Možnosti vyhledávání
Home Média ECB vysvětluje Výzkum a publikace Statistika Měnová politika Euro Platební systémy a trhy Kariéra
Návrhy
Třídit podle
António Dias Da Silva
Principal Economist · Economics, Supply Side, Labour and Surveillance
Antonella Fabrizio
Matthias Mohr
V češtině není k dispozici.

Recent country-specific and sectoral developments in labour productivity in the euro area

Prepared by António Dias da Silva, Antonella Fabrizio and Matthias Mohr

Published as part of the ECB Economic Bulletin, Issue 5/2024.

A combination of various adverse shocks has contributed to productivity growth being suppressed in the euro area over the last four years. The pandemic, along with disruptions in global supply chains and the energy price increases from 2021, which were aggravated by the Russian war in Ukraine, have all contributed to the slowdown in productivity growth. These factors have exerted a particularly significant impact on the industry, wholesale and retail trade, and construction sectors. As a result, productivity dynamics have been weaker than in the past, with average productivity per person employed declining by 0.2% on average per year since the fourth quarter of 2019 compared with average growth of 0.8% per year before the pandemic. The average growth rate of productivity per hour since the fourth quarter of 2019 amounted to 0.2% per year, compared with average growth of 1.2% per year before the pandemic. In the first quarter of 2024 productivity per person employed was 0.7% lower than in the fourth quarter of 2019 and productivity per hour worked was higher by just 0.7% (Chart A).

While the productivity slowdown reflects cyclical factors, structural factors may also have played a role.[1] Productivity is inherently procyclical, rising in booms and falling in recessions. In the euro area, labour market institutions and social preferences give prominence to employment protection over flexibility, as reflected by the job retention schemes put in place during the pandemic, for instance.[2] However, it is not yet possible to assess whether the extensive usage of job retention schemes and the impacts of the rise in energy prices from 2021 will only have a cyclical effect or whether they will add to existing structural weaknesses.[3]

Among the five largest euro area economies, France and Spain stand out as recording the largest decline and the largest increase in productivity per hour worked respectively. In France, labour input, in terms of both hours worked and persons employed, increased about twice as fast as GDP, in part driven by an increase in the number of apprenticeship contracts offered. As new apprentices are on average less productive than experienced workers, this may have contributed to the sharp short-term decline observed in productivity (Chart A). Spain recorded robust growth in labour productivity per hour, in part related to a sharp decline in average hours worked, as productivity growth per person employed was negative. Growth in labour productivity per person employed was negative in all five of the largest economies, except for Italy which recorded a strong increase in average hours worked. Sectoral differences are key to explaining these developments. For example, the construction sector supported productivity growth in Italy, whereas its effect was negative in the other four largest economies.[4] The public sector made a negative contribution to productivity per person in all five countries.

Chart A

Labour productivity growth by country

(cumulative percentage changes Q4 2019-Q1 2024)

Sources: Eurostat and ECB calculations.

At the euro area level, the slowdown in productivity growth has been broad-based, albeit with differences across sectors.[5] The construction sector stands out as showing the largest cumulative fall in productivity in the period between the fourth quarter of 2019 and the first quarter of 2024, driven by a decrease in gross value added and a large increase in employment and hours worked (Chart B). These two factors together contributed to a decline of about 8% in labour productivity in this sector. In larger sectors, such as trade and transport and the public sector, productivity per person declined, while productivity per hour broadly stagnated.[6] Information and communication technology services recorded a substantial increase in productivity, driven by strong growth in gross value added. However, compared with the change in the four years preceding the pandemic (to use a similar time frame), this sector recorded the most significant deceleration in productivity growth, second only to construction. For some sectors, the four-year period comprises two distinct phases. The manufacturing sector, for example, showed cumulative growth of 3.7% in productivity per person and 5% in productivity per hour worked up to mid-2022. However, the energy price shock helped cause productivity growth to turn negative, which meant that compared with the period before the pandemic, cumulative growth became negative when measured per person and increased by only 1% when measured per hour worked. Contact-intensive service sectors recorded a 0.3% rise in productivity per person and a 0.7% gain in productivity per hour worked from the fourth quarter of 2019 to the second quarter of 2022. This was followed by a 1.5% drop in productivity per person and a 0.5% decline in productivity per hour worked from the third quarter of 2022 to the first quarter of 2024.

Chart B

Labour productivity growth by sector

(cumulative percentage changes Q4 2019-Q1 2024)

Sources: Eurostat and ECB calculations.
Notes: The NACE Rev. 2 codes on the x-axis refer to the following economic sectors: Total: Total economy; A: Agriculture, forestry and fishing; B-E: Industry; F: Construction; G-I: Trade, transportation and accommodation; J: Information and communication; K: Financial and insurance activities; L: Real estate activities; M-N: Professional, scientific, technical, administrative and support service activities; O-Q: Public service activities; R-U: Arts, entertainment, recreation and other service activities. Data for the Information and communication sector (J) are off scale. The actual values are 23.1% for gross value added, 15.8% for total hours worked and 17.3% for employment.

The weak growth in productivity is the result of declines within sectors rather than a reallocation of labour across sectors. The reallocation of labour from low to high-productivity sectors has had a positive impact on the cumulative change in productivity since the first quarter of 2020. Without this positive effect, labour productivity per hour worked (Chart C, panel a) and per person employed (Chart C, panel b) would have been even lower. However, this positive reallocation effect was outweighed by the negative impact of the pandemic and fell back to close to zero thereafter. Up to the first quarter of 2021, the share of less productive contact-intensive services sectors declined, while the share of high-productivity sectors such as industry, information and communication technology and professional services increased and remained at a higher level during the recovery. Looking at year-on-year changes, the reallocation effect reversed from mid-2021 to mid-2022, reflecting reopening dynamics, and was close to neutral after that. Thus, the pandemic did not induce a substantial structural change in the sectoral composition of the economy: compared with 2019 sector shares have remained broadly stable in terms of both total hours worked and value added.

Chart C

Shift-share analysis of productivity developments

(year-on-year and cumulative percentage changes Q1 2020-Q1 2024)

Sources: Eurostat and ECB calculations.
Notes: In panel b), data for the second quarter of 2020 and the second quarter of 2021 are off scale. For the second quarter of 2020 the “within” component was -11.9% and the total change was -11.3%. For the second quarter of 2021 the “within” component was 12.2% and the total change was 11.7%. The calculation follows the shift-share analysis in Denis, C., McMorrow, K. and Röger, W., “An analysis of EU and US productivity developments (a total economy and industry level perspective)”, European Economy – Economic Papers, No 208, European Commission, July 2004, p. 78.

Overall, the slowdown in productivity is largely the result of adverse shocks affecting GDP growth. The pandemic and the energy price shock have weighed on euro area GDP, which has resulted in a broad-based decline in productivity, given its procyclical nature. Higher profit margins, coupled with lower real wages, strong growth in the labour force and lower average hours worked have all helped support employment growth, while raising procyclicality.[7] As some of these factors unwind, with weakening profits and rising real wages, further improvements in the labour market will become increasingly more difficult to achieve if they are not supported by stronger productivity growth.

  1. See Arce, O. and Sondermann, D., “Low for long? Reasons for the recent decline in productivity”, The ECB Blog, ECB, 6 May 2024.

  2. The share of firms hoarding labour was significantly elevated during the pandemic and the post-pandemic