Mag 7 at 34% of S&P: Worse Than 2000, Bigger Than Nifty Fifty
The Mag 7 now represent roughly 34% of the S&P 500. That is more concentrated than the Nifty Fifty peak and the 2000 tech peak. The scenario math.
The Magnificent Seven (Apple, Microsoft, NVIDIA, Alphabet, Amazon, Meta, and Tesla) now represent roughly 34% of the S&P 500 by market capitalization, per the S&P 500 Factsheet published by S&P Dow Jones Indices. That is the highest index-level concentration of a small cohort of names in the modern history of the index.
Two historical comparisons anchor the discussion. At the March 2000 tech peak, the top seven information-technology names represented roughly 22% to 24% of the S&P 500, per S&P historical concentration work referenced in J.P. Morgan’s Guide to the Markets Q2 2026. In the early 1970s, the “Nifty Fifty” cohort of one-decision growth stocks peaked at roughly 42x forward earnings against a broader S&P forward P/E near 19x, per Jeremy Siegel’s work in Stocks for the Long Run. The Nifty Fifty was a valuation-concentration event more than a market-cap-concentration event; those 50 names were at most a quarter of the index by weight.
In our view, the Mag 7 concentration is not a bubble call on the underlying businesses. The Mag 7 members are profitable, cash-generating, and fundamentally different from the speculative names of 1999 or the conglomerates of 1972. But concentration is about the distribution of portfolio outcomes, not the quality of the underlying holdings, and in our view the risk that index-level returns become dominated by a small group’s multiple path is the part of the 2026 setup investors should understand.
This is a Research piece. We are putting historical concentration episodes next to today’s data and walking through the scenario math on what multiple compression in the Mag 7 would mean for the S&P 500.
The Concentration Table
Here is how the top cohort has stacked up at three historical extremes and today.
| Regime | Date | Top cohort | Share of S&P 500 | Top cohort forward P/E | Rest of S&P forward P/E |
|---|---|---|---|---|---|
| Nifty Fifty peak | Dec 1972 | Top 50 growth names | ~24% | ~42x | ~19x |
| Tech peak | Mar 2000 | Top 7 tech | ~22-24% | ~56x | ~24x |
| Mega-cap retracement | Nov 2021 | FANMAG | ~24% | ~30x | ~20x |
| Mag 7 today | Apr 2026 | Top 7 | ~34% | ~28x | ~18x |
Sources: S&P Dow Jones Indices, J.P. Morgan Guide to the Markets Q2 2026, and FactSet Earnings Insight.
Two observations. First, the Mag 7 share of the index is meaningfully higher than the top-cohort share at either 2000 or 1972. In weight terms, today’s setup is more concentrated. Second, the Mag 7 multiple is more reasonable than the 2000 cohort’s multiple (roughly 28x forward versus 56x), and closer to the Nifty Fifty’s 42x. The 2026 picture is a very-concentrated-but-more-reasonably-valued version of prior episodes.
The Earnings Contribution
Concentration of earnings matches the concentration of weight. FactSet Earnings Insight has tracked Mag 7 contribution to S&P 500 EPS growth for the last several quarters. In 2024, Mag 7 members contributed an estimated 55% of the index’s year-over-year EPS growth. In 2025, that share was roughly 45%. The group is still driving the marginal earnings dollar, but the contribution share is declining as other sectors recover.
That is a soft piece of good news. The rest of the S&P is no longer dormant. But the cap-weighted math is still dominated by the 7 names.
Three Historical Deconcentration Drawdowns
What has happened to concentrated cohorts when they mean-reverted? Three episodes.
Nifty Fifty, 1972-1974. The top cohort lost 60% to 90% in the 1973-74 bear market, with names like Xerox, Polaroid, and Avon falling well over 60% from peak despite continued earnings growth, per Siegel’s Stocks for the Long Run. The broader S&P 500 fell 42% over the same window. The Nifty Fifty underperformed the index by a meaningful margin even though the reason for the decline was multiple compression, not fundamentals.
Tech 2000-2002. The top 7 tech names (MSFT, CSCO, INTC, ORCL, IBM, LU, GE) declined an average of roughly 74% from peak to trough, while the S&P 500 as a whole fell 45%, per FRED and S&P data. The difference, about 29 percentage points, represents the concentration premium paid on the way up, unwound on the way down.
Mega-cap 2021-2022. The Mag 7 and FANMAG cohorts experienced a 32% drawdown in 2022, while the S&P 500 fell 18%. The episode was brief (roughly 10 months peak to trough) and reversed completely by early 2024. This was a rates-driven multiple compression, not a fundamental break, and the group recovered.
In all three cases, the concentrated cohort underperformed the broader index during the deconcentration. The magnitude of the underperformance scaled with the starting concentration premium.
Scenario Math: What Multiple Compression Would Mean
Here is the arithmetic that anchors the Research question. Hold earnings constant. Apply a multiple compression to the Mag 7 alone. Keep the rest of the S&P 500 at its current forward multiple of roughly 18x. What does the index do?
Assumptions for the calculation:
- Mag 7 weight: 34%
- Rest-of-index weight: 66%
- Mag 7 current forward P/E: 28x
- Rest-of-index current forward P/E: 18x
- Earnings held constant across scenarios
The index-level forward P/E today, blended, is approximately: (0.34 × 28) + (0.66 × 18) = 9.52 + 11.88 = 21.4x.
Now the three scenarios.
