You’re paying $41.50 for every dollar of inflation-adjusted earnings in the S&P 500 right now. That’s using the Shiller CAPE ratio — a 10-year average that smooths out booms and busts so you can see what you’re actually paying for stocks. And in 145 years of data, this number has only been higher once.

December 1999. The CAPE hit 44.2. Everyone knew the internet was going to change the world. They were right about that. They were wrong about the price they paid for it. Over the next decade, the S&P 500 lost 24% of its value. Investors who bought at CAPE 44 didn’t break even for more than 12 years.

Now... you’ll hear the usual explanations for why this time is different. Profit margins are higher. Tech companies earn more per dollar of revenue. AI changes everything. Maybe. But those were the same kinds of arguments people made in 1999 about the internet. And in 1929 about radio and consumer credit. The story changes. The math doesn’t.

Here’s what most people get wrong. The CAPE doesn’t predict crashes. It never has. What it predicts — with remarkable consistency — is the next decade of returns. When you start from a CAPE above 40, history says you’re looking at low single digits... at best. Invesco’s research puts the average 10-year annualized return from these levels at slightly negative.

That doesn’t mean sell everything tomorrow. It means the easy money of the last three years — just buying the S&P and watching it go up 20%-plus a year — probably isn’t coming back for a while. The median CAPE since 1881 is 17.3. We’re sitting at 2.4 times that. Gravity doesn’t care about your feelings. It just works slower than you expect.

None of this means the market falls tomorrow. It could run higher for another year. It did in 1999. But it does mean this is a good week to look at what percentage of your portfolio is riding on the S&P 500 and ask yourself: if the next decade delivered 2% a year instead of 20%, would I need to change anything? That’s worth 20 minutes with your advisor. Not next month. This week.