Guest Post by Peter Reagan
Today, I’m going to start with some good news.
(I can’t always do that, so it’s nice to be able to sprinkle a little sunshine your way.)
The U.S. economy grew at a 3.4% annual rate in the third quarter of 2026, according to the Bureau of Economic Analysis’s advance estimate. That’s a substantial acceleration from the 2.2% rate in the second quarter. The long-term average GDP growth rate is 3%, by the way, adjusted for inflation.
The Associated Press noted that economists had forecast 3% growth on average, so the report was quite a nice surprise!
Of course, GDP has its limitations. I’ve written before about the difference between an economy that looks healthy in the aggregate and one that actually feels healthy day-to-day.
Still, 3.4% real economic growth is good news.
So, I’m sure you’re wondering: Where did that growth come from?
Let’s dig a little deeper and find out…
Two major forces helped drive the third-quarter economic expansion.
First, consumer spending rose at a 3.8% annual rate. Remember, consumer spending makes up the vast majority of economic activity, about two-thirds on average.
Second, business investment surged – particularly in areas associated with artificial intelligence.
According to the Bureau of Economic Analysis, business investment rose 12.7% during the quarter overall. Here are some of the biggest areas of investment:
Those are enormous numbers!
Now, I know what you’re thinking… But no, the increase in consumer spending can’t be dismissed as inflation. These are real, inflation-adjusted GDP figures. Americans genuinely bought more goods and services.
You might think that means folks are suddenly feeling flush with cash. However, the latest Bureau of Economic Analysis report on household finances paints a more complicated picture.
In September:
Meanwhile, consumer prices were still 3.7% higher than a year earlier in September, according to the Bureau of Labor Statistics.
So yes, consumers are spending. But income isn’t keeping pace with that spending – and so families are saving a smaller portion of their income. The personal savings rate fell to 3.8%, well below the long-term average of 8.4%. Especially when you remember the rule of thumb for saving for retirement is to set aside at least 10% of your income…
The result? A much less cheerful picture than the GDP headline alone might suggest.
So let’s take a look at the today’s major engine of growth:Artificial intelligence.
There’s an easy trap to fall into when discussing the AI boom.
We see hundreds of billions of dollars pouring into data centers, chips, software, electrical infrastructure and research, and we ask:
But does all that spending really count as economic growth?
Yes! It absolutely does.
Building a data center means buying or leasing land, hiring construction workers, buying or renting equipment, buying and installing computer racks (and everything else a building needs: Fire suppression systems, bathroom fixtures, electrical wiring, windows…) That’s real work.
Developing software counts as economic activity. Manufacturing computers and servers counts as economic output. Capital investment is one of the components of GDP. So when businesses spend huge sums building AI datacenters, same as when they build a factory or open a mine, that genuinely increases measured economic output.
Some people ask whether AI spending really “counts” as economic activity. That’s absurd! Of course it is!
I think the much more important question is whether all that investment eventually produces a payoff.
There’s a major difference.
Imagine a farmer buying a new tractor…
Buying the tractor contributes to economic activity today. But the farmer didn’t buy it just to own an expensive machine. He bought it because he expects the tractor to help him produce more, with less labor, for years afterward. (That’s why we call it “capital investment,” by the way, to separate it from “consumption.”) If the numbers work out as predicted, the farmer is spending a lot of money now in exchange for a more profitable future.
AI infrastructure works the same way.
The spending is happening now.
The promised productivity gains are supposed to come later.
And that handoff is where things get interesting…
Goldman Sachs Research has been asking exactly this question. In May, its researchers noted that while consumer adoption of AI had been spectacular, successful adoption by businesses would ultimately decide whether the economics of the sector work out.
Goldman also observed that, while a handful of semiconductor companies had made substantial revenue and profits from the buildout, most companies in the AI ecosystem had yet to make a profit. That’s concerning. But it’s not unheard of for start-up companies in cutting-edge industries to run at a loss for years. So long as they make a profit eventually.
Just one month later, Goldman Sachs Research’s Jim Covello put the issue even more plainly: Companies make investments because eventually those investments will generate returns. He went on to explain that the big spenders, the AI model developers and the companies building massive data centers, all of them had yet to demonstrate profits proportionate to the enormous amount they’ve spent.
