# Unemployment Rate / Jobs Reports The unemployment rate is the Bureau of Labor Statistics' monthly measure of the share of the labor force that is jobless and actively seeking work. The companion **nonfarm payrolls** report (from the Establishment Survey) counts how many jobs the economy added or lost. Together, they form the most politically ritualized economic data release in America — the first Friday of every month, a number drops, and the entire political apparatus spins. The unemployment rate matters for the KB because it demonstrates **how the same statistic can be credible or "fake" depending on who occupies the White House** — a pattern Trump made explicit when he called the number "phony" under Obama and "very real" once he took office. ## Latest reading *As of 2026-10-07 · refreshed monthly · the cross-indicator read lives at [[State of the US Economy]].* - **September 2026: 4.2%** (up from 4.1%). It was **4.4% in December 2025** and 4.2% in June 2026. - **U-6: 7.6%** (August 7.7%). **Employment-population ratio: 59.2%.** [[Labor Force Participation Rate]]: **61.8%**. - **Low-hire, low-fire.** Initial jobless claims were **197,000** for the week ending Sept 26 (4-week average 200,000), near multi-decade lows. Payroll growth meanwhile has nearly stalled ([[Nonfarm Payrolls]]: +29K in September). JOLTS openings were **7.1M** in August, and the quits rate held at **1.9%**. - **Gap in the record:** the **October 2025 household survey was never collected** because of the shutdown, so the unemployment series has a permanent hole there. - **Next:** October jobs report, early November. *Sources:* [BLS Employment Situation](https://www.bls.gov/news.release/empsit.nr0.htm) · [Reuters via Virginia Business](https://virginiabusiness.com/us-job-growth-slows-september-unemployment-rises/) · [BLS Dec 2025 release](https://www.bls.gov/news.release/archives/empsit_01092026.htm) · [DOL weekly claims](https://www.dol.gov/ui/data.pdf) · [TheStreet on claims](https://pro.thestreet.com/uncategorized/initial-jobless-claims-remain-near-57-year-lows-continuing-claims-a-3-year-low) · [BLS JOLTS, Aug 2026](https://www.bls.gov/news.release/archives/jolts_09292026.htm) · [Fortune on the cancelled October survey](https://fortune.com/2025/11/19/government-cancels-october-jobs-report-late-unemployment-rate) ## Methodology The BLS produces employment data from **two separate surveys**: **Household Survey (Current Population Survey):** - ~60,000 households surveyed monthly - Produces the unemployment rate (U-3) and alternative measures (U-1 through U-6) - Counts self-employed, agricultural, and household workers - Used to derive the labor force participation rate **Establishment Survey (Current Employment Statistics):** - ~131,000 businesses and government agencies covering ~670,000 individual worksites - Produces nonfarm payrolls (the "jobs added" number) - More stable (larger sample) but doesn't capture self-employment or agricultural work - Subject to revisions — preliminary numbers are updated twice (1 month and 2 months later), then benchmarked annually ### The Six Measures of Unemployment | Measure | Definition | Typical Level (2023–24) | |---------|-----------|----------------------| | **U-1** | Unemployed 15+ weeks | ~1.2% | | **U-2** | Job losers and temporary job completers | ~1.8% | | **U-3** | **The official rate** — total unemployed as % of labor force | ~3.5–3.8% | | **U-4** | U-3 + discouraged workers | ~3.9% | | **U-5** | U-4 + marginally attached workers | ~4.4% | | **U-6** | U-5 + part-time for economic reasons ("underemployment") | ~6.8–7.2% | **The political game:** U-3 is the headline number. When a party wants to argue the economy is worse than it appears, they cite U-6 ("real unemployment"). When they want credit for a strong economy, they cite U-3. Trump's "42% unemployment" claim during the 2016 campaign was effectively citing something beyond U-6 — counting retirees, students, stay-at-home parents, and anyone not working for any reason as "unemployed." ## Historical Context | Period | U-3 Rate | Context | |--------|---------|---------| | 1953 (Eisenhower) | 2.9% | Post-WWII low | | 1982 (Reagan recession) | **10.8%** | Post-war high, Volcker shock | | 2000 (Clinton) | 3.9% | Dot-com peak | | Oct 2009 ([[Great Recession]]) | **10.0%** | Financial crisis peak | | Feb 2020 (pre-COVID Trump) | 3.5% | 50-year low | | Apr 2020 (COVID) | **14.7%** | Highest since [[Great Depression]] | | Jan 2023 (Biden) | **3.4%** | 53-year low | | Dec 2024 (Biden exit) | ~4.1% | Still historically strong | | Early 2025 ([[Trump 2.0]]) | ~4.0–4.2% | DOGE layoffs + tariff uncertainty building | | Dec 2025 | 4.4% | October 2025 household survey never collected (shutdown) | | Sep 2026 | 4.2% | Low-hire, low-fire; payrolls +29K | ## The Birth-Death Model Conspiracy The BLS uses a "birth-death model" to estimate the net effect of new business creation (births) and closures (deaths) on employment between surveys. Because new businesses aren't yet captured in the establishment survey sample, the model adds estimated jobs that probably exist but haven't been counted. **The conspiracy version:** Right-wing media and commentators regularly claim the BLS "makes up" jobs through the birth-death model, inflating employment numbers under Democratic presidents. This is a perennial claim that intensifies whenever the jobs report is stronger than expected. **The reality:** The birth-death model is methodologically transparent, has been used since 2003 (under both parties), and is corrected during annual benchmark revisions. Those revisions sometimes adjust numbers down (2024 benchmark revision reduced 2023–24 jobs by ~818,000) and sometimes up. The model introduces noise, not systematic bias — but the noise provides ammunition for narrative warfare. **Why the annual benchmark revision matters politically:** When the BLS revised 2023–24 job creation down by ~818,000 in August 2024, right-wing media treated it as evidence of deliberate Biden-era manipulation. The revision was actually routine and the methodology unchanged from prior administrations. But "they've been lying about jobs" is a far more politically potent frame than "statistical sampling has inherent uncertainty that gets corrected through periodic benchmarking." ## The [[Labor Force Participation Rate]] **LFPR** = (employed + actively seeking work) / civilian noninstitutional population age 16+ This rate has been on a structural decline since peaking at 67.3% in 2000, driven by demographics (Baby Boomer retirements), rising educational enrollment, disability trends, and changing social patterns. **January 2000:** 67.3% **February 2020 (pre-COVID):** 63.4% **April 2020 (COVID trough):** 60.2% **Late 2024:** ~62.5–62.7% **The political exploitation:** "The unemployment rate only looks low because people have given up looking for work." This argument has a kernel of truth (the U-3 rate doesn't count discouraged workers) but dramatically overstates the case by ignoring the structural drivers. Most of the LFPR decline since 2000 reflects demographics, not discouraged workers — but "10 million Americans have given up" makes a better headline than "Baby Boomers are retiring on schedule." **Immigration connection:** The LFPR discussion connects directly to the KB's immigration entries. Immigrants have a *higher* labor force participation rate than native-born Americans (~66% vs. ~62%). The anti-immigration argument that immigrants "take American jobs" runs directly counter to the data — immigrants expand the labor force and fill gaps that demographic decline (aging population, declining native-born LFPR) would otherwise leave unfilled. This is the empirical basis behind the CBO's finding that immigration added ~0.5% to GDP growth annually during 2022–2024. ## Trump's Relationship with the Jobs Number The unemployment rate provides the KB's cleanest case study of **indicator credibility as partisan variable**: | Period | Trump's Position on the Unemployment Rate | |--------|----------------------------------------| | **2012–2016 (Obama era)** | "The unemployment rate is a hoax." "Don't believe those phony numbers." "The real number is 42%." | | **Jan 2017 (inauguration day)** | White House Press Secretary [[Sean Spicer]]: "They may have been phony in the past, but it's very real now." | | **2017–2019 (Trump first term)** | Cited the unemployment rate constantly as proof of his economic success | | **2020 (COVID)** | Blamed COVID, not his administration, for unemployment spike | | **2025 (second term)** | Resumed citing employment numbers when favorable; DOGE mass layoffs complicated the narrative | This trajectory — the same statistic goes from "fake" to "real" to "my achievement" based solely on political utility — is a case study in Goffman's impression management. The front stage shifts; the methodology never changes. ## How Jobs Data Gets Weaponized | Technique | Description | |-----------|------------| | **Seasonal adjustment disputes** | Claim the BLS seasonal adjustment is biased (it's mechanical and unchanged across administrations) | | **Full-time vs. part-time** | "Sure, jobs were added, but they're all part-time" (claim is intermittently true but systematically overstated) | | **Ignoring revisions that go the wrong way** | Loudly trumpet downward revisions; ignore upward revisions | | **Government jobs vs. private jobs** | Dismiss public sector job growth as "not real" job creation | | **Immigration framing** | "Immigrants got all the new jobs" (conflates immigrant labor force entry with native-born job displacement) | | **DOGE paradox** | Trump 2.0 simultaneously claims credit for a strong economy while DOGE fires tens of thousands of federal workers who then appear in unemployment statistics | ## The DOGE Employment Paradox (2025) The mass federal layoffs under DOGE in early 2025 created a novel political-economic situation: - An estimated 100,000+ federal workers were fired, put on leave, or had positions eliminated in the first 100 days - Many affected workers were in probationary status and may not have been captured in traditional unemployment claims - The administration argued these layoffs represented "efficiency" while simultaneously claiming credit for overall job market strength - State-level unemployment claims in Washington, D.C. and federal employment hubs surged This is the Starve the Beast logic made literal: cut government employment, then point to the resulting economic disruption as evidence that government was bloated. The unemployment data becomes both the instrument and the measure of the ideological project. ## See Also - [[Consumer Sentiment Index]] — Sentiment diverged from strong employment data in 2022–2024 - [[Consumer Price Index]] — The Phillips Curve tension between unemployment and inflation - [[GDP]] — Employment as a component of growth - [[Gas Prices]] — Visceral price signals override employment statistics in voter perception - [[Stock Market Indices]] — Market performance during full employment - [[DOGE Ecosystem]] — Mass federal layoffs as ideological project - [[Starve the Beast]] — The fiscal theory connecting government employment cuts to broader economic doctrine - [[Erving Goffman Self-Presentation Theory]] — Impression management: the same statistic goes from "fake" to "real" based on political utility