Transcription
If 398 out of 400 Wall Street professionals agreed the market was about to crash, why didn't a single one of them say anything?
Jeremy Grantham called the dot-com bubble, the housing bubble, and the Japanese asset bubble. Each one before it burst. He co-founded GMO, a firm managing $35 billion. And what he revealed about the system that's supposed to protect your money changes how you should read every analyst rating, every price target, every bullish headline from here on.
Grantham did something no one had ever tried at the biggest investment conference in America. He asked the room a question. And the answer changed everything.
But in fact, one of the things that you talk about is how really everybody knew. Everybody knew, but they couldn't take the career risk or the benchmark risk of accepting the truth.
One of the most important thing I ever did, and and one of the keys in the book, um no one ever takes any interest in. And that is when I was debating the bulls. Yeah. I I would I would ask the audience a few questions. And these were This was The critical one was the national the annual meeting of the financial analyst society. And it was in Los Angeles, I think. Certainly in California. And there was a cast of thousands in there, 1,500 people or so. Yeah. And um before I did my debate, I was allowed to go first by Jeremy Siegel, now dean of Wharton. And I eagerly accepted, and then I played a dirty trick by asking the audience hands up who considered themselves full-time professional stock players, institutional stock players. And 400, according to me and my friends counting, about 400 put their hands up. And I said, "Right, I have two two questions only for you. One, if the current PE, which is 31, comes down to 17 anytime in the next 10 years, will it guarantee a major bear market?" And the vote was 100%. They said, "Yes, it would guarantee." So I said, "Right, now we get to the jackpot. How many of you think it will come down?" And I was so shocked I had to ask the question three times, rephrase it. Um only two people in the audience uh of that 400 thought it would not come down to 17. And of course it did. Yes.
But 398 people in the engine room of the Goldman Sachs's and the JP Morgans and so on, Morgan Stanley's, they believed uh in data that guaranteed a major bear market in the future. And the people on the podium, however, their bosses or their marketing bosses, let's put it that way, were saying, "Jeremy, Jeremy, let's be serious. Don't get so carried away. The market is okay and we'll muddle through." I think that's a fair statement for the top of the market. Yeah. People don't get that. They thought, because they only hear the bosses, they thought that intelligent opinion was that the market would keep going. And in fact, intelligent research opinion, the real experts, would have said, "We guarantee a major bubble breaking here." Interesting. So >> one knew that. And no one cared when I pointed it out. And in the end, we had over a thousand people polled who were claiming to be professionals. Yeah. So while you looked like an outlier with your opinions there in the late '90s, in fact, the engine room agreed with you. Absolutely. And what we found then, and ever since, uh is the obvious fact that it's not a business strategy to be a bear in a bubble.
So 398 professional money managers, the people running your pension, your 401k, the engine room of Goldman Sachs and JP Morgan, all agreed the market would crash. And their bosses went on television and said everything was fine. Think about that for a second. The people doing the actual research knew. The people talking to you on CNBC didn't care. This wasn't some room full of doomsayers or gold bugs. This was the annual meeting of the financial analyst society. CFA charter holders, portfolio managers, quants. The most credentialed investors in the country. Grantham didn't ask a trick question. He asked, "If the PE drops from 31 to 17 anytime in the next decade, will that guarantee a bear market?" Every hand went up. Then he asked, "How many of you think it will drop?" Only two people out of 400 thought it wouldn't. The conviction rate was 99.5%.
And it didn't stop there. At the peak of the dot-com bubble in mid-2000, 74% of all analyst ratings on Wall Street were buy. 2% were sell. 74 to two. Let that sink in. For every analyst brave enough to say, "Get out," there were 37 saying, "Buy more."
Hong and Kubik, two finance professors at Cornell and Syracuse, published a study in the Journal of Finance in 2003. They tracked what happened to analysts after they made correct but unpopular calls. The results were ugly. Analysts who stayed optimistic got promoted, even when they were wrong. Analysts who were accurate, who actually called the crash correctly, got punished. They were reassigned to smaller stocks. Some were let go entirely. The career incentive on Wall Street points in one direction. Up. Always up. Accuracy doesn't matter. Consensus does. The research team writes one thing in their private models. The marketing department publishes something completely different. And the only version that reaches your brokerage account is the marketing version. That's the machine Grantham exposed in that room. And it runs exactly the same way today.
