The discrepancy between QCEW and the BLS jobs report is rising.

What is QCEW?

QCEW stands for Quarterly Census of Employment and Wages

The Quarterly Census of Employment and Wages (QCEW) program publishes a quarterly count of employment and wages reported by employers covering more than 95 percent of U.S. jobs, available at the county, MSA, state and national levels by industry.

QCEW vs Nonfarm Payrolls

The above image from QCEW Concepts.

CES stands for Current Establishment Survey, the BLS nonfarm payroll report.

The QCEW consists of 12.2 million establishments in the first quarter of 2024. The BLS nonfarm payroll report has a sample size of 629,000 individual worksites in 2024.

QCEW excludes proprietors, the unincorporated self-employed, unpaid family members, certain farm and domestic workers, and railroad workers covered by the railroad unemployment insurance system. Also excluded are elected officials in the executive or legislative branch, members of the armed forces or the Commissioned Corps of the National Oceanic and Atmospheric Administration.

Neither QCEW nor CES include self-employed persons in their counts. Both BLS and QCEW may double count individuals with multiple jobs.

BLS Response Rates

The above image is from BLS Response Rates.

Comparison: QCEW consists of 12.2 million establishments with about 95 percent coverage.

The Nonfarm payroll report surveys ~629,000 establishments with a response rate of 42.6 percent as of March 2025.

Large establishments likely have someone responsible for handling government surveys.

Small businesses, especially struggling businesses, may not take the time to respond to surveys.

To make up for its small sample size, the CES report uses a Birth-Death model to estimate the numbers of employees at new businesses vs businesses going out of business. At turns, these estimates may be seriously wring.

Timely Garbage

The latest QCEW stats are from December of 2024. The latest BLS nonfarm payroll report is May of 2025.

Because of sampling errors, non sampling errors, and questionable birth-death models, the BLS numbers may best be considered timely garbage,

Does the Discrepancy Between CES and QCEW Matter?

The answer is no. Rather, it’s the direction and rate of change that matters.

For several quarters, I tracked QCEW vs BLS. But it was the same discrepancy over and over again.

Now we see evidence that the discrepancy is growing.

I downloaded the monthly data from the QCEW and produced the lead chart and the following chart.

QCEW vs Nonfarm Payrolls Change from Year Ago in Thousand

QCEW Chart Notes

• Between July 2023 and December 2023 the discrepancy was generally narrowing.
• Between January 2024 and May 2024 the discrepancy was generally stable.
• Between June 2024 and December of 2024 the discrepancy was generally increasing.

Those are year-over-year numbers.

When making year-over-year comparisons, it makes little difference if one compares seasonally-adjusted numbers or unadjusted numbers.

We need month-over-month comparisons.

However, one cannot accurately compute month-over-month changes from unadjusted numbers, and QCEW data is not seasonally adjusted.

Lack of month-over-month analysis makes QCEW even less timely.

But what if we could seasonally adjust QCEW numbers?

I do not have a way to seasonally adjust QCEW, but I do have permission from John “Jake” Bush at Piper Sandler & Co. to post his seasonally-adjusted QCEW data.

QCEW vs Nonfarm Payrolls Seasonally Adjusted

QCEW SA by John Bush, Nonfarm Payrolls SA by the BLS

The seasonally-adjusted trends by Bush show the increasing discrepancy between the surveys better than my unadjusted charts.

In 2024, QCEW flattened dramatically whereas reported nonfarm payrolls accelerated to the upside in the fourth quarter of 2024.

The next chart provides better visualization.

BLS Nonfarm Payrolls Minus QCEW Month-Over-Month

QCEW SA by John Bush, Nonfarm Payrolls SA by the BLS, calculation by Mish

The seasonally adjusted discrepancy between the reports was stable for 2023 but is now increasing, noticeably so in the last three months.

The discrepancy is now 3.6 million. More importantly, it’s rising.

This strongly suggests bigger negative revisions to CES data will come in 2025.

Questions of the Day

Given Piper Sandler knows how to seasonally adjust data, and so does the New York Fed, the Philadelphia Fed, ADP, and the BLS itself ….

• Why doesn’t the BLS seasonally adjust QCEW data?
• Why does the BLS rely on proven wrong birth-death models instead of using faster QCEW data with a 2 or 3 month lag instead of 5 months?
• Why can’t we have the QCEW monthly instead of quarterly, even if it’s preliminary?

It’s clear to everyone that the monthly jobs report is totally screwed up and that Covid distortions and Birth-Death nonsense are still in play.

Perhaps Covid permanently disrupted data collection, perhaps it’s immigration, perhaps both, or perhaps it’s something unknown, but the BLS had made no serious attempt to fix the problems.

The CES model is flawed. What should we do about that?

Mish Suggestion

Rather than tweak a broken model, I suggest starting all over with preliminary MCEW (monthly) reports instead of QCEW (quarterly) reports.

Note that even the first two months of CES are considered preliminary, and they are based on tiny samples with low response rates, massaged by a broken Birth-Death model.

How could my proposed MCEW be worse than that?

Looking Ahead

The payroll and QCEW trends will be increasingly negative due to Trump’s deportations, fear of immigrants to show up on the job, tariffs costs, and tariff uncertainty.

It is difficult to see how we avoid recession.

Related Posts

June 5, 2025: Fed Beige Book Shows Only 3 of 12 Regions Growing, 6 Declining

This report reeks of stagflation, defined as rising prices and recession simultaneously.

June 6, 2025: Nonfarm Payrolls Rise by 139,000 Employment Declines by 696,000

Can anyone believe these reports anymore?

June 4, 2025: Trump Demands Fed Rate Cut After Weakest ADP Payroll Report in 2 Years

ADP reported a slim 37,000 private payroll rise for May.

Large- and medium-sized employment has peaked and is now rolling over while small-business employment has stagnated.

Businesses with 20-49 employees are shedding jobs rapidly.

There is no driver for job growth. Tariff uncertainty alone will halt much business expansion.

More By This Author: