenagers wdym you’re pregnant? (it’s just a skit)

Tiktok failed to load.

Enable 3rd party cookies or use another browser

In a year where the politics of kids went to extremes and have the potential to polarize, it’s good to know the best memes are able to navigate and communicate across these large boundaries. But I’m still not going to watch Love Island.

I am deeply worried that President Trump has brought about a collapse in the governing capacity of American political institutions. The post-1970s balance of power rested on a Congress interested in defending its powers, political parties not beholden to a single individual, and courts that were not fully polarized. Trump stepped into that vacuum of countervailing power and filled it with personalized executive rule, systematically dismantling agencies and weaponizing them under his control. What we took for granted about the state has fractured, and I haven’t seen anything close to finding an alternative.

But I am no longer worried that President Trump has created a durable political economy that realigns the electorate. At the beginning of the year I worried we were in a 1936 or a 1984 moment where, without the electoral blowout, Trump would execute an agenda that addresses the challenges of this stage of capitalist development while securing a new economic and political center of gravity.

That hasn’t happened. It now feels borderline absurd to say it could have happened after this past year. It’s not just the bad sentiment and polling numbers, or the weakening labor market. There’s no theory of the case for what they are doing and why they are doing it. The centerpieces of this administration, the tariffs and the deportations, are unpopular. The tariffs are politically radioactive. The justifications for the tariffs change month by month. President Trump has somehow been outmaneuvered by the Democrats into political ownership of rising health care costs in 2026. And that’s even before the trillion dollars in Medicaid cuts, which the smarter Republicans realize will fall disproportionately on the aspirational coalition they had hoped to assemble.

What Went Wrong?

Why is this? Trump’s own ego, grievance, and transactional corruption are no doubt part of it. Trump coins, Trump battleships, The Trump-Kennedy Center. Trump is building a new Mar-a-Lago on top of the White House’s East Wing while CEOs bring him personal gifts. It’s also been a mean and grim year, a parody of right-wing wokeness, under President Trump. Trump began his first presidency with the idea of “American carnage,” but the viciousness and vindictiveness the phrase invokes hasn’t fully materialized until now.

The administration’s staffing decisions reflect a generational and ideological shift, the groyperfication as John Ganz describes it, that prioritizes cultural grievance over policy development. Who cares about coming up with a GOP plan for health care anymore? DOGE letting a bunch of kids loose with AI, thinking they’d reboot the government and eliminate the deficit, turned out to be a failure except for the lulz along the way. But groyperfication goes further. Their vibes bubble has them believing their mission is to engage in a culture war whose outlines are confusing even to the most internet-brain-melted, far afield from the non-based, normie goals of trying to secure the coalition that elected Trump in 2024 on a permanent basis.

But beyond those, I think this is the inevitable result of a coalition that dreams to be several contradictory things at once: a libertarian’s dream of a hollowed-out state, a protectionist’s managed economy, a global hegemon unburdened by obligations, and a hardliner’s deportation police state. Politics is trying to manage these contradictory dreams at the same time, but in 2025 it seemed like it was trying to go in every direction at once.

The Hyundai-ICE raid is the example that stands out to me. In September 2025, ICE agents raided the Hyundai electric vehicle battery plant in Ellabell, Georgia, detaining 475 workers, mostly South Korean nationals, over allegations of visa fraud and unlawful employment. Here’s a clear conflict between two priorities. Some in MAGA want to reindustrialize America and build strategic regional allies to isolate China. Some want no immigrants at all and reject the idea that we should use visas to bolster our productivity and knowledge.

Instead of picking a path, the administration did both. They did the raid, humiliating those running the plant and the South Korean government, causing an international scandal. They then turned around and reversed course, with Trump saying he was “very much opposed” to the raid. It could have been handled more easily. If the administration wanted some action on the immigration front while keeping the factory running, they could have picked up the phone. Hyundai and South Korean officials would have answered. But they didn’t choose. They ended up trying to do both and did neither, causing major problems instead.

Everything Bagels

The inability to direct among different pulls of their coalitions calls to mind the writer Ezra Klein’s 2023 description of “everything-bagel liberalism,” where “sometimes [the government] tries to accomplish so much within a single project or policy that it ends up failing to accomplish anything at all.” This description didn’t particularly hit the intended target of the CHIPS semiconductor grants. There’s an excellent overview of the process of CHIPS by the people who executed it, a Substack called Factory Settings hosted by the Institute for Progress, and I’m very happy they are doing it. Knowing a bit second-hand about that process, it accurately reflects it, all the strengths and weaknesses. They are dishing the dirt.

The “everything-bagel” term was a contentious one inside liberal politics, because of the implied attack on high-road employment and the coalitional politics required to pass bills into laws. You can read rejoinders by Ben Beachy here and here. But I think the concept is very useful to abstract up a level from that: what are the higher-level conflicts within your coalition, and how do they get settled and prioritized?

Because “everything-bagel authoritarianism” does make sense to me. Let’s consider what some of the goals the MAGA Trump coalition wants to accomplish:

• Mass deportations beyond securing the border or targeting criminal offenders.

• Reindustrialize America as a matter of national security, innovation, and employment.

• A libertarian attack on the administrative state, to strip the government of any power to regulate, especially for Silicon Valley.

• Isolate China as a global rival, decoupling from Chinese supply chains.

• Stop peer nations from ‘free-riding’ on the United States, and have a more aggressive stance against traditional allies and the rules-based liberal international order.

These goals are in direct tension. One can, with leadership and smarts, navigate the tensions here. But nobody appears to be trying to do that.

One place we see this conflict is between the libertarian attack on the administrative agencies versus the interest in reindustrializing America. Attacks on and firings at agencies like the Department of Energy, Commerce, and the EPA come into direct conflict with ambitions around reshoring supply chains, accelerating domestic manufacturing, and outcompeting China in strategic sectors like batteries, semiconductors, and advanced materials.

Right out the door in March DOGE made major cuts of the staff responsible for the CHIPS implementation of semiconductors in the U.S. Commerce Department. Half of the Department of Energy’s Loan Program Office (LPO) took deferred retirement alongside employee caps being announced. As The American Prospect’s Ryan Cooper noted, there’s a dark irony to Elon Musk attacking LPO after he himself got an LPO loan for Tesla, one of its major success stories. Clean energy initiatives, which are where China is taking the global lead, are being eliminated, often in the middle of the process.

Beyond that, if we’re going to upskill the domestic labor force into manufacturing, who will do all the jobs those workers were previously doing? We were near full employment, so it is difficult to discuss moving some workers into fields unless we understand where they were coming from. There’s hand-waving assumptions that deportations will increase automation, or bring native-born workers out of the sidelines. I don’t see any reason to believe that, largely because the proponents aren’t offering grounded empirical arguments for them.

The other place is on the global stage, something I’ll say I follow the dynamics of less closely. But perhaps because I don’t, the dissonance of trying to use economic statecraft to both isolate China while also attacking peer countries stands out even more to me. If the goal is to actually try and build alliances to keep China’s exports in check, why spend the last year mocking Canada as a potential 51st state of the USA and threatening to purchase Greenland? You see this in the National Security Strategy (NSS), which shifted from China in the first term to the “civilizational erasure” of Europe in the second.

Consider India. As a result of Trump’s actions this year, India now faces the highest tariffs when exporting to the United States (50%) of anyone else in the world. This is higher than China (47%). Do we want companies to decamp from China to India or not? This has sent the Modi government to meet with China for the first time in seven years to try and normalize relations, directly against decades of U.S. foreign relations interest. I understand anti-India sentiment is normalized among the younger groyper staffer class, but what is even the goal here? And who is trying to navigate it?

These goals are genuinely in conflict. You can’t simultaneously gut the Commerce Department while executing complex semiconductor policy, or deport workers while reshoring labor-intensive manufacturing, or isolate China while alienating Europe and India.

It can’t be said enough: if you want to dismantle a liberal, pluralist democracy by empowering a strongman who promises to purge corrupt elites, bypass a paralyzed legislature, and subjugate civil society and the business class to personal loyalty tests administered by the Executive, you need to be able to tell parts of your far-right coalition “no.” That is the core of the everything-bagel analysis. But they aren’t doing that. Which means the libs might have a chance?

This isn’t just the general malaise surrounding the economy right now. The Trump administration came in with four plans to reorganize the labor market. None of them came to pass. Next week I’ll sketch out why these ideas failed. For now, it’s worth laying them out clearly, because the gap between the theory and the reality explains a lot of what we’re seeing.

First: Rebalancing Health Care Jobs

At the start of 2025, the Trump administration argued that the economy was far weaker than it appeared. The headline job numbers, they claimed, were misleading, propped up almost entirely by health care and government employment.

Stephen Miran, then chair of the Council of Economic Advisers, put it bluntly in March:

“73% of all jobs created in [the last] two years were due to government and government-adjacent sectors [...] like education, sectors like health care [and] the short-term pain is coming from the reorientation of the economy, from the government to the private sector.”

Treasury Secretary Scott Bessent echoed these sentiments in February. According to Reuters, he argued that “growth in the past 12 months has been concentrated in public and government-adjacent sectors such as health care and education.”

