Are Stock Markets Overheated?

Is It Time to Take Profits?

When the World’s Most Careful Investor Moves to Cash

In the world of finance, there are only a few moments when it really pays to stop following the headlines for a moment and watch what the people who have survived every kind of market cycle for seven decades are actually doing. We are living through one of those moments right now.

Berkshire Hathaway, the company Warren Buffett ran until the end of 2025, is sitting on a record 397 billion dollars in cash and short-term US Treasury bills. This is not just a big number. It is the largest cash position in the history of any investment company in the world. It is bigger than the entire market value of Bank of America. It is bigger than Coca-Cola. And it has been built up over fourteen straight quarters in which Buffett sold more stocks than he bought.

In May 2026, at the annual Berkshire Hathaway meeting, Buffett sat in the audience for the first time in sixty years. Even so, his comment defined the whole event. He described today’s American stock market with one harsh word: a casino. He added that he has never seen people in such a gambling mood. This is not the opinion of a man who has lost his sense of time. This is the warning of someone who closed his investment partnership back in 1969 because he could no longer find fair prices, and who also raised cash in 1999 and 2007, well before the crashes that followed.

History teaches us one clear lesson. When Buffett builds cash, it does not mean the market will crash tomorrow. It does mean that the tools he developed over seventy years of investing can no longer find enough fair value to justify a purchase. That is a signal no serious investor should ignore.

The Shiller Index: Second Highest in 145 Years

The second warning sign is even more concrete. The Shiller CAPE index, developed by Nobel laureate Robert Shiller, measures the price of the S&P 500 compared to the average earnings of companies over the past ten years, adjusted for inflation. In May 2026, it stood at around 40, more than twice its long-term average of about 17 since 1881.

What makes this number truly worrying is that in the entire history of US stock markets, the CAPE has only crossed 40 once before. That was in December 1999, just before the dot-com bubble burst. Back then it peaked at 44, and the S&P 500 then lost about half of its value over the next two and a half years.

Today’s reading is well above the level before the 1929 crash, when the CAPE was around 32. It is also far above the level before the 2008 financial crisis, when it reached about 27. Every time in history that the Shiller CAPE has moved between 35 and 40, a market drop has followed. Every single time, without exception.

There is a common counterargument. Today’s S&P 500 is not the same as the one from 1999. It is dominated by big technology and software companies with more stable earnings than the cyclical industrial giants of earlier eras. Some analysts argue this justifies a permanently higher CAPE. But Shiller himself warns that elevated readings have always preceded periods of lower returns over the next ten to twenty years. The implied future return at today’s CAPE is just 1.3 percent per year, by some calculations. That is well below the cost of capital, below inflation, and below the returns from many alternative investments.

Artificial Intelligence and the Money Going in Circles

Almost all of the stock market growth over the past two years has been driven by one single story: artificial intelligence. The Magnificent Seven, the seven biggest US tech companies, today represent about one third of the entire S&P 500. Together they are worth around 21.5 trillion dollars. That is roughly 16 percent of the value of all stocks in the world.

When the S&P 500 price-to-earnings ratio reached 30.9 at the end of 2025, every member of the Magnificent Seven was trading at a multiple above 30. Nvidia briefly crossed 5 trillion dollars in market value. Its price-to-sales ratio went over 30 in November, a level experts traditionally associate with a bubble.

But the numbers themselves are not the most worrying part. The structure of how AI is being financed is even more concerning. Goldman Sachs estimates that capital spending on AI will hit 390 billion dollars this year, and rise by another 19 percent next year. Bank of America projects that AI capital expenditure will reach 1.2 trillion dollars per year by 2030.

So who is actually paying for all this spending? A research note from Morgan Stanley revealed that the ten most important AI companies are tied together in an increasingly circular way. OpenAI, Nvidia, Oracle, Microsoft, CoreWeave and AMD are passing billions of dollars between each other through equity stakes, revenue-sharing deals, vendor financing and repurchase agreements.

This is not healthy economic activity. This looks very similar to the telecom bubble of the late 1990s, when network operators and equipment makers financed each other in circles until the music suddenly stopped. The real question that is being asked more and more loudly on Wall Street is when the AI giants will finally show that all this investment is actually producing profits. Right now the money is flowing into buying graphics chips and building data centers, not into stable earnings from people and businesses actually using AI services.

The Dollar Is Weakening and the Bond Market Is Cracking

While stock markets keep climbing, something far less friendly is happening in the background. The US dollar index, known as DXY, has dropped to around 97 in May 2026. That is a ten-week low. More importantly, the dollar lost about 10 percent of its value during 2025, and most forecasts expect it to weaken further to between 90 and 96 by the end of 2026.

This is happening in an environment that, by every classic rule, should be making the dollar stronger. There are geopolitical tensions in the Middle East. Oil prices are elevated. The Federal Reserve cannot cut interest rates because inflation is too high. And yet the dollar keeps falling. The reason is simple but profound. The world is quietly moving its reserves away from US financial assets.

