Layoffs Dataset 2019–2026: Global, India, US, UK

Key findings

1. 19,208 layoff events at 13,303 companies in 169 countries, January 2019 to June 2026. 57% of events with a known country happened outside the United States.
2. Layoff events peaked in 2020 (3,780 events); 2023 had 3,391 and 2025 had 2,977, against 1,759 in 2019.
3. India: 971 layoff events, peaking in 2023 (219); the United Kingdom has 2,198. 9,541 events state a verified headcount, together 10,929,045 jobs; the median cut is 195.
4. The largest single permanent cut is Amazon (99,000 jobs, July 2022, United States); 540 events are furloughs or temporary lay-offs, flagged in is_furlough.
5. 2,768 companies had more than one layoff round; 371 had five or more.
6. Retail has the most layoff events (1,008), followed by Food and Agriculture (795) and Consumer Goods (773).

Layoffs by year

2026 covers January to June. Jobs cut sums only verified headcounts.

Layoff events by quarter

Layoffs by month, July 2024 to June 2026

News volume thins after May 2026, so the last months may undercount.

Top 50 largest layoffs worldwide, 2019–2026

Verified headcounts only; one row per layoff round; furloughs and temporary lay-offs excluded.

Layoffs by country worldwide (top 40)

Layoff events by country and year (top 30)

Layoffs by sector (top 30)

Companies with repeated layoffs (top 50)

Companies with two or more separate layoff rounds.

Top 50 largest layoffs in India, 2019–2026

India layoffs by sector (top 25)

India layoffs by company HQ state (top 25)

HQ location from the company database.

Top 50 largest layoffs in the United States, 2019–2026

US layoffs by sector (top 25)

US layoffs by company HQ state (top 25)

HQ location from the company database.

Top 50 largest layoffs in the United Kingdom, 2019–2026

UK layoffs by sector (top 25)

UK layoffs by company HQ city (top 25)

HQ location from the company database.

Column dictionary

20 columns in the sample CSV.

Method

1. Extraction. Radar's pipeline reads each news article, finds sentences reporting job cuts and extracts company, headcount, share of workforce and country. Upstream, reports carrying the same event fingerprint are grouped and retrospective mentions are dropped.
2. Window and dates. The date is the publication date of the earliest report, read in Indian Standard Time; January 2019 to June 2026. Bulk archive back-fill days (over five times the usual volume) are dropped.
3. Companies. Descriptor prefixes are removed ("Tata Motors-owned Jaguar Land Rover" → Jaguar Land Rover). Rows naming a person, an unnamed or generic employer ("the company") or a whole sector ("IT companies") are dropped. A website is kept only when the company record matches the reported name.
4. One row per layoff round. Reports of the same company within 45 days of the round's first report (and not in a different stated country) are one event. Exact repeat sentences are removed first.
5. Headcount and %. A headcount is kept only when the same number (±2%, including "1.2 lakh", "10k", "a dozen") appears in the sentence; the earliest verified figure of the round is used. Headcounts above 150,000 are blanked. The share of workforce is kept only when the sentence states it (a percentage, "half", "a third").

6. Countries. country is the country where the cut happens as reported, else the company's home country; names are normalised (USA / US → United States).

7. Sample. Each company key is ranked by sha1(key). Keys are taken in rank order until the file holds max(25%, 5,000) rows (capped at 100,000).

8. Profile links. Links are built from Fliar's company index and checked against it. A row whose company has no public profile has an empty link.

Limitations

• Derived from news: unreported layoffs are missing, and coverage favours large companies and English-language media; India and Europe are well covered, many emerging markets thinly.
• ClickHouse shards were partially unavailable at build time (shard 2 down; queries skip unavailable shards), so some layoffs may be missing. Article links resolve for roughly half of the rows because some article-store nodes were down.
• Many reports give no headcount, or a figure that could not be verified; jobs-cut totals sum only verified headcounts and understate the true total.
• Merging is automatic: two separate rounds at one company within 45 days count as one, and a long-running restructuring may be split.
• Sector and HQ location are known only for companies matched to the company database.
• News ingest thins after May 2026; 2026 covers January to June only.

Cite this dataset

@misc{fliarbi_layoffs_2019_2026_2026,
  author    = {{Fliar BI}},
  title     = {{Layoffs Dataset 2019–2026: Global, India, US, UK}},
  year      = {2026},
  publisher = {Fliar BI},
  url       = {https://fliarbi.com/datasets/layoffs-2019-2026/},
  note      = {Sample release. Licence CC BY-NC 4.0}
}