Revelio’s Realistic Assessment Of The US Labor Market And Jobs – Sinking Fast

By Mish Shedlock of MishTalk

Friday, November 7, 2025 9:20 AM EST

Kudos to Revelio for providing an excellent set of jobs-related data.

Revelio Labs employment year over year, chart by Mish

Year Over Year Change

January 2022: 4.585 million
January 2023: 3.466 million
January 2024: 1.436 million
January 2025: 561 thousand
October 2025: 203 thousand

Revelio Press Release

Revelio reports Public Labor Statistics (RPLS) Reports U.S. Economy Shed 9,100 Jobs in October as BLS Shutdown Persists

Revelio Labs today released its October edition of Revelio Public Labor Statistics (RPLS), a monthly data release delivering transparent and reliable insights into the U.S. labor market. The latest data indicate a loss of 9,100 jobs in October, signaling flat employment growth and underscoring a broader slowdown across the labor market.

In terms of leading indicators, active job postings declined by 1.9% month-over-month, while salaries from new job postings increased by 0.7%. Both hiring and attrition rates also declined, extending the downward trend seen in recent months.

“Our October RPLS numbers show that the labor market has continued to cool as expected. The slightly negative job growth is driven by job losses in the Government sector. Education and Health is the only sector to meaningfully add jobs last month. Our job postings, hiring, and attrition numbers are also all pointing downwards. The Fed made the right call in their last meeting and may want to look at our numbers ahead of the December meeting,” said Chief Economist Lisa Simon.

RPLS is a freely available macroeconomic labor market set of statistics built from 100+ million U.S. profiles to provide a clear view of workforce dynamics. It follows a format similar to the U.S. Bureau of Labor Statistics (BLS), tracking employment levels, wages, and job transitions at a scale that traditional surveys cannot, offering a continuous picture of the labor market. RPLS intends to close the growing information gap and deliver unbiased data on the U.S. workforce for policymakers, businesses, and the public.

Revelio Labs Revisions vs BLS

Revelio Labs Employment in Thousands

Revelio Labs national employment level, chart by Mish

Revelio Month-over-Month Change by Industry

Revelio Labs employment month-over-month change, chart by Mish

Month-Over-Month Change in Thousands by Industry

Agriculture: 0.7
Construction: 1.6
Manufacturing: -5.1
Wholesale Trade: -2.8
Retail Trade: -8.5
Transportation and Warehousing: 1.7
Financial Activities: 9.6
Education and Health Services: 21.9
Leisure and Hospitality: -0.6
Other Services: -2.3
Government: -22.2

The total including small numbers not shown is -9,100 for October.

In the last two months, government shed 43,100 jobs. This is expected. However, last month seems early.

Laid off government workers were paid through September.

Year-Over-Year Change by Industry

Year-Over-Year Change in Thousands by Industry

Agriculture: 1.4
Mining: -14.2
Utilities: 9.0
Construction: 48.1
Manufacturing: -40.8
Wholesale Trade: -31.9
Retail Trade: -72.6
Transportation and Warehousing: 59.6
Information: -8.8
Financial Activities: 172.2
Professional and Business Services: 18.6
Education and Health Services: 403.6
Leisure and Hospitality: -199.3
Other Services: -29.1
Government: -89.6
Unclassified: 2.3

That is quite the mix.

The only two big bright spots were Education and Health Services (related to aging demographics and illegal immigration), and Financial Activities (related to AI and stock market deals).

The latter seems poised to cool.

Construction is related to data center builds, not residential or office. Warehousing is related to tariff front running and storage.

Based on the lead chart, tariffs, and small business woes, and immigration, I expect the year-over-year net number to soon be negative.

FAQs

What data sources are used? Powered by a dataset representing close to the whole population of employed people in the United States, Revelio Public Labor Statistics (RPLS) will draw from 100+ million U.S. profiles that mirror the national workforce and cover two thirds of all employed individuals, compared to an estimated 27% from the BLS establishment survey and 0.03% from the BLS household survey.
What metrics are covered? Revelio Public Labor Statistics (RPLS) will publish (a) changes in employment, (b) job openings (c) employee hiring and attrition rates, and (d) salaries from new openings. Note that we will not be covering unemployment rate and labor force participation rate. All metrics will be available by occupation (SOC 2-digit), sector (NAICS 2-digit), and state.
What is the release schedule? Revelio Labs will publish its data monthly, releasing Revelio Public Labor Statistics (RPLS) updates the day before Jobs Friday. These monthly releases are designed to complement the BLS’s data series with more timely and granular insights.

About Revelio

Revelio Public Labor Statistics (RPLS) is a freely available macroeconomic labor market dataset built from 100+ million U.S. profiles to provide a clear, transparent view of workforce dynamics. It follows a format similar to the U.S. Bureau of Labor Statistics (BLS) and is designed to be used alongside it, tracking employment levels, wages, and job transitions at a scale that traditional surveys cannot, offering a continuous picture of the labor market. Our hope is that RPLS helps close the growing information gap and delivers unbiased and trustworthy data on the U.S. workforce for policymakers, businesses, and the public.

Methodology

RPLS provides a set of employment statistics derived from over 100 million professional profiles sourced from professional networking websites. After deduplication, adjustments for reporting lags, and reweighing to ensure that the data resembles the national distribution of the workforce, these data yield timely and detailed measures of employment dynamics. The resulting data series captures the level of total employment (headcount), which informs monthly job gains, and the flows of workers into and out of jobs. This data enables us to track hiring, separations, and net employment changes with a scope and frequency comparable to official labor statistics

Online professional profiles are not a representative sample of the U.S. workforce. Profiles on these platforms disproportionately reflect white-collar occupations, larger firms, and workers in metropolitan areas, while under-representing blue-collar jobs, smaller companies, and rural locations. To correct for these biases and produce nationally representative measures of employment, we construct sampling weights using two complementary models at the micro and macro sides.

Comparison to CES

The BLS tracks changes in employment via two main surveys. The establishment survey that is used to generate the Current Employment Statistics (CES), surveys approximately 121,000 businesses and government agencies, covering roughly 631,000 individual worksites, and serves as the source of the monthly “headline” change in payroll jobs. The household survey, covering around 60,000 households, captures broader labor market trends, including payroll jobs, self-employment, and multiple job holdings. Together, these surveys provide complementary snapshots of U.S. employment.

When we compare Revelio Labs’ estimated employment series to CES estimated employment, we find that Revelio Labs captures the same broad trends while offering more timely observations and flexibility in disaggregation.

Comparison to other employment data sources

One prominent source of employment data is produced by , drawing on payroll records covering more than half a million companies and over 25 million employees. Unlike Revelio Labs’ profile-based data, which captures both private- and public-sector workers, ADP’s measure is limited to private payroll employees. ADP draws from two sources: payroll transactions, which record when individuals are paid and how much, and administrative records that identify who is currently on the company payroll, even if not paid in a given cycle. This provides detailed business-level insight into private employment trends, but differs in scope and design from Revelio Labs’ approach, which relies on individual-level professional profiles and offers a broader view of the workforce, including public-sector employment.

Revisions

RPLS revisions has been roughly half that observed in CES revisions in both the positive and negative directions, reflecting the stability and timeliness of our real-time pipeline. Furthermore, RPLS revisions are less procyclical and serially correlated compared to CES revisions.

Thanks to Revelio for making their data public.

Anyone for replacing the BLS with Revelio?