Moving workers was a broader strategy. Here’s how the Department of Government Efficiency (DOGE) put it in their ‘Fork in the Road’ materials encouraging federal workers to take deferred resignation: “We encourage you to find a job in the private sector as soon as you would like to do so. The way to greater American prosperity is encouraging people to move from lower productivity jobs in the public sector to higher productivity jobs in the private sector.”

I did not find this 2024 labor market story convincing at the time and wrote about it here. But they did. So their first order of business was reducing the proportion of jobs coming from government and health care, with the assumption, spelled out more in a minute, that they would create enough jobs to replace them. How did that go?

With health care jobs accounting for more than 100 percent of net job growth in some months, clean comparisons become tricky. But looking at private education and health care, the bugaboo of the early second Trump term, in 2019 they were 29% of job growth. For 2023 and 2024 it was 45% and 50% of job growth. In 2025 it has been 113% of job growth. If you include government jobs as well, the share rises from 73% in 2023–24 to 88% in 2025.

Health care continued to do the heavy lifting, while the rest of the labor market collapsed. The promised handoff from government-adjacent employment to private-sector dynamism never materialized.

Second: Tariffs to Reindustrialize

Tariffs were supposed to reindustrialize the country.

The administration argued that higher tariffs would reverse decades of decline in manufacturing and goods-producing employment. This was framed not just as an economic adjustment, but as a historic restoration of American industry.

The Liberation Day tariffs materials made this explicit: “The President’s reciprocal trade agenda means better-paying American jobs making beautiful American-made cars, appliances, and other goods.”

Commerce Secretary Howard Lutnick pushed the vision even further. In a widely circulated interview, he described a coming wave of reshored production:

“[An] army of millions and millions of human beings screwing in little screws to make iPhones -- that kind of thing is going to come to America. It’s going to be automated and great Americans, the tradecraft of America, is going to fix them, is going to work on them [...] it’s going to have the greatest resurgence of jobs in the history of America to work on these high-tech factories, which are all coming to America.”

This was always framed as a longer-term project. But even on its own terms, the question is straightforward: did the administration at least begin to bend the trend lines?

No. That’s not happening.

Here is the data for 2025 with trend lines from December 2022 to 2024. I incorporate the preliminary annual BLS benchmark revision, which aligns monthly payroll estimates with more accurate administrative employment counts, to the earlier data, reducing 2024 job gains. (Data pulls and estimates here.) There is no visible break from the preexisting trajectory, let alone the beginnings of a resurgence. There’s actually further deceleration. This might be for the best, as goods-producing jobs pay, on average, less than private-sector service jobs. But whatever is meant to happen, it hasn’t started yet.

Three: Deportations for Native-Born Benefits

Beyond creating new jobs, the administration argued it could reclaim jobs through mass deportation. By removing foreign-born workers from the labor market, native-born workers would supposedly see their job prospects improve.

Vice President Vance has been especially explicit about this theory. In August, he tweeted: “I was told six months ago that Americans losing jobs and the foreign-born gaining jobs was an irreversible demographic fact.”

As poor Jed Kolko has had to spend 2025 explaining, the reported levels Vance references aren’t correct. But Vance is definitely right about one thing: that pattern isn’t irreversible. Another possibility is that both groups lose at the same time.

As shown above, it’s a worse year for native-born workers with their unemployment up. The administration believes that employment is a zero-sum competition between groups. But the evidence points to a shared exposure, to both prosperity and, as we’re now experiencing, to rising joblessness amid weakening demand and policy confusion.

Four: The Breadwinner Economy

I think the last is less discussed but is one of the more important things, and that’s the gender theory of MAGA economics. The idea that bringing back manufacturing jobs with high wages for native-born male workers would make them more attractive suitors to women, which in turn would raise the marriage rate, and then fertility rate, is hanging in the background here. I think the gender and culture war aspect of this helps explain why the broader conservative movement is willing to risk tariffs.

However, that’s not happening. With goods weakening and health care holding the labor market together, Trump has made it a worse environment for male employment. Starting in May, right after Liberation Day, men have gained 4,000 out of a total of 119,000 new jobs. Overall in 2025, women have gained 67% of jobs, compared to 51% in 2023 to 2024. The labor market is moving in the opposite direction of the one this theory requires.

I wish I could tell you that the reason people in their twenties can’t find jobs and the reason many of us are likely to be poorer over the next few years was more sophisticated than “we’re going to turn all the girlbosses into tradwives once they see all the manly men at the USA iPhone-screwing factory.” But I don’t think it is.

The bleak irony is that even after sacrificing real prosperity to chase this 4chan-level political economy, they still won’t achieve their goal. The jobs aren’t coming back, the wages aren’t rising, and family formation won’t be rescued by trying to rewind the labor market to a world that never existed in the first place.

Much of the coverage has focused on how divided the Fed appears. Joe Weisenthal had a column noting it is as if “that implicitly the Fed has already abandoned its target” and Matthew Klein wonders if the Fed has submitted to political demands.

A Target Range

Here’s a theory of what the Fed is thinking. I’m not entirely sure whether this reflects what Fed officials actually believe, a useful way to model how they behave, or simply what I think they should be doing. It’s just a take.

My theory is that the Federal Reserve acts as if they have a target range of between 1.6 and 2.4 for inflation, and their dovishness and hawkishness alternate as they get to either end of the range. Why is that? Here’s a chart I watch closely on PCE inflation:

And here it is marked up to follow a few points:

This is total PCE inflation, what the Fed targets at 2 percent, along with four contributors that account for the total. Start at the upper-right green-colored box. Across 2018 to 2019, inflation was 1.7% a year. When inflation was much lower earlier in the 2010s, they did a lot to try and boost economic activity, including Operation Twist and Quantitative Easing (QE). But when inflation was 1.7% they did not stress much about it being too low. They instead defaulted to having a dovish lens on events as they happen, such as letting unemployment drift into a rate below 4 percent alongside strong real wage growth, while giving speeches on how unknowable r* is.

Let’s jump forward to now, and discuss the red-colored box. Over the six months prior to September, inflation was 2.7%. Yet 0.44% of that is in core goods, which was zero in 2024 and zero in 2018-2019, and is largely attributable to the one-time impact of Trump’s tariffs. That puts underlying inflation right at 2.26%. 2.26% is not 2.0%. But 2.26% is not something you should cause a recession to end.

However, being at the top-end of the range does put a hawkish lens on new events as they happen, mirroring the dovish lens on being at the low-end in 2018-2019. So when the tariffs, for instance, show up, the Fed is rightfully concerned about them breaking expectations, or becoming more persistent, or also causing a shift in services to bring their respective price levels into balance. It’s too close to the upper-bound of the range to assume it’ll act like a tax and nothing else, because if something breaks here you are well outside the range. You’ll keep more of the headline inflation rate in your Taylor Rule calculations.

So what about this divisive vote? One implication of operating near the top end of an implicit range is that it creates sharp discontinuities in how different policymakers read the same data. A slightly different read puts you either inside or outside the range. Take the purple-colored box in the chart above, where over the past three months inflation is still in the ~2.8% range, but now it’s all in services, not goods. You can see this in the chart below. Normally you wouldn’t over-index on any three-month period. But we’re at the top-end of the range, making us more nervous about these short-term movements, so that’s why many want to hold.

Yet, we’ve also seen the labor market weaken in the last several months, with, as you can see below, negative job growth in two out of the last four months. Other labor market measures are going south. It’s unclear whether this is because demand is falling or potential is, because of deportations and tariffs.

Meanwhile there’s a third way to read the data. Some still work for the White House but are on leave, and others are aiming to get Powell’s job next year, pushing them in the more dovish direction in reading the data.

Ranges are Good

The macroeconomist Greg Mankiw had a good argument in 2024 that “A target of 2 for inflation is better than a target of 2.0.” From the piece:

I feel strongly that a target of 2 percent is superior to a target of 2.0 percent. The difference between these targets, of course, is the number of significant digits. If you recall some science class you had in high school, you likely learned that the number of digits a person reports should reflect the precision of his or her estimate. Central bankers often forget that lesson. They sometimes speak as if they are targeting an inflation rate of 2.000 percent.

It would be better if central bankers admitted to the public how imprecise their ability to control inflation is. They should not be concerned if the inflation rate falls to 1.6. That comfortably rounds up to 2. And they should be ready to declare victory in fighting inflation when the inflation rate gets back to 2.5. As the adage goes, that is good enough for government work.

Maybe the Fed should even ditch a specific numerical target for inflation and instead offer a range. It could say, for example, that it wants to keep the inflation rate between 1 and 3. Doing so would admit that central bankers are not quite as godlike as they sometimes feign.

I like ranges. Back at Team Macro at the Roosevelt Institute, the economist Justin Bloesch put out an excellent paper with the case for an inflation range of 2 to 3.5 percent. Explicit inflation ranges are used by Canada (“The Bank of Canada aims to keep inflation at the 2 per cent midpoint of an inflation-control target range of 1 to 3 per cent”) and Australia (“Australia’s inflation target is to keep annual consumer price inflation between 2 and 3 per cent”), early adopters of inflation targeting that are widely viewed as having strong macroeconomic track records. Ranges also help resolve some of the persistent “catch-up” questions that complicate strict point targets. It’s good that the Fed might be using it, at least in my imagination.

