Manufacturing Economy
Two Productivity Gains Barely Recovered Manufacturing’s Previous Decline
Compounding the published annualized quarterly productivity rates of −3.60%, +2.20% and +2.40% gives about +0.22% across Q4 2025–Q2 2026. Unit labor costs rose about 2.91% across the same three quarters.
Editorial evidence cutoff: September 9, 2026. Published September 29, 2026. Observation periods are stated throughout; older figures are retrospective evidence.
Manufacturing productivity rose in the first two quarters of 2026, but those increases barely repaired the preceding decline. Compounding the published annualized changes of minus 3.60%, plus 2.20% and plus 2.40% produces an approximate cumulative gain of just 0.22% from the third quarter of 2025 to the second quarter of 2026. Unit labor costs rose approximately 2.91% over the same span.
This calculation uses the historical series available by September 9, including the September 3 BLS productivity release. It offers a more useful September perspective than counting positive headlines. The direction of the latest change matters, but a recovery assessment also needs to show how far the level has moved from its earlier starting point.
THREE HEADLINES DO NOT ADD UP TO A CUMULATIVE RESULT
The rates in the quarterly productivity series are annualized. A 2.40% annualized quarterly increase does not mean productivity rose 2.40% during that quarter. It expresses the pace as though the quarter's change continued through a full year. Recovering the quarterly factor requires taking the fourth root of one plus the annualized rate.
For each quarter, the calculation uses the factor (1 + rate / 100) raised to one quarter. Multiplying the three quarterly factors then gives the cumulative change. This procedure treats the decline and subsequent gains consistently. Adding minus 3.60, plus 2.20 and plus 2.40 would produce a number that is neither the cumulative change nor a correctly calculated annual growth rate.
A RECONSTRUCTED LEVEL EXPOSES WHAT THE HEADLINES HIDE
Set manufacturing labor productivity to an illustrative index of 100 in the third quarter of 2025. The annualized fourth-quarter decline pushes that index below 100. The first-quarter increase recovers part of the loss, and the second-quarter increase brings the reconstructed level only slightly above its starting value, to approximately 100.22.
The sequence matters more than the count of positive quarters. Two increases can follow a sufficiently large decline and still leave the level below its original position. Here they just exceed it. That is a different description from suggesting that two positive reports established a substantial cumulative productivity advance. It also does not diminish the positive direction of the last two observations; both statements can be true.
THE UNIT-LABOR-COST PATH DID NOT RETURN TO ITS START
The corresponding annualized unit-labor-cost changes were plus 8.90% in the fourth quarter of 2025, plus 3.30% in the first quarter of 2026 and minus 0.30% in the second quarter. Applying the same compounding method gives an approximate cumulative increase of 2.91%. An illustrative cost index beginning at 100 would therefore finish near 102.91.
The second-quarter decline in unit labor costs was real within the published estimate, but it did not erase the earlier increases. A September budget that interprets the latest negative rate as a return to the old cost level would confuse a change in direction with a full reversal. The cumulative calculation makes that distinction visible without requiring a prediction about the following quarter.
PRODUCTIVITY IS AN OUTPUT-PER-HOUR MEASURE
The aggregate measures output relative to labor hours. It is not a direct score of individual effort or diligence. Changes in production volume, capital, technology, organization and the composition of activity can influence the result. A national increase does not prove that every factory worker became more productive by the reported percentage.
Unit labor costs link compensation to output rather than reporting an hourly wage alone. The series therefore cannot be replaced by a payroll wage series without changing the question. A business may experience rising hourly compensation alongside a different cost-per-unit path if output per hour changes. Interpreting the productivity and cost measures together is more informative than assuming one is a direct proxy for the other.
ROUNDING LIMITS THE PRECISION OF THE RECONSTRUCTION
The 0.22% and 2.91% cumulative results are calculated from published rates rounded to one decimal place. They are approximate reconstructions, not exact official level changes carried to arbitrary precision. Official index observations are preferable when an exact cumulative comparison is required. Reporting two decimals makes the calculation inspectable; it does not eliminate uncertainty introduced by rounded inputs.
The endpoint also matters. This comparison begins in the third quarter of 2025, before the reported fourth-quarter decline. It does not describe all of 2025, the entire recovery since the pandemic, or a year-over-year change. A different starting period answers a different question. The selected window is useful specifically for testing whether the two subsequent gains recovered the immediately preceding loss.
AN AGGREGATE RECOVERY CANNOT VALIDATE EVERY FACTORY TARGET
A plant manager should not use the national 0.22% cumulative result as a target for a specific line without examining its products, labor hours and operating conditions. The broad manufacturing series combines industries with different processes and different movements. A site's output mix or hours can move unlike the aggregate even when its internal performance improves.
The historical record nevertheless provides a useful challenge to a broad planning assumption. A budget justified by claims of rapid manufacturing productivity improvement needs to specify the actual period and level comparison behind that claim. If the supporting evidence consists only of the last two annualized headlines, it has not yet established the cumulative improvement the budget assumes.
THE SEPTEMBER CONCLUSION IS ABOUT LEVELS, NOT A FORECAST
By the September 9 cutoff, the available three-quarter sequence showed a modest net productivity gain and a larger cumulative rise in unit labor costs. It did not show whether the following quarter would accelerate, reverse or remain close to the earlier level. That forecast requires evidence beyond this arithmetic.
What the calculation does establish is simple and consequential. Recovery is a question about levels as well as directions. The latest positive observation can be encouraging while the cumulative improvement remains small. Keeping both facts in view produces a more defensible account of manufacturing costs and performance than allowing a succession of favorable headlines to stand in for the underlying path.
Sources and evidence
Evidence period: 2025Q3 base=100 through 2026Q2. The frozen evidence record lists the source files and verified hashes available September 9, 2026. Source revision: 5e4fb7726c3d40060c0151c086c903baae856cab. Later live-data updates do not alter the historical evidence in this article.
bls.gov/news.release/archives/prod2_09032026.htm
Published 2026-09-29.