The +178K Nobody Should Take at Face Value

The Bureau of Labor Statistics reported this morning that the US economy added 178,000 nonfarm payroll jobs in March — a number that tripled the Wall Street consensus of 57,000 and which financial media will spend the weekend describing as a blowout. It is not a blowout. It is a statistical artifact that happens to be wearing blowout clothes, and it will be traded as a blowout on Monday when markets reopen from Good Friday, which means the confusion will have real consequences.

Here is what the report actually shows.

The strike math

Healthcare alone accounted for 76,000 of the 178,000 total — a sector running at 2.6 times its trailing twelve-month average of 29,000 per month. The reason is not that healthcare hiring accelerated. The reason is that a physicians strike at Kaiser Permanente, which had pulled approximately 37,000 workers off payrolls in February, resolved in March. Those workers returned. The establishment survey counts returns as additions. Offices of physicians added 35,000; hospitals added another 15,000 separately.

Strip out healthcare, and the organic print is approximately 102,000. That is still above the 57,000 consensus — the underlying data is not bad — but it is not a blowout. It is a respectable month in a soft labor market.

The distinction matters because the methodology is the message. The establishment survey does not distinguish between workers who were newly hired and workers who resolved a dispute and went back to existing jobs. Both register as additions. A reporter who reads the BLS press release as "the labor market is on fire" has misread the instrument. (Source: [BLS Employment Situation, March 2026](https://www.bls.gov/news.release/archives/empsit_04032026.htm); [Verified Investing breakdown](https://verifiedinvesting.com/blogs/us-economic-metrics/us-nonfarm-payrolls-march-2026-jobs-report))

The number that actually matters to the Fed

Average hourly earnings rose 0.2% month-over-month in March — half the pace of January and February. The year-over-year rate dropped to 3.5%, down 30 basis points from February's 3.8%. That is the most significant single observation in this report, and it is buried beneath the headline.

Wage growth of 3.5% year-over-year, with the personal consumption expenditures deflator running near 2.5%, implies real wage growth of roughly 1%. That is not inflationary. That is actually the kind of wage growth the Fed has said it wants to see — compensation rising without stoking the price pressures that would require additional tightening. The March print reduces, not increases, the case for a rate hike.

The Fed's April 28-29 FOMC meeting is the next scheduled decision point. There will not be another major employment report before it. The March data — properly read — is dovish. ([Samuel Tombs, Pantheon Macroeconomics, via Reuters](https://www.reuters.com/world/us/us-employment-growth-rebounded-march-unemployment-rate-falls-43-2026-04-03/): "Payrolls boosted by one-time factors; the trend still looks weak.")

The revision story that changes the narrative

The February figure was revised from -92,000 to -133,000. That is a 41,000-unit downgrade, and it changes the character of February from "a bad month, explainable" to "the worst single month of job losses in the current cycle, full stop."

January was revised upward, from 126,000 to 160,000. The two revisions net out to 7,000 fewer jobs than previously estimated. More usefully, they give us a cleaner picture of the three-month trend: the average across December, January, and February — adjusted for all known revisions — is approximately 3,000 jobs per month. Not 3,000 per month seasonally adjusted. Three thousand. Effectively flat.

A headline March print that triples the consensus looks different when the baseline it's recovering from is a cycle-worst February, and the three-month trend underneath it is stagnant. The strong March and the weak February are not separate facts. They are a single fact about a labor market that is oscillating around zero, distorted by one-time events in both directions.

Federal government: the contraction continues

Federal employment fell another 18,000 in March. Since October 2024, the federal workforce has contracted by 355,000 — a reduction of 11.8 percent from peak. The DOGE-driven downsizing has now run for six months and shows no sign of plateauing.

To put that number in historical context: 355,000 federal job losses in six months represents roughly the equivalent of eliminating the entire civilian workforce of the Department of Defense, twice. These are positions that typically do not come back on the same timeline they were lost; federal hiring operates on civil service timelines, not market-speed adjustments. The structural reduction in federal employment capacity has implications — for agency function, for the communities near federal installations, and for the broader labor market — that will be measured in years, not quarters.

Manufacturing showed little change in March, consistent with the ISM Manufacturing Employment sub-index at 48.7 — a reading that indicates sector-level contraction. Construction added 26,000 on weather normalization, not on a trend reversal. Transportation and warehousing added 21,000 (mostly couriers), but the sector is still down 139,000 from its February 2025 peak.

The Monday problem

US equity and bond markets were closed today for Good Friday. The March jobs report released into a void. By the time traders can react — Monday morning, April 6 — they will be processing two things simultaneously: a payroll number that appears, on its face, to be a strong beat; and the administration's April 2 tariff announcement, which this report's reference period does not capture at all.

The tariff announcement and the jobs report tell opposite stories about economic trajectory. The jobs report is backward-looking: it describes what happened in March, before the new trade regime took effect. The tariff announcement is forward-looking: it remakes the cost structure for imported goods, with effects on inflation, consumer spending, supply chains, and business investment that have not yet propagated through to any available data. Absorbing both simultaneously is exactly the condition that produces volatile gap opens and poor price discovery. The financial system is being asked to trade on incomplete information about an economy in transition.

Watch May 8. The April employment report — covering the reference period that begins after April 2 — will be the first report whose data reflects the labor market impact of the new tariff structure. That print, not this one, is when the accounting begins.

What the data does and does not show

The March employment report shows: a labor market that produced a respectable organic monthly gain of roughly 102,000 once the strike-reversal distortion is removed; wage growth that is cooling materially, which is exactly what the Fed needs to see; a federal workforce that continues to contract at a historically significant pace; and an underlying three-month trend that is effectively flat.

What it does not show: any data from after April 2. Any effect of the Iran war on business investment and hiring. Any response to the new tariff structure. The March report is a snapshot of an economy whose major near-term determinants had not yet occurred during its reference period. This is not a criticism of the BLS methodology — it is an inherent limitation of any retrospective survey. But reading the March report as a forecast of the second quarter is not a reading of the data. It is an act of hope.

The actual economic conditions that will govern the next six months — oil above $4 nationally, a global tariff structure that is still being negotiated, a Supreme Court-constrained executive acting at the edge of its authority, an ongoing war in the Middle East with no visible end — are present in none of these numbers. They are present in the numbers that have not yet been collected.

Sources: [BLS Employment Situation Summary, March 2026](https://www.bls.gov/news.release/archives/empsit_04032026.htm) (USDL-26-0580); [Verified Investing, NFP March 2026 breakdown](https://verifiedinvesting.com/blogs/us-economic-metrics/us-nonfarm-payrolls-march-2026-jobs-report); [Reuters, "US job growth accelerates by the most in 15 months in March"](https://www.reuters.com/world/us/us-employment-growth-rebounded-march-unemployment-rate-falls-43-2026-04-03/), April 3, 2026; [BLS Employment Situation: Sector detail, March 2026](https://www.bls.gov/news.release/empsit.nr0.htm); Samuel Tombs, chief US economist, Pantheon Macroeconomics, via Reuters, April 3, 2026.