It’s the holiday season, let’s do a fun post that won’t be for everyone. But if you make it to the end, you can see how proper economic microfoundations can clarify the debate over pronatalism on the left. Last week, we dug into the fertility data and started to build a take that worked in practice. But does it work in theory?

The take I’m exploring is that we should be thinking of the slowdown in fertility in terms of marginal penalties that would be best served by focusing on the concrete, near-term barriers facing people in their late 20s and early 30s. Others have written along these lines. Rachel Cohen Booth writes about how the “economy isn’t built for the biological clock.” Suzanne Kahn notes, in a recent book review of After the Spike by Spears and Geruso, that a goal should be “lowering the opportunity cost of children” especially “by giving parents more time.”

But does this take survive contact with rational expectations? We stress test all potential takes around here, including hitting them with the question of microfoundations. Our take must remain valid even when we assume people are forward-looking actors who will actively change their behavior to anticipate and react to changing constraints and policies. I don’t want to be doing my new take on a podcast or a panel, and suddenly get blindsided by the Lucas Critique.

This is extra relevant as we discussed issues with total fertility rate (TFR), which you can find more about from Our World in Data and The Conversation, a measure that assumes current timing stays fixed and people don’t shift births around.1 Put differently, TFR doesn’t survive rational expectations, it’s a statistic that can’t keep up with how real people optimize against changing constraints. But how well does a theory of penalties survive this test?

Model Setup

We also have this graphic from our dive last week:

Between the 1980 and 1986 cohorts, we see a notable slowdown in when people start families, even though they eventually arrive at the exact same destination. They both start with ~0 kids at age 15 and end up with 1.83 kids by age 38. But the 1986 cohort takes a very different path, delaying those births.

Whether or not it has forgotten how to answer basic questions about the business cycle, current macroeconomics excels at modeling how an ethereal representative agent allocates decisions across an infinite future under perfect information. Though young people would probably say, if not phrased exactly this way, “I don’t understand the parameters of how to date on apps and raise a family in this moment,” that isn’t going to stop us here.

There’s a literature around how future-looking agents decide to have children, built around models by Becker and Barro (not that one). Usually, those models are trying to solve complex problems like intergenerational wealth, cross-country comparisons, or human capital accumulation. I want to keep our model of a penalty very simple, a toy model we could build on later. What does our hypothetical “penalty” do?

We imagine a representative agent for the yearly cohort, who can choose a fraction of a child each period. Today’s choice to have a kid, bₜ, adds to tomorrow’s number of kids, Bₜ₊₁. Now you love your children equally, of course. But your love, U(Bₜ), is twice-differentiable, with the first derivative strictly positive and <looking around to make sure the kids are out of earshot> the second derivative strictly negative. And, last, there’s a penalty, pₜ, to having a kid at time t, that can change over time. The penalty is quadratic, which means the more you have kids in a period, the more the penalty bites. The agent starts with no kids, knows all the things, and also knows at 38 they’ll have exactly 1.83 kids.

All pretty straightforward? Let’s write that out.

Baby Bellman

Now we can solve this baby equation with a baby Lagrangian or a baby Bellman equation. I want to use the baby Bellman equation.

I actually think this recursive structure is a great way to think about fertility. Don’t get lost in the equation. V(B) is the value of having kids to our agent. How does it model when to have kids? A decision today, that’s the left part, and a decision tomorrow, that’s the right part. But what’s the decision tomorrow? It’s also V(…)! It’s the same equation, the same decision, but just repeated tomorrow.

Many often think “when’s the right time to have kids?” In the Bellman world, there’s just two time periods: today, and tomorrow, and tomorrow is just today all over again. But a little more exhausted, as you have to discount. You can literally have infinite time but you’ll still just have two times: today, and tomorrow, which is just today’s decision repeated again. It’s today all the way down. (There’s no good time, just have the kids.)

Baby Euler Equation

Yet people are shifting fertility around, and we have a way of determining how to assess that trade-off: the Euler equation. We take the baby Bellman equation, and take the baby first-order conditions. From there we apply the baby envelope theorem, which gives us the baby Euler equation:

Let’s focus on the Euler: The left side is b at time t, which is the decision to have a kid now. The thing that makes this unique is that b at time t+1 is on the right side of the equation; it is reflecting the decision today explicitly against the decision tomorrow. The logic is that you balance your decision-making so you are indifferent between moving just a fraction of having a kid between today and tomorrow.

Let’s talk about the term labeled “timing penalty,” which is the penalty tomorrow divided by the penalty today, that is added to the decision to have kids now. I’m going to assume you hate math and are mad you got here and take this slow, because it’s important. When the penalty today grows larger than the penalty tomorrow, the denominator gets bigger than the numerator; that number goes down. And you are adding it to b, the decision to have a kid, so it decreases having a kid now. If the penalty tomorrow gets larger than the penalty today, the opposite happens, the numerator grows faster and the number of kids today increases.

Or, in terms of life, when young women face higher penalties today, like unstable jobs, thin safety nets, career costs, expensive childcare, relative to what they expect tomorrow, they delay births. And when anticipating higher future penalties, like infertility or balancing aging parents, they shift births earlier.

Note if you don’t hate math you can just take the derivative of having a kid today with respect to the two penalties (noting p is positive and b is non-negative) and get the same result, a higher penalty today lowers today’s fertility and a higher penalty tomorrow increases it today:

With very simple assumptions, a rational agent will spread the timing to have children seeking out when the penalty is lowest. One can even try to estimate the penalty, which is an exercise for the footnotes.2 I believe my new take survives rational expectations, so I will continue to build on it.

Pronatalism Under Microfoundations

Currently there’s a lot of interest in “pronatalism,” though it’s often difficult to pin down the politics of it, especially on the left. But this is where microfoundations can clear the confusion, by examining the penalty in the baby Euler equation.

We can identify two different approaches to this equation. There are those who want to alleviate the penalty term, to reduce pₜ in those critical late-20s and early-30s years when the penalties are currently highest. This approach takes the equation seriously and asks what makes the denominator spike for young adults today, then works to address those constraints directly. Childcare, paid family leave, housing policy that builds, child allowances, healthcare, and workplace flexibility, all to move the marginal penalty. This path respects the optimization problem, acknowledging that people are rationally responding to real constraints, and seeks to lower those constraints so the penalty is smaller during these years.

Then there are those who want to suppress the ability to time, to remove b as a choice variable entirely. (Choice as in both the reproductive and Bellman sense.) This is happening with anti-choice campaigns to eliminate access to abortion and contraception. It’s also happening with the Trump administration broadly, most recently moving to reclassify nursing and other female-dominated fields as non-professional degrees, raising financial barriers to access. This is based on the conservative argument that preventing women from accessing higher education can increase fertility, one that, as Darby Saxbe shows, is wrong on its own arguments. You can’t make the timing choices that works for you if you have no control over your life or your economic future.

Both approaches seek to change fertility patterns, but through fundamentally different mechanisms. One lowers the barriers people face in having the families they want when they want them. The other removes the ability to choose, to time at all. One side trusts women to make decisions for themselves when constraints are eased. The other side sees women’s decision-making itself as the problem to be suppressed. If there’s a question of pronatalism in front of you, ask what side it’s on. And always ask yourself, what side are you on?

2

I’m going to pause here and say I am not confident that this next section is correct, and it may just get deleted. Dynamic macro numerical methods is not something I’ve allocated skill points toward. Outside the academy it seems to mostly allow you to work at the Fed, and there’s a small chance the Supreme Court is going to say that the new job of the central bank is to just do whatever random 24-year-olds in the West Wing tell them to do, just as the Founders intended. If any serious macro people want to read the code and give feedback I’d love it!

But we came this far. We have the recursive model, and we have two historical fertility paths that end up in the same place but with different timing. We don’t have anything random to keep it simple. Let’s take log utility and set β to 0.92 (it’s every two years for a period) and back out the implied penalty for each period. This isn’t really estimation, it’s just walking the implied penalty backwards.

I think this is correct, that this is the penalty sequence {pₜ} that creates the the birth pattern for birth year 1980 and 1986 for the model above. If so, notice that the penalty from 28 to 32 is higher for 1986. But note that this higher penalty is not that much higher. In fact, it’s as if it just doesn’t get easier at 30 than it was at 26. But many of those likely increased penalties (building more housing, above all) are addressable, if we want it.

With the government shutdown, there’s been no new economic data, and it’s been very boring without any new numbers to investigate. So I decided to get to the bottom of this fertility crisis everyone is always arguing about. I’ll say up front that I didn’t have a strong take on the topic. I’m not experienced in these debates, but I am careful with data and can share what I find.1

I decided to dig into recent microdata from IPUMS: the 2024 CPS Fertility Supplement. The survey is fielded every two years with Census, with the most recent round in 2024. Unlike some other statistics, which only show births in a given year, the supplement captures cumulative births across the life cycle, making it one of our better windows into long-run patterns in fertility. I find I can replicate the aggregate Census numbers from the microdata, and then we’re off.

One thing stands out: the eldest millennials have the same completed fertility rates as the youngest boomers. There’s been no drop within the 21st century of the number of kids a woman has once she reaches 44 years old. That stability might surprise you given the popular narratives about collapsing family formation, and it’s worth diving into the data.

