All In: America Has Never Leaned This Hard on Stocks

For the most part, I’ve kept my mouth shut on artificial intelligence.

I’ve stuck to my corner of the investment world, focused entirely on resources.

But given how hard US households now lean on the technology sector as an investment class, the risks are too important to ignore.

33 Times Earnings and Climbing

As of late August, the Nasdaq 100 traded at a trailing price-to-earnings ratio of about 33 times.

That is not, on its own, proof of a bubble.

In fact, at its March 2000 peak, just before the dot-com bust, the Nasdaq 100 traded at a forward P/E of nearly 60. An extreme multiple by any historical measure.

Back then, the unwind wiped close to 78% off the Nasdaq Composite over the following two and a half years. It took the index until 2015, some fifteen years, to reclaim its old high.

Bottom line: the Nasdaq 100 was exceptionally overvalued 26 years ago. And today’s valuation numbers don’t mirror that, yet.  

So, does that mean it’s time to jump into the AI trade for some additional speculative gains? Perhaps for a short while.

But here’s a good reason to be extra vigilant right now…

According to Goldman Sachs, American families now have nearly half their financial wealth riding on stocks, the highest concentration since records began in 1945, including the dot-com bubble.

According to the latest Federal Reserve data, US households allocated a record 48.23% of their financial assets to corporate equities.

And that sits well above the 27% peak reached during the dot-com era.

So, why is that concerning?

Simple, America is all in on the AI trade. And that limits this boom’s capacity to withstand negative news.

In other words, there’s no room for error in this historical melt-up.

A major earnings downgrade or a capex cut by one of the tech giants could trigger panic.

Similarly, further escalation in the Middle East, putting more pressure on bond yields and inflation risks, could be the final nail in this record market melt-up.   

I’ll say it again: with 48% of US household wealth sitting in a market trading at 34 times earnings, concentrated in just a handful of technology names, there’s no room for error.

If you’re still hanging onto the belief that the AI investment theme has plenty of runway for long-term sustainable growth, well, may you live in interesting times.

But for the rest of us, its time to think about alternatives…

Can resources offer a safe haven?

Most assume that rising commodity prices are linked to massive forecast demand for AI data centre construction.

Meaning that if the AI trade unwinds hard, metals like copper will take a hit, given how much of the recent bull case for the metal has been built on data centre demand forecasts.

That may be partly true.

However, what if the biggest story for resources over the next few years is not AI at all?

But something that’s been sitting in the background behind this relentless AI trade… What do I mean?

Across Europe and the United States, governments are reopening nuclear plants, lifting defence budgets, and pushing tariffs to discourage foreign imports.

Call it what you like, but these are all signs that the West is looking to re-industrialise its economies.

In my mind, this is the most important theme to watch; a trend quietly building behind AI mania.

And if it evolves, reindustrialisation of the West will require far more steel, copper, aluminium, and energy than any data centre ever will.

I will get into exactly why in the next edition.

For now, the takeaway is simple: don’t assume a resource portfolio automatically sinks or swims with the Nasdaq. The drivers are not the same, even if the headlines like to pretend they are.

Extreme valuations, record household exposure to stocks, and a market that has never leaned this hard on a handful of technology names…

These are the key risks, but that’s not a reason to stay out of the markets entirely.

Instead, it’s a basis to know exactly what you own and why you own it.

We built our model portfolio around resource businesses with real earnings, real production, and genuine demand tailwinds, not narrative-driven multiples.

If you’d like to see how we’re positioning for what comes next, I suggest checking it out here.

Until next time.