Evidence auditCompanion analysis · 20 August 2026

Orders / composition / realized turnover

Germany’s order boom is smaller than it looks.

Manufacturing orders rose 3.1% in June. Remove major transactions and they fell 0.5%. Real turnover fell 1.3%. The headline is correct; the recovery story is too broad.

Follow the decomposition
June release / month on monthOne month. Three readings.
June 2026 manufacturing signalsTotal manufacturing orders increased 3.1%, orders excluding major transactions declined 0.5%, and real turnover declined 1.3% from May.-2%-1%0%+1%+2%+3%+4%TOTAL ORDERS+3.1%EX-MAJOR ORDERS−0.5%REAL TURNOVER−1.3%

Volume indices · X13 adjusted · 2021 = 100

Total orders / June
+3.1%

Official headline

Orders ex-major
−0.5%

+3.5 pp large-order impulse

Real turnover / June
−1.3%

Activity realized now

The claim

A large order is real. It is not the same thing as broad demand.

Germany’s June release appears to deliver a clean result: real manufacturing orders increased +3.1% from May. That number is neither wrong nor meaningless. Large projects create workload, investment, and future production. Yet the same release shows orders excluding major transactions falling −0.5% and real turnover declining −1.3%. Three correct statistics imply three different diagnoses.

The distinction is composition. Total orders answer how much indexed demand arrived, including unusually large projects. The ex-major series asks whether the underlying flow improved without those projects. Turnover asks whether factories realized more activity now. Calling the first result a manufacturing recovery assumes that all three move together. In June, they did not.

The order boom exists in the aggregate. The broad recovery does not yet exist in the data.

This is the companion question to yesterday’s order-conversion audit. That article tested whether backlogs became less predictive of production after 2021 and found no clear structural deterioration. Today’s question comes one step earlier: how much of the incoming demand is broad enough to require conversion in the first place?

Order breadth

June improved the headline much more than the underlying path.

Interactive 01 / order breadth

The headline and the broad order path are not the same series.

Both indices are rebased to January 2019. The impulse view subtracts monthly broad-order log growth from total-order growth.

Total and broad German manufacturing orders, January 2019 to June 2026Total orders and orders excluding major transactions, indexed to January 2019. The total series is more volatile and rises above the broad series in June 2026.506070809010011012020192020202120222023202420252026
Total ordersOrders excluding major transactions
View the latest 12 months as a table
MonthTotalEx-majorImpulse
Jul 202590.288.0−3.2 pp
Aug 202588.884.7+2.3 pp
Sep 202593.390.3−1.6 pp
Oct 202594.589.8+1.8 pp
Nov 202599.590.2+4.7 pp
Dec 2025105.189.6+6.2 pp
Jan 202693.187.8−10.2 pp
Feb 202694.690.8−1.7 pp
Mar 202698.894.9+0.0 pp
Apr 202695.691.7+0.1 pp
May 202695.991.2+0.9 pp
Jun 202698.990.8+3.5 pp
Momentum through June 2026
Series3m / prior 3m6m annualizedYear over year12m volatility
Total orders+1.4%−5.8%+4.4%17.5%
Excluding major transactions+0.1%+2.9%+3.3%10.6%
01

Quarterly breadth was flat

Ex-major orders rose only +0.1% against the prior three months.

02

The headline is noisier

Annualized volatility is 17.5% versus 10.6% ex-major.

03

Geography also split

Euro-area orders fell 14.0%; non-euro-area orders rose 10.2%.

Sector split

Five sectors produced five different versions of June.

Sector divergence / June 2026Orders booked versus turnover realized
June sector orders and turnoverA dumbbell chart comparing monthly order growth and monthly turnover growth across chemicals, electronics, machinery, motor vehicles, and other transport equipment.-40%-20%0%+20%ChemicalsChemicals orders: −3.2%Chemicals turnover: −2.3%GAP −0.9 PPElectronicsElectronics orders: +22.7%Electronics turnover: −2.3%GAP +25.1 PPMachineryMachinery orders: +12.7%Machinery turnover: −0.3%GAP +13.0 PPMotor vehiclesMotor vehicles orders: +3.8%Motor vehicles turnover: −1.3%GAP +5.0 PPOther transport equipmentOther transport equipment orders: −41.7%Other transport equipment turnover: +18.5%GAP −60.2 PP
View the sector values as a table
June 2026 monthly and annual order and turnover growth by manufacturing sector
SectorOrders m/mTurnover m/mOrders y/yTurnover y/yMonthly gap
Chemicals−3.2%−2.3%−0.7%+2.4%−0.9 pp
Electronics+22.7%−2.3%+29.5%−7.6%+25.1 pp
Machinery+12.7%−0.3%+19.7%−8.9%+13.0 pp
Motor vehicles+3.8%−1.3%+6.0%−1.1%+5.0 pp
Other transport equipment−41.7%+18.5%−1.2%+26.9%−60.2 pp

Electronics and machinery supply the intuitive recovery story: orders increased +22.7% and +12.7%. Yet turnover fell in both. Chemicals weakened on both measures. Other transport did the reverse: orders collapsed −41.7% while turnover jumped +18.5%.

These are not contradictions. Large projects have milestones, delivery schedules, and recognition dates. A sector can book an aircraft or machinery order today and realize turnover months later; it can also deliver an old order while current bookings fall. The point is not that orders are useless. The point is that sector timing prevents a one-month aggregate from describing the median factory.

Execution gap

The latest mismatch is large. That still does not prove a broken mechanism.

Electronics

+3.15σOrders +20.5% / turnover −2.4%

Machinery

+2.23σOrders +12.0% / turnover −0.3%

The execution proxy reaches +3.15 standard deviations in electronics and +2.23 standard deviations in machinery. That makes June unusual. It does not make the divergence permanent.