Figure 1 shows the average number of births by age, taken as a cross-section across two years. In the graphs here and elsewhere, I use three-year averages around each age, though the results look the same either way. The underlying variable is “Number of live births ever had” and the question is “Altogether how many children (have/has) (name/you) ever given birth to?” For women at each age, we sum up all the births, and divide that number by the total number of women to get the average.

Completed Fertility

In the year 2000, the green line above in Figure 1, a 43-year-old woman, who is among the younger baby boomers, averaged 1.92 births. In the year 2024, the navy line, a 43-year-old woman, who is among the eldest millennials, averaged 1.92 births. (For age context, a voice of a generation, the fictional core millennial Hannah Horvath from HBO’s Girls, turns 36 this year.)

You have probably heard of total fertility in the news; more on that in a bit. But the term here is completed fertility, the number of children a woman has by the end of her reproductive years, usually measured as an average for the early-to-mid 40s. This stability is well known; other analyses confirm that completed fertility has been essentially flat since 2000 (Census 2024, Pew 2018, BGSU 2023). But it’s worth emphasizing that this holds all the way through the 2024 data.

What has changed is the timing of births. Remember this is a cross-section based on the year we are observing. In 2000, a 30-year-old woman had, on average, 1.34 children. In 2024, that number had dropped to 0.98. The two cross-years ended up in the same place, but they follow very different paths to get there.

Before you ask, is this all immigration? No. We can track nativity status in this data. For women 41 to 44 who have at least one native-born parent, the weighted average number of children was 1.88 in 2000 and 1.87 in 2024.

Now completed fertility used to be higher than it is today. So when did the drop happen? We can look at the average number of births for women ages 41 to 44 in each CPS Fertility Supplement, going back every two years to 1976, as shown in Figure 2:

The big drop happened some time ago. There has been some variation since 2000. It hits a low of 1.86 in 2006, and a high of 2.07 in 2016. (What’s going on with those born around 1974?) But the headline is how stable completed fertility has been for more than two decades.

The Biggest Generation Happening

One reason you’ve probably heard about a fertility crisis is that the total fertility rate (TFR), which is the average number of children a woman is expected to have over her lifetime based on today’s birth rates, has declined. But TFR is sensitive to and misled by timing shifts like the one above. As Matt Bruenig once wrote, it’s “kind of a nonsense statistic.” If births shift later, even if the total number doesn’t ultimately change, TFR mechanically falls for a period. A birth disappears for a specific year, and TFR can’t observe the later birth that offsets it.

Demographers separate fertility into quantum (the eventual number of children, like completed fertility) and tempo (the timing of childbearing, like TFR). Completed fertility is a quantum measure, and that can be misleading in real time as well because we only know the final tally once a cohort finishes childbearing. If there were a genuine collapse in fertility underway, the completed fertility metric would reveal it only after the fact.

This is relevant because the largest single-year cohorts in the United States right now are those ages 34 and 35. These are the 1990 and 1991 birth cohorts that are the later millennials. To get a sense of this, Figure 3 is a chart of the age distribution in the United States for 2024 from Census data:

Jeanna Smialek, who is in that cohort, had a great piece last year on this generation as an economic-distorting force moving through the life-cycle in the New York Times (titled “It’s Me, Hi, I’m the Problem. I’m 33”). What happens to them is probably going to be an important part of demographics going forward. We can follow 34-year-olds in 2024 to get a sense in Figure 4, and compare them to slightly older cohorts:

Figure 4 plots completed fertility on the y-axis for people born in 1980, 1986, and 1990, from the point of view of the year 2024. So 34-year-olds in 2024 (born 1990) have a completed fertility of 1.46. This is comparable to 38-year-olds now, who, when they were 34, had a completed fertility of 1.51. Note however, that they were lower than the 44-year-old cohort, who were at 1.74 at that age. But the 38-year-olds caught up to the trend of the 44-year-olds. The 34-year-olds seem to be catching up as well?

Take a second to understand Figure 4, because we’re going to extend it to every year born between 1980 and 1991. Figure 5 provides that spaghetti-style graphic:

This might be the graph of the post. You can see a shift happening here, as if the entire profile were shifted outward to the right. Yet it appears to still be forming a consistent line and converging. If true, it’s not resulting in fewer kids, but just having kids later. Using some estimates, it looks like, for a given average number of children, the age has shifted to the right by between 2-3 years.

By Generations

So there was a shift. How can we visualize how this happened? Let’s go big, and look at the change in fertility at each age within the 21st century. In Figure 6, I’m going to color-code the lines by generation. Some quick age bands: Baby Boomers are born through 1964; millennials begin in 1981. Generation Z begins in 1997, making the oldest Zoomers 27 in 2024.

In Figure 6, I’m rescaling each y-axis to highlight the relevant shape of each series, while keeping the x-axis consistent across all plots: the years 2000 through 2024, with data from every even year in the CPS Fertility Supplement. This lets us watch how the tempo slowdown is evolving across time. But it does mean paying attention to the scale of each y-axis, which varies.

How should you read this? Take the box for age 30. For a 30-year-old, from 2000 through roughly 2010, the pattern is stable: they consistently had between 1.3 and 1.4 children by that age. Then, right after 2010, the floor drops out. By 2020, the completed births at age 30 land around 1.0. That’s a big shift in a short window.

There’s a lot here. Where to begin?

• There’s a break between 2008 and 2014 that looks especially clear from ages 28 to 33. The Great Recession casts a long shadow. A recovery that may have been slower than the Great Depression isn’t a great way to keep demographic trends stable. Real median household income falls over 5% during this time, so it’s probably not higher wages delaying children. But smartphones also became ubiquitous during this period. This mix is one reason I’m always hesitant on “the slowdown is across peer countries, what can you do?” arguments. Whether it’s that recession or tech, there are things that link countries that are distinct events and policy choices.

• It appears that there’s been some flattening of the drop in 2022 to 2024. See the ages 26 to 32 for some sense of this, where it’s pretty flat 2020 through 2024. I try not to over-index on any single year, but it’s worth people investigating this more fully. Why is this? Perhaps hybrid work has allowed for more flexibility in building a family (see Bailey, Currie and Schwandt (2023)). Is this the “vibe shift”? Maybe income support and wage compression during the recovery played a role. Did the inflation wave make having a perpetual youth of low-cost services seem less attainable than starting a family? Having the cause of something that may or may not have happened already be overdetermined is part of the fun of this work.

• The decline in teenage pregnancy is a major story here and it rarely gets discussed. We have, as measured here, the average number of children for 17-year-olds around 0.13 in 2000. That’s around one in eight, and that goes to 0.024 by 2024. That’s a near-total disappearance of teen childbearing over a single generation. Any model of tempo or quantum that ignores this shift is missing a central piece of the puzzle.

You probably have your own questions and interest in the numbers. I wanted to see how well I can incorporate vibecoded apps into blog posts. So I decided to vibecode a visualization app to display the data multiple ways using Google AI Studio with Gemini 3, which was surprisingly easy (about 20 minutes start to end). However, I can’t display it here, but you can click here or on the graphic below to see it yourself at my website. Click on your age and year if you so dare!

Tools and Measures

For something that is so often framed as a priority among elites, it’s remarkable how little high-quality data we have on the full lifecycle of births and family situations. Even straightforward questions like “controlling for education, does income or student debt correlate with fewer children?” are hard to answer.

One reason we end up using total fertility is that it is a very well-recorded event when children are born. You aren’t leaving the hospital after having kids until the Social Security documents are filled out. Births are logged immediately, and the age of the mother is a clean, consistent datapoint. Interviewing women later in their lives about how many children they’ve had (as Census does here) is a completely different, and more difficult, exercise.

One genuinely bipartisan effort would be to expand and strengthen the CPS Fertility Supplement. We could make it more regular, more in-depth, and add questions the experts studying this think would be helpful. But right now, sadly, the Bureau of Labor Statistics is under political assault, exactly when we need this data to be better, not worse.

Conclusion

For anyone who wants the United States to keep growing, 1.92 children is below the replacement level of 2 or 2.1. The shift to later parenthood carries real tradeoffs, which will reshape family life and intergenerational ties for both individuals and society. And this analysis is specific to the United States.

But spending time in the data leaves me less alarmist than the prevailing rhetoric. The shift in births is recent and likely shaped by shocks like the Great Recession and technology. The shocks are common across countries, and that means there is real policy space to act. We could begin by not wrecking the economy in service of the crudest mercantilist fantasies. But my new take is, instead of framing declining birth rates as a civilizational crisis, we would be better served by focusing on the concrete, near-term barriers facing people in their late 20s and early 30s.

“A lot of young people are saying, ‘Housing is way too expensive.’ Why is that? Because we flooded the country with 30 million illegal immigrants who were taking houses that ought by right go to American citizens.”

That claim is incorrect. The ‘30 million’ number is wrong. But beyond that, as Figure 1 shows, the rate of rent and home value increases peaked in 2022, along with the rest of inflation, and slowed from 2023 to 2024 even as immigration accelerated.

But maybe deportations could still make a difference and help renters. How much would mass deportation lower your rent? The Trump administration relies on a specific 2007 study, a standard in the field, “Immigration and housing rents in American cities,” by economist Albert Saiz. The Trump team points to it, and the Trump White House official on the Federal Reserve’s FOMC, Stephen Miran, cited it in his first Fed speech.

