The S&P 500 now sits just beneath its all-time high. Bitcoin and gold are pressing up against theirs too. Investment-grade credit spreads are as tight as they were when Bill Clinton was supposedly only smoking cigars in the Oval Office. The Treasury market is serene to the point of somnambulance. As the Bard might have asked, is something wicked this way coming?
Unprecedented concentration is a worrying sign
Several features of today’s markets look undeniably unusual. First is the unprecedented level of concentration. A small number of technology juggernauts continue to generate an outsized proportion of returns, leaving all other equities in their wake. While there are tentative signs of a broader rally taking shape, with market participation improving modestly in recent weeks, the top line performance since the April nadir has been overwhelmingly down to a handful of names.
According to Bloomberg data, the 10 largest companies in the US — Nvidia, Microsoft, Apple, Amazon, Alphabet, Meta, Broadcom, Tesla, Berkshire Hathaway and JPMorgan Chase — now account for more than 40% of total S&P 500 market capitalisation, and a staggering 56% of returns since US President Donald Trump’s disastrous Liberation Day announcement on 8 April 2025.
What is perhaps even more remarkable is not just their dominance in terms of performance, but also in sheer profit growth. Over the past 12 months they have generated 55% of the growth in net income across the US index. Older readers will be alarmed to hear that market concentration, by the metric of percentage of total market capitalisation, comprised by the largest companies, surpasses even the extreme levels seen at the peak of the dotcom bubble in 2000.
Strong get stronger
Yet, market concentration alone is not an indication of an impending crash. Perhaps these companies are just superior to any others, and we live in a sad new form of capitalism where the strong only get stronger and to the victors go all the spoils.
Today’s tech leaders, Tesla aside, are highly profitable cash machines that are also reinvesting in growth at a previously unimaginable level. Their valuations, from 23 times future earnings like Alphabet and up to 57 times for Broadcom, look expensive but far removed from the levels in the early 2000s when Cisco traded at 85 times earnings and Oracle was at 90 times. Back then, too, it was the blow-up of unprofitable bubble stocks (immortalised by pets.com) that decimated many.
Such a dynamic does not exist today. The companies now leading the market are genuine industrial powerhouses with real earnings, obscene free cash flows and monopolistic (or at least oligopolistic) economic moats. Could we see Nvidia, which has a market cap of nigh on $4.5-trillion (roughly nine times SA’s GDP), double again?
Reasons to worry
Nevertheless, it would be complacent to assume that these businesses are impenetrable. Should the US enter a recession and their cash flows be affected, multiples could rerate downwards. After all, much of their valuations are related to vast AI spending that largely consists of the Magnificent 7 simply buying and selling from each other. In this regard, the market looks rather like an AI Ponzi scheme. Were these stocks to sell off to simply their historical mid-cycle averages at only two-thirds of current levels, the impact on the broader index would be brutal. This is the risk from so much of the market’s current value being down to a handful of companies.
The second red flag is the reappearance of exuberance among retail investors. Meme-stock behaviour, so emblematic of 2021, is back. Retail trading volumes have surged with retail trading on margin calls, an especially worrying indicator, at all-time highs and now exceeding $1-trillion for the first time, according to Financial Industry Regulatory Authority (Finra) data. If it looks like a bubble, and sounds like a bubble, it probably is a bubble.
Finally, and perhaps most concerning, is that markets no longer seem to be a useful or relevant gauge for the economy. The signs are clear that the US economy, and therefore probably the global economy, is slowing, even before the effects of sharply higher tariffs feed through. Yet equity markets, driven by the tech leviathans, simply shrug it off.
Historically the markets have been a leading indicator; where cyclicals (capital goods, industrials, consumer, banks) lead, the economic data would follow. Now one must ask whether the two can live in total separation. Could we go into recession without markets blinking? Is the market now just 10 stocks, three themes (AI, banks, European defence) and one factor (momentum), and therefore not really a market at all? How sustainable is this, and if not, what will be the event that causes Wile E. Coyote, now hovering with legs spinning, to plunge into the abyss?
South African equities are potentially in an even more vulnerable position than the US and European markets. They are also at record highs and exhibit even greater concentration. Bloomberg data shows that the top 10 names on the JSE now make up an extraordinary 55% of total market capitalisation, tied largely to Chinese tech (Naspers and Prosus), gold (Gold Fields and AngloGold) and financials (notably Capitec, on a peaky 30 times earnings). A global downturn would almost certainly cause those valuations to rerate, and this is before one even considers the dismal domestic economic growth outlook.
Why the market is not about to crash
And yet, these concerns, while valid, do not mean one should necessarily panic sell. A couple of provisos are worth remembering. First, periods of speculative ebullience tend to endure for far longer than sceptics expect. As Keynes famously remarked: “The market can stay irrational longer than you can stay solvent.”
For example, equity markets felt overheated in the mid-1990s, yet the dotcom boom thundered relentlessly until early 2000. Bubbles do not simply run a little too far, they run orders of magnitude further than seems even remotely plausible. The last meltdown was in 2020, or possibly 2022. That was not very long ago.
Furthermore, anecdotal signals from the everyday world remain muted. During previous bubbles, right before they were about to collapse (2001, 2007, 2021), frenzied market chatter spilled into everyday conversation; taxi drivers offered their Robinhood trading tips, Reddit feeds fizzed with Fomo, hairdressers debated the merits of bitcoin and Ethereum and Sotheby’s auctioned Bored Ape NFTs. Yet today a strange calm persists. That absence of mania, which is admittedly almost impossible to quantify, is perhaps the most encouraging sign of all.
Equity markets are peculiar, concentrated and — at these valuations — not without their risks. But while they may have advanced far, if the earning power of the tech hegemons continues, and in the absence of full-blown retail hysteria, they also have room to run further.
In that sense, perhaps then the market is less a raving lunatic and more an ageing rock star; still capable of delivering a blowout performance, even if the moves are getting riskier, the outfits more ridiculous and the chance of an on-stage collapse increasingly hard to ignore. One need not stay for the encore, but it would take a heartless cynic to declare the show already over. DM
To read more business stories, click here.