back
248 comments
Not the Atlanta Fed, the Atlanta Fed’s GDPNow model [1].

It’s pretty good [2]. And we are seeing a flattening of 2024’s aggregate wage growth of 4.15%. But the difference in wages is like 0.42% which is indistinguishable from noise. (GDPNow predicted a phantom recession in 2022.)

In this case, the model is probably recording a surge in January imports without “an offsetting increase in inventories,” as “that is a lagging indicator” [3].

[1] https://www.atlantafed.org/-/media/documents/research/public...

[2] https://caia.org/blog/2024/08/15/increased-accuracy-gdp-mode...

[3] https://www.calculatedriskblog.com/2025/03/a-comment-on-gdpn...

> GDPNow is an excellent tracking model, however, the January surge in imports - especially for gold - caused the model to move negative. As the Atlanta Fed noted: "the contribution of net exports to first-quarter real GDP growth fell from -0.41 percentage points to -3.70 percentage points".

Anyone know what's up with the surge of gold imports? We don't have gold tariffs, right? Are people hedging against the dollar's instability?

These models all seem like they are lagging. Are there any models that are more leading?
That's important because it's a pure mathematical model which may be wrong as it cannot take in all influences that drive it.
All due to exports collapse (scroll to the bottom):

https://www.atlantafed.org/cqer/research/gdpnow#Tab3

May still hold true, the 2s10s has recently deinverted, which is typically the last stage pre-recession.

After decades of seeing various administrations get blamed for economic downturns that were set up by conditions well preceding their terms, I guess it will be interesting to see what happens now that we're finally heading towards one that actually was caused by a discrete, identifiable policy, all within a couple of months.
Looking at that graph, though, while exports collapse is certainly the biggest factor, it's not the only one. That is, looking at the March 3rd entry, "Residential Investment" is now negative, and if I'm reading it correctly (I hate it when colors on a graph are too similar) consumer spending is now zero when it was recently quite positive.
> May still hold true, the 2s10s has recently deinverted, which is typically the last stage pre-recession.

Perhaps worth noting that Harvey's original paper was about 3-month and 10-year Treasuries, but 2-year is now used by some folks:

> To determine whether the yield curve is inverted, it is a common practice to compare the yield on the 10-year U.S. Treasury bond to either a 2-year Treasury note or a 3-month Treasury bill. If the 10-year yield is less than the 2-year or 3-month yield, the curve is inverted.[4][5][6][7]

* https://en.wikipedia.org/wiki/Inverted_yield_curve

* https://people.duke.edu/~charvey/Term_structure/Harvey.pdf

> Harvey: Flat or inverted yield curves are historically associated with slow economic growth or recessions. I did notice that the yield curve inversion of the 10-year Treasury bond and the 3-month Treasury bill yield curve preceded all four recession since the 1960s. My dissertation committee at the University of Chicago was concerned that this might be a fluke given there were only four recessions. Frankly, I was nervous too because it is well known in science that strong findings become weaker after publication -- or sometimes vanish. However, in my case, this did not happen. Yield curve inversions preceded each of the next three recessions, including the important global financial crisis.

* https://www.linkedin.com/pulse/yield-curve-inversion-explain...

> May still hold true, the 2s10s has recently deinverted, which is typically the last stage pre-recession.

For those of us not in the know, could you give some more detail? What is 2s10s, what does it mean that it's deinverted, and why is that typically the last stage pre-recession?

Easy fix for that! Just change the definition of GDP by excluding government spending: https://apnews.com/article/trump-gdp-economy-government-spen...
There are easier ways. I'll write a poem and sell it to you for $50 trillion and you draw a picture and I'll buy it for $50 trillion from you. We could get this done today and bump GDP by $100 trillion.
Not relevant. The effect is almost entirely from net exports collapsing.
Does anybody know how that would work? I know of the 3 ways of measuring GDP in econ basics (All income, All expenditure, All value add). I assume they only use one method and just pull out government? I assume it's expenditure method, but what would it mean for income method. Ignore incomes of gov employees? But then what about contractors. At what point do we start to view the number shifting as untrustworthy like China and Russia.
This surprises nobody. People import goods before tariffs because the price is lower, and that money going to trade isn’t going to GDP. Greater economic uncertainty and higher savings rates. Difficult labor market.

The only question is the depth and the duration of the dip.

It's interesting that the Great Depression was triggered by tariffs.
> People import goods before tariffs because the price is lower, and that money going to trade isn’t going to GDP

It does, in inventories. But 34 have 77% reporting on inventories and 85% on imports [1]. To the extent this model is saying something, it’s to watch inventories build and deplete.

[1] https://www.ismworld.org/supply-management-news-and-reports/...