Saiz (2007) finds “An immigration inflow equal to 1% of a city’s population is associated with increases in average rents and housing values of about 1%.” Note what’s in the denominator of this elasticity: it’s the initial total population, not just the number of immigrants.

This gives us what we need to run the numbers. Let’s say there are 342 million people in the U.S., and the average rent is $1,500. Let’s define mass deportation as the government putting into motion an operation that results in 1 million immigrants leaving. Using the Saiz (2007) estimate, we get:

If they pulled this off, renters would save roughly $4.39 per month. To put that in context, economists have estimated the cost of RealPage’s AI algorithms using anticompetitive practices to determine your rent is around $70 a month. $4.39 is not going to make a meaningful difference, or really any noticeable difference, in the rent people face. And there are multiple issues here.

Beyond the Initial Estimate

Note that the Saiz estimate isn’t a low one across the literature. Ottaviano and Peri (2012) find an elasticity of 0.6 to 0.82 overall, though it can vary up to 2.3 depending on skill. Sharpe (2019) finds a much lower value of 0.3 to 0.4.

Let’s take a moment to understand why this is a difficult thing to measure. The intuition seems simple: immigrants move to an area, increasing housing demand. Since housing supply is fixed in the very short term, prices go up. But immigrants don’t move to random zip codes. They move to booming cities where jobs are plentiful and wages are rising. Those cities were already seeing rents go up because of that economic heat. Separating “rent growth caused by immigrants” from “rent growth caused by a booming economy that attracted immigrants” is the challenge of the literature. That’s why estimates differ. Even so, the range of estimates is not large enough to generate meaningful savings for people.

Maybe there was a break in this recent period? I don’t believe so. I downloaded Census 1-year ACS data on median gross rent, both foreign-born and overall population, and median household income, for all core-based statistical areas (the standard in the literature). This is quick-and-dirty. I’m not going to try to control for the reverse-causation issue here as I want to see if there’s a break in the estimate, and, since immigrants tend to move to booming cities where prices are already rising, I assume the reverse-causation biases upward making this a ceiling estimate.

Code, including to download the data from Census, is here; modeling choice details in the footnotes.1 Model 1 finds the standard result that, controlling for income, a 1% increase in immigrants correlates with a 1% increase in rents for 2010-2019. Model 2 finds the same number for 2019-2024. This is a quick glance, but I’m not seeing any kind of break that would cause this number to have skyrocketed in the post-pandemic period. Note that in 2019-2024, both the constant and the effect of incomes increase.

I want to make sure that this denominator issue really matters. Model 3 uses the (log) percent increase in immigration rather than increase divided by total population. Immigration is still significant but is now a much lower value, which makes sense because it’s a much smaller denominator. This reminds us that it is important to check how people describe the value: if anyone differs from total population in the denominator we know that’s off.

Though it’s a heated debate, I haven’t seen this mistake in the broader discourse. With one exception. Stephen Miran, in his first FOMC speech, says:

”Work by Albert Saiz finds an elasticity of rents with respect to the number of renters of about 1, identified from a large quasi-random immigration shock. Net immigration averaged roughly 1 million per year in the decade leading up to the pandemic. Given that roughly 100 million Americans rent, net zero immigration going forward would imply 1 point lower rent inflation per year.”

I was going to leave figuring out the error here as an exercise to the reader but, as ChatGPT 5.1 pointed out to me while fact-checking this post, the study author was forced to jump into this. Imagine being Saiz himself and having to explain to reporters that someone sitting on the Fed messed up this elasticity question from research you published eighteen years earlier.2 Here’s Saiz speaking with Ann Saphir from Reuters, confirming our analysis above (my bold). Reuters:

Miran, in his debut speech as the Fed’s newest policymaker, latched onto that idea but didn’t follow Saiz’s formula - choosing an estimate of the national renter population of about 100 million people as his denominator rather than the far-larger total U.S. population of 340 million. The result was an imputed impact on inflation about three times larger than using Saiz’s approach.

“If you did the calculation using the right magnitudes, you get 1 divided by 340 million - that’s about 0.29 percent a year,” Saiz said in an interview. Given that the share of housing in the consumer price index is about one third, the overall impact on consumer inflation would be at most 0.1 percentage points, he said. “Obviously population growth does impact the price of housing, but the magnitude isn’t big enough to justify major changes in monetary policy.”

Additional

But there’s more. Three additional notes:

1. This administration is unlikely to reduce the immigrant population by 1 million people. There’s an interesting debate on how much of the 2025 immigration slowdown is due to Trump’s immigration actions, rather than steps taken in 2024 and Trump’s collapsing labor market in 2025. But the overall immigration level is unlikely to drop.

2. Over time, supply adjusts. But the loss of workers is immediate. Deporting people means losing construction crews, plumbers, and maintenance staff, which directly constrains housing supply. If the price of construction increases, it’ll more than offset these nonexistent savings for renters. Worse, reporting suggests deportation quotas are being met by diverting federal agents away from serious criminal investigations, creating social costs that go far beyond housing. So the more you try to scale up deportations, the bigger the longer-term costs to everyone.

3. Saiz (2007) and other papers study immigration as a whole. But it’s also understood in the field that the undocumented population is different from the rest of immigrants. They are more geographically concentrated: Pew Research Center finds that “In 2023, the top six states were home to 56% of the nation’s unauthorized immigrants.” And they may put less pressure on rental markets, since they have less access to the formal legal and credit systems for renting. Huang, Li, and Xu (2025) find no state-level relationship between unauthorized immigrants and rental prices.

Immigration isn’t the reason housing is expensive. We have a strong empirical consensus on what drove skyrocketing rental and housing-price increases in 2021-2022. As Riordan Frost summarized, demand was propelled by pent-up household formation among native-born millennials, pandemic-era demand for more housing and larger homes (especially for remote work), historically low mortgage rates, and it ran straight into inadequate and sluggish supply. We have enough stories about what caused a rise in rents, without needing claims that do not hold up to basic scrutiny.

Over the summer I argued that the independence of the Federal Reserve allows officials to have frank discussions of the data, where the White House is always playing some communications role. The Fed can have a serious discussion on whether the labor market is slowing and we’re losing jobs, while the White House keeps on talking about how well everything is going. The Fed doesn’t get subsumed into comms talking points.

Another thing I just realized this past week was that genuine outsider takes on the FOMC can be examined as such, without having to worry that they are carrying water for the White House. The Fed can be subject to groupthink, so having a few people to shake up the consensus can be welcome, so long as they are independent. If they are not, it’s hard to determine what’s good faith pressure on conventional wisdom, and what’s ad hoc justification for the President’s priorities.

Is r* at zero percent?

Take Stephen Miran, on leave from his job as President Trump’s chief economist at the Council of Economic Advisers to join the FOMC. Many have worried about the obvious conflict of interest in his simultaneously holding one of the most senior White House roles while voting on the committee. And he just gave a speech, Nonmonetary Forces and Appropriate Monetary Policy, arguing for significant rate cuts based on the idea that President Trump’s policies have quickly reduced the neutral rate of interest (hence r*) to zero percent. Miran:

I get a new real r* that is 1 to 1.2 percentage points lower, or near zero. [...] Leaving short-term interest rates roughly 2 percentage points too tight risks unnecessary layoffs and higher unemployment.

It’s worth reading and discussing. As Claudia Sahm notes, there is nothing in the aggregate data to suggest that rates are that restrictive. Just so you don’t think this is all lib stuff, Wall Street isn’t buying it. JPMorgan’s Michael Feroli wrote “We find some of his arguments questionable, others incomplete and almost none persuasive,” and, as Molly Smith reports at Bloomberg, that’s a pretty consensus take.

But there’s a bigger problem that goes to central bank independence. There’s no polite way to say this so let’s just do it: it is difficult for all of us to determine if this speech is his genuine outside-the-box thought, or an attempt to shoehorn in what the President (his boss, whom he still serves at the pleasure of even while on leave) wants to see. Without true independence, this worry will always be with us.

Thought Exercise - Do This Backwards

To show the conflict, I want to motivate this with a quick thought exercise. Let’s say you work for the President, like Miran does, and have been ordered to come up with a justification for what Donald Trump wants. Over the summer President Trump told Maria Bartiromo in an interview, in a response to a question about the federal debt, that “when you get somebody into the Fed who’s going to be able to lower the rates, we should be at 1% or 2%.” How could you justify a 2 percent rate on solid economic grounds?

Now you can’t say the obvious and base the case for rate cuts on a weakening economy. Miran doesn’t mention the slowing labor market at all in his speech. Whether it’s a “somewhat softer labor market” (Powell) or “a labor market on the edge” (Waller), the new debate for cuts is predicated on a bad labor market. However you just saw someone get ended simply for showing that job growth was slowing, like a movie where the villain just knocks off advisers for saying factual things. So you aren’t going to say that.

There are other constraints. You sensibly don’t want to say inflation is below target, that doesn’t seem credible. And you don’t want to propose a brand new framework, explicitly declaring fiscal dominance, because you don’t want to freak people out. But you also need some math, ideally in the Taylor Rule framework.