Scenario 1: 10% multiple compression on Mag 7 only. Mag 7 forward P/E falls from 28x to 25.2x. Blended index P/E becomes (0.34 × 25.2) + (0.66 × 18) = 8.57 + 11.88 = 20.45x. Implied SPX decline (holding earnings constant): (20.45 − 21.4) / 21.4 = -4.4%.
Scenario 2: 20% multiple compression on Mag 7. Mag 7 forward P/E falls from 28x to 22.4x. Blended index P/E becomes (0.34 × 22.4) + (0.66 × 18) = 7.62 + 11.88 = 19.50x. Implied SPX decline: (19.50 − 21.4) / 21.4 = -8.9%.
Scenario 3: 30% multiple compression on Mag 7 (reverts toward 2021 mid-cycle levels at ~19.6x). Mag 7 forward P/E falls from 28x to 19.6x. Blended index P/E becomes (0.34 × 19.6) + (0.66 × 18) = 6.66 + 11.88 = 18.54x. Implied SPX decline: (18.54 − 21.4) / 21.4 = -13.4%.
If earnings also compress (a realistic assumption during a deconcentration event), the implied index decline is larger. A 10% earnings cut on the Mag 7 combined with the 20% multiple compression in Scenario 2 implies an index decline closer to -16% to -17%.
These are not forecasts. They are the translation of “concentration + multiple compression” into index-point math.
What This Means for Passive Allocators
A retail investor with the broad S&P 500 via VOO, SPY, or similar cap-weighted vehicles holds a portfolio that is, in practice, 34% in seven names. That is not a diversification failure of the index construction. That is simply what happens when seven companies become very large.
The implication is about expected portfolio behavior more than what to do. A passive S&P 500 holder should expect:
- Higher correlation with AI-capex sensitivity than at any prior point in the index’s history.
- Lower diversification benefit from adding more S&P 500 exposure on top of existing Mag 7 exposure.
- Greater sensitivity of the index to a small number of earnings reports in the Apr 29-30 and Jul 29-30 windows.
Investors seeking genuine diversification away from this concentration typically look at equal-weight US equities (the Invesco S&P 500 Equal Weight ETF (RSP) is the most common vehicle), mid-cap and small-cap sleeves, and international equity. The trade-off is that those diversified sleeves have underperformed during the concentration build. Positioning for deconcentration means accepting that the group you are diversifying away from may keep working for a while.
What We’re Watching
Four signals that would shift the probability weighting between “continued concentration run” and “deconcentration event.”
Q1 2026 earnings from Tesla and Alphabet, Apr 23. Tesla has the highest multiple in the cohort (around 80x forward per consensus). Alphabet has the lowest (around 21x). Earnings variance at these endpoints is the first read on how wide the Mag 7 distribution is becoming.
Microsoft, Meta, and Apple earnings, Apr 29-30. These three, plus Amazon and NVIDIA later in the cycle, will either validate or disturb the capex-and-margins thesis we discussed in our AI capex piece.
Guidance changes on 2026 capex from any hyperscaler. A reduction in forward capex from any of the four hyperscalers would compress the forward growth assumption embedded in the Mag 7 multiple.
Rates path. The Mag 7 multiple is rates-sensitive. Any meaningful shift in the June-meeting cut probability from the current market pricing would flow through the group’s valuation faster than it flows through the rest of the index.
What Would Change Our View
We would update our read toward more benign if:
- The earnings contribution from the rest of the S&P 500 continues to recover and Mag 7 share of EPS growth falls below 30% sustainably.
- Mag 7 forward multiple compresses toward 22x to 24x without meaningful earnings damage, cleaning up the valuation premium organically.
- Sector breadth in the S&P 500 improves measurably, with the equal-weight index outperforming the cap-weighted index for two or more consecutive quarters.
We would update toward more concerned if:
- Mag 7 forward multiple expands further above 30x without commensurate earnings revisions.
- AI capex grows faster than AI-attributable revenue for another two quarters.
- The top-10 S&P 500 weight rises above 40%, the highest level in at least 50 years of data.
In Our View
In our view, the Mag 7 concentration is not a bubble on the underlying businesses. It is a concentration risk on the index. A passive S&P 500 investor in 2026 owns a portfolio that is 34% in seven highly correlated, AI-capex-sensitive names. The scenario math says a 20% multiple compression on that sleeve translates to a roughly 9% decline in the index, holding earnings constant. A 30% compression plus a modest earnings cut translates to mid-to-high-teens.
Each cycle differs. The Mag 7 setup has better earnings quality than 2000 and better balance sheets than 1972. But concentration is concentration, and the distribution of potential outcomes from here, in our view, is wider than it was two years ago.
For the related question of whether the AI capex cycle that drove the concentration is priced for perfection, see our AI Capex piece. For the Fed-policy backdrop that matters for multiple compression, see our Fed Pause-to-Cut Playbook.
Ferrante Capital LLC is a registered investment adviser. Information presented is for educational purposes only and does not constitute investment advice, a solicitation, or a recommendation to buy or sell any security. All investing involves risk, including the possible loss of principal.
FC and its principals may hold positions in GOOG, TSLA. This analysis is for educational purposes only and does not constitute a recommendation to buy, sell, or hold any security.
Forward-looking statements reflect Ferrante Capital’s current analysis and involve assumptions and estimates. Actual results may differ materially. Past performance is not indicative of future results.
Please consult a qualified financial professional before making investment decisions.
