Unemployment rate, explained

What it measures

The unemployment rate is the share of the labor force that is jobless and actively looking for work. The labor force includes everyone 16 and older who is either employed or actively seeking employment in the four weeks before the survey; it excludes retirees, students not seeking work, stay-at-home parents not seeking work, and people too discouraged to keep looking. A person who has given up the job search is counted as "not in the labor force" rather than as unemployed, which means the unemployment rate can fall even when the actual employment picture is worsening (if discouraged workers leave the labor force faster than the jobless find work).

The ACS publishes unemployment rates at the same geographic granularity as the rest of its data, down to census tract, using a 5-year average. The monthly unemployment rate you see in the news comes from a different source (the BLS Current Population Survey) and refers to the nation; sub-state monthly figures from BLS use a smaller sample and are not directly comparable to the ACS rate.

Why it matters

Local unemployment rates drive everything from federal disaster recovery aid to commercial-real-estate cap rates. The 2009 Recovery Act, for example, weighted block-grant funding by county unemployment rate. Bond-rating agencies use metro-level unemployment as one of the primary inputs to municipal credit ratings. For homebuyers and renters, sustained local unemployment above 7% is a warning that the local job market is structurally weak; for employers, sustained unemployment below 3% means hiring will be expensive and slow.

How the Census measures it

ACS Table S2301 reports employment status for the civilian non-institutional population 16 and older. The Census Bureau classifies each respondent as employed (worked for pay or profit in the reference week, OR temporarily absent from a job), unemployed (didn't work, looked for work in the last four weeks, available to start), or not in the labor force. The unemployment rate is unemployed / (employed + unemployed) × 100. The 5-year ACS averages 60 months of responses, so rates lag the BLS monthly series in recognizing recessions and recoveries.

How to read the numbers

The US unemployment rate, ACS basis, has ranged from about 3% (tight labor markets) to over 10% (post-2008 trough) over the past two decades. State rates rarely exceed 7% even in downturns, but county and city rates can spike much higher in places with concentrated industry exposure. Below about 4%, unemployment becomes a misleading metric, places at "full employment" can still have high underemployment (people working part-time who want full-time hours) or labor-force-participation gaps that the unemployment rate doesn't capture.

Caveats and limitations

The ACS rate is a 5-year average, so it smooths over short-term spikes and lags rapid changes. It also undercounts the actually-jobless because it excludes discouraged workers and the underemployed. For a complete picture of labor-market health, look at unemployment alongside the labor-force participation rate, a place with falling unemployment AND falling participation is often weaker than the headline rate suggests. Small-place unemployment estimates carry wide confidence intervals; differences of 1-2 percentage points between adjacent ranks may not be statistically meaningful.