Now, I need to be honest. When I first pulled up the current data, part of me wanted to push back on Grantham. Because here's the thing. That was the dot-com era. The S&P 500 in 2000 was stuffed with telecom companies burning cash and dot-com retailers with zero revenue. Today's market looks different. Actually, scratch looks different. It is different. Apple prints over $100 billion a year in profit. Microsoft's cloud business throws off margins north of 40%. These aren't pets.com.
Jeremy Siegel, the Wharton professor who debated Grantham at that very financial analyst society meeting, has made this case for years. His argument, corporate earnings quality improved. The S&P shifted toward asset-light, high-margin tech that structurally deserves higher multiples. The Shiller CAPE ratio, that's the cyclically adjusted price-to-earnings ratio, which smooths out earnings over 10 years to filter out short-term noise, sitting at 25 to 30 might be the new normal. Not because of irrational exuberance, but because today's companies are genuinely better businesses. And there's data behind it. Strip out the Magnificent Seven, and the S&P's CAPE drops from about 40 to 33. Still above average, but well below the bubble threshold Grantham uses. If the overvaluation is concentrated in just seven names, maybe this isn't a market bubble. Maybe it's a sector bubble. And sector bubbles historically don't take down the whole index. And even then, there's genuine debate to be had. Howard Marks, for example, has made the argument that the Magnificent Seven are great businesses and deserve to trade at a premium to the broader market. So maybe Grantham's wrong this time. Maybe the system really has changed. Maybe 2026 bears are making the same mistakes they made in 2013, 2016, 2019, calling a top that never came.
I sat with that for about an hour. Then I listened to what Grantham said next, and it knocked the wind out of my sails. Grantham says the answer has nothing to do with intelligence. It has everything to do with survival. Listen to this.
Yes, and I think this is the the biggest failing of ordinary investors is a failure to realize that if you've done your homework, you've looked at the data, you really can be right and the authorities can be wrong. At turning points, the authorities are nearly always wrong. And and I don't think it's it's a bug. I think it's part of the system that they will be wrong for the following reason. If you have a large organization, you will typically be led by someone with substantial political skills, right? >> Uh-huh. And if you have political skills, you understand Keynes in chapter 12 anybody. Never be wrong on your own. Or you'll not receive much mercy. Uh-huh. Right? Just make sure that all your mistakes have plenty of company, and you'll do fine. And so what happens at a turning point is turning points are lethal. No one can accept the career risk of a turning point. So they all predict it will keep going. Mhm. Mhm. They all look around nervously at each other, but they keep going. As long as the music's playing, they're going to be dancing. Doesn't matter that they know the market is silly, they're still dancing and they've confessed to it and we know that that's how it works. And then when one of them jumps, another jumps and pretty soon every last one of them jump. Uh because last one out uh is a is a donkey. Yeah. Um
And what do you think the institutions, the authorities are mainly wrong about at the moment? Um Well, you you just have to say to yourself, what what is likely to be a great turning point? Yes. And um I think um I think it's obvious that uh internet investing near term, regardless of how well it does in the long term, that near term it will be overdone and it will have a bust. Uh-huh. And I expect the authorities to be completely relaxed about that and not mention it. Yeah. In round numbers. They all look around nervously at each other, but they keep going as long as the music's playing.
That sentence hit me in the chest because I've seen the proof. Not in theory, in a real name, a real year, a real stock price. July 2007, Chuck Prince, CEO of Citigroup, at the time the largest bank in the world by assets, he sat down with the Financial Times and said, on the record, as long as the music is playing, you've got to get up and dance. We're still dancing. He used Grantham's exact metaphor, word for word, the music, the dancing, the nervousness, all of it. And here's the part that should make your stomach turn. Prince wasn't clueless. He wasn't some mid-level trader who didn't see the CDO exposure. He was the CEO. He had every piece of data Citigroup produced. He knew the subprime loans were rotting. He said so publicly, on camera, using a metaphor that basically admitted, I know this ends badly, but I can't stop.
So why didn't he stop? Because the board would have replaced him. Citi's competitors, Lehman, Bear Stearns, Merrill Lynch, all still dancing. If Prince had pulled Citi out of mortgage-backed securities in July 2007, the stock would have dropped. The board would have fired him for underperformance. And some other CEO would have danced right back in. The system doesn't reward the person who stops first. It punishes them. 4 months after that interview, Prince was forced out anyway. Citi stock went from $55 to under a dollar by March 2009. Prince walked away with a $38 million exit package. The shareholders, they lost 98 cents on every dollar. And the next Citigroup CEO, he kept dancing too because the music hadn't stopped yet. That's career risk. That's what Grantham described. And it operates on autopilot in every bull market, including this one.
So what does all of this mean right now, today, for your money? Remember that CAPE ratio we talked about earlier? The 10-year smoothed measure of how expensive the market is? It sits at 40.19 as of February 2026. That's the second highest reading in 140 years of data. The only time it was higher was December 1999 at 44.19. We all know what happened in the 3 years that followed. The S&P fell 49%. The Nasdaq fell 78%. Amazon dropped 94%.
But remember, Grantham himself says timing a bubble is nearly impossible. So instead of pretending I can tell you the exact month it breaks, here are three signals worth watching.
First, earnings delivery versus capital spending. The hyperscalers, Amazon, Alphabet, Meta, Microsoft, they spent nearly $300 billion on AI CapEx in 2025. That's 1.3% of total US GDP funneled into data centers and chips. Grantham calls it a prisoner's dilemma. Every company has to spend because if their competitor cracks AI first, they're finished. But if everyone spends and the revenue doesn't show up, the collective hangover will be massive. Meanwhile, OpenAI projected $12 billion in revenue against an $8 billion operating loss for 2025. Loss is expected to double to 17 billion in 2026. The question isn't whether AI changes the world. It probably does. The question is whether these stocks can earn back what they're spending before investors lose patience.
The second signal worth watching is market breadth. Back in 2023, the S&P gained for the year, but strip out the Magnificent Seven and the index actually fell for 11 straight months. Since then, breadth improved. By mid-2025, over half the S&P 500 was outperforming the Mag Seven median. The rally broadened out. But here's why that doesn't settle the argument. As of early 2026, just 10 stocks still account for 40% of the entire index. The Mag Seven alone make up nearly a third. If earnings disappoint in even two or three of those names, the whole index gets dragged down regardless of what the other 493 are doing. Watch for the narrowing pattern to reappear. If the rally concentrates again, it means the average stock is already in retreat.
The third signal to watch would be credit spreads. When the gap between corporate bond yields and treasury yields start widening, it means the bond market smells trouble. The bond market doesn't care about narrative or hype cycles. It didn't buy the quote, "Housing is fine" story in 2007. Credit default swaps on subprime started blowing out months before the stock market noticed. Credit traders are usually the first to run. They don't wait for earnings calls or analyst upgrades. They price risk in real time. That's your canary.
Now, I want to be fair to Grantham's track record because it cuts both ways. He called the dot-com bubble correctly. He called the housing bubble correctly. But he called them early. During the late '90s, while Grantham was warning about overvaluation, his clients were watching the S&P rip higher without them. GMO's assets under management went from 30 billion to 20 billion. They lost a third of their client base. Those clients got tired of waiting for a crash that hadn't arrived yet. Grantham was right about the destination, but he was wrong about the departure time. And in asset management, being early and being wrong feels identical, especially when your bonus depends on this quarter's performance.
So what's the range of outcomes here? Let's run two scenarios using the CAPE ratio. Scenario one, the bull case. The CAPE reverts to Siegel's new normal of 25. That would mean today's multiples are too high, but not catastrophically so. Even in this friendly scenario, you're looking at a 37% decline from current levels. Call it a repricing rather than a crash, but it still turns a million-dollar portfolio into 630,000. Scenario two, the historical case. The CAPE reverts to its 140-year median of 16. That's a 60% decline. A million-dollar portfolio becomes 400,000. That's the house down payment your kid was counting on, just gone. And this has happened before. It happened after 1929. It happened after 2000. The math doesn't require a recession or a banking crisis. It's just the iron law Grantham describes. The higher price you pay today, the lower the return you collect tomorrow.
Now, I'm not saying sell everything tomorrow. I don't know when this cracks. Nobody does. Grantham has been wrong on timing before. AI might genuinely change the earnings trajectory in ways that justify these multiples. But the 398 knew the dot-com bubble would break and they couldn't act on it. You can. The real lesson from Grantham has nothing to do with predicting crashes. What matters is understanding who's talking to you and why. The analysts writing those price targets have bosses. Those bosses have revenue goals. And those revenue goals depend on you staying bullish. The analysts in the engine room see data. The executives on the ship's bridge, the ones who actually talk to clients and go on television, sell the story. So next time you see a price target on CNBC, ask yourself, am I hearing from the engine room or the marketing department? Your job as an investor is to know the difference.
If you found this useful, make sure you're subscribed because we break down the world's greatest investors every single week to help you make smarter decisions with your money. And if you want to see why legendary investor Monish Pabrai thinks we're entering into a lost decade for stocks, check out this video next. I will see you over there.