The only thing to do to meet the President’s demand is to argue about r*. It’s the only free variable. Specifically, with some algebra in Figure 1 above, the only thing you can do is to find a way to argue it is zero percent. Trying to discuss an unobservable and theoretically debated idea like r* is going to involve a lot of hand-waving and balanced guesswork. But that this conclusion so naturally fits what the administration wants should give us pause.

Reversing Course

Now, of course, people with outside-the-consensus arguments tend to be consistent about them. So has Miran been arguing r* was quite low for a while? No. Many (e.g. Jonathan Levin) are flagging that Miran is contradicting his own previous arguments, as he had spent 2024 arguing that r* had increased significantly. See, for instance, his March 2024 Barron’s article, The Fed Is Facing a Changed World. The Case Against Cuts, with Sander Gerber.

There are five reasons he gives for why r* had been increasing. We’ll discuss two, immigration and the budget, in a minute. But three are quite obviously still with us and have likely accelerated. It’s worth quoting at length, my bold:

[First] Household debt fell from about 100% to 75% of gross domestic product between 2010 and 2019, but has been stable since. [Second] Wars and the proliferation of sanctions and tariffs incentivize firms to invest in supply-chain resilience over efficiency, which requires capital expenditures and boosts neutral. [Third] Finally, while software investment last decade was capital-light and suppressed rates, the 2020s seem to be more demanding on hardware for national security purposes and new AI technologies.

All three are still in play and have accelerated. Household debt has continued to fall in 2025, trade wars are far more vicious than one would have predicted in 2024, and the economy is running even more on capital-intensive AI investments.

And to back up, Miran wasn’t alone in arguing this in 2024. The FOMC and financial markets recently moved their estimate of r*. This is a whole other giant discussion, but in the same way people argued r* was down in the 2010s, they argue it is creeping back up now. But those are debates about big ongoing forces, like an aging population or productivity growth. Will those old secular stagnation forces still drive fundamentals (e.g. Minneapolis Fed 2024), or are they already burning out (SF Fed, 2025)? It is not a debate centered on administration talking points about The One Big Beautiful Bill.

The Actual Macro Impact of Deportations

But Miran does bring up immigration and budgets as reasons to drop r*. Miran makes two arguments about why a slowdown in immigration and deportations should reduce r*.

• “it is plausible to me that 2 million illegal immigrants will have exited the country by year-end,” and the lower population growth slows r*.

• Following Saiz (2003), a case study of the Mariel boatlift, Miran argues that removing immigrants will lower rents, a disinflationary impulse capable of getting inflation back to target.

The first is likely an incorrect overstatement based on best estimates, a misreading of the BLS data, see this important general post from Jed Kolko and this thread by Stan Veuger.

The second misses a whole literature that has evolved far later than Saiz (2003) on immigration and inflation. First, in a broad level, higher immigration is associated with lower price level for services, especially immigration-intensive ones (e.g. Cortes 2008). Given the strong remaining impulse in non-housing services this can’t simply be waived away.

This channel is particularly important when it comes to rentals and housing construction, as immigrants disproportionately work in construction. Using the rollout of ICE’s Secure Communities [SC] program, which deported 300,000 people, as an instrument, Howard, Wang, and Zhang 2024 find:

(i) when undocumented workers are deported, domestic labor only partially fills vacated construction jobs; and an apparent complementarity between immigrant and domestic labor leads to net job losses for US-born workers, (ii) residential construction output is highly sensitive to these declines in labor supply, (iii) the resultant reduction in homebuilding leads to higher home prices, with demand-driven declines in prices being small and highly transient. [...] “Three years after SC rollout, the average county of approximately 500,000 residents has foregone the equivalent of an entire year’s worth of additional residential construction.”

As decades of complementarities of immigration research would predict, other studies found SC reducing immigrant workers “also decreased the employment and hourly wages of US-born individuals” (East et al 2023), something relevant to what’s coming now. Other studies, like that of the Mexico Peso Crisis (Monras, 2020), bear this out.

More generally, the idea that the macroeconomic effect of deporting 2 million people will only show up in lower rents strikes me as implausible, especially after a period with big labor market convulsions. Reasonable estimates of deportations have a drop of 0.3 to 0.4 percentage points in 2025 (Edelberg, Veuger and Watson 2025), and that’s before we get to any new bottlenecks forming in specific labor-intensive industries.

I do find it funny that after a long period of having to get conservatives to think that immigrants don’t just take jobs, they also buy things and create jobs, we now have a conservative analysis where immigrants just buy things like housing but don’t actually take any jobs making things. General equilibrium is hard from both angles.

Admin Talking Points in the FOMC

I was interested in whether or not the absurd estimates coming out of Trump’s budget process or CEA would show up in the Survey of Economic Projections. (You can read Jared Bernstein on those estimates here.) They weren’t reflected in real GDP projections for the next few years, but they apparently did in the large number of cuts reflected in the dot plot, and here in the argument now.

I find this to be a conflict and a compromise of central bank independence. The President’s mid-budget review and associated CEA materials are a marketing gimmick, a forecast so outside conventional analysis we should understand it as just a justification for the White House to say they are meeting the deficit reduction metrics they promised. There’s a couple different threads here, but to give a sense of how much of an outlier the CEA is, see this graphic from CRFB:

But where to start on the idea that Trump’s tariffs, tax law, and deficit impact are driving r* to zero? The tariff revenue may not be legal, the Supreme Court will take that up in arguments that conservative legal analysts say may be difficult for the Court to simply ignore. It’s true that OBBB reduces deficits on a current policy baseline, which is relevant for the fiscal impulse, but the numbers for OBBB itself assumes that tax cuts end at certain deadlines.

The best analysis I’ve seen says that the tariffs and tax cuts roughly balance each other out through 2026, the time the FOMC needs to be thinking about; most of the deficit reduction is backloaded in OBBB. Miran assumes the loans and loan guarantees from other countries are actually going to happen at the scale promised. I assume that never happens but even if you do, we aren’t sure of what it actually looks like enough to drive monetary policy. There’s a reason financial markets don’t take the President’s own talking points for granted, and the FOMC shouldn’t either.

I enjoy reading outside-of-conventional wisdom on macroeconomics. Most financial sector people do, it’s why there’s such a flourishing newsletter network. And I enjoyed reading and reflecting on this speech. I just wish I had the confidence it was coming from a place of independence. While Miran’s boss no doubt likes the conclusion, I’m not sure everyday people will enjoy the consequences.

If real GDP matches or exceeds the 2.4% rate currently projected by the Atlanta Fed’s GDPNow model, we can expect the White House and conservative media to ramp up their pressure on Chair Powell, arguing that a strong economy warrants a rate cut. (This is, of course, the exact opposite of conventional economic logic.) Conversely, a lower-than-expected GDP print would prompt many economists to worry about potential economic weakness and advocate accelerating rate cuts at the July meeting.1

All of this makes preregistering the dovish case for a cut next week especially urgent. Luckily, Federal Reserve Governor Chris Waller did that in his recent speech The Case for Cutting Now. It opens: “My purpose this evening is to explain why I believe that the Federal Open Market Committee (FOMC) should reduce our policy rate by 25 basis points at our next meeting.”

Some critics dismissed Waller's speech as mere positioning for the Fed Chair job.2 I disagree. It’s a clear, specific, and well-articulated argument. One particular section merits attention in light of the White House seeking greater control over monetary policy (emphasis mine):

With the data in hand, estimates suggest that real GDP increased at an annual rate of about 1 percent in the first half of this year, compared with 2.8 percent in the second half of 2024. That comparison is important not only for the extent of the slowdown, which is considerable, but also because it is well below most estimates of the potential growth rate of the economy. […] The slowdown in GDP is evident in consumer spending. […]

Looking across the soft and hard data, I get a picture of a labor market on the edge.

Waller argues that inflation expectations are steady and that inflation, properly measured, is near target.3 Yet his most compelling point is the suggestion that the economy is slowing more sharply than the general consensus believes. Notably, as the White House pushes for greater control over monetary policy, this is precisely the argument they won't make.

The Republican Case for a Rate Cut

I’m not exaggerating. Aaron Rupar captured a funny moment on CNBC where House Speaker Mike Johnson argued “we should reduce interest rates, the American economy is hot.” The CNBC hosts had the awkward task of explaining that “if it’s too hot you don’t want to cut rates […] normally you think of a rip-roaring economy as not a good time to cut rates.” The hosts are right.

Or consider this case from CEA Chair Stephen Miran, who argued in my mentions yesterday that rates can come down because “The supply side economic policies of President Trump will bring inflation materially lower” and Trump’s policies “that expand the supply side of the economy by aggressive deregulation, providing tax incentives for increased labor supply and capital stock, and achieving energy abundance, are enormously disinflationary.”

This is not economic analysis but political messaging. These individual policies may be good or bad but their effect on overall nominal spending growth is uncertain. The most immediate thing Trump has done that we can see in the data is increase tariffs and decrease immigration, both of which reduce supply in the short-term and justify the Fed holding steady.

And it’s fine that it’s political messaging! As long as there is a space for actual analysis necessary to set monetary policy.

The Labor Market Now Versus 2024

Because there is a genuine debate here. How would you characterize the labor market of the first six months of Trump’s second term? Let’s compare it to the last six months of the labor market in 2024.

The unemployment rate remains about the same, but job growth has slowed noticeably. A central debate today is whether this slowdown reflects reduced immigration (which moderates job growth to maintain a steady unemployment rate) or signals actual economic weakening that the unemployment rate is missing or masking. The answer remains unclear.

Additionally, as Figure 1 above illustrates, nonfarm payroll employment revisions have become more negative, with the three-month average shifting from +16 in the second half of 2024 to -52 for 2025 so far.

I bring that up because negative revisions were a major Republican talking point against the 2024 labor market. Here’s then Senator Marco Rubio tweeting that the BLS was cooking the books for the Biden administration through subsequent negative revisions (an example of contemporary coverage of this from Courtenay Brown at Axios).

Negative revisions naturally occur at the end of recoveries and during economic turning points due to the complexities of estimating business births, deaths, and seasonal adjustments. The current uptick in negative revisions is concerning. The Trump White House is not going to make that case. But independent members of the FOMC can and do.

Yet More on Fed Independence

This highlights an understated benefit of Federal Reserve independence. No administration, and especially not the Trump administration, is eager to announce, "We need to cut rates because the labor market is weakening on our watch." Politically, that's a tough headline. Independent Fed officials, however, can cite labor market weakness to justify rate cuts in ways that communicate clearly to financial markets without political baggage.

It isn’t the most dramatic rationale for central bank independence; you can read former Chairs Yellen and Bernanke on that. But it is a feature. Losing transparency about what truly informs policy decisions is another potential cost of increased White House influence over Fed policy.

2

The case for Waller as Fed Chair is going to be awkward because he’s the best at being a Fed Governor, but Trump might want someone who will torch the place for him. It’s a no-win scenario, and when dealing with a Kobayashi Maru, you need to change the rules of the game.

Here’s a suggestion: Waller uses FACTS and LOGIC to DESTROY Democratic economists. Specifically, Waller crushed all the leading Democratic economists who debated him on the question of the slope of the Beveridge Curve in 2022. It’s some of Trump’s opponents too, be it a former President of Harvard who said the OBBB made him ashamed for the country, or a European globalist who ran the IMF. That’s a pitch that might help Waller, and I even tried to get a Ben Garrison MAGA political cartoon from AI for it:

It's a dangerous time for the Federal Reserve. Last week, President Trump’s administration began building a case to fire Fed Chair Jay Powell, as Trump surrogates and right-wing media escalated direct attacks. OMB is probing the Fed, the GOP House Judiciary Chair has signaled willingness to investigate Powell, former Trump NEC director and Fox Business host Larry Kudlow is using his show to attack Powell, the Vice-President is posting, and Trump economic validator Oren Cass has declared “something is rotten” at the Fed. Reporters are scrambling to see if Powell will be fired.

The details of monetary policy are complicated. But at this moment of potential political takeover, it’s important to see that the conservative case against Powell’s monetary policy is built on four sleight-of-hand moves, four tricks where the conservative arguments don’t survive scrutiny. If Trump follows through on removing Powell, we should know just how flimsy the case was.

1. Trump Doesn’t Want One or Two Cuts. He Wants a Dozen Cuts and a Different Fed.

As a reminder, the Fed rate is set at 4.25% to 4.5%, and it cuts and raises in 0.25% increments. If you look at the financial community, there’s an active debate about whether the Fed should cut once, maybe twice, with others saying the Fed should stay steady or even increase rates.

But President Trump isn’t calling for 1-2 rate cuts, he’s demanding 10-14. He has said that interest rates should be between 1% and 2%, and just this week called for a cut of 3 percentage points (equivalent to 12 cuts). He even sent a note to Powell stating this, shown in the press room and reprinted above.

There’s no economic case for a dozen rate cuts. But there’s a political one. Trump wants the Fed to cover for the debt explosion under his tax cuts. This is often called “fiscal dominance,” and it generally goes with weak central banks and political dysfunction.1

Even if you lower the stakes, asking the Fed to manage both the business cycle and the fiscal balance violates the Tinbergen Rule: for every n policy objectives, you need n independent tools. The Fed controls interest rates to manage the business cycle; adding a second objective, managing the debt load, requires a second independent tool, or else the Fed will fail at both.

It’s likely to fail. Large politically pressured cuts could raise the longer-term rates consumers and businesses face, if only from a political-instability premium. Or rates could become extremely accommodative into this stagflationary environment, risking more inflation. This, in turn, will have President Trump demanding more control.

An ordinary argument about whether the Fed should change rates at the margins is being used to mask the White House pressuring the Federal Reserve to take on a fundamentally different role. The White House is blowing out the fiscal deficit, and it wants the Fed to cover for it, no matter the cost to independence or long-term economic stability. This goes beyond the usual case for central bank independence, where the concern is politicians juicing short-term stimulus at the expense of longer-term inflation. This goes to whether the central bank should exist at all.

2. They Make a Cut Seem Obvious, When It’s Very Much Up for Debate.

I find it strange that the conservative movement, which is meant to have a temperament that avoids tearing down the fence of central bank independence without knowing why it was built, is willing to risk generational economic damage because the Fed might be off 0.28%.

You wouldn’t know that from their rhetoric. It’s not just the dozen rate cuts the President has called for. The Vice-President has said the Fed is “totally asleep at the wheel” and “they’re TOO LATE.” Others have argued that the Fed should have obviously already cut rates many times, and the only reason they wouldn’t have is fear of tariffs.

But is that true? Is it obvious that multiple cuts are warranted right now?

Let’s walk through the data. A straightforward way is to use a Taylor Rule, which is built on the intuition that the Federal Reserve should set rates based on a neutral rate of interest while balancing both deviations of inflation from the Fed’s target and deviations of actual output from its potential. Note the disagreement this method can produce, given you need to estimate two different unknowable values of potential output and neutral rates.

Let’s start with a standard Taylor Rule, adapted from the Fed’s own framework, along with two sets of assumptions, one normal and one conservative (temperamentally, not the actual movement):

The following bit is unfortunately technical, but it’s important. The current Fed funds effective rate is 4.33%. We take inflation from year-over-year core PCE. In the first version we take the neutral rate and natural rate of unemployment from the Fed’s FOMC SEP long-term values. This gives us a Taylor Rule of 4.05%, off by 0.28% or one cut. In the second, we take the neutral rate from the higher L-W model, as many people wonder if the Fed is too optimistic on its neutral rate, and we get 4.42%, right around where we are.

I’m on the dovish end of this spectrum, but think it’s telling of what a power grab this is to take over the institution based on 0.28% difference when many reasonable assumptions get you current rates.

But to take this out of my estimates, we can look at regional Federal Reserve banks that have their own Taylor Rule approach, and they validate this finding. Here’s the heatmap produced by the Federal Reserve Bank of Atlanta’s Taylor Rule calculator, using a slightly different formula.

The horizontal spread is the different rates depending on the output gap, and the vertical spread differs on the neutral rate. As you can see from the superimposed blue box, using the FOMC’s longer-run neutral rate, there’s a range from 3.92% to 4.57% for rates depending on the output gap you choose. And assuming the Fed’s unemployment gap for output, the red square shows that different neutral assumptions can range from 3.71% to 4.92%. Reasonable assumptions can get you from 2 cuts to 1 increase. None get you to a dozen cuts.

You can also hit the Taylor Rule approach with a battery of modeling and forecasting assumptions, as the Federal Reserve Bank of Cleveland cleverly does at their site. They use 7 rules and 3 sets of data assumptions to get and display a range of values. For 2nd quarter of 2025 the 25th to 75th range is 4.1% to 4.5%, with the median value at 4.4%. This is in line with where we are.

The Taylor Rule isn’t the end-all of policy, but it gives us a baseline range and also provides us with ways for conservatives to justify the Fed being off. The output gap could become larger than we expect. But, after the immediate withdrawal of the maximum April tariffs, it’s unclear how much unemployment will increase. Even in the immediate aftermath of Liberation Day the worry was 4.7% unemployment, in which case forecasters anticipated rapid cuts, and that has backed down dramatically. The same administration pointing out unemployment hasn’t gone up is at the same time weakening the case for cuts.

It could also be that inflation is far lower than we realize. Certainly that’s their talking point, with an example of Larry Kudlow saying “I think inflation is 1.4% at an annual rate.” But this is wishful thinking. PCE remains elevated in the mid-2% range. You can rely on CPI, which is unusually lower than PCE for the time being, but the Fed doesn’t do that. Even in recent months, core PCE inflation was 2.7 percent annualized from January to May, 2025. Stylized metrics over short time-horizon don’t give us the full inflation picture that remains above target.

Rates are within a reasonable range given the fundamentals. People can argue at the margins, they always do about monetary policy, but there’s no evidence the Fed is far outside where they should be.

3. The Same Argument Providing Restraint Now Called For Cuts Last Year

There were cuts last year in 2024. Some conservatives are arguing that this reflects political bias, that Powell cut for President Biden but won’t for Trump. As Vice-President Vance tweeted, “I’d love to hear an argument for why Powell cut rates 50 points right before an election but can’t do it now with inflation lower.”

I can give one. Let’s use simplest version of the Taylor Rule above, the first one that gave 4.05% now, and run it backwards with the same assumptions:

Here we can see that rates were far more restrictive, for much longer, in 2024 (the blue Fed rate was higher than what was implied in the red Taylor Rule), than what we have right now. Moreover, we saw the “Sahm Rule”, which is an important recession warning based on the acceleration of the unemployment rate, trigger by going above 0.5, as unemployment went from 3.5% to 4.2% in a short period, leading to worries unemployment was increasing too fast.

These are two distinct but overlapping conditions. In 2024, rates were both obviously restrictive for a long time, and something in the economy, the Sahm Rule, was worrying in a way that reasonably predicted future weakness. In 2025, rates are not obviously restrictive now, and haven’t been for a while, and there’s no indication the economy is slowing at a pace to trigger action.

The 2024 comparisons are completely off unless someone can give me something more specific.

4. The Fed is Correctly Concerned With Second-Order Tariff Impacts

Critics claim the Fed is panicking over tariffs simply because they raise prices. That’s a caricature. The actual concern is about how tariffs are affecting expectations and feeding second-round inflation pressures.

Oren Cass attacks Powell for not understanding inflation based on Powell’s House Financial Services Committee testimony in June, where he responded to a question by saying “some significant inflation will show up from tariffs. And we can't just ignore that.” Cass says that a one-time increase in prices is not “cognizable for monetary policy.” But here’s what Powell says clearly in his opening:

The effects on inflation could be short lived—reflecting a one-time shift in the price level. It is also possible that the inflationary effects could instead be more persistent. Avoiding that outcome will depend on the size of the tariff effects, on how long it takes for them to pass through fully into prices, and, ultimately, on keeping longer-term inflation expectations well anchored.

The FOMC’s obligation is to keep longer-term inflation expectations well anchored and to prevent a one-time increase in the price level from becoming an ongoing inflation problem.

This isn’t just Powell, many FOMC participants believe this. And there’s reasons to take it seriously. The University of Michigan survey’s inflation expectations for 5–10 years ahead just spiked. That didn’t happen during the 2021–22 inflation wave. It’s happening now. That’s a red flag.

Of course this could just be noise; it’s not reflected in the hard financial data. Sentiment trackers are no longer as predictive of economic activity in the post-pandemic period (Hoke, Feler, Mitchell, Chylak, 2025). But if we were to wait for this to end up in the financial markets or professional forecaster community, then it’s probably too late and you’d need a recession to get it out of the system.

I’d go further than this. Cost-push shocks can be looked through, but a tariff brings extra liabilities as a cost-push shock. They lower GDP in a way that pressures the output gap, and their “inherently stagflationary” nature complicates straightforward monetary policy analysis compared to other shocks (Auclert, Rognlie, Straub, 2025). More, that so much of the current tariffs are on intermediate inputs risk more persistent inflation in a way they wouldn’t if they just faced final goods (Cuba-Borda, Queralto, Reyes-Heroles, Scaramucci, 2025).

My weak baseline is that these won’t pan out in a substantial way. But given the unique situation, and with the labor market stable, why should the Fed completely ignore the potential risks?

The conservative attempt to take over the Federal Reserve isn’t about Powell being off by one cut. It’s about whether the Federal Reserve can continue to exist as an independent institution. Trump and his allies aren’t arguing for better policy. They’re arguing for a Fed that does what they say, because they’re running up deficits and want someone else to take the hit. If Powell is removed under these pretenses, the consequences will outlast any single rate decision.

Subscribe now

Are they? There’s excellent work being done on how this budget will impact everyday people, whether it is Groundwork Collaborative explaining how the bill will raise the cost of living or the Hamilton Project showing how Medicaid will get cut for workers with unstable hours. But I was curious about whether these cuts stand out historically.

Fortunately, Stan Dorn, Director of the Health Policy Project of UnidosUS, and someone who has been working on health care policy for decades, has done the work. He put together old CBO analyses, historical knowledge, and data analysis to be able to compare those cuts to now, both in terms of 2025 dollars per year and as a percentage of total spending.1 Here is his chart for Medicaid:

Let’s put the percentage of total Medicaid spending cuts into its own charts:

As Dorn summarizes (my bold):

The $70 billion average annual Medicaid cut under the current legislation would be more than 10 times the size of the largest previous cut, the Deficit Reduction Act of 2005’s average annual reduction of $5.8 billion in March 2025 dollars.

The current bill’s 9.6% drop in projected Medicaid spending would be roughly twice the size of the largest previous percentage drop that resulted from two sequential budget reconciliation bills added together—namely, the 5.0% Medicaid spending reduction that resulted from the Omnibus Budget Reconciliation Acts (OBRA) of 1981 and 1982. According to the Reagan Administration’s official evaluation, the latter legislation caused a 13% drop in total Medicaid coverage. 

And it’s not just Medicaid that stands out historically. Dorn and Hannah Garelick find similar extremes for SNAP:

In dollar amounts, it’s more than three times the size of any prior cut. Ronald Reagan cut food stamps by 12.6 percent and Bill Clinton, ending welfare as we knew it, cut it by 18.6 percent. This budget would cut it by 26 percent.

The idea that the largest Medicaid cut in U.S. history is being used to fund tax breaks for pass-through entities and high earners doesn’t square with Trump’s plan to realign the electorate. And it can’t. Medicaid expansion supports the working-class families conservatives want to realign into the GOP. And as Senator Hawley and others have noted, the ability of conservatives to secure this realignment depends on making peace with social insurance, and especially Medicaid. Medicaid also backstops rural hospitals, and provides care to roughly half of new births.

But when push comes to shove, Trump and conservatives are going to cut Medicaid by twice as much as President Reagan did when he formally brought libertarianism into the political mainstream and the federal budget.

Consider three major Republican tax cuts from this century: (1) the George W. Bush cuts of 2001 and 2003, (2) Trump’s TCJA of 2017, and (3) the 2025 proposed Big Beautiful Bill (BBB) tax cuts working their way through Congress. All three massively increase the deficit by lowering taxes disproportionately for the wealthiest, throwing off any kind of long-term balance of revenues and spending.

But there’s something new this time. Let’s compare distributional tables:

As you can see, previous tax cuts showed the smallest gains in after-tax income for the bottom quintile, in dollar amount and percentage, compared to much larger gains for higher-income groups. That’s not controversial. But what people would fight about is what to take away from it. Liberals argue this is regressive; conservatives counter, pointing out that everyone’s taxes went down and that this is the direct result of progressive taxation. It generates a lot of discourse.

But here are the distributional tables for the proposed BBB working its way through Congress:

Even though conservatives are still blowing out the deficit, this time they aren’t even bothering to bring everyone along. Instead, those at the bottom are outright worse off. The cuts to spending programs, especially Medicaid and SNAP, are severe. Estimates suggest at least the bottom two quintiles, 40%+ of Americans, will experience a clear loss in income. This isn't a subtle debate over which basis to judge the proportionality of tax cuts: millions will simply have less money, even as the bill adds $3.8 trillion to the deficit.

And this does not even include tariff revenues. We know that tariffs will fall disproportionately on lower-income Americans. They spend more of their money in general and more of their money on goods. Estimates vary, but we could see tariff revenue landing between $270 and $450 billion a year, so upwards of 1.5% of GDP. Tariffs at this level function effectively like a consumption tax on imported goods, but instead of doing things that benefit everyone, this broadly borne tax is going to fund very regressive tax cuts.

Between direct spending changes and tariff impacts, the bottom half of Americans will become poorer while the deficit explodes. Why would Republicans embrace this?

As has been well documented, the thin majority in the House means that there are a ton of veto points, and those who want more cuts and those who want less of an increase in the deficit are using poor and working people’s lives as bargaining chips. There are numerous conflicting justifications being given for the tariffs, from revenue to strategy to compensation for being the world’s global currency.

Lucky Ducky

But at the risk of Boomer-ing out, I’m going to suggest an important element comes from something I and all other economic blog readers remember from 2002. Back then, the Wall Street Journal published a criticism of the proposed Bush tax cuts, in an infamous op-ed piece called The Non-Taxpaying Class (my bold):

But as the Republicans construct their tax plan, there is a large and under-appreciated fact they would do well to keep in mind. Over the past decade or so, fewer and fewer Americans have been paying income taxes and still fewer have been paying a significant percentage of income in taxes. […]

Who are these lucky duckies? They are the beneficiaries of tax policies that have expanded the personal exemption and standard deduction and targeted certain voter groups by introducing a welter of tax credits for things like child care and education. When these escape hatches are figured against income, the result is either a zero liability or a liability that represents a tiny percentage of income.

“Lucky Ducky” became a meme at the time, popularized by the cartoonist Ruben Bolling. The notion that the bottom half of income earners were politically suspect because they paid no income taxes, and they needed to be paying more in taxes even as taxes came down for the top half, showed up from time to time, but was rarely spoken out loud outside conservative think tanks and private donor meetings.1 I expected it to be buried in the Trump years, with his right-wing populism and appeals to working people. But apparently not.

This tax and tariff plan is what you’d propose if your goal was to wage war on the lucky duckies. The law doesn’t do the politically radioactive thing that the Wall Street Journal wanted, which is directly raising income taxes on low earners while cutting them for high earners. Instead, it opts for their second-best policies of a stealth consumption tax through tariffs and cuts to essential social and income insurance. But the effect is the same. It causes immediate suffering while making the big long-term challenges we face - inequality, security, revenues - all worse.