The drama in the bond market is even more dramatic. The yield on the ten-year US Treasury bond reached 4.60 percent in late May 2026. That is the highest level in a year. The thirty-year US Treasury was sold at auction with a yield above 5 percent. That is the first time this has happened at an actual auction since 2007. In just one week, the US government sold 691 billion dollars in bonds, and buyers are demanding ever higher yields to absorb the supply.

This means one thing only. Confidence in the US debt system is cracking. The gap between the thirty-year bond yield and the Federal Reserve’s effective interest rate has grown to 149 basis points. The long-term bond market is completely ignoring the Fed’s signals about cutting rates. It sees only inflation and fiscal fragility ahead.

Private Credit: The Quiet Bubble That Is Starting to Burst

The third weak point in today’s financial system is private credit. This is a market that has ballooned from about 500 billion dollars to 1.7 trillion dollars in just a few years. It operates largely outside the reach of traditional banking regulation.

According to Fitch Ratings, the default rate in the US private credit sector reached a record 6 percent in April 2026. That is the highest level since the index was created in 2024. Morgan Stanley forecasts that default rates could climb close to 8 percent in coming months. The worst-case scenario from UBS even points to a 15 percent default rate this year.

Behind this lies what experts call the maturity wall. In 2026 alone, 162 billion dollars of private credit must be refinanced. And there is more. The broader commercial real estate market faces 875 billion dollars in loans maturing this year. Major firms like KKR are already being forced to inject fresh capital into their own funds just to prevent panic from spreading.

The software industry, which makes up about 40 percent of all private credit loans tied to private equity, has been hit especially hard by the disruption from artificial intelligence. This is an important paradox. AI is at the same time inflating the value of a handful of stocks and destroying the business models of companies that were once considered untouchable software giants.

Four Warning Signs Pointing in the Same Direction

It is worth pausing for a moment to notice something important. Each of the four warning signs we have looked at comes from a completely different corner of the financial system.

The world’s most respected investor moving into cash is a behavioral signal. The Shiller CAPE is a valuation signal. The circular financing among AI companies is a structural signal about how the boom is being funded. The weakening dollar combined with rising bond yields is a signal about confidence in the entire US debt system. The private credit defaults are a signal about real stress in the corporate world.

These are not four versions of the same story. They are four different parts of the financial machine flashing red at the same time. That is rare. In quiet markets, you usually see one or two warning lights. When four major systems all start blinking together, history tends to deliver some kind of correction. Not always in the same month. Not always in the same way. But the pattern is consistent enough that ignoring it has been very expensive for investors in the past.

So Is It Time to Take Profits?

Each of these elements, looked at by itself, may not mean an immediate crash. Buffett is building cash, but that does not mean tomorrow. The Shiller CAPE is just a warning indicator, not a timer. The circular financing of AI could continue for years. A weakening dollar can be survived. Private credit problems are geographically and sector-specific.

But when you put them all together, they create a picture we have not seen since 1999. Stock markets are historically overheated by every measure available, while the foundations of the financial system are quietly cracking in the background.

This does not mean that investors should sell everything in a panic. That kind of behavior has almost always led to wrong decisions in history. It does mean that this is the moment when every thoughtful investor should ask themselves one simple question:

If I had cash today, would I buy these stocks at these prices?

If the answer is no, then it makes sense to think about gradually reducing exposure. About taking some profits off the table. About spreading risk into asset classes that are not part of the AI story. Real assets. Precious metals. Energy companies with real cash flows. Selected investments in emerging markets. Not as an all-or-nothing decision, but as a step-by-step rebalancing of a portfolio that may have become too dependent on a single story over the past few years.

There is also a practical side to all of this that often gets overlooked. In times like these, holding some cash is not a sign of weakness or fear. It is optionality. Buffett did not build a 397 billion dollar cash pile because he is pessimistic about the world. He built it because he knows that when corrections come, the best opportunities appear suddenly and disappear just as fast. Those who have cash ready will be able to buy quality companies at fair prices. Those who are fully invested will only be able to watch.

Buffett’s message from this year’s meeting is simple. When prices no longer offer a margin of safety, waiting itself becomes an investment decision. It is not the most exciting choice. It will not give you bragging stories over coffee with friends. But it is often the choice that preserves capital when markets remind everyone that gravity also applies to stocks.

2026 will probably be remembered as one of those years when history repeated itself in the same old pattern. First euphoria. Then realization. Then a sharp move that separates the prepared from the unprepared. Who ends up on the right side will not be decided by courage. It will be decided by caution and diversification. As it always is.

Matjaž Štamulak
Independent financial advisor, investor and lecturer
www.cresus.si

This article is for informational purposes only and does not constitute investment advice or a recommendation to buy or sell any financial instruments.