The performance of the Magnificent Seven stocks has been astonishing in recent years, strongly outperforming the rest of the market. On 29 October last year, their combined value surpassed USD 20 trillion (!), more than a third of the total value of all 500 stocks in the S&P 500. Some have been uneasy about this remarkable run, yours truly included. Since hitting that milestone, Mag7 stocks have fallen, while the rest of the US market has risen. Does this mean the Magnificent Seven’s glory days are over, that they are no longer so magnificent — and what comes next for the market?
I admit it: like so many others in financial markets, I have been fascinated by the performance of the Magnificent Seven — Alphabet (‘Google’), Amazon, Apple, Meta (‘Facebook’), Microsoft, Nvidia and Tesla. They have done exceptionally well and have dominated the market so clearly that, in effect, they have been driving it.
Percentage performance
Since 2019 — to choose a starting point — the combined market value of the Magnificent Seven stocks has increased by 500 per cent, as Figure 1 shows.
What about the rest of the market? While these have been good years for the broader market also, its performance pales in comparison. The remaining 493 stocks in the S&P 500 have risen by 130 per cent.
Figure 1. Total market value of Magnificent 7 companies and the remaining 493 companies in the S&P 500, both normalized to “1” on January 1, 2019. Daily data, Jan. 2019 – 12 March 2026. Source: Datastream via Refinitiv and J.Rangvid.
The aggregate value of the 493 non-Mag7 stocks have grown by around 12 per cent per year on average (continuous compounding) since 2019 — in fact, a very strong performance.
But again, this pales in comparison with the value of the Mag7 stocks, which have grown at an average annual rate of around 30 per cent for seven consecutive years. It is crazy. Or rather, it is fascinating.
Dollar performance
In 2019, the combined value of the Mag7 stocks was around USD 3 trillion. Today, it stands at roughly USD 18 trillion — the 500 per cent increase referred to above.
The 493 non-Mag7 stocks were worth around USD 18 trillion in 2019. Today, they are worth approximately USD 40 trillion. Figure 2 illustrates these developments.
Because the Mag7 have grown so rapidly, they now account for around 30 per cent of the S&P 500, compared with about 14 per cent in 2019.
Figure 2. Total market value of the Magnificent Seven companies and the remaining 493 companies in the S&P 500, USD billions. Daily data, January 2019 – 12 March 2026. Source: Datastream via Refinitiv and J.Rangvid.
Cracks starting to appear?
I have been fascinated by this performance, but I have also been uneasy. I have questioned its long-term sustainability. I have never shouted ‘run away’, because I know it is dangerous to fight the market, but I have warned about the hype.
In November last year, I wrote a couple of posts about the risks surrounding the highly prized AI stocks (link, link). The timing was not bad. The Mag7 have been underperforming since then.
(I have written other articles about the Mag7 — link, link — but my November posts were my clearest warnings.)
If you look very closely at Figure 2, you may be able to spot the Mag7 peak on 29 October last year. On that day, the combined market value of the Magnificent Seven reached — and briefly surpassed — USD 20 trillion.
I tend to believe in largely rational markets. Still, there is something symbolic about the much-followed and much-discussed Magnificent Seven just reaching USD 20 trillion — and then the decline beginning.
Figure 3 zooms in on the period since the peak on 29 October 2025. The Magnificent Seven stocks are down by more than 10 per cent. By contrast, the remaining 493 non-Mag7 stocks are up around 1 per cent, although they have borne the brunt of the Iran war’s consequences. The Mag7 have thus underperformed by more than 10 percentage points since autumn. Given their extraordinary outperformance in recent years, this is a remarkable turn of fortunes.
Figure 3. Total market value of Magnificent 7 companies, the remaining 493 companies in the S&P 500, and all companies in the S&P 500, all normalized to “1” on 29 October 2025. Daily data, October 2025 – 12 March 2026. Source: Datastream via Refinitiv and J.Rangvid.
Since the non-Mag7 stocks account for roughly two-thirds of the S&P 500 and the Mag7 about one-third, a 1 per cent gain in the former combined with a 10 per cent decline in the latter results in an overall market decline of about 3 per cent. As Figure 3 shows, this drop in the total market has occurred since the outbreak of the war.
