Datasets:

APProjects
/

warn-act-notice-type-codes-crosswalk

pretty_name: >-
  WARN Act notice-type codes crosswalk - every raw state string and what it
  means
license: cc-by-4.0
language:
  - en
task_categories:
  - tabular-classification
  - text-classification
tags:
  - warn-act
  - data-cleaning
  - entity-resolution
  - normalization
  - crosswalk
  - codebook
  - data-dictionary
  - layoffs
  - layoff-notices
  - plant-closings
  - labor
  - labor-market
  - public-records
  - government-data
  - daily-updated
  - united-states
size_categories:
  - n<1K
configs:
  - config_name: crosswalk
    data_files: data/notice_type_crosswalk.csv

WARN Act notice-type codes — the crosswalk

Every US state publishes WARN Act layoff notices with a free-text column saying what kind of event it is. The statute recognises two: a plant closing and a mass layoff. Across 48 states that column contains 552 distinct exact strings (531 once you fold case).

This dataset is the crosswalk: one row per raw string, how many notices carry it, which states emit it, and what it normalizes to.

The finding that matters

521 of the 552 strings (94%) are used by exactly one state. Only 31 are shared across states at all.

There is no common vocabulary. CL means closure in Indiana and Wisconsin; Illinois fills the column with State; five states write WARN — the name of the statute — where the event type belongs. 372 strings appear on a single notice in the entire archive.

This is why "just parse the state files" does not converge: there is nothing to parse toward until someone reads all 48 vocabularies and keeps reading them as states change their exports.

Vocabulary by what it maps to

maps to	distinct strings
layoff	185
closure	147
mixed (both terms present)	23
unknown (deliberately unmapped)	197

unknown is a real category, not a failure to try. It covers blanks, strings that name which statute applies rather than what happened (State, WARN, numeric record codes), and strings that name a cause (loss of contract, sale of company, COVID-19) rather than an event. Guessing these would manufacture thousands of false rows that no downstream user could detect.

The 12 most common strings

raw value	maps to	notices	states	used by
(blank)	unknown	13,759	29	AK CO CT DC FL HI ID KY LA MA MN MO MT ND…
`Layoff Permanent`	layoff	7,812	2	CA TN
`Closure`	closure	6,077	15	AK AL CO KY MI MS NC NE NV NY OH PA VA WA WV
`Layoff`	layoff	5,943	15	AK AL KY MI MO MS NC NE NV NY OH PA VA WA WV
`State`	unknown	4,845	1	IL
`Closure Permanent`	closure	4,164	2	CA TN
`Layoff Temporary`	layoff	3,657	2	CA TN
`WARN`	unknown	1,658	5	AZ DE KS ME VT
`Plant Closing`	closure	1,398	4	MI NY OK PA
`Closing`	closure	1,196	8	CO CT IA MN MO PA RI WV
`CL`	closure	862	2	IN WI
`Mass Layoff`	layoff	838	8	CO IA MD MI NY OK PA WV

Files

data/notice_type_crosswalk.csv — 552 rows: raw_value, kind, permanence, reason, notices, n_states, states.

Covers all 61,304 notices in the archive as of 2026-09-13; rebuilt daily in the same pipeline run as the notices themselves, so the two can never disagree.

Limits — read before citing

These are notices, not verified job losses. A WARN notice can be amended, rescinded, or never carried out.
The mapping is conservative: anything that does not clearly say closing or layoff is left unknown rather than guessed.
permanence is only populated where the state says so; most states do not.
Counts are of notices, not workers.

Daily US WARN notices (flagship) — the full normalized feed
Plant closings vs mass layoffs — this mapping applied
Site · Source and methodology

License CC-BY-4.0. Built from public state government records.

A layoff record you can audit, not just download

This dataset is one cut of a single daily rebuild: 61,330 US WARN Act layoff notices from 48 state agencies, 1988 to today, one schema, no login, no delay, CC BY 4.0. Snapshot as of 2026-09-17; the files above are rebuilt every day, so the live count is the truth.

Several projects publish a current WARN scrape and two of them carry more rows than we do. None of them publish what the records used to say:

617 observed changes to already-published notices, logged daily since 2026-08-31. data/revisions.csv records every field that differed between two consecutive daily builds — employee counts, effective dates, notice types, company names — with the old value, the new value and the date we saw it. We publish the observation and not the cause: a change is equally explained by the agency amending the notice or by our own parser improving, and we do not guess which (see data/revisions.README.txt). A scrape that starts tomorrow cannot backfill any of it; it only exists if someone was watching.
6,799 notices whose state agency page no longer lists them. Agencies take notices down. We keep them, flagged as archive-only, so a count you ran last year still reconciles.
Point-in-time employer identity. The ticker crosswalk resolves a filer to the company as it existed at the time of the notice — Kmart, Sears Holdings, Symantec — not to whatever is on today's ticker file.

If you have to defend a number to an editor, a referee or a compliance reviewer, that provenance layer is the part you cannot rebuild yourself. How to cite this dataset →

Look something up right now — free, no signup, nothing to install. Check any employer or state against the last 180 days → It runs in your browser against these same files.

Building something with it? The same files are a free HTTP API — JSON and CSV, no key, no signup, access-control-allow-origin: * so fetch() works from a browser: endpoints, schema and curl examples →

Need one industry only? The same filings, cut by an auditable employer-name rule (each row keeps the rule that fired): tech companies · hospitals & healthcare · retail store closings · restaurants & hotels · factory & plant closings · banks, insurance & finance · warehouses, trucking & logistics · all 20 sectors.

Or have it watch a list for you. Coming back to look is the part a CSV cannot do. WARN Watch — $49 for a year, one payment, nothing auto-renews, 14-day refund, no login: up to 500 employer names plus whole states, matched on every daily refresh, delivered to a private alert page + calendar (.ics) + RSS + an optional Slack / Discord / Teams webhook. Every alert carries that employer's whole filing history from the archive, which a keyword rule on an RSS feed cannot see. There is no built-in email — we do not claim one.

Not deciding today? Join the update list → — one email when a new dataset or tier is published; nothing promotional. A state added or a column renamed ships in the daily release instead, no address needed. The list is shared across APProjects datasets, holds an email address only, is run by Gumroad, and any message unsubscribes you. Rather give no address at all? Watch the repo's releases — GitHub notifies you on every daily republish, and a new state or changed field is in those notes the day it lands.

Reaching a human. WARN Feed is published by APProjects, an automated data publisher — that is stated plainly rather than dressed up. Corrections, coverage gaps, schema questions and refund requests all go here and are read: open an issue. Payments are handled by Gumroad as merchant of record, so an invoice can carry your company name.

Source, scrapers and methodology · the 48-state site

_{Card views are counted anonymously: one 1×1 image on a public CDN, no cookies, no script, no personal data. The count is public: jsDelivr stats.}