The 2nd Most Expensive Market in History — and Where the Money Actually Are

I've spent this week staring at one number, and it quietly takes apart the most common reply I get from you: that stocks always come back, so just hold and don't overthink it. They do come back. That was never the question. The question is what you earn from here — and here is the second-most-expensive the US market has been in 155 years.

One measure cuts through the noise on this. It's called CAPE — the price of the market divided by ten years of earnings, adjusted for inflation, so a single blockbuster year can't flatter the picture. Today it sits near 41. It has been higher exactly once in a century and a half: December 1999, the peak of the dot-com bubble. Not 1929. Not 2007. Only 1999.

Stay with me and I'll show you the machine that turns a number like that into a forecast, the three times in history it ran and how each one ended — and the part almost nobody puts next to it: the handful of places where the same math still points up.

Here's the engine underneath that number, and it's simpler than Wall Street likes it to sound. When you buy the market, you're buying a claim on its earnings. The price you pay decides your yield — exactly the way the price you pay for a rental property decides your rent-to-price return.

Pay 20 times earnings, and you've bought yourself a 5% earnings yield. Pay 40 times — roughly where we are — and you've bought a 2.5% yield. Same businesses. Half the return. For no reason other than that you paid twice as much to own them.

For a year or two, The price can go wherever the sentiment wants it to go. However, over 10 years, pricing tends to go back to the mean. It's called mean reversion. What you pay at the start sets what you earn. Pay a high price, and it will take you years to get back your money.

And the tape is brutal in how consistent it is.

December 1999. CAPE hit 44 — the only reading in history above today's. If you bought the S&P that month, a decade later you were still underwater in real terms. A full lost decade. The NASDAQ, where the story was loudest, fell 82% before it found a floor.

Japan, 1989. The most expensive major market on earth at the time. A generation of investors who bought that top waited more than thirty years just to get back to where they started. Not to get rich — to break even.

September 2007. One investor wrote that US housing was in "genuine bubble territory," months before it broke and took the banking system with it. The same man had called Japan in '89 and the dot-com top in 2000 — and then, unlike the perma bears who only ever see doom, he turned around and bought aggressively in March 2009, near the exact bottom. His name is Jeremy Grantham, and his firm exists to do one thing: price today's valuations against every peak in history. He is the co-founder and chief investment strategist of GMO LLC, a Boston-based asset management firm. GMO had more than US$118 billion in assets under management as of March 2015.

Grantham's rule is blunt: he's tracked 26 bubbles that stretched two standard deviations above their long-run trend, and all 26 fell back to earth. Not most. All. We're the 27th test.

Grantham's firm publishes a seven-year forecast for every major asset class, and it's public. As of its latest read this July, US large-cap stocks land around minus 6% a year in real terms — the scary number that's making headlines. But keep reading down the same table and it turns green. International value, emerging markets, and Japanese small-value are all forecast to earn positive real returns — the cheapest corners of the world running low-to-mid single digits a year. Even high-quality bonds now clear a positive real return, something they flatly couldn't offer four years ago when yields were on the floor.

The money didn't disappear when the S&P got expensive. It moved. It's sitting in the boring, cheap, unloved parts of the market that a decade of American mega-cap dominance trained everyone to ignore.

This is consistent with what I've been mentioning about the 18-year real estate cycle, where emerging markets tend to do really well in the last few years of the cycle. However, Grantham's forecast is for the next seven years, but as per our cycle, we don’t see it extending that long, perhaps in the next two years max.

And this is where the cycle backs the math up. We're late — around year 14 of the roughly 18-year rhythm I track — and the readings agree: asset prices are near the top of their range while credit spreads are compressed to levels that only ever show up when the market has stopped pricing risk at all. Rich prices and nobody worried. That's not the setup for the next great decade of US index returns. It's the setup for a handoff. I keep those readings live in Smart X Terminal so I'm reading the market against its own history, not against last week's headlines.

I don't think the next ten years belong to the S&P 500. I think they belong to what's cheap — value over growth, international over domestic, real businesses over the index everyone already owns. Not because a crash is coming next Tuesday. I genuinely don't know about next Tuesday. But because from a starting CAPE of 41, the arithmetic has never once — not in 155 years — paid the buyer well over the decade that followed.

Could I be early? Easily. Expensive gets more expensive, and 1999 proved a bubble can run two more years past "this is insane." Timing the top is a mug's game and I won't play it. But this was never about timing. It's about what you own when the arithmetic finally asserts itself — and that's a decision you can make today, calmly, without predicting a single date.

That's the free half — the why. Below the line is the what: the specific map of where those positive-real returns actually sit right now — the three regions and two sectors where GMO's math and my own screen agree — how I've tilted my own money toward quality-at-a-discount instead of growth-at-any-price, and the exact filter I run in Smart X Terminal to tell a genuinely cheap, high-quality business apart from a value trap that's cheap for a reason. If you've been sitting in an S&P index fund quietly wondering whether "just hold" still works at the second-most-expensive valuation in history, the next section is the one written for you.