EU hourly labour costs, €12.0 to €56.8: what the Eurostat measure counts

Bulgaria recorded the largest rise in hourly labour costs in the EU between 2024 and 2025: 13.1 per cent, against 4.1 per cent for the EU as a whole. Over the same year the distance between the two, measured in euro rather than in percentages, did not move at all.

Eurostat’s prior-year release put 2024 at €33.5 for the EU and €10.6 for Bulgaria, a difference of €22.9 an hour. The 2025 figures are €34.9 and €12.0 — a difference of €22.9 an hour. Convergence that is unmistakable in the growth column is invisible in the currency an employer actually pays in. That is the first thing to know about this series: it is a level, published to one decimal place, and most of the interesting questions about it are questions about the arithmetic done on top of it.

What the 2025 series says

Average hourly labour cost, whole economy, 2025

Territory	€ per hour	× EU-27
Bulgaria	12.0	0.34
Romania	13.6	0.39
Hungary	15.2	0.44
Poland	19.1	0.55
Czechia	19.8	0.57
Slovakia	19.8	0.57
EU-27	34.9	1.00
Euro area, 21 countries (from 2026)	38.2	1.09
Euro area, 20 countries (2023–2025)	38.6	1.11
Germany	45.0	1.29
Austria (provisional)	46.3	1.33
Netherlands	47.9	1.37
Denmark	51.7	1.48
Luxembourg	56.8	1.63

Two features of that table deserve a pause. Czechia and Slovakia return the same value to the published decimal place, and Poland sits €0.70 below both: the central European cluster is much flatter than the headline range implies. And the euro-area line is ambiguous in a way the EU line is not. The same API response that returns €38.6 for the twenty-country euro area returns €38.2 for the twenty-one-country euro area, and it is the second of these that Eurostat quotes in its own news article. One aggregate, one reference year, two correct answers forty cents apart.

What the number counts

It is a cost to the employer per hour, not a wage. Eurostat publishes the split: in 2025 non-wage costs were 24.8 per cent of total labour costs in the EU and 25.6 per cent in the euro area on its current composition, which includes Bulgaria. Wages and salaries are the remainder.

The spread in that component is wider than most cost models allow for. France at 32.3 per cent and Sweden at 31.7 per cent sit at the top; Romania at 4.8 per cent, Lithuania at 5.5 per cent and Malta at 5.8 per cent at the bottom. France’s share is roughly 6.7 times Romania’s. A non-wage uplift calibrated in one member state and carried into another is not approximately right. It is wrong by a multiple.

What it does not count

It does not count output. Nothing in the published series reports what an hour produces, and no productivity term appears anywhere in the figures verified for this piece.

Nor is it what workers are paid. Eurostat publishes median gross hourly earnings as a separate statistic: €14.9 in the EU in 2022, from €4.1 in Bulgaria to €29.8 in Denmark, with Luxembourg at €24.0. That is a different reference year, a different statistic — a median rather than an average — and a different concept. Subtracting one from the other does not produce a non-wage margin, and the three-year gap between the two reference periods is not incidental.

Coverage has edges as well. The sectoral figures for 2025 describe a “mainly non-business economy excluding public administration”, so a general government hourly cost is not what the headline range ranks.

Using it, and not

Three uses survive scrutiny. Sizing an unknown: where a business case needs a figure for a country the firm has never operated in, the published level beats an assumption, provided it travels as a level and not as a rate quotation. Checking a quotation for plausibility: a proposed rate implying an hourly cost far outside the national figure is worth a question. And budgeting the non-wage component using the destination country’s own share rather than a rule of thumb imported from the home market.

Three do not. Ranking countries on cost with no output term, as above. Reading small differences, when the euro-area ambiguity alone is forty cents and Austria’s value is still provisional. And quoting a figure that has simply aged: Germany’s series runs €37.3 in 2021, €39.6 in 2022, €41.4 in 2023, €43.5 in 2024 and €45.0 in 2025, a rise of 20.6 per cent across four years. Quoting €43.5 for Germany today does not look wrong. It is last year’s number.

What to watch

The faster series is the quarterly labour cost index, which put year-on-year growth at 3.2 per cent in the euro area and 3.6 per cent in the EU in the first quarter of 2026, against the 3.8 per cent and 4.1 per cent recorded between the 2024 and 2025 annual levels. Those are different collections on different cycles and should not be spliced together, but the quarterly series is where a turn would show first. Its database is refreshed each quarter, and countries must send data within seventy days of the end of the reference period.

The slower clock is the four-yearly Labour Cost Survey. Its listed reference years end at 2020, with 2024 named as the next, and Eurostat states that survey data become available about two years after the reference period ends. Planning that depends on the structure of labour costs rather than their level should be pinned to that calendar.

The third thing to watch is the currency. Growth in this series is measured in euro, and the euro area covered twenty of the EU’s twenty-seven member states in 2023 to 2025 and twenty-one from 2026 — leaving seven, then six, outside it. For those countries the published growth rate is a single euro number in which a wage movement and an exchange-rate movement arrive together.