The S&P 500 is treated as “the market,” but it is a curated list. A committee of S&P Dow Jones Indices staff decides who is in it, the weights come from a float-adjusted rule rather than raw size, and the index now leans heavily on a handful of names. This note covers the mechanics of inclusion, how weights are set, and the risks that follow from the design. For the broader vehicle that most people actually buy, see Index Fund.
A short history
The lineage runs back to the Standard Statistics Company, which built a 233-stock weekly index in 1923 and a 90-stock daily index in 1926 (50 industrials, 20 rails, 20 utilities). On March 4, 1957, after Standard merged with Poor’s Publishing, the index expanded to 500 names: 425 industrials, 60 utilities, 15 railroads, covering roughly 90% of US market value. A Wealth of Common Sense traces the timeline.
The other structural change came in 2005, when S&P shifted from total market-cap weighting to float-adjusted weighting in two steps: half-float in March, full float in September. Before that, a company’s full share count drove its weight even if most shares were closely held and never traded.
How weights are set
Weight is float-adjusted market cap, not raw market cap. Each company’s tradable share count is multiplied by price, where “tradable” excludes blocks held by insiders, governments, and other corporations holding strategic stakes. The Investable Weight Factor (IWF) captures that count; a company must have an IWF of at least 0.10, meaning at least 10% of shares are public.
Because weight tracks price, money flowing into the index buys more of whatever has already risen. That mechanic, together with the megacaps’ actual earnings growth outrunning the rest of the index, has concentrated the index over the past decade. As of August 2025 the top 10 holdings were about 40% of the index, roughly double the ~20% level of 2015. The “Magnificent Seven” (Apple, Microsoft, Nvidia, Alphabet, Amazon, Meta, Tesla) sat near 34% as of December 2025, up from 12% in 2015. A buyer of an S&P 500 fund is making a much larger bet on seven companies than the word “diversified” implies. This is the concentration limit of diversification: adding the other 490 names does little once a few dominate the weight.
How stocks get in and out
The index is maintained by the U.S. Index Committee, a group of full-time S&P Dow Jones staff that meets monthly. Selection is discretionary. Meeting the criteria makes a company eligible, not entitled.
To be eligible a company must:
- be US-domiciled and listed on an approved US exchange,
- have a market cap above a periodically updated floor (about $22.7 billion as of 2025),
- have positive GAAP earnings in the most recent quarter and summed over the trailing four quarters,
- have a 12-month trading history and an IWF of at least 0.10.
Closed-end funds, ETFs, ADRs, limited partnerships, and SPACs are ineligible. The profitability screen is why SpaceX and other large but unprofitable firms stay out, and why the committee rejected a 2026 proposal to relax it.
The committee also manages sector balance, comparing each sector’s index weight against the broader S&P Total Market Index and sometimes passing over a qualified candidate to avoid overweighting a sector. Removals are mostly mechanical, not punishments: the common trigger is a corporate action, an acquisition that delists the shares, a spin-off, a domicile change, or bankruptcy. Changes are announced at least three business days ahead.
The index effect, and its fade
When a stock is added, index funds must buy it, a forced demand shock against fixed supply. Historically this produced a measurable jump. Greenwood and Sammon’s The Disappearing Index Effect measured the average price impact around inclusion at 3.4% in the 1980s, rising to 7.4% in the 1990s, easing to 5.2% in 2000-2009, and falling to about 1.0% in 2010-2020.
The effect shrank as the rebalance became predictable. Arbitrageurs front-run the announcement, buying additions early and pulling the return forward, which compresses the premium left at the reconstitution date. The flip side is a documented execution cost for passive holders: trade volume on reconstitution dates spikes sharply (the close can see volumes many multiples of normal), and funds that mechanically trade at that moment pay for the crowding. Greenwood and Sammon also find discretionary deletions tend to outperform additions in the year after the change, the opposite of what naive demand pressure would predict.
Risks and criticisms
Concentration. A cap-weighted index has no cap on how large its largest members get. At ~40% in the top 10, the index’s return is increasingly a bet on a few stocks and, given how correlated the megacap tech names are, effectively a bet on one factor. The diversification a broad index is supposed to provide thins out exactly as the top gets heavier.
Passive flows and price discovery. As more capital indexes, more buying is driven by inflows rather than by judgments about value. Mike Green’s argument, built on the inelastic markets hypothesis of Gabaix and Koijen, is that flows now move prices more than fundamentals do, because a dollar into the index buys shares regardless of price. See Inelastic Markets Hypothesis. The market starts to behave like a low-float stock, where small demand changes cause large price swings, and the information content of prices erodes. This is contested: index funds are a minority of total trading, and active managers still set prices at the margin. But the direction is not disputed, only the magnitude and the breaking point.
Governance and discretion. A small committee decides membership for a benchmark that trillions of dollars track. That is a single point of judgment, and additions move prices, so the decision has real market power. Tesla’s November 2020 inclusion was one of the largest single additions ever and was telegraphed weeks ahead, with heavy front-running into the event.
It is not the whole market. The S&P 500 is large-cap US equity. It excludes small caps, most mid caps, and all foreign listings. “I own the S&P 500” is a concentrated large-cap US position, not the diversified global portfolio the phrase implies. The distinction between spreading risk and the kind that actually offsets a loss is the subject of Diversification vs Hedging.
Try it
Rebuild the weights and see the concentration (1-2 hours, Python). Pull current S&P 500 constituents and their float-adjusted market caps (a free constituent list plus yfinance for marketCap gets you close). Sort descending, compute the cumulative weight share, and plot the Lorenz-style curve. You are looking for how few names cross 40% of the total. Then compute the index return two ways for the last year, cap-weighted versus equal-weighted, and compare. The gap is the concentration effect: when the top names lead, cap-weighting wins; when they lag, equal-weighting does. The Invesco RSP (equal weight) versus SPY (cap weight) spread is the same comparison if you would rather use real fund data.
See also
- Index Fund — the broader vehicle: what index funds are, why they won, and the structural critiques.
- Systematic and Idiosyncratic Risk — why adding names stops cutting risk once a few dominate the weight.
- Diversification vs Hedging — owning the index spreads risk but does not hedge a market-wide drawdown.
- Systems Thinking — price-tracking weights and mechanical inflows form a reflexive feedback loop, not a one-way reaction.
Sources
- The Disappearing Index Effect — Greenwood and Sammon (HBS working paper) on the inclusion premium by decade and why it faded.
- S&P U.S. Indices Methodology — the primary source for eligibility criteria, IWF, and committee process.
- Inside the S&P 500: Float Adjustment — S&P’s own explanation of the 2005 weighting change.
- Is there a rising concentration risk in the S&P 500? — Guinness Global Investors on top-10 weight over time.
- Why Passive Investing Is Not Passive — Mike Green’s case built on the inelastic markets hypothesis.