TL;DR. When a stock is added to the S&P 500, index funds are forced to buy it, and its price typically rises between the after-close announcement and the effective date. Academic studies going back to 1986 have measured this “index effect” at anywhere from a few percent to nearly ten percent in some periods — but the premium has visibly shrunk in the last decade as arbitrageurs, hedge funds, and index-fund trading desks anticipate the moves earlier. Deletion produces the mirror image: a temporary sell-off followed by partial reversion.
The core idea
The S&P 500 is not just a benchmark. Trillions of dollars in index funds and ETFs track it mechanically, meaning they own each of its 500 constituents in proportion to the index’s weighting. When the index committee announces a change, every mechanical tracker must:
- Buy the incoming stock, in enough size to match its float-adjusted market-cap weight in the index.
- Sell the outgoing stock, closing out its previous weight to zero.
Both trades must be substantially complete by the close on the day before the effective date, because that closing price is what the index provider uses to compute the new index level. That creates a concentrated, one-sided burst of demand for the added stock and supply for the deleted stock. The index effect is the price signature of that concentrated flow.
How the S&P 500 actually picks stocks
The S&P 500 is not a rules-only index. A committee inside S&P Dow Jones Indices makes the call, and it applies these baseline eligibility screens (with committee discretion on top):
| Criterion | Threshold |
|---|---|
| Market capitalisation | ≥ US$22.7 billion (as of July 1, 2025) |
| Profitability | Positive net income from continuing operations for the most recent quarter and for the sum of the last four consecutive quarters |
| Liquidity | Annual dollar volume traded / float-adjusted market cap ≥ 0.75; minimum 250,000 shares in each of the six months before evaluation |
| Exchange | NYSE (incl. Arca/American), Nasdaq (Global Select, Select, Capital) or Cboe BZX/BYX/EDGA/EDGX |
| Domicile & security type | U.S. company; common equity only (no MLPs, no closed-end funds, no ETFs, generally no OTC) |
| Selection authority | Discretionary decision of the S&P U.S. Index Committee |
Two consequences flow from the committee’s discretion. First, satisfying every criterion does not guarantee inclusion — Tesla waited months after it qualified before finally being added in December 2020. Second, the committee’s announcement is genuine news: markets cannot perfectly forecast which of many eligible candidates will be picked and when. That uncertainty is the fuel for the index effect.
The lifecycle of an inclusion: what actually happens
A typical S&P 500 addition follows this timeline:
- Announcement (usually Friday after close, about 5–10 trading days ahead of the effective date). S&P DJI publishes a short press release naming the incoming and outgoing companies.
- Overnight and next-day pop. The stock gaps up in after-hours and opens sharply higher the following session. This is the “announcement return” that Harris & Gurel (1986) first quantified at roughly +3 % in their 1976–1983 sample.
- Run-up to the effective date. Over the following days, arbitrageurs, hedge funds and eventually the index funds themselves accumulate the stock. Petajisto (2011) measured the average total price impact from announcement to effective day at +8.8 % for S&P 500 additions in a 1990–2005 sample, versus +4.7 % for Russell 2000 additions.
- Effective-date close and post-effective decay. Once the mechanical trackers finish rebalancing, the buying pressure disappears. A meaningful fraction of the run-up typically reverses over the following weeks, which is what Chen, Noronha and Singal (2004) called the “transient” component of the effect.
A worked example
Suppose stock XYZ trades at $100, has 200 million shares outstanding, and is announced for inclusion. Its float-adjusted market cap is $20 billion, giving it an initial S&P 500 weight of roughly 0.04 % (against a ~$50 trillion float-cap index total).
If passive S&P 500 assets sit around $10 trillion (a rough figure across index mutual funds and ETFs like VOO, IVV and SPY combined), then index funds collectively need to own:
0.04 % × $10 trillion ≈ $4 billion of XYZ
That is 20 % of XYZ’s free float, all of it demanded in a compressed window. If average daily dollar volume is $250 million, the passive demand alone equals sixteen days of normal trading — enough to move the price meaningfully even before hedge funds, sector-rotation ETFs and closet-benchmark active managers add their own flows.