Some of you may be wondering who the main culprit is. Have all seven Mag7 stocks fallen, or only a few? Figure 4 shows that most of the decline is attributable to Microsoft (down by around USD 1 trillion), Nvidia (down by roughly USD 700 billion), and Amazon (down by about USD 400 billion). All staggering sums.
Figure 4. Market value of the Magnificent Seven companies, USD billions. Daily data, October 2025 – 12 March 2026. Source: Datastream via Refinitiv and J.Rangvid.
Nvidia also reached its peak on 29 October. On that day, it was valued at more than USD 5 trillion, also a symbolic number and the most valuable company ever.
Shortly after surpassing that symbolic threshold, the decline began. At the time of writing, Nvidia is valued at around USD 4,400 billion. Still an enormous company, of course — but about 15 per cent below its peak.
A scary comparison with the dot.com bubble
These are still early days. The downturn in the Mag7 stocks has lasted only a few months. Perhaps they will recover.
Or perhaps they will not.
Figure 5 shows CAPE, the cyclically adjusted price–earnings ratio. As you will recall, CAPE tells us what it costs to buy one dollar of average real earnings per share over the past ten years. Today, it stands at 40 — close to an all-time high. Only at the peak of the dot.com bubble did CAPE reach a higher level, at close to 45. As you can also see in Figure 5, after that peak CAPE fell dramatically, driven by a sharp decline in share prices. Historically, CAPE has been a useful indicator of major turning points in the US stock market.
Figure 5. CAPE (cyclically adjusted price-earnings ratio). Monthly data, Jan. 1881 – March 2026. Source: R.Shiller’s webpage and J.Rangvid.
That is one comparison with the dot.com bubble. Another lies in the run-up to it.
In Figure 6, I show how the S&P 500 behaved in the build-up to the bubble at the turn of the millennium, and how it performed in the aftermath. I overlay this with the performance of the S&P 500 since it bottomed out following the 2022 correction triggered by the post-pandemic surge in inflation and interest rates.
Figure 6. Development in the S&P 500 since January 1997 for the dot.com period and since October 2022 (upper x-axis), “Today”, both normalised to “1” in the first month. Monthly data. Source: R.Shiller’s webpage and J.Rangvid.
As you can see, the behaviour of the S&P 500 in the late 1990s and since October 2022 is strikingly similar:
- In the late 1990s, the S&P 500 rose by 90 per cent over a little more than three years. Over the past three-plus years, it has risen by roughly the same amount.
- In the late 1990s, the strong equity market was driven by optimism about a new computer-based technology — the internet. Today, it is driven by optimism about another transformative technology — AI.
- Then, as now, valuations were very high (see Figure 5).
The worrying part, of course, is what followed in the summer of 2000: share prices fell by around 50 per cent. The key question is whether history will repeat itself — that is, whether the S&P 500 will continue to track the pattern of the late 1990s and early 2000s.
Fortunately, there are important differences, not least that some companies during the dot.com era traded at truly extreme valuations. The Mag7 also trade at elevated valuations, but not at the extraordinary levels seen 25 years ago. Put simply, the Magnificent Seven generate substantial profits, whereas many dot-com companies did not. Moreover, Figure 6 begins at the trough of the 2022 pull-back. In a previous analysis, I showed that if one starts the comparison in 1995 rather than 1997, the current rally appears to be less dramatic in comparison with that of the late 1990s (see Figure 6 here: link).
On the other hand, the market is more concentrated today, meaning that the remaining companies must carry a heavier share of the load, as they have done since 29 October 2025.
Conclusion
The Magnificent Seven companies have performed outstandingly in recent years. They surged to reach the symbolic milestone of USD 20,000 billion on 29 October 2025, with Nvidia alone touching the equally symbolic USD 5,000 billion mark. Since achieving these peaks, their combined value has fallen by around 10 per cent.
The broader market has not tanked, however, as other stocks have delivered solid performance. What lies ahead? No one knows. There are certain flashbacks to the dotcom bubble in all this, though there are also notable differences. For that reason, a reasonable assumption is that we avoid the dramatic crash that followed the bursting of the dotcom bubble. That said, it would not be surprising if valuations were to ease through several years of more subdued returns.