Hire clowns, get a circus.
it's more like the circus hired psychotic villains because they thought it would be funny. The state of things is a direct result of a functioning democracy.
Are his voters at least in marginal numbers realizing whats happening and possibly about to happen, to US itself and globally? Or its all fine and going on as expected for them.

We get 0 on-the-ground info here in Europe on such topics (at least not in mainstream media), its all bombastic shit like he still runs his reality show and not directly affecting lives of hundreds of millions with each tweet/outburst. I'd expect massive difference in above between cities like SF, LA, NYK compared to midwest or bible south for example.

I believe this is due to a surge in imports in response to looming tariffs - supply chains are trying to stockpile before they hit. I am skeptical that this stockpiling is significantly displacing real investment.

The way that the figures are calculated views imports as a negative factor to GDP (because NET exports is an input to the model). Please correct me if I am wrong.

In any event, view the headline with suspicion.

Imports are neutral to GDP. The reason they're subtracted in the standard formula for GDP is that makes it easier to count.

GDP = Consumption + Investment + Government Spending + Exports - Imports

The reason that imports are subtracted is because Consumption, Government Spending and Exports all have a domestic and imported component. So instead you could have GDP = (Domestically produced consumption) + Investment + (Government spending on domestic products) + (Domestically produced Exports) and not subtract imports.

But that's a lot harder to measure than measuring totals and subtracting imports.

> this is due to a surge in imports in response to looming tariffs

“GDPNow is an excellent tracking model, however, the January surge in imports - especially for gold - caused the model to move negative. As the Atlanta Fed noted: ‘the contribution of net exports to first-quarter real GDP growth fell from -0.41 percentage points to -3.70 percentage points’.

Usually there would be an offsetting increase in inventories, but that is a lagging indicator. This is a short-term distortion and will balance out over the next month or so. I don't expect negative GDP in Q1.”

[1] https://www.calculatedriskblog.com/2025/03/a-comment-on-gdpn...

I think, this named deferred demand - people known before about new tariffs and buy really need things before new policy taken effect, so now will be some time without new buys.

It will normalize in few years, may be for some markets in few months, and for some will need decades, from EU practice.

This looks like the work of a team of stable geniuses.
Just wondering: If we magically made the federal government 20% more efficient would that decrease GDP substantially?
The question is underparameterized.

20% more efficient happens when you accomplish 20% more work with the same amount of money.

It also happens when you accomplish the same amount of work with 17% less money.

Right now it looks like neither option is happening.

Not necessarily. Hypothetically speaking, if the federal government becomes more efficient, i.e., it produces the same services for less money, the FED could stimulate the economy with low interest rates, so that the private market employs the people released from the federal government and we will have higher production and higher GDP.

Practically, speaking if you just fire a bunch of federal employees and close departments randomly, you are not making the government more efficient you are just making it less productive. Then if, at the same time, you put in a bunch of sudden arbitrary tariffs that cause inflation across the board, then you tie the hands of the FED and the FED cannot lower rates to preserve employment. So in that case, yes GDP will decrease.

In the short term, yes. Long-term, no (it would have the opposite effect). Productivity gains accrue compounded over time, spending cuts show up immediately.

Government spending is included as part of GDP, so a 20% reduction in spending would have an immediate effect on this number.

That's why a lot of economists think GDP is a bad metric, since a debt-fueled spending spree (like the US government loves to do) shows up as GDP growth which makes it hard to compare GDP numbers since nobody ever adjusts for debt-to-GDP ratios.

If you look at this chart, you can see what I'm talking about. For the sake of argument let's say both Sweden and the UK had the same GDP growth rates and similar levels of government spending. Sweden would be the more productive economy because they'd be doing that with far less debt: https://en.wikipedia.org/wiki/Debt-to-GDP_ratio#/media/File:...

It's funny that this question isn't answered anywhere in the news!

The question is a bit vague, let's split it up into different options:

1. The government does exactly the same work as it does today, but with 20% fewer employees. The US spends $270B on civilian employees. So 20% of that is $54B. US GDP is 27.72T. 54B is totally irrelevant.

2. The government spends 20% less on everything it can. Most of what the government spends on cannot be cut, it's fixed. Social security, medicare, defense, healthcare, veterans benefits, interest in existing loans, etc. https://fiscaldata.treasury.gov/americas-finance-guide/feder... If you take away the parts that cannot be cut, you're left with discretionary spending. https://usafacts.org/articles/how-much-of-the-federal-budget... That's about $300B (because we need to leave out defense and things like veteran's benefits and income security which are discretionary but must be paid). What remains is education, parks, research, etc. If we cut 20% of that about $300B which is left over, we're still talking $60B.

