Temporary help services employment by county

The number everyone asks for is an administrative headcount, not a survey estimate. Under NAICS industry 5613 - Employment Services - sit the businesses that sell labor for a living: temporary help services, employment placement agencies and professional employer organizations. Every employer covered by state unemployment insurance law reports employment and wages as a condition of that tax system, so a county-quarter cell is a count of actual staffing branches rather than a projection with a margin of error.

The Q1 2025 slice for that one industry code alone holds 3,623 published records across counties, metropolitan areas, states and the nation, inside an archive window that opens in 1990. That is thirty-five years of county-grain staffing history - depth almost nothing else in the labor-data world can match at this geography.

Datadory delivers BLS QCEW - Employment Services (NAICS 5613) as one joinable table: establishments, month-one-through-three employment levels, quarterly payroll, average weekly wage, plus the location quotients and over-the-year changes already computed upstream of you.

Two records carry the county question, and they answer different halves of it.

BLS QCEW - Employment Services NAICS 5613 is the anchor: one record per county x ownership x industry x quarter, cut to every U.S. county, MSA, state and the nation, scored Quality 10 on Datadory's rubric - the top of the nineteen-record human resource and employment services shelf.

Census LEHD Quarterly Workforce Indicators - NAICS 56 supplies the worker-flow half QCEW lacks: quarterly hiring, separations, turnover and earnings, reaching county, metro/micro and Workforce Investment Board areas, observed from 1991 Q1 forward.

Around them sit the complements: BLS JOLTS for monthly openings, hires and separations since December 2000; BLS OEWS for what the blended county wage is made of; ASA Staffing Industry Research for the trade body's own counts of staffing employment and penetration; California EDD WARN Notices for county-identified layoff notices refreshed weekly.

For discovery past the shelf, data.gov Employment Services datasets indexes the wider federal-state pool - a relevance-ranked subset of a 552,271-dataset catalog.

One row per area, per ownership, per industry, per quarter. Two real records from the first quarter of 2025, both private ownership, both NAICS 5613:

area   own naics year qtr  estabs emplvl_m1 emplvl_m3 qtrly_wages avg_wkly_wage
01000  5   5613 2025 1    3021   47890     50658     495427517   765
01001  5   5613 2025 1    12     24        22        882848      2996

Both behaviors matter when you size a market: the state row benchmarks, the county row locates. And because the monthly triple rides inside every quarterly record, a branch-opening surge never averages away the way it would in a single quarter-end snapshot.

Twenty-seven fields carry each record, and three groups do most of the modeling work.

Suppression arrives flagged, not silent: cells withheld for confidentiality land marked in disclosure_code rather than blanked without explanation, so any total you cannot re-derive announces itself. Annual-average rows extend the quarterly core with annual employment, total wages and average pay, folding four quarters into one comparable figure.

Treat NAICS 5613 as the official county proxy for the segment: it groups temporary help with placement agencies and PEOs, so the series measures staffing-and-placement together rather than temp help alone.

Files, feeds, or straight into your warehouse. Daily, weekly, or hourly - your call.

Extraction, typing and schema stability happen upstream of you. Suppressed cells stay flagged, revisions propagate, and nobody on your side maintains a downloader. The sample comes first: name the counties, get real rows, then decide.

Three faster series bracket the cycle between prints, all delivered through Datadory:

• BLS JOLTS reports openings, hires and separations monthly, all 50 states plus DC at total nonfarm, with national estimates by industry and size class - the current file runs December 2000 through June 2026, roughly 376,000 rows.
• ASA Staffing Industry Research moves weekly: its Staffing Index sits beside monthly index reporting, quarterly employment-and-sales benchmarks and an annual sales series spanning 2004-2022.
• California EDD WARN Notices refreshes weekly with about 138 mass-layoff notice rows carrying employee counts per event, each identified by county - one state, but the template for county-grain early warning anywhere you expand.

Against that speed sits the payoff of patience: the county archive opens in 1990, giving three and a half decades of seasonality and trend history at geography most labor series never reach.

Close substitutes exist for Canada and Europe, though none match county precision - and every one ships through Datadory in the same typed-row format as the U.S. ledger.

Statistics Canada - NAICS 56 data tables tracks administrative and support employment monthly across Canada's 14 provinces and territories; Table 14-10-0201 alone holds 17,598 series and about 5.3 million datapoints, running January 2001 through May 2026 as of the August 2026 release.

Destatis Germany Labour Market Statistics publishes monthly German employment and business-services series with Bundesland breakdowns, some series reaching back to 1970 or 1991.

ONS People in Work organizes the UK labour market into hundreds of series - its A01 summary workbook alone carries about 25 sheets, and the latest Labour market overview landed 18 August 2026.

Above the series level, World Employment Confederation Statistics sizes the private employment services market across roughly 40 countries and six service segments, with placements, penetration rates and agency counts in a 41-page annual edition.

The table below lines the ledgers up side by side as of August 2026.

Raw headcount flatters big counties, so run the ledger beside two complements.

Wage context comes from BLS Occupational Employment and Wage Statistics: the May 2025 all-data file holds 413,528 rows and 32 columns across roughly 530 metropolitan and nonmetropolitan areas, with nine wage percentiles for about 830 occupations including HR managers, HR specialists and recruiters - the people a branch would actually hire. Payscale Research adds crowd-reported 10th-90th percentile bands with bonus and commission components; its HR Manager page alone cites 15,237 salary profiles.

Concentration comes free with the ledger itself: because location quotients arrive precomputed, ranking counties by staffing intensity relative to the national baseline is one sort, not a join tree. A county can post modest absolute headcount and still be one of the most staffing-dependent labor markets in the country - exactly the signal a site-selection or territory model needs before capital commits.

Downside risk surfaces in California EDD WARN Notices, whose weekly sweep shows what county-grain early warning looks like in practice - a pattern worth replicating wherever you expand.

• Market researchers read location quotients as concentration made numeric - which counties run hot on staffing and which are white space; the sizing workflow lives on the market researchers use cases page.
• Investors and quant researchers treat county staffing payrolls and their over-the-year changes as ground-truth reads on contingent-labor demand, joined against operator revenue; the panel assembly is on the investors quants use cases page.
• Sales and growth teams turn establishment counts into territory math - how many placement agencies, temp-help operators and PEOs sit inside a county or metro, and what they pay on average.
• Competitive intelligence teams watch establishment churn and wage drift by geography to see where staffing capacity forms before it reaches headlines; the play is on the competitive intel product teams use cases page.
• Journalists and academics cite an administrative count they can explain in one sentence - a census of UI-covered employers, not a poll.

County-level staffing employment is one question inside a nineteen-record shelf. The human resource & employment services data guide walks all seven ledgers - vacancy feeds, wage benchmarks, skills taxonomies and trade-association research - the full inventory with quality scores sits on the human-resource-employment-services hub, and the QCEW employment and wages glossary entry defines the vocabulary the field dictionary uses.