Yesterday morning I tweeted the statistic that 84% of the stocks in the S&P 500 Index have underperformed the return of the index itself over the past year. That's a pretty shocking level of concentration within the index, which has been a hot topic of conversation around here lately, given its implications for some of our portfolio holdings.
Shortly after, I found myself listening to the following discussion with Rob Arnott, founder and chairman of the board of Research Affiliates, on the Monetary Matters with Jack Farley podcast. In it, Arnott discusses the similarities between the dot-com bubble of the late 1990s and the AI bubble he believes is unfolding today. Importantly, he sees several of the same arguments being applied to today's market-leading stocks.
“When I say troublesome, I mean the fact that the Magnificent Seven are highly profitable and are in fact very big, successful businesses is a positive if you want to bet on cap-weighted markets and if you want to bet on growth. These are good, solid companies that are truly path-breaking.
"The downside is that the concentration in the market is without any historical precedent. The allocation to the five most valuable stocks in the country and in the world is just under 30% of the S&P 500. Where was it at the top of the dot-com bubble? 16%.
"We have almost twice the concentration that we had at the highly concentrated top of the dot-com bubble, which means these companies had better not be knocked off their perch. And the simple fact is that history is full of technological revolutions where pioneers and disruptors get disrupted by new pioneers. We've seen just two examples that I alluded to that are more or less this year, where people are dumping Google in favor of ChatGPT as their search engine, and in which ChatGPT is being dumped in favor of DeepSeek, that sort of thing.
"So the simple fact is disruptors get disrupted, and change happens slower than people think."One anecdote that I think is relevant here, the chairman of Cisco in March of 2000 said, "I don't see any reason that Cisco's sales and profits can't grow 40% a year as far as the eye can see. Now, what is 40% growth? After five years, you're six times the size. That's a bunch.
"Where is Cisco today? It's six times the size that it was in 2000. It had 8% growth, not 40% growth. 8% growth is terrific. For a quarter century, wonderful. But it was priced to reflect those lofty expectations with the result that to this day, Cisco has never exceeded its March 2000 peak price.
"Qualcomm is an even better example. Qualcomm was the biggest gainer in 1999. It was up 2700% in a single year. The biggest gain worldwide. It entered 2000 at a valuation multiple of 280 times earnings. How has it done since then as a business? It has grown 60-fold.
"Its sales and its profits are up roughly 60-fold in the last quarter century. That is astonishing. And that's enough that the share price has exceeded the 2000 peak, but not exceeded S&P returns. So it's had returns two-thirds that of the S&P over the last quarter century, even though it's grown 60-fold."That's the issue with bubbles. You have Palantir, fantastic company, visionary leadership, and yet it peaked at, I think it was 140 times trailing 12-month sales.
"Scott McNally, in the dot-com bubble, testifying before Congress in the year 2002, I think it was. He was being challenged for, why did you sell a bunch of your stock at the top in 2000? What did you know?
"And his response was, 'We were trading at 10-time sales. How does a company deliver a shareholder's money back over, let's say, the next 10 years if you're priced at 10-time sales? Assuming a steady state, you're going to have to return 100% of revenues to your shareholders. You'll have to pay zero for cost of goods and services. You'll have to pay no taxes, which is sort of illegal,' I think he said. And that's simply not plausible. What were people thinking?"So, he basically said, 'I sold because 10-times sales makes no sense.' Well, what does 100-time sales tell you? It tells you either that they're going to see 40% growth as far as the eye can see for a decade or more, or that they're a frothy bubble.
"Look, the outperformance of these companies has slowed drastically. And it's the opposite of bottom-bouncing. It's kind of top-bouncing. That's kind of dangerous. And so I would steer clear."Whoever coined the expression 'Magnificent Seven' clearly didn't see the movie. Four of the seven are dead at the end of the movie. I don't want to pick which of the four is going to be dead 10 years from now.
"No, actually, I don't think any of them will be dead, but I think five or six of the seven will have been gravely disappointing to investors 10 years hence.”
Valuation as the altitude of the plane
Options guru Cem Karsan often compares market valuation to the altitude of a plane, saying that valuations are like altitude: They don't really impact the market as long as liquidity (i.e., the market's engines) is still firing and driving prices higher. As long as the engines run, the plane can stay aloft even at very high altitudes, with expensive valuations not causing immediate problems.
The impact of that high altitude isn't felt until liquidity dries up and the market engine stalls (recessions are a key risk). At that point, valuations (altitude) suddenly matter — a plane at a higher altitude faces a more dangerous descent if engines fail. That's why a bear market starting from extreme valuations, as was the case in 2000-2002, is likely to be more painful than one starting from more modest valuations.
If that sounds vaguely familiar, it may be because SMI has long taught something similar: Current valuations tell us a lot about what type of returns we should expect over the next 7-12 years, but they're essentially useless as a short-term timing indicator. High valuations can persist indefinitely without seeming to matter, only to then matter a lot once a bear market finally commences.
Holding conflicting thoughts in our head
One of the hardest parts of being a successful long-term investor is having to sometimes hold conflicting ideas in your head. On the one hand, it's hard to look at the current market's concentration, the valuations and projections of some of its leading companies, and the typical historical pattern of previous technological breakthroughs, without feeling some trepidation about the possibility that we're experiencing a market bubble.
On the other, exiting the market because of such concerns typically doesn't work out well, because bubbles tend to go on longer and inflate far bigger than most investors expect.
We've written numerous times this year that the economy seems sound, and that with the Fed poised to start an interest rate cutting cycle on top of the already significant fiscal deficit spending, it's tough to be bearish about the current near-term setup. This bubble could definitely end badly — eventually. But could the market keep climbing another few quarters or even years? Absolutely.
This is especially true given the debasement element of the current setup. As a currency gets debased, its purchasing power declines. However, real assets, including stocks, tend to increase in value to offset that debasement of the currency. In other words, in such a scenario, you may be losing purchasing power even as stock prices continue to climb.
SMI's approach to dealing with the possibility of bubble conditions is to use trend-following signals that help us recognize when market trends shift and we need to pivot our portfolios. This doesn't necessarily mean selling stocks and going to cash, although that can be an element at extreme points. During the 2000-2002 bear market that lasted nearly three years, our systems helped us rotate from growth stocks to value, and from large companies toward more small-caps. Both of those moves helped reduce losses. Our Dynamic Asset Allocation strategy employs a different process to steer us out of stocks during bear markets while moving into asset classes such as cash, bonds, and gold that typically hold up better.
Best of all, we don't have to anticipate (i.e., guess) when an inflating bubble might burst. That's a losing game. We just have to keep following our strategy trend signals. If we are indeed witnessing an inflating bubble, this approach will let us harvest all of its gains, while likely avoiding at least some of the ensuing bear market losses when the party ends.
