Western Europe & Türkiye · Updated 22 September 2026
ManufacturingML
Manufacturing data · Western Europe · Eurostat, extracted 22 September 2026

Western European manufacturing in 2026: the data

Ten countries, twelve Eurostat tables, one extraction date: what the published figures say about output, cost, power, hiring and AI in Germany, Italy, Spain, Portugal, the Netherlands, Belgium, Austria, Sweden, Denmark and Finland.

In shortWestern European manufacturing is not one market. Value added per person employed in 2025 runs from EUR 233.7k in Denmark to EUR 47.1k in Portugal (Eurostat nama_10_a10 and nama_10_a10_e). An hour of manufacturing labour costs EUR 55.0 in Denmark and EUR 16.5 in Portugal. Industrial electricity costs 3.2 times more in Denmark than in Finland. The region's largest manufacturing sub-sector is machinery, not automotive.

How to read this page

Every figure below was read from the Eurostat dissemination API on 22 September 2026 and carries its dataset code, year and unit. Nothing is interpolated or carried over from an earlier year. Where Eurostat publishes no value the cell says n/r, with the reason under What is not in this data.

Units"Manufacturing" is NACE Rev.2 section C throughout. Value added is at current prices (B1G, CP_MEUR) and is not adjusted for price levels, so euro comparisons between countries mix real output with local prices. The one PPS-adjusted series here is tesem160, whole-economy productivity per hour, EU-27 = 100.

Ten numbers that describe the region

The productivity spread inside Western Europe is wider than the East–West gap

Manufacturing value added per person employed in 2025 runs from EUR 233.7k in Denmark to EUR 47.1k in Portugal (nama_10_a10 B1G CP_MEUR ÷ nama_10_a10_e EMP_DC, NACE C, nominal). The Danish figure is a pharmaceutical artefact: pharma alone was EUR 24,423m of Denmark's EUR 58,463m of manufacturing value added in 2023 (nama_10_a64). Portugal sits below Czechia (EUR 49.8k). Denmark's number describes a product mix, not a way of running a line.

−153,000

Germany shed seven of every ten manufacturing jobs the EU lost in 2025

German manufacturing employment fell from 7,415.0 to 7,262.0 thousand persons in 2025, −2.1% (nama_10_a10_e, EMP_DC, THS_PER, NACE C). The EU-27 lost 214,900 net, so Germany is about 71% of it. Austria (−2.1%) and Belgium (−2.2%) contracted at the same rate on far smaller bases. Spain is the only clear increase, +3.5%.

+0.5%

Output grew almost everywhere; it barely moved where most of the workers are

German manufacturing value added rose 0.5% in nominal terms in 2025, to EUR 797.2bn, against +2.0% for the EU-27 (nama_10_a10). Denmark grew 8.3%, the Netherlands 5.1%, Austria 5.0%. Spain added 3.4% of value added on 3.5% more people — growth by headcount, not by output per head. On tesem160 Spain slipped from 96.1 to 95.6, Italy from 98.8 to 95.9 and Portugal from 68.2 to 66.7 against the EU-27 index.

+13.0%

Labour cost rose fastest where it is lowest, but nothing here is a CEE-style shock

Hourly manufacturing labour cost rose 13.0% in Portugal between 2023 and 2025 (EUR 14.6 → 16.5), 11.0% in Austria and 10.0% in Sweden, against +9.0% for the EU-27 (lc_lci_lev, lcstruct D1_D4_MD5, NACE C). Germany (+6.7%), Denmark (+6.6%) and Finland (+5.1%) rose more slowly than the EU. None of it resembles Poland's +29.5% or Bulgaria's +27.5% over the same two years: in CEE the case for measuring a line is that wages are catching up fast, in the West the slope is flat and the case rests on the level.

EUR 55.0/h

The Western argument is the absolute cost of an idle hour

Denmark pays EUR 55.0 an hour in manufacturing, Austria 51.3, Germany 49.5, Sweden 47.2 — against 35.0 for the EU-27 and 16.5 for Portugal (lc_lci_lev, 2025). Arithmetic on those rates: a twelve-operator line idle for one eight-hour shift carries about EUR 4,750 of direct labour in Germany and EUR 1,580 in Portugal, before machine, order or penalty effects. That ratio decides whether capturing stop causes at the machine pays for itself.