Let’s go back to our farmer, for example. A mid-range combine harvester costs about $500,000. So, at a minimum, he needs to see $50,000 extra profit over the next 10 years to break even on the transaction.
If he goes for a top-of-the-line, GPS-guided self-driving machine? That’ll run him $1.2 million. Say it’s more efficient, and it makes him an extra $60,000 in annual profits. Now, though, he’s looking at a 20-year break-even date.
My point is simple: Spending more money doesn’t necessarily improve return on investment.
So does this mean AI is a dud? Not at all! In fact, Goldman Sachs Chief Economist Jan Hatzius still expects AI to generate huge productivity improvements. Over the longer term.
Those benefits are likely to arrive more gradually than the more enthusiastic forecasts claim.
And this creates a timing problem…
Now, Goldman Sachs is not predicting that AI investment suddenly collapses next year.
In fact, its August research estimated U.S. AI investment at about $581 billion in 2026. They projected that AI investment could grow from 1.8% of GDP this year to 2.5% in 2027 and 2.8% in 2028.
So the AI buildout may continue getting larger. What may peak sooner is its contribution to economic growth.
Hatzius recently estimated that the growth impulse from the AI investment boom could crest in late 2026 or early 2027.
That sounds contradictory at first. But think about it this way.
Suppose you increase your spending from $100 to $200.
That’s a huge increase! It’s doubled, or rose 100%.
Then the next year you go from $200 to $220. You’re spending more than ever – but the rate of increase is just 10%.
That’s what we’re talking about with AI infrastructure. Hundreds of billions of dollars can continue flowing into data centers and computing equipment even while that spending adds less additional momentum to GDP than it did before.
That creates what I think is the central question for the AI economy:
Will productivity gains arrive quickly enough to replace the boost from the construction and equipment spending boom?
I don’t know. And frankly, nobody does. We aren’t sure whether those productivity gains will arrive at all.
And that’s a real concern for me…
New technologies frequently require enormous upfront investment before the benefits spread throughout society.
Railroads had to be laid before they could transform commerce. The electrical grid had to be built before factories could reorganize themselves around electric motors. The internet required fiber-optic cables, servers and data centers years before businesses figured out all the ways they could use it.
Sometimes that process produces extraordinary gains.
Often, investors build too much infrastructure too quickly.
And sometimes both happen.
The technology changes the world – while many of the people who financed the first wave still lose money.
For regular families, though, I think there’s a simpler lesson: We shouldn’t confuse optimism about technology with certainty about the economy.
Because AI isn’t the only enormous economic experiment underway.
Even if AI eventually delivers the massive productivity gains its boosters predict, the federal government still faces a difficult outlook.
The Congressional Budget Office’s 2026 baseline projected federal spending of $7.4 trillion.
Revenues were projected at $5.6 trillion.
That leaves a projected deficit of roughly $1.9 trillion.
Here’s the problem: The economy grows at about 3% a year, on average. But the accumulation of federal government deficits, the “national debt,” has been growing at 7.8% per year.
There are a number of ways to describe a person whose spending grows more than twice as fast as their income…
To be clear, spending might change. Economic growth might surprise us. And a genuine productivity revolution could certainly improve the situation by making the economy larger. (Remember, when the economy grows, so does tax collection.)
But faster growth doesn’t automatically close the gap between what the government spends and what it collects.
That math still has to work. Somehow.
Which means AI is being asked to do an awful lot.
It’s supposed to make companies more productive. Increase economic growth. Help businesses earn back hundreds of billions of dollars in infrastructure investment, datacenter construction and so on. Even soften the strain of an aging population and slower labor-force growth? Remember when Elon Musk said AI would make us all so rich we wouldn’t need to worry about money anymore, like a scene out of Star Trek?
On top of all that, AI is supposed to fix the federal debt burden, too…
Maybe it will. I genuinely hope it does. But we all know hope isn’t a plan...
This is where all of this becomes relevant to you.
You don’t have to believe there’s an AI bubble.
You don’t have to believe AI is a pipe dream destined to fail.
You don’t even have to have an opinion about AI at all!
The important point is just that the future is uncertain.
Today, a meaningful share of total economic growth is coming from an enormous investment in a technology whose benefits have not yet appeared.
