In late May, JPMorgan’s Marko Kolanovic argued that the S&P 500 was significantly overvalued, defending his stubbornly negative stance on U.S. stocks and predicting that the large-cap index would fall 24% from then-levels to 4,200 by the end of the year. At that point, Kolanovic was the last bear market strategist at a major Wall Street bank, and other strategists, even some previously pessimistic, were predicting further gains by the end of the year, believing the incredible recovery that began in early 2023 would continue.
Kolanovic was wrong when he predicted another strong stock market in 2022, and the market crashed. Two and a half years of back-to-back wildly wrong predictions likely cost him his job. But was his analysis of market fundamentals wrong? Not at all.
In fact, the circumstances surrounding his departure contain two lessons. First, very short-term predictions about stock direction are worthless. Second, history shows that while periods of high prices can last for long periods, gravity, which dictates earnings growth and price-earnings ratios over the long term, always eventually takes effect and prevails.
Despite the bull market and the exit of previously wrong skeptics like Kolanovic, stocks look dangerously overpriced.
In his latest report, Kolanovic said stocks are simply pricing in much higher earnings growth than he considers appropriate, based on their most modest historical performance and a high starting point. This is a good argument. In the first quarter of this year, the S&P 500’s net earnings, based on the last four quarters, were $192. This is 38% higher than the pre-pandemic record of $140 set at the end of 2019. The famous CAPE model, developed by Yale economist and Nobel laureate Robert Shiller, strongly suggests that $192 is unsustainable and that the earnings “norm” is somewhere around $155.
The problem is that even as earnings remain sky-high, investors are also paying a lot for each dollar of those irreproducible earnings. The S&P is currently trading at 28.9x. Excluding the financial crisis periods when earnings collapsed, this is the highest quarterly PE since the tech bubble of the late 1990s and early 2000s. And in the past 36 years, the S&P’s PE has only been this high for five months, all during this century’s most infamous boom.
This episode is a reminder that prices can break free from what has been justified by fundamentals for years, and so while that is happening, predictions of an imminent decline are always wrong—until they are eventually and inevitably right. Simply put, the outcome is predictable, but the timing is anything but. By June 2017, the Shiller model was warning that the S&P had entered the ultra-rich zone. But in the three years to August 2000, the index rose another 79%, from 847 to 1517. By February 2003, the S&P had effectively round-tripped back to 841, and would not breach 1500 again until a decade later.
Adjust the numbers according to basic market math and they’re just as frightening as what the JPMorgan bears predicted.
Let’s say the market PE falls from the current 29 to 22, which is still high by historical standards. That would send the index down 25% from current levels, roughly the same as the correction predicted by Kolanovic. Alternatively, EPS would likely trend downward (it has been flat for several quarters already) and settle at $155, which seems reasonable given Shiller’s data. In that case, even with a high PE of 27, the S&P would be 25% overvalued at $5,567, the close recorded on July 5th.
Remember, low inflation-adjusted interest rates should be good for stocks because they increase the present value of future earnings, and rising real interest rates reduce the present value of earnings, thus lowering stock prices. Indeed, “real yields” have soared from negative to over 2%, and the S&P has soared. The combination of historically high PEs, presumably inflated earnings, and real interest rates that have gone from extremely favorable to normal does not inspire any confidence that stocks will continue to rise.
Again, the bulls’ predictions may be correct for a long time to come. But over the longer term, science shows that starting at such high prices will only produce modest returns over a decade, or even seven years. It’s guesswork to predict where we’ll be this time next year, but the more the market booms, the more it sows the seeds of its own downfall.