3.2×

Industrial electricity costs three times more in Denmark than in Finland

Band ID (2,000–19,999 MWh/year), excluding recoverable taxes, 2025-S2: Denmark EUR 0.2732/kWh, Germany 0.2369, Netherlands 0.2192, Italy 0.2177, Austria 0.2159, Belgium 0.1901, EU-27 0.1920 (nrg_pc_205, siec E7000, I_TAX). Finland pays 0.0846 and Sweden 0.1065. Denmark is the most expensive industrial electricity market in these ten countries and in the seven CEE ones too — above Hungary (0.234) and Slovakia (0.219). Energy-per-part reporting earns its keep in DK, DE, NL, IT and AT; in the other Nordics it is close to noise.

C28

Machinery, not automotive, is the region's largest manufacturing sub-sector

Machinery n.e.c. (NACE C28) is the largest sub-sector by value added in Italy (EUR 46,649m), the Netherlands (22,576m), Austria (11,328m) and Finland (6,908m), and second in Germany, Sweden and Denmark (nama_10_a64, 2023). Motor vehicles lead in only two of the ten; pharma leads in Denmark and Belgium. That mix is discrete, high-mix and low-volume: many part numbers, frequent long changeovers, short runs, losses sitting in set-up and material rather than cycle time. A plan written around a high-volume automotive assumption mis-targets most of this region.

26.6% / 5.1%

The Netherlands is short of people; Italy is not

26.6% of Dutch manufacturers named labour as a factor limiting production in 2026-Q3, against 5.1% in Italy and 18.0% for the EU-27 (ei_bsin_q_r2, BS-FLP3-PC, NSA). The Dutch manufacturing vacancy rate was 3.9% in 2025, Italy 1.5%, Spain 0.6%, EU-27 1.5% (jvs_a_rate_r2). The two series disagree in direction — vacancy rates fell year-on-year in eight of ten countries while the survey share rose in nine of ten — and sit on different time bases. Quote both or neither.

9.9% → 39.8%

AI use in manufacturing spans a factor of four, and the 2025 series has a break in it

Manufacturers using at least one AI technology in 2025: Belgium 39.8%, Denmark 38.6%, Sweden 32.9%, Austria 32.5%, Netherlands 28.6%, Germany 24.4%, Spain 17.1%, Italy 14.7%, Portugal 9.9%, EU-27 17.3% (isoc_eb_ain2, E_AI_TANY, PC_ENT, 10+ employees, NACE C). Eurostat broadened the AI question battery for the 2025 wave, so part of every 2024→2025 jump is methodological. Treat the two years as two measurements, not a growth rate. Portugal sits in the CEE band, between Hungary (7.8%) and Croatia (12.9%).

3 gaps

Three series here are not published, and nothing was substituted for them

Finland's manufacturing AI figure for 2025 is withheld under a low-reliability flag; the all-sector figure (37.8%) measures something else and does not stand in for it. Dutch 2025 hourly labour cost is confidential and the Belgian 2025 observation is absent, so both change columns cover 2023→2024 only. Sweden's 2023 chemicals and pharma value added is suppressed — EUR 10,177m is known only as a combined residual — so the Swedish third place is uncertain.

Output, employment and productivity

Manufacturing value added, employment and value added per person employed, 2025
CountryMfg GVA 2025 (EUR bn)Change 2024→25Mfg % of total GVAEmployment 2025 (thousand)VA per person employed (EUR k)Productivity/hour, PPS, EU-27=100
Germany797.2+0.5%19.4%7,262.0109.8124.2
Italy338.7+2.0%16.8%3,922.686.495.9
Spain178.8+3.4%11.7%2,279.278.495.6
Netherlands122.9+5.1%11.7%757.0162.3126.0
Austria78.5+5.0%17.1%673.9116.5115.9
Denmark75.0+8.3%20.2%321.0233.7137.7
Sweden72.2+1.0%13.4%565.4127.8108.8
Belgium71.7+1.1%12.3%485.2147.8131.0
Finland39.9+1.8%16.2%342.9116.4105.6
Portugal34.7+1.8%13.0%736.147.166.7
EU-272,668.8+2.0%15.8%29,901.689.3100.0

Eurostat nama_10_a10 (B1G, CP_MEUR, NACE C and TOTAL) and nama_10_a10_e (EMP_DC, THS_PER, NACE C), 2024–2025; last column tesem160 (NLPR_HW, PC_EU27_2020_MPPS_CP), 2025. VA per person employed = mfg GVA ÷ mfg employment, nominal. Extracted 22 September 2026.

Five of the ten are more manufacturing-intensive than the EU-27's 15.8%: Denmark (20.2%), Germany (19.4%), Austria (17.1%), Italy (16.8%), Finland (16.2%). Intensity and productivity do not move together — the Netherlands is second on value added per worker and joint-last on intensity.

