The S&P 500 Is More Expensive Than Before the Dot-Com Crash — The Lost-Decade Risk

TL;DR With the Buffett Indicator and CAPE ratio near record highs, the risk that the S&P 500 delivers almost no real return over the next decade is rising. Market-to-GDP is roughly 140% above its historical average, and every time markets started at this level, the average 10-year forward return was negative.

The S&P 500 is more expensive today than right before the dot-com crash

Let me lead with the conclusion I reached after digging back through the data myself: the S&P 500 is more expensive right now than it was in 2000, just before the dot-com crash.

Don't get me wrong. I'm not here to trash the index. The 500 largest U.S. companies by market cap make up one of the greatest wealth-building tools ever created. Warren Buffett and nearly every value investor has recommended it to regular people as the way to match the market and build wealth over a lifetime. For most people, most of the time, buying it and holding forever is a fantastic plan.

The problem is right now. This is not one of those ordinary times.

The single most important rule in investing is this: the price you pay decides the return you achieve. Buy a wonderful business when it's expensive and your future returns tend to be weak. Buy an ordinary asset at a discount and your future returns tend to be great. And right now, we're all being told to buy at some of the highest valuations in history.

The Buffett Indicator and CAPE, two measuring sticks

There are two famous ways to gauge whether the market is expensive.

The first is the Buffett Indicator, which compares the total value of the stock market to the size of the U.S. economy. Today it sits at essentially its highest level ever.

The second is the CAPE ratio — the cyclically adjusted P/E, a long-term price tag that measures prices against ten years of earnings. It, too, is near the highest levels in history.

The key point is that both are flashing historic extremes at the same time. And this isn't just my read. Paul Tudor Jones recently said that if you buy the S&P at a P/E of 22, the 10-year forward return is negative — that's what history shows.

Howard Marks of Oaktree Capital cited the exact same statistic in a recent memo. According to a JP Morgan chart, over the only window with data — 1987 to 2014 — every time you bought the S&P 500 at 23 times the coming year's earnings, your average return over the next ten years landed between +2% and -2%, every single time. To the extent that P/E history is relevant, it bodes poorly for the S&P from here.

We've already seen this movie

I want to stress that this isn't theory — it actually happened.

From 2000 to 2012, the S&P 500 went essentially nowhere. That's the so-called lost decade. The Nasdaq (QQQ) was worse: it went sideways for a full 16 years. In 2000, QQQ was the must-own investment because the internet was reshaping the world. And that obvious growth story turned into 16 years of stagnation.

That is exactly the scenario the quietly worried crowd is worried about today. Not because the growth story is wrong, but because the starting price was too high.

What the market-to-GDP ratio predicts

I pulled a hundred years of market-to-GDP data and sorted it from undervalued to overvalued. Note that the returns below exclude dividends, which historically added roughly 3–3.5% a year (less lately).

Valuation at the starting point	Average 10-year forward return (ex-dividends)
30%+ undervalued	+10.7%
Undervalued	+9.0%
Fair	+2.1%
Overvalued	-1.2%
Heavily overvalued	-2.0%
50%+ overvalued	-2.4%

The historical 10-year average return is about 6.4%. So where are we now? Roughly 140% overvalued versus history — well past the bottom of that table, deep in negative territory.

The counterargument, and where I stand

There is a rebuttal, of course. The claim is that because these companies earn so much overseas, the market-to-GDP ratio doesn't mean what it used to. I understand that logic. There's something to it.

But does it really justify 140% overvaluation? My honest take is that we're closer to inventing reasons because we don't want to admit the market is massively expensive.

One thing I'll say plainly: I've called this market overvalued for about four years, and it kept climbing anyway. Valuation is not a timing tool. "Expensive" does not mean "down tomorrow." It means a lower expected return over the next ten years. Today looks a lot more like 2000 than 2009. If history rhymes even a little, betting everything on the expensive index deserves a careful second look.