Yesterday’s longer backlog analysis matters here. It found that post-2021 order stocks were not clearly less predictive of future production than before the pandemic, although controlled coefficients were imprecise. The two results fit together: June contains a severe execution mismatch, while the historical record does not establish that German industry has permanently lost the ability to convert demand. Episode and structure are different claims.

Predictive test

The headline does not become a crystal ball when it enters a regression.

Interactive 02 / predictive check

Neither order measure becomes a reliable turnover forecast.

Coefficients are standardized. Intervals use Newey-West HAC uncertainty with lag length equal to the forecast horizon.

electronics2015-2019
Broad orders
−0.06
electronics2015-2019
Large-order impulse
+0.02
electronics2021-2026
Broad orders
+0.03
electronics2021-2026
Large-order impulse
+0.27
machinery2015-2019
Broad orders
+0.14
machinery2015-2019
Large-order impulse
+0.31
machinery2021-2026
Broad orders
−0.37
machinery2021-2026
Large-order impulse
−0.01

Every displayed 95% interval crosses zero. The models explain between 1% and 11% of turnover growth variation across all sectors, periods, and horizons.

The predictive check asks whether aggregate broad-order growth or the large-order impulse forecasts future turnover in electronics and machinery after controlling for energy-supply producer-price growth. It does not. Across one-, three-, and six-month horizons, before 2020 and after 2021, every 95% interval for both order coefficients crosses zero.

This is not evidence that orders never matter. The samples are small, the predictors are aggregate while turnover is sector-specific, and monthly realization is noisy. The defensible conclusion is narrower: neither the broad series nor the large-order wedge supports a precise short-run turnover forecast in these two sectors. The release is descriptive evidence, not a scheduling model.

Revisions and method

The vintage is part of the result.

May changed from +1.9% to +0.3%.

Destatis revised the previous month after amended price adjustment. The change was −1.6 percentage points, large enough to alter the apparent trajectory. This publication records that release-documented revision; it does not claim a complete real-time revision distribution from one frozen vintage.

Design

  1. 1. Separate composition. Total orders are compared with orders excluding major transactions from January 2015 through June 2026.
  2. 2. Align measurement. All index comparisons use real X13 JDemetra+ calendar- and seasonally adjusted series with 2021 = 100.
  3. 3. Measure momentum. The analysis calculates adjacent three-month growth, annualized six-month growth, annual growth, and rolling volatility.
  4. 4. Add realization. Aggregate turnover and five-sector order/turnover paths use static Destatis exports covering January 2025–June 2026.
  5. 5. Test prediction. Two matched historical sectors support standardized 1/3/6-month turnover models with Newey-West HAC intervals.
  • Index series by market direction are not additive; the article reports growth rates, not contribution shares.
  • The five-sector static cross-section covers 18 months, not the full 2015–2026 history.
  • Historical execution and predictive tests remain limited to electronics and machinery.
  • A registered GENESIS token is required for the complete aggregate turnover and revision-vintage panel.
  • Regression coefficients are predictive associations and do not identify causal conversion effects.
  • Large transactions are not allocated to individual sectors in the ex-major aggregate table.
Germany may have won valuable projects. But one large contract can fill an order book; it cannot, by itself, fill an economy.

Data and provenance

Every claim travels with the series, models, and hashes.

The historical backbone reuses the eight-table reviewed package from 19 August. Three additional static Destatis exports add aggregate turnover and the five-sector June cross-section. The generator verifies all input hashes, reproduces official release values, and records every public output hash.

Release contract: 138 historical months, 90 five-sector rows, 12 turnover models, data through June 2026, and one frozen 20 August source manifest.

Complete analysis

JSON · headline, momentum, directions, sectors, execution proxies, models, and sources

Download analysis JSON

Monthly order breadth

CSV · 138 months of total, ex-major, impulse, momentum, and volatility measures

Download monthly CSV

Sector orders and turnover

CSV · 90 observations across five manufacturing sectors

Download sector CSV

Turnover models

CSV · standardized coefficients, HAC intervals, horizons, and samples

Download model CSV

Publication manifest

JSON · deterministic input and output bytes with SHA-256 hashes

Download manifest
Additional static inputs
TableMeasureRelease stampSHA-256
42152-0004Monthly manufacturing turnover volume index for aggregate manufacturing and main groups06.08.2026 / 19:49:46f01e7bf0c41f292cb771a8d0a25c806321c54ae6cbed3ee833258e6fe372c3ec
42151-0005Monthly manufacturing new-orders volume index by WZ 2008 sector and market direction06.08.2026 / 19:49:07c4084e015b2502c7256571026b00f19f2d2384ffc97d9a1be3874dfbaf1add26
42152-0005Monthly manufacturing turnover volume index by WZ 2008 sector and market direction06.08.2026 / 19:49:559f4e7f619125326969fef09e1ec2021a4a0bde7290fa680330d9bed7912cafb8

08 / Sources

References

Official data are licensed under Data licence Germany – attribution – Version 2.0. Release values are cross-checked against the cited Destatis publication.

One historical package · three new static snapshots
  1. 01

    New orders in manufacturing in June 2026: +3.1% on the previous month. Destatis press release No. 276, 6 August 2026.

  2. 02

    Manufacturing new-orders volume indices. GENESIS-Online tables 42151-0004, 42151-0005, and 42151-0008.

  3. 03

    Manufacturing turnover volume indices. GENESIS-Online tables 42152-0004 and 42152-0005.

  4. 04

    The German Order-Conversion Gap. Companion evidence audit published 19 August 2026.

  5. 05

    Data licence Germany - attribution - Version 2.0. Official German open-data licence.