Deletions: the mirror image, but not symmetric
Chen, Noronha and Singal (2004) found that additions and deletions do not produce equal and opposite reactions. Additions get a larger, more persistent price bump; deletions get a sharper but more mean-reverting drop. Their leading explanation: additions carry an information signal — the committee is publicly certifying that a company meets the profitability, size and liquidity bar — while deletions are usually mechanical (a merger, a spin-off, a company that has simply fallen below thresholds), which the market has often already priced in.
Why the effect has shrunk
Three forces together explain why the effect is a fraction of what it was in the late 1990s:
- Arbitrage capital has grown enormously. Once the effect was documented, hedge funds and prop desks began accumulating candidates before the announcement and unwinding into the effective-day auction. Their front-running smooths the demand curve and captures much of what was once left on the table.
- Index funds themselves have become better traders. Modern S&P 500 trackers use auction-on-close, portfolio-trade sweeps and off-index sourcing to build positions over several days instead of dumping them into a single closing print.
- The candidate pool is more predictable. After decades of committee decisions, systematic screens can shortlist likely candidates months in advance. Uncertainty is what the arbitrage was pricing; less uncertainty means less premium.
Greenwood and Sammon’s 2025 Journal of Finance paper, aptly titled “The Disappearing Index Effect,” documents this decline even as the share of U.S. equity held in passive vehicles has kept climbing — a genuine puzzle for anyone who assumed that more mechanical demand should mean bigger index effects.
Common mistakes people make about the index effect
- Confusing the Dow with the S&P 500. The Dow Jones Industrial Average is price-weighted and tracked by a much smaller pool of assets, so its inclusion effect is far weaker than the S&P’s. Alphabet joining the Dow on June 29, 2026 — replacing Verizon — is a governance and prestige story more than a mechanical-flows story. (The Dow is decided by a joint committee of three S&P DJI representatives and two Wall Street Journal editors.)
- Treating the announcement return as free money. By the time you can read the press release, the after-hours market has already priced most of the pop. The realised return from “buy on announcement, sell on effective” is meaningfully smaller than the raw abnormal return figures in academic papers, and any strategy that tries to capture it faces borrow costs, closing-auction slippage and adverse selection.
- Assuming deletions are symmetric. They are not. Deletion drops mean-revert faster and more completely than addition premia, because deletions carry little new information beyond “the company shrank, merged, or de-listed.”
- Ignoring the deletion candidate. When index-effect trading was more profitable, betting on which stock would be removed was often as lucrative as betting on the addition — and easier, because the smallest, weakest names in the index were the obvious candidates.
Related concepts to learn next
- Reconstitution effect in the Russell 2000 — a bigger, noisier version of the same phenomenon, because Russell’s annual rebalance is mechanical and the shortlist is knowable.
- Free-float adjustment — why an index weight depends on shares available to trade, not total shares outstanding.
- Passive ownership share — the still-open debate over whether ~50 % passive ownership distorts price discovery.
- Sector reclassification — when S&P/MSCI move a company between GICS sectors, sector ETFs generate their own mini-index effect.
Sources
- Wikipedia summary of the S&P 500 index — eligibility criteria, committee governance and index-effect research citations — https://en.wikipedia.org/wiki/S%26P_500
- Wikipedia entry on the Dow Jones Industrial Average, including the Alphabet-for-Verizon change effective June 29, 2026 and Dow governance — https://en.wikipedia.org/wiki/Dow_Jones_Industrial_Average
- Harris, L. and Gurel, E. (1986). “Price and Volume Effects Associated with Changes in the S&P 500 List: New Evidence for the Existence of Price Pressures.” Journal of Finance. Citation index — Google Scholar
- Chen, H., Noronha, G. and Singal, V. (2004). “The Price Response to S&P 500 Index Additions and Deletions: Evidence of Asymmetry and a New Explanation.” Journal of Finance — Google Scholar
- Petajisto, A. (2011). “The Index Premium and Its Hidden Cost for Index Funds.” Journal of Empirical Finance, 18(2). — Google Scholar
- Greenwood, R. and Sammon, M. (2025). “The Disappearing Index Effect.” Journal of Finance — Google Scholar
- Bender, J., Nagori, R. and Tank, M. (2019). “The Past, Present and Future of the Index Effect.” Journal of Index Investing — Google Scholar
Disclosure: This article was produced with AI assistance and reviewed before publication. It is for informational purposes only and is not investment advice.