Historical S&P 500 membership, including the companies that left.
A clean historical membership and turnover file covering 2004–2025, built from Persistfolio’s certified Historical Cohort reconstruction.
S&P 500 Membership & Turnover History, 2004–2025
US$49 one time — retail license for one named purchaser (personal, academic or internal research)
Organization-wide internal use: the Historical Cohort Business / Commercial license (US$499)
The sample is the same retail file restricted to the 29 securities of the Historical Cohort’s public sample (31 membership periods), with exactly the same columns. It contains: sample.csv, README.md, LICENSE.txt.
What You Get
- sp500_membership_turnover_2004_2025.csv (994 rows) — one row per membership period, including companies that left the index
- security_id (Persistfolio’s own identifier) and the issuer/security name, with its name history
- Membership start and end dates, with how certain each date is and why
- The index provider’s published index-effective dates, where evidenced (blank otherwise — never filled from other dates)
- The turnover event type behind each start and end
- Existing successor/lineage links (merger successor, spin-off, acquisition, holding company, bankruptcy successor) where available
- README.md (column guide and how to read the dates), LICENSE.txt and SHA256SUMS
What This Does Not Contain
- Price history, returns or delisting returns
- Historical (point-in-time) ticker mapping, current tickers, FIGI or CUSIP history
- SEC CIK mapping
- Identity tables, adjudication evidence, reconciliation and the certification package
Need the full research package with CIK periods, identity tables, adjudications, reconciliation and certification evidence? View the Persistfolio Historical Cohort — US$149 Individual / Academic or US$499 Business / Commercial →
Coverage & Product Facts
- Coverage
- 2004-09-30 – 2025-09-30
- Source
- Derived, without changing any value, from Persistfolio’s certified Historical Cohort Release 1.1 (historical-cohort-2025-09-30-r1.1): its membership periods, constituent events and lineage tables
- Observation Frequency
- One row per continuous membership period, with event-driven start and end boundaries
- Release
- sp500-membership-turnover-2025-09-30-r1 (2025-09-30)
- Formats
- CSV
- Update Policy
- Fixed, dated release (cutoff 2025-09-30). One-time purchase; no lifetime updates.
Methodology & QA
Every row is one certified membership period of the Historical Cohort Release 1.1, copied unchanged with its date-certainty fields. The turnover event type is joined from the release’s own event ledger through each period’s start and end event; successor/lineage links are attached where the release records one taking effect during the period. Nothing is researched, inferred or filled in: a blank published effective date stays blank. The full reconstruction method, QA and certification belong to the Historical Cohort research package.
Data Dictionary
| File | Field | Description |
|---|---|---|
| sp500_membership_turnover_2004_2025.csv | security_id | Persistfolio's own stable security identifier (`PFHC-` + 10 hex characters). Not a ticker, CUSIP, FIGI or CIK. |
| sp500_membership_turnover_2004_2025.csv | security_name | The latest legal issuer name evidenced during this membership period. A security with two periods can carry two different names. |
| sp500_membership_turnover_2004_2025.csv | name_history | Every distinct name the security carried during membership, oldest first, separated by semicolons. |
| sp500_membership_turnover_2004_2025.csv | membership_start_date | The date this membership period begins, as a reconstructed membership boundary. Not always the index provider’s published effective date - read `start_date_certainty`. |
| sp500_membership_turnover_2004_2025.csv | membership_start_basis | Why the start boundary has this date (same meaning as `start_date_certainty`, in the parent release’s basis vocabulary). |
| sp500_membership_turnover_2004_2025.csv | start_date_certainty | How certain the start boundary is (values below). |
| sp500_membership_turnover_2004_2025.csv | index_effective_start_date | The index-effective date the index provider published for this start, where one is evidenced. Blank otherwise - never filled from a corporate, announcement, SEC, roster or trading date. |
| sp500_membership_turnover_2004_2025.csv | start_earliest_known_date | For a start that is only bounded above by a later anchor: the earliest known date (the announcement date), i.e. the lower side of the window. |
| sp500_membership_turnover_2004_2025.csv | membership_start_event_id | The turnover event that began this period. Blank for securities already in the index when coverage opens on 2004-09-30. |