The cost of an hour and a kilowatt-hour

Labour cost, industrial electricity price and labour availability
CountryLabour cost/hour 2025 (EUR)Change 2023→25Electricity band ID 2025-S2 (EUR/kWh)Vacancy rate 2025 (%)Labour limiting production 2026-Q3 (%)
Denmark55.0+6.6%0.27321.98.6
Austria51.3+11.0%0.21592.811.6
Germany49.5+6.7%0.23691.717.0
Sweden47.2+10.0%0.10651.412.2
Finland43.6+5.1%0.08460.915.0
Netherlandsn/r+5.6% (2023→24)0.21923.926.6
Belgiumn/r+3.6% (2023→24)0.19012.913.3
Italy32.8+7.5%0.21771.55.1
Spain28.2+8.9%0.14310.611.1
Portugal16.5+13.0%0.14821.111.9
EU-2735.0+9.0%0.19201.518.0

Labour cost: lc_lci_lev (EUR, D1_D4_MD5, NACE C), 2023–2025 — NL 2025 confidential, BE 2025 not published, so those two changes are one-year and not comparable with the column. Electricity: nrg_pc_205 (E7000, MWH2000-19999, I_TAX), 2025-S2. Vacancies: jvs_a_rate_r2 (NACE C, AVG_A), 2025. Labour limit: ei_bsin_q_r2 (BS-FLP3-PC, NSA), 2026-Q3, a 2026 quarter and not on the annual time base. Extracted 22 September 2026.

Labour cost and electricity price are not correlated. Finland pays a high wage and the cheapest power in the set; Portugal the lowest wage and mid-priced power; Denmark tops both. A business case built on "European energy costs" needs the country's own row.

Where the value added sits

Three largest manufacturing sub-sectors by gross value added, 2023 (EUR m)
Country#1#2#3
GermanyMotor vehicles 155,935Machinery n.e.c. 119,739Fabricated metal 67,158
ItalyMachinery n.e.c. 46,649Fabricated metal 45,395Food, beverages, tobacco 36,708
SpainFood, beverages, tobacco 34,193Fabricated metal 15,417Motor vehicles 14,993
NetherlandsMachinery n.e.c. 22,576Food, beverages, tobacco 18,769Pharma 9,921
AustriaMachinery n.e.c. 11,328Food, beverages, tobacco 9,071Fabricated metal 7,350
DenmarkPharma 24,423Machinery n.e.c. 6,074Food, beverages, tobacco 4,939
SwedenMotor vehicles 14,261Machinery n.e.c. 11,518Wood/paper/printing 6,190 (uncertain)
BelgiumPharma 14,579Food, beverages, tobacco 12,122Chemicals 9,831
FinlandMachinery n.e.c. 6,908Wood/paper/printing 4,246Computer/electronic 4,226
PortugalFood, beverages, tobacco 5,505Textiles, apparel, leather 4,423Fabricated metal 3,580

Eurostat nama_10_a64 (B1G, CP_MEUR), 2023, extracted 22 September 2026. Sweden: chemicals (C20) and pharma (C21) are suppressed individually for 2023 and are known only as a combined residual of EUR 10,177m, so the Swedish #3 could be displaced; #1 and #2 are firm.

Portugal is the only country here whose profile — food, textiles, fabricated metal, low value added per worker — matches the Central and Eastern European pattern.

AI and the 2025 survey break

Manufacturers using at least one AI technology (10+ employees, % of enterprises)
CountryManufacturing 2024Manufacturing 2025
Denmark21.838.6
Finland20.8n/r
Sweden19.532.9
Belgium23.239.8
Netherlands18.028.6
Austria22.732.5
Germany16.124.4
Spain9.817.1
EU-2710.617.3
Italy8.014.7
Portugal6.89.9

Eurostat isoc_eb_ai (all sectors, C10-S951_X_K) and isoc_eb_ain2 (NACE C), indicator E_AI_TANY, unit PC_ENT, size class GE10, 2024–2025, extracted 22 September 2026. Finland × NACE C × 2025 carries flag u with the observation withheld: n/r, not substituted and not carried forward.

The difference between the two manufacturing columns is not an adoption rate. The 2025 column supports a ranking only: Belgium, Denmark, Sweden and Austria at roughly twice the EU rate; Germany above the EU average and behind the Nordics; Italy and Portugal below it. On the same basis Romanian manufacturing was 3.2% and Polish 7.7%.