So no, a magically 20% more efficient federal government won't do anything to GDP, because it won't do anything for government spending. Pretty much all government spending is in direct payments to help people and in defense. That's why DOGE and others cannot possibly make any difference at the large scale, they can only hurt people while providing nothing meaningful to the country.

Definitely, no. After Laffer, in normal open market, all government spending (equal to sum of all taxes), should be less than 30% (less then optimum, so will have gap for emergency cases).

Why it is more effective to limit taxes, because by definition, private business is most effective form of production, and gov't entities are least effective form, and with tiny taxes people will have more money to reinvest into economy grow (via investments into existing and new private businesses), which is definitely more effective than spend money by government or just use government to redistribute money to people.

So in ideal case, fed gov should be zero size, and only in extreme cases appear and save world, then immediately disappear and return zero taxes.

https://en.wikipedia.org/wiki/Laffer_curve

PS what's also funny, usually money redistribution bureaucracy spend more money to their functioning than distribute to people, even in cases of very large systems with millions participants.

PPS yes, exist number of cases, where concentrated spending via government is beneficial, because of size factor. But problem is, many of such cases are only seen post factum, and it is not easy to predict, if something is such big thing.

Examples of cases benefit from fed size, are: railway from west to east on early 20th century; nuclear power ~80 years ago; space scale rockets in 1960s. - Now all these cases will be more effective handled at private business.

For now we have perspective cases of AI and quantum computers, but at the moment we don't know, which approaches will deliver value and which will just gather low hanging fruits.

It depends on what the broader economy is doing

Is government spending crowding out private investment? (Is the private sector competing with the government for employees?). If so then GDP should increase

Is unemployment high? Then probably it would hurt GDP, but that depends on interest rates. If interest rates are very low, then there's very little cost to having excess employees being paid by the government: indeed presumably you can find something else useful for them to do. If interest rates are high, then the government is paying a heavy rate to subsidize these now-redundant employees. That's stagflation, which is essentially economic hard mode.

For example, what a waste it was during the Great Recession, when interest rates were basically zero, to have people unemployed instead of doing something useful, like maintaining or upgrading infrastructure. Alas!

> If we magically made the federal government 20% more efficient would that decrease GDP substantially?

Not in the long run. (Particularly not if being done in the current deficit-increasing way.)

I don't know. If we had a time machine, we could go back to the 1950s and defund government-sponsored education and research programs instead of increasing spending on them, and see how things turn out, GDP-wise.

Trouble is, that could easily do enough damage to prevent the invention of our time machine in the far future, or at least seriously delay it. Maybe not such a good experiment after all.

You ain't seen nothing yet.

In a few short weeks, Trump imploded the fundamental US brand from good to um, pretty much pure evil. We are untrustworthy backstabbers. Rightly so, people hate us now; they are literally burning the American flag all over the world. USA products and services are toxic items. Not that we make much money off the travel industry, but you'd have to be a complete idiot to vacation in (or really even travel to) the US now.

It's going to get really ugly; I don't think people get it yet.

Trump wants lower interest rates. The road to getting there will inevitably involve some pessimistic forecasts. That benefits Trump too because he can say "they all said the economy was going to be bad, but look what I accomplished" when the higher rates heat things up.
Hey e/acc subset who were pro-Trump because of how much he was recruiting from Silicon Valley, is this the whole massively revitalized economy we were supposed to be getting?
I'm seeing a lot of "one easy answer" type posts for this data, and I thought I'd contribute my own hypothesis as to why an economic downturn of some sort seems inevitable. Bearing in mind this is grossly oversimplified, and an eensy bit hostile in tone, and written from one of the internet's multitudes of "armchair theorists", so take it all with a healthy dose of skepticism.

---

On the one hand, you've got a tech industry so addicted to ZIRP that they've actively been trying to engineer a recession since COVID's interest rate hikes. They don't want to adapt to a new norm of low interest rates, they want zero interest rates so they can take out all the debt they need to justify share buybacks, AI and Quantum investments, and further industry consolidation around infinite services rather than tangible products. To those types of leaders, the pain is the point, and a means to their end of depressing wages and fueling more artificial growth.

That said, they're ultimately a drop in the current bucket. Once the current President got elected, businesses immediately began bulk-importing ahead of tariffs to preserve margins, in the hopes they could lobby to get them dropped again like last time. That is not happening, partly because one of their own is President de facto if not President de jure, and this man is rampaging like a petulant toddler through the ranks of the Civil Service. Laying off and outright firing a bunch of workers - surprising absolutely nobody with a basic grasp of economics - has knock-on effects on the larger economy. Those people have bills to pay, and often took lower-paying Civil Service gigs for the stability of the role - something the economy adapted to as dependable and reliable income streams. That image has been irreparably shattered, and Civil Servants are viewed as the same unstable debtor as private sector workers, surviving not even admin-to-admin anymore. This means employers are nervous about their hiring practices, eliminating open roles (the "Job Market Freeze" as it's being called) and not backfilling others, with a prime example being the tech sector refusing to hire developers and claiming AI will replace them.

Anyway, so we have tariffs squeezing already-declining consumer demand as COVID surpluses have dried up, a demolished civil servant base (the Federal Government is the single largest employer in the country, and possibly the Earth inclusive of its multitude of other, oft-excluded branches), and an unstable Executive Branch more focused on agendas of hatred and vengeance than sound economic policies.

That still only scratches the surface.

Compounding the above are asset prices and inflation, both of which I'm going to grossly oversimplify and lump into the "infinite growth" problem category. The only thing holding back the human species from stripmining the entire planet is policy, and that policy has been globally manipulated and hollowed out to funnel cash upward from the working class worldwide. It's not an American problem, and it's not a Capitalism problem (Communist and Feudalist countries have had the exact same issue). By funneling more wealth into fewer hands, there's less avenues for production of goods and services other than "rental" markets (like streaming, or XaaS) - a market segment that's been infamously toxic with bad returns in the long run relative to other investments, though always buoyed by better-than-expected returns in the short-to-mid terms as investors seek market capture through "disruption". Paradoxically, giving consumers the ability to own actually increases economic output to a degree, especially if products are well-made and repairable, by propping up local craftspersons and small businesses; perpetual "rental" services focus that capital into very few hands, and deter such knock-on economic expansion, which ultimately slows growth.

And that growth is the problem every country faces right now. The past century (post-WW2 in particular) has been strongly focused on growth at the expense of all else, and that was never sustainable in the long run. Until and unless we actually have (practically) infinite resource extraction, refinement, and re-utilization, infinite growth is functionally impossible - and even then, growth would be limited to the sum total of the value of resources effectively exploited in a closed-loop supply chain. When growth halts or slows, we get recessions as the investor class, greed impossible to satiate even in the best of times, withdraws from markets until such time that new industry or technological innovation creates the illusion of infinite growth yet again. With population growth stagnating (due to wealth inequality - go compare birth rates to wealth inequality ratios historically to see how neatly those two inverse one another), this also threatens systems built with the presumption of infinite growth forever - like government welfare programs based on low taxes and high population/wage growth, rather than higher taxes and fixed benefits.

So now we circle all the way back to the beginning, and my hypothesis on the potential recession:

* Consumer sentiment is low because people keep getting laid off, wages remain flat, RTO mandates eat away at time and money savings the pandemic created, and asset prices remain unaffordably high for the 90%

* Business confidence is low because higher tariffs disproportionately impact American businesses who import most goods, and a dysfunctional Federal Government more focused on tantrums, authoritarianism, and identity politics than effective governance weakens that confidence further since lobbying is no longer a guarantee of outcome

* International confidence in American institutions (government and private alike) is decreasing as a result of highly-public meltdowns of both the President de jure and the President de facto, forcing many developed economies to reconsider their business and political relationships with the world's largest economy.

* A hollowed out economic core that focused exclusively on services (which can and are continuously outsourced) in lieu of diversity of industry, making it incredibly vulnerable to outside market and political forces

* A capital class that believes it can escape any harm by simply relocating elsewhere

And that's my position. I'm definitely oversimplifying complex issues for the sake of brevity (economic diversification, asset valuations, the housing crisis, etc), but I think my core position is pretty sturdy.

Uh-oh.

I wonder how all the Ayn Rand acolytes in power today will react to the bad news.

Their approach to macroeconomics can be summarized as: first, they always blame bureaucrats for bad economic conditions; and second, they believe that if they can get the bureaucrats out of the way, they need only to inspire people, to get everyone's animal spirits roaring, so the economy can grow, because they believe "growth is a choice." I'm exaggerating, but only slightly.

Well, all these Rand devotees are successfully destroying the federal bureaucracy, or at least preventing it from functioning, as they have always dreamed, and they're constantly chanting about making things great again, to inspire people. So far, they're getting everything they want, and yet... here we are, seemingly headed for a recession.

Could it be they don't understand things as well as they think they do? It sure could. Alas, they won't want to change their minds. As J. K. Galbraith wrote, "faced with the choice between changing one's mind and proving that there is no need to do so, almost everyone gets busy on the proof."

If we indeed have a recession, my best guess is that we'll see them repeatedly calling for more optimism, or something like that.

You know things are bad when you have negative GDP!