| sp500_membership_turnover_2004_2025.csv | start_event_type | The type of that event (values below). Blank when there is no start event. |
| sp500_membership_turnover_2004_2025.csv | membership_end_date | The date this membership period ends, as a reconstructed membership boundary. Blank if the security was still a member on 2025-09-30. |
| sp500_membership_turnover_2004_2025.csv | membership_end_basis | Why the end boundary has this date (same vocabulary as the start, plus `STILL_ACTIVE_AT_CUTOFF`). |
| sp500_membership_turnover_2004_2025.csv | end_date_certainty | How certain the end boundary is (values below). |
| sp500_membership_turnover_2004_2025.csv | index_effective_end_date | As `index_effective_start_date`, for the end of the period. |
| sp500_membership_turnover_2004_2025.csv | end_earliest_known_date | As `start_earliest_known_date`, for the end of the period. |
| sp500_membership_turnover_2004_2025.csv | membership_end_event_id | The turnover event that ended this period. Blank if still a member at the cutoff. |
| sp500_membership_turnover_2004_2025.csv | end_event_type | The type of that event (values below). Blank when there is no end event. |
| sp500_membership_turnover_2004_2025.csv | confidence | HIGH / MEDIUM / REVIEW - Persistfolio’s classification of the evidence behind the events bounding this period (REVIEW: weak evidence or a bounded date). |
| sp500_membership_turnover_2004_2025.csv | review_status | CLEAR / REVIEW - REVIEW if an event bounding this period carries an open, disclosed review item. |
| sp500_membership_turnover_2004_2025.csv | lineage_relationships | Successor/lineage links from this security that take effect within this period, separated by ` | ` (values below). Blank if none. |
| sp500_membership_turnover_2004_2025.csv | lineage_to_security_ids | The linked (successor) security_id for each link, in the same positions. |
| sp500_membership_turnover_2004_2025.csv | lineage_to_names | The linked security’s name for each link, in the same positions. |
| sp500_membership_turnover_2004_2025.csv | lineage_effective_dates | The effective date of each link, in the same positions. |
Full data dictionary (JSON) → · Complete dictionary with enumerated values and date semantics (Markdown) →
Pricing & License
Individual / Academic: US$49 one time — retail license for one named purchaser (personal, academic or internal research)
Commercial / Business: Organization-wide internal use: the Historical Cohort Business / Commercial license (US$499)
Retail data license for one named purchaser (sp500-membership-turnover-2025-09-30-r1) — see full license terms
Known Limitations
- Membership boundary dates are not always the index provider’s published effective dates; read the certainty columns, and use index_effective_* where populated.
- Members already in the index on 2004-09-30 start at the dataset’s coverage start, not their index entry date.
- No point-in-time ticker history, price data or delisting returns; joining to price data means matching on name and dates yourself.
- Lineage links are shown only where they take effect during a membership period; 15 of the parent release’s 93 links take effect after the security had left the index and are not in this file.
- Independent reconstruction, not an official S&P dataset.
FAQ
Does this include companies that left the S&P 500?
Yes. Every membership period in the 2004-09-30 to 2025-09-30 coverage is included, including securities that were removed, acquired or merged away.
Can I use it for a backtest on its own?
It tells you which securities were members and when. It has no prices, returns, delisting returns or ticker history, so a backtest also needs your own price data, matched by name and dates.
Are the dates official S&P effective dates?
Not always. Each start and end is a reconstructed membership boundary with a certainty classification; the published index-effective date is given where it is evidenced and left blank otherwise.
Can my company use it?
This $49 license is for one named purchaser’s personal, academic or internal research. Organization-wide internal use needs the Historical Cohort Business / Commercial license.
Will I get updates?
No. It is a fixed, dated release (cutoff 2025-09-30) with no lifetime updates.
Is it official S&P data?
No. Persistfolio is independent; this is derived from its own historical reconstruction.
Persistfolio is independent and is not affiliated with, sponsored by, endorsed by, or authorized by S&P Global or S&P Dow Jones Indices. S&P 500® is a trademark of S&P Dow Jones Indices LLC and/or its affiliates. This product is an independent research reconstruction and is not an official S&P dataset.