Now, maybe those benefits show up next year. Or five years.
They could exceed expectations. And it’s possible they disappoint everyone.
If you’re 25 years old, you may have decades to ride out those possibilities. You could still go to a trade school and learn how to be a plumber or an electrician, or one of those few jobs AI can’t do. You have time.
If you’re my age – or already retired – time matters more. We have fewer working years available to recover from financial setbacks.
And once you begin relying primarily on your savings, preserving purchasing power becomes increasingly important.
That’s why I don’t think retirement planning should depend entirely on any single optimistic assumption about the future.
Not an AI productivity miracle, or permanently low inflation, or endless economic growth, or Congress balancing the budget. I don’t think we should believe in Santa Claus, either.
Diversification exists precisely because nobody knows which forecast will be right.
That’s what I love about physical precious metals. Assets that do not depend on a technology company meeting its growth projections or policymakers getting every economic decision right.
That doesn’t mean precious metals are guaranteed to rise in value. They aren’t. Prices fluctuate, and diversification cannot eliminate risk.
But if your goal is to preserve what you’ve spent decades accumulating, it may be worth learning how physical gold and other precious metals can fit into a broader diversification strategy.
The dollar buys less today than it did a year ago, a decade ago, a generation ago. That slide is the predictable result of endless printing and borrowing. Dr. Ron Paul believes the FED has the answer. But they’re keeping it to themselves. Fortunately, a new financial chart has publicly exposed their secret moves. See it for yourself: https://freekit.birchgold.com/lf/ro...D_v02a_article&placement=article&cid=rp_media
To learn more about how physical gold could help protect your retirement portfolio, click here to get your FREE info kit on Gold IRAs from Birch Gold Group. And now introducing a Crypto IRA to capitalize on the fastest growing market in the world.
Tweet
Continue reading...
[ H/T The Burning Platform ]
Today, I’m going to start with some good news.
(I can’t always do that, so it’s nice to be able to sprinkle a little sunshine your way.)
The U.S. economy grew at a 3.4% annual rate in the third quarter of 2026, according to the Bureau of Economic Analysis’s advance estimate. That’s a substantial acceleration from the 2.2% rate in the second quarter. The long-term average GDP growth rate is 3%, by the way, adjusted for inflation.
The Associated Press noted that economists had forecast 3% growth on average, so the report was quite a nice surprise!
Of course, GDP has its limitations. I’ve written before about the difference between an economy that looks healthy in the aggregate and one that actually feels healthy day-to-day.
Still, 3.4% real economic growth is good news.
So, I’m sure you’re wondering: Where did that growth come from?
Let’s dig a little deeper and find out…
The economy is growing – but look underneath the headline
Two major forces helped drive the third-quarter economic expansion.
First, consumer spending rose at a 3.8% annual rate. Remember, consumer spending makes up the vast majority of economic activity, about two-thirds on average.
Second, business investment surged – particularly in areas associated with artificial intelligence.
According to the Bureau of Economic Analysis, business investment rose 12.7% during the quarter overall. Here are some of the biggest areas of investment:
- Information-processing equipment: +37.4%
- Software: +9.6%
- Research and development: +7.5%
Those are enormous numbers!
Now, I know what you’re thinking… But no, the increase in consumer spending can’t be dismissed as inflation. These are real, inflation-adjusted GDP figures. Americans genuinely bought more goods and services.
You might think that means folks are suddenly feeling flush with cash. However, the latest Bureau of Economic Analysis report on household finances paints a more complicated picture.
In September:
- Personal income: +0.2%
- Disposable personal income: +0.2%
- Consumer spending: +0.5%
- …as a result, inflation-adjusted disposable income: -0.3%
Meanwhile, consumer prices were still 3.7% higher than a year earlier in September, according to the Bureau of Labor Statistics.
So yes, consumers are spending. But income isn’t keeping pace with that spending – and so families are saving a smaller portion of their income. The personal savings rate fell to 3.8%, well below the long-term average of 8.4%. Especially when you remember the rule of thumb for saving for retirement is to set aside at least 10% of your income…
The result? A much less cheerful picture than the GDP headline alone might suggest.
So let’s take a look at the today’s major engine of growth:Artificial intelligence.
