S&P 500 Index: What It Is, How It Works, and Why It Moves Markets in 2026

NEW YORK — The S&P 500 is not a stock. It is not a fund. It is a calculation — a proprietary, committee-governed methodology that distills the combined market value of 503 of the largest publicly traded companies in the United States into a single number that, on August 26, 2026, stands at 5,618.

That number moves every second the New York Stock Exchange and Nasdaq are open. It has moved 15.2% higher since January 1, making 2026 one of the stronger first-eight-months performances of the post-pandemic era. And it moves in ways that affect not just the investors watching it, but anyone with a 401(k), a pension, or a savings account in a fund that tracks it.

Here is what the index actually is, how it is constructed, and why the structural choices embedded in its design matter more in 2026 than they have in years.

The selection criteria

S&P Dow Jones Indices, a division of S&P Global, maintains the index through a committee that meets quarterly and reviews constituents against a fixed set of criteria. To be eligible, a company must be U.S.-domiciled, have a market capitalization above $20.5 billion, trade on a major U.S. exchange, and have reported positive as-reported earnings over the most recent quarter and over the trailing four quarters combined. Annual dollar trading volume relative to float-adjusted market cap must exceed 1.0.

That last requirement — the profitability test — is why certain high-profile companies have waited years for inclusion. It is why Palantir Technologies Inc. (PLTR), despite a market cap well above the threshold, only gained entry after its earnings trajectory became consistently positive. The committee does not chase momentum. It applies criteria.

Why it is cap-weighted

The S&P 500 weights each constituent by its float-adjusted market capitalization — the value of its freely tradable shares. Larger companies have larger influence. This is not arbitrary. A cap-weighted index reflects where actual investor capital has been placed. It is, in a narrow sense, the aggregate judgment of every market participant about what the largest companies are worth.

The consequence, in 2026, is that five companies — NVIDIA Corp. (NVDA), Apple Inc. (AAPL), Microsoft Corp. (MSFT), Amazon.com Inc. (AMZN), and Alphabet Inc. (GOOGL) — account for approximately 27% of the index’s total weight. NVIDIA alone, after its dramatic appreciation over the past two years, accounts for roughly 6.5%. The NVIDIA earnings report after Wednesday’s close is therefore not merely a company story. A 10% move in NVIDIA’s stock would, all else equal, shift the S&P 500 by approximately 65 basis points.

This concentration is historically unusual. The five largest constituents’ combined weight has not been this high since the dot-com era, a comparison that makes some analysts uncomfortable and others dismissive, on the grounds that today’s largest companies generate real and growing earnings rather than speculative projections.

The equal-weighted alternative

For context, the S&P 500 Equal Weight Index — which assigns the same 0.2% weighting to every constituent regardless of market cap — has risen roughly 9% year-to-date, compared to the cap-weighted index’s 15.2%. The 6.2-percentage-point divergence is a precise measurement of how much of 2026’s gains have been driven by mega-cap technology appreciation versus the broader market’s performance. Most of the 503 companies in the index have not had a particularly exceptional year.

How it differs from the Dow and the Nasdaq Composite

The Dow Jones Industrial Average, the oldest and most-quoted index in financial media, is price-weighted — a methodology that gives higher-priced stocks greater influence regardless of company size. It tracks only 30 companies, hand-selected by a committee, with minimal transparency in the selection rationale. The S&P 500 is far more representative and far more methodologically sound. Institutional investors rarely benchmark to the Dow.

The Nasdaq Composite, by contrast, tracks all equities listed on the Nasdaq exchange — over 3,300 companies — with a heavy technology and growth-stock bias. It captures a different slice of the market: earlier-stage, higher-volatility, sector-concentrated. The Nasdaq-100, a subset, covers only the 100 largest non-financial Nasdaq names. The S&P 500 spans all 11 GICS sectors and includes financial companies the Nasdaq-100 excludes.

How it is calculated

The index level is calculated continuously during market hours using the following formula: divide the aggregate float-adjusted market cap of all 503 constituents by a divisor. That divisor, maintained by S&P Dow Jones Indices, is adjusted whenever corporate actions occur — stock splits, additions, deletions, mergers — to ensure continuity of the index level. The divisor is not published in real time but is available in S&P’s methodology documentation.

The result is a price-return index: it captures changes in constituent stock prices but does not include dividends. The total return version of the S&P 500, which reinvests dividends, is what most financial comparisons use when evaluating long-term investment performance.

Where 5,618 sits in historical context

The S&P 500’s all-time closing high stands at 5,878, reached in February 2026. Wednesday’s 5,618 is 4.4% below that peak — technically a pullback, not a correction (which requires a 10% decline from the high). The 52-week low is 4,607, recorded in October 2025 during a brief Federal Reserve-driven selloff.

From its March 2009 bear market low of 676, the index has returned approximately 730% — or roughly 16% annualized including dividends. That compounding is why the index has become the default benchmark for almost every professional investor and retirement plan in the United States. According to S&P Dow Jones Indices, assets passively benchmarked to the S&P 500 exceed $11 trillion.

The S&P 500’s current session performance heading into NVIDIA’s earnings will determine how the next chapter of that record reads.