Unemployment rates in the United States have never been uniform across racial and ethnic groups. The Bureau of Labor Statistics (BLS) has tracked unemployment by race and ethnicity for decades, and the gaps it documents are among the most consistent patterns in U.S. labor market data — persistent across economic expansions, recessions, and recoveries alike.
The BLS collects unemployment data through the Current Population Survey (CPS), a monthly household survey conducted in partnership with the U.S. Census Bureau. Respondents are asked about their employment status, job search activity, and availability to work during a specific reference week.
The BLS publishes unemployment rates broken down by four major racial and ethnic categories:
These are the groups for which the BLS reports consistent monthly data. Other groups — including American Indian or Alaska Native, and multiracial individuals — appear in less frequent supplemental reports due to smaller sample sizes.
The most documented gap in U.S. unemployment data is between Black and White unemployment rates. Over the past 50-plus years, the Black unemployment rate has consistently run roughly twice the White unemployment rate — a ratio that holds during both high-unemployment periods and low-unemployment periods.
Some historical reference points:
| Period | White Unemployment Rate | Black Unemployment Rate | Ratio |
|---|---|---|---|
| 1972 (earliest consistent BLS data) | ~5% | ~10% | ~2:1 |
| 1983 recession peak | ~8.8% | ~19.5% | ~2.2:1 |
| 2009 recession peak | ~8.7% | ~16.5% | ~1.9:1 |
| 2019 (pre-pandemic low) | ~3.3% | ~5.5% | ~1.7:1 |
| April 2020 (pandemic peak) | ~14.2% | ~16.7% | ~1.2:1 |
| 2023 (post-pandemic) | ~3.1–3.5% | ~5.3–6.0% | ~1.7:1 |
Figures are approximate BLS monthly averages. Exact figures vary by month and year.
Hispanic or Latino unemployment rates have historically fallen between White and Black rates, though they respond sharply to construction and service-sector contractions given the concentration of Hispanic workers in those industries.
Asian unemployment rates have generally tracked at or below White rates in recent decades, though the COVID-19 pandemic produced a notable spike in Asian unemployment — particularly in sectors like hospitality, food service, and retail.
BLS unemployment statistics measure who is actively looking for work and not finding it. They don't measure:
Researchers point to multiple structural factors when analyzing racial unemployment gaps, including differences in industry and occupational concentration, geographic labor market conditions, educational attainment distributions, access to professional networks, and documented hiring discrimination. These are areas of ongoing academic and policy debate — the BLS data describes the outcome but does not assign a single cause.
Unemployment rates and unemployment insurance (UI) claims are related but distinct measures.
The unemployment rate (from the CPS) counts everyone who is jobless and looking for work — whether or not they file for benefits. UI claims data, published separately by the Department of Labor, tracks only people who have applied for and are receiving state unemployment insurance benefits.
The BLS does not publish UI claims broken down by race. What is known from research is that take-up rates — the share of eligible unemployed workers who actually claim benefits — vary by demographic group. Factors affecting take-up include awareness of the program, language access, immigration status concerns, prior work history meeting wage thresholds, and the administrative burden of filing.
This means racial unemployment rate data and UI claims data capture overlapping but not identical populations. 🔍
One consistent pattern in the historical data: recessions widen racial unemployment gaps, and recoveries close them unevenly. Black unemployment typically rises faster and falls more slowly than White unemployment during and after economic downturns.
The 2020 pandemic recession was a partial exception — unemployment spiked across all groups at unusual speed, temporarily compressing the ratio. But as the labor market recovered through 2021 and 2022, racial gaps largely re-emerged at historical proportions.
Aggregate unemployment statistics describe population-level patterns. They say nothing about whether any individual worker qualifies for unemployment insurance benefits, what their weekly benefit amount would be, or how their claim will be adjudicated.
Those outcomes depend on state-specific rules: base period wage requirements, the reason for separation from an employer, whether the employer contests the claim, and how a state's UI agency applies its eligibility standards. A worker in one state with a given work history and separation type may have a very different outcome than a worker in another state with nearly identical circumstances.
The racial unemployment data tells a story about the labor market at scale. What happens with a specific unemployment claim depends entirely on the facts of that individual situation and the rules of the state where the claim is filed.