What is not in this data

  • Finland, manufacturing AI use 2025. Withheld by Eurostat, flag u. No substitute used.
  • Netherlands and Belgium, hourly labour cost 2025. NL confidential, BE not published; their changes cover 2023→2024 only.
  • Sweden, 2023 chemicals/pharma split. Suppressed; EUR 10,177m known only as a residual, so the Swedish #3–#5 order is uncertain.
  • Enterprise counts by size class. The sbs_ovw_act totals (Italy 352,446, Germany 201,177, Spain 167,401, 2023) include micro-firms and sole traders and are not a count of factories. No 10+ or 50+ breakdown was retrieved or estimated.
  • Digital Intensity Index and robot density. Not pulled for these ten countries; nothing stands in their place.
DisclosureThis site is published by TEEPTRAK SAS, which sells production-monitoring and OEE software. That is the bias to read the page with. The figures are Eurostat's, not ours, and they do not point at a purchase: where the binding constraint is the order book or material supply, a monitoring system measures something it cannot move. A month of stop-cause logging on paper, a local integrator, or an MES module already paid for are all legitimate first steps.

Country profiles

Each profile runs the same sections in the same order: value added and its trend, employment, value added per person employed against the EU-27, the three largest sub-sectors, labour cost and electricity, hiring, AI adoption, and what the combination means for a plant there.

Free data
Western Europe Manufacturing Data Pack 2026

Every indicator on this page for 17 European countries and EU-27, each row carrying its Eurostat dataset code and link. Excel.

Get the data pack →

By country

Questions

Which Eurostat tables are behind these figures?
Twelve: nama_10_a10, nama_10_a10_e, nama_10_a64 (sub-sectors, 2023), sbs_ovw_act (enterprise counts, 2023), lc_lci_lev (hourly labour cost), nrg_pc_205 (electricity), jvs_a_rate_r2 and jvs_q_nace2 (vacancies), ei_bsin_q_r2 (labour as a limiting factor), isoc_eb_ai and isoc_eb_ain2 (AI use), tesem160 (productivity). All extracted 22 September 2026 and re-checkable in the Eurostat Data Browser.
Why is Danish value added per worker so much higher than everyone else's?
Product mix. Pharmaceuticals were EUR 24,423m of Denmark's EUR 58,463m of manufacturing value added in 2023 (nama_10_a64), and pharma carries very high value added per hour. Denmark's +8.3% growth in 2025 comes from the same place. EUR 233.7k is not a benchmark for a Danish machine shop.
Can I compare the 2024 and 2025 AI adoption figures?
No. Eurostat broadened the AI question battery for the 2025 wave, so part of every increase between the two years is methodological rather than real. Use the 2025 column to rank countries against each other and against the EU-27 manufacturing average of 17.3%. Do not present the difference between columns as growth.
Why does Portugal appear alongside Czechia and Poland rather than alongside Spain?
On the productivity measures it belongs there. Portuguese manufacturing value added per person employed was EUR 47.1k in 2025, below Czechia's EUR 49.8k; whole-economy productivity per hour was 66.7 against EU-27 = 100, below Slovenia (86.2), Slovakia (81.1) and Czechia (79.6); manufacturing AI use was 9.9%, between Hungary (7.8%) and Croatia (12.9%).
Are the euro figures adjusted for price levels?
No. Value added, labour cost and electricity are in nominal current euros, so a Denmark-to-Portugal comparison mixes real output with local price levels. The only PPS-adjusted series here is tesem160, which covers the whole economy. Use the euro figures for what a plant pays and the PPS index for relative efficiency.

Sources

  1. Eurostat nama_10_a10 — gross value added, NACE C and total (2024–2025)
  2. Eurostat nama_10_a10_e — employment, domestic concept, NACE C (2024–2025)
  3. Eurostat nama_10_a64 — value added by manufacturing sub-sector (2023)
  4. Eurostat sbs_ovw_act — number of enterprises, NACE C (2023)
  5. Eurostat lc_lci_lev — hourly labour cost, NACE C (2023–2025)
  6. Eurostat nrg_pc_205 — electricity prices, non-household consumers (2025-S1, 2025-S2)
  7. Eurostat jvs_a_rate_r2 — annual job vacancy rate, NACE C (2024–2025)
  8. Eurostat jvs_q_nace2 — quarterly job vacancy rate, NACE C (2025-Q3, 2025-Q4)
  9. Eurostat ei_bsin_q_r2 — factors limiting production, labour (2026-Q2, 2026-Q3)
  10. Eurostat isoc_eb_ai — enterprises using AI, all sectors (2024–2025)
  11. Eurostat isoc_eb_ain2 — enterprises using AI, NACE C (2024–2025)
  12. Eurostat tesem160 — labour productivity per hour worked, PPS, EU-27 = 100 (2024–2025)

Published by TEEPTRAK SAS, which makes production-monitoring and OEE software. Every figure is sourced on the page. Funding rules, standards and reporting duties change: check the official documents before you budget or commit.