AI infrastructure spending IS economic growth
There’s an easy trap to fall into when discussing the AI boom.
We see hundreds of billions of dollars pouring into data centers, chips, software, electrical infrastructure and research, and we ask:
But does all that spending really count as economic growth?
Yes! It absolutely does.
Building a data center means buying or leasing land, hiring construction workers, buying or renting equipment, buying and installing computer racks (and everything else a building needs: Fire suppression systems, bathroom fixtures, electrical wiring, windows…) That’s real work.
Developing software counts as economic activity. Manufacturing computers and servers counts as economic output. Capital investment is one of the components of GDP. So when businesses spend huge sums building AI datacenters, same as when they build a factory or open a mine, that genuinely increases measured economic output.
Some people ask whether AI spending really “counts” as economic activity. That’s absurd! Of course it is!
I think the much more important question is whether all that investment eventually produces a payoff.
There’s a major difference.
Imagine a farmer buying a new tractor…
Buying the tractor contributes to economic activity today. But the farmer didn’t buy it just to own an expensive machine. He bought it because he expects the tractor to help him produce more, with less labor, for years afterward. (That’s why we call it “capital investment,” by the way, to separate it from “consumption.”) If the numbers work out as predicted, the farmer is spending a lot of money now in exchange for a more profitable future.
AI infrastructure works the same way.
The spending is happening now.
The promised productivity gains are supposed to come later.
And that handoff is where things get interesting…
Where is the AI payoff?
Goldman Sachs Research has been asking exactly this question. In May, its researchers noted that while consumer adoption of AI had been spectacular, successful adoption by businesses would ultimately decide whether the economics of the sector work out.
Goldman also observed that, while a handful of semiconductor companies had made substantial revenue and profits from the buildout, most companies in the AI ecosystem had yet to make a profit. That’s concerning. But it’s not unheard of for start-up companies in cutting-edge industries to run at a loss for years. So long as they make a profit eventually.
Just one month later, Goldman Sachs Research’s Jim Covello put the issue even more plainly: Companies make investments because eventually those investments will generate returns. He went on to explain that the big spenders, the AI model developers and the companies building massive data centers, all of them had yet to demonstrate profits proportionate to the enormous amount they’ve spent.
Let’s go back to our farmer, for example. A mid-range combine harvester costs about $500,000. So, at a minimum, he needs to see $50,000 extra profit over the next 10 years to break even on the transaction.
If he goes for a top-of-the-line, GPS-guided self-driving machine? That’ll run him $1.2 million. Say it’s more efficient, and it makes him an extra $60,000 in annual profits. Now, though, he’s looking at a 20-year break-even date.
My point is simple: Spending more money doesn’t necessarily improve return on investment.
So does this mean AI is a dud? Not at all! In fact, Goldman Sachs Chief Economist Jan Hatzius still expects AI to generate huge productivity improvements. Over the longer term.
Those benefits are likely to arrive more gradually than the more enthusiastic forecasts claim.
And this creates a timing problem…
The AI boom doesn’t have to end for its GDP boost to fade
Now, Goldman Sachs is not predicting that AI investment suddenly collapses next year.
In fact, its August research estimated U.S. AI investment at about $581 billion in 2026. They projected that AI investment could grow from 1.8% of GDP this year to 2.5% in 2027 and 2.8% in 2028.
So the AI buildout may continue getting larger. What may peak sooner is its contribution to economic growth.
Hatzius recently estimated that the growth impulse from the AI investment boom could crest in late 2026 or early 2027.
That sounds contradictory at first. But think about it this way.
Suppose you increase your spending from $100 to $200.
That’s a huge increase! It’s doubled, or rose 100%.
Then the next year you go from $200 to $220. You’re spending more than ever – but the rate of increase is just 10%.
That’s what we’re talking about with AI infrastructure. Hundreds of billions of dollars can continue flowing into data centers and computing equipment even while that spending adds less additional momentum to GDP than it did before.
That creates what I think is the central question for the AI economy:
Will productivity gains arrive quickly enough to replace the boost from the construction and equipment spending boom?
I don’t know. And frankly, nobody does. We aren’t sure whether those productivity gains will arrive at all.
And that’s a real concern for me…
We’re paying for AI before we see the profits
New technologies frequently require enormous upfront investment before the benefits spread throughout society.
Railroads had to be laid before they could transform commerce. The electrical grid had to be built before factories could reorganize themselves around electric motors. The internet required fiber-optic cables, servers and data centers years before businesses figured out all the ways they could use it.
Sometimes that process produces extraordinary gains.
Often, investors build too much infrastructure too quickly.
And sometimes both happen.
The technology changes the world – while many of the people who financed the first wave still lose money.
For regular families, though, I think there’s a simpler lesson: We shouldn’t confuse optimism about technology with certainty about the economy.
Because AI isn’t the only enormous economic experiment underway.
AI can’t repeal financial arithmetic
Even if AI eventually delivers the massive productivity gains its boosters predict, the federal government still faces a difficult outlook.
The Congressional Budget Office’s 2026 baseline projected federal spending of $7.4 trillion.
Revenues were projected at $5.6 trillion.
That leaves a projected deficit of roughly $1.9 trillion.
Here’s the problem: The economy grows at about 3% a year, on average. But the accumulation of federal government deficits, the “national debt,” has been growing at 7.8% per year.
There are a number of ways to describe a person whose spending grows more than twice as fast as their income…
To be clear, spending might change. Economic growth might surprise us. And a genuine productivity revolution could certainly improve the situation by making the economy larger. (Remember, when the economy grows, so does tax collection.)
But faster growth doesn’t automatically close the gap between what the government spends and what it collects.
That math still has to work. Somehow.
Which means AI is being asked to do an awful lot.
It’s supposed to make companies more productive. Increase economic growth. Help businesses earn back hundreds of billions of dollars in infrastructure investment, datacenter construction and so on. Even soften the strain of an aging population and slower labor-force growth? Remember when Elon Musk said AI would make us all so rich we wouldn’t need to worry about money anymore, like a scene out of Star Trek?
On top of all that, AI is supposed to fix the federal debt burden, too…
Maybe it will. I genuinely hope it does. But we all know hope isn’t a plan...
What happens if the payoff arrives later than expected?
This is where all of this becomes relevant to you.
You don’t have to believe there’s an AI bubble.
You don’t have to believe AI is a pipe dream destined to fail.
You don’t even have to have an opinion about AI at all!
The important point is just that the future is uncertain.
Today, a meaningful share of total economic growth is coming from an enormous investment in a technology whose benefits have not yet appeared.
Now, maybe those benefits show up next year. Or five years.
They could exceed expectations. And it’s possible they disappoint everyone.
If you’re 25 years old, you may have decades to ride out those possibilities. You could still go to a trade school and learn how to be a plumber or an electrician, or one of those few jobs AI can’t do. You have time.
If you’re my age – or already retired – time matters more. We have fewer working years available to recover from financial setbacks.
And once you begin relying primarily on your savings, preserving purchasing power becomes increasingly important.
That’s why I don’t think retirement planning should depend entirely on any single optimistic assumption about the future.
Not an AI productivity miracle, or permanently low inflation, or endless economic growth, or Congress balancing the budget. I don’t think we should believe in Santa Claus, either.
Diversification exists precisely because nobody knows which forecast will be right.
That’s what I love about physical precious metals. Assets that do not depend on a technology company meeting its growth projections or policymakers getting every economic decision right.
That doesn’t mean precious metals are guaranteed to rise in value. They aren’t. Prices fluctuate, and diversification cannot eliminate risk.
But if your goal is to preserve what you’ve spent decades accumulating, it may be worth learning how physical gold and other precious metals can fit into a broader diversification strategy.
The dollar buys less today than it did a year ago, a decade ago, a generation ago. That slide is the predictable result of endless printing and borrowing. Dr. Ron Paul believes the FED has the answer. But they’re keeping it to themselves. Fortunately, a new financial chart has publicly exposed their secret moves. See it for yourself: https://freekit.birchgold.com/lf/ro...D_v02a_article&placement=article&cid=rp_media
To learn more about how physical gold could help protect your retirement portfolio, click here to get your FREE info kit on Gold IRAs from Birch Gold Group. And now introducing a Crypto IRA to capitalize on the fastest growing market in the world.
Tweet
Continue reading...
[ H/T The Burning Platform ]
