Is the stock market overvalued right now?

By the Buffett Indicator, yes, and by a wide margin. The ratio of total US market value to GDP sits near 236% as of August 2026, per Longtermtrends. That is the highest reading in the measure's recorded history. It is well above the roughly 153% peak of the dot-com bubble and above the 2021 high near 195%. Overvalued does not mean a crash is scheduled. It means you are paying more for each dollar of economic output than investors have paid at any prior point.

Does a high Buffett Indicator mean a crash is coming?

No. The indicator has no useful timing ability, and treating it as a sell signal has cost people years of returns. What high readings have historically predicted is lower average returns over the following decade, and deeper drawdowns when a decline does arrive. Those are two different claims. One is about direction over ten years and has decent support. The other is about next month and has almost none.

What should I actually do if the market is this expensive?

Mostly keep going, with your risk sized honestly. Selling everything to wait for a correction is a timing bet that usually loses. What high valuations justify is checking that your allocation matches your real time horizon, keeping enough cash that you are never a forced seller, and lowering the return you assume in your own planning. That last one costs nothing and changes your math more than people expect.

Why is the Buffett Indicator so much higher than it used to be?

Part of it is genuine overvaluation and part of it is a measurement problem. US listed companies now earn a large and growing share of revenue overseas, while GDP only counts domestic output. Profit margins are also structurally higher than they were in the 1970s. Both effects push the ratio up without requiring a bubble. This is why the level alone is a weaker signal than the direction and the speed of the change.

How far do markets fall from record valuations?

History gives a range, not a number. The dot-com unwind cut the NASDAQ by about 78% from its 2000 peak, while the broader market fell far less. The 2007 to 2009 decline took the S&P 500 down roughly 57%. Neither began the moment valuations peaked. Both took months to develop and both fully recovered for investors who kept contributing.

Is the Buffett Indicator still useful if it keeps hitting records?

It is useful as a gauge of what you are paying, not as a clock. An indicator that has been elevated for years is telling you something real about expected returns, even though it has told you nothing at all about timing. Treat it the way you would treat a fuel gauge that cannot tell you where the next station is.

Is the Stock Market Overvalued? Buffett Indicator Hits 236%

Is the stock market overvalued? The Buffett Indicator sits near 236%, a record. What that means for future returns.

The most cited valuation gauge in investing is at a record, and it is not close. Total US stock market value divided by US GDP, the ratio Warren Buffett once called the best single measure of where valuations stand, sits near 236% as of August 2026 according to Longtermtrends.

So is the stock market overvalued? By this measure, more than at any point on record. Higher than the dot-com peak. Higher than 2021. The harder question is what that actually buys you as a piece of information, because the honest answer is narrower than the headlines suggest.

Is the stock market overvalued at 236%?

Yes, by this measure, and by the widest margin ever recorded. A 236% reading means US stocks are valued at nearly two and a half times the annual output of the entire US economy. The dot-com peak was roughly 153%. The 2021 high was near 195%. Every previous record has been broken and then broken again.

What the number actually measures

The math is simple, which is most of its appeal. Take the total market value of US stocks. Divide by annual US GDP. That is the whole calculation.

The logic behind it is that corporate profits ultimately come from economic activity. Over a long enough period, the value of all companies cannot grow faster than the economy that supports them without something eventually giving.

We wrote a fuller breakdown of the mechanics in our complete guide to the Buffett Indicator, including its history and Buffett’s own comments about it. This piece is about what a record reading does and does not tell you.

The part most coverage gets wrong

Here is where we want to be careful, because this is where most articles about this number go badly astray.

A high Buffett Indicator has historically predicted lower average returns over the following ten years. That relationship holds up reasonably well across long stretches of market history. Buy when the number is low and the following decade tends to be generous. Buy when it is high and the following decade tends to disappoint.

A high Buffett Indicator has not predicted when a decline starts. Not once, not reliably. The indicator crossed its dot-com peak and kept climbing for years. Anyone who sold in 2017 because the number looked crazy has now watched it go far crazier while missing an enormous run.

Those are two completely different claims, and blurring them is how people talk themselves into expensive mistakes. One is a statement about the next decade. The other is a statement about next quarter. Only the first has evidence behind it.

Why corrections look more likely from here

There is a version of “corrections are more likely” that is defensible, and a version that is nonsense. The defensible version goes like this.

Valuation does not tell you when a decline begins. It does tell you how much room there is to fall when one does. When prices already reflect optimistic assumptions, there is less cushion. Ordinary disappointments that would normally cause a mild pullback have more distance to travel.

That is a statement about the size of a potential drawdown, not its timing. From 236%, a return to even the elevated 2021 level of roughly 195% would be a decline of about 18% in the ratio. A return to the dot-com peak near 153% would be a drop of around 36%. Neither of those is a forecast. They are just arithmetic about how far above prior extremes we are sitting.

The nonsense version is “the indicator is at a record, so a crash is imminent.” That claim has been made every year for most of a decade and has been wrong every time.

The measurement caveat that matters

The ratio is not comparing what it compared in 1975, and pretending otherwise makes it look scarier than it is.

US listed companies now earn a large share of revenue outside the United States. Apple, Microsoft, and Nvidia sell globally, but GDP only counts domestic output. The numerator captures worldwide earnings power while the denominator does not. That gap alone pushes the ratio structurally higher.

Profit margins are also durably wider than they were decades ago. Software and services businesses simply keep more of each dollar than industrial ones did.

So part of that 236% is genuine expensiveness and part is the yardstick bending. We do not think this makes the number useless. We think it means the level is a weaker signal than the direction and the speed, and that anyone quoting the raw figure without this caveat is selling alarm rather than analysis.

What we actually do about it

Nothing dramatic. That is not a dodge, it is the honest answer.

Lower the return you assume. This is the single most useful response, and it costs nothing. If your retirement math assumes 10% annual returns, run it again at 6% and see what breaks. High starting valuations are precisely the condition under which optimistic assumptions fail. Our portfolio stress test tool will show you what a bad decade does to a real balance.

Keep enough cash that you are never forced to sell. The people genuinely hurt by the 2008 decline were largely those who had to liquidate at the bottom. An emergency fund is not a market call, it is what turns a crash into a paper loss instead of a permanent one.

Keep contributing. If a decline arrives, ongoing contributions buy at lower prices. That mechanism does more work over a career than any attempt at getting out first.

Check your allocation against your real horizon. Money you need in three years should not be in equities at any valuation. Money you need in thirty can ride this out. Record valuations are a good excuse to be honest about which bucket things belong in.

What we do not do

We do not sell out and wait. We have watched people do this since the indicator passed 150%, and the cost of being early has been enormous.

We do not treat 236% as a date. It is a price level, not a calendar.

And we are not certain about any of this. Anyone who tells you they know what happens next is guessing with more confidence than the evidence supports. What we are reasonably confident about is narrower: paying more for each dollar of output means expecting less back, and having less room beneath you when something goes wrong.

The bottom line

The Buffett Indicator at 236% is a real signal about a real thing. It says future returns are likely to be lower than the last decade trained us to expect, and that the drop from here has further to travel than it did from lower levels.

It says nothing about when. It has never said anything about when.

Plan for lower returns. Keep cash so you are never a forced seller. Keep buying. That combination works whether the record breaks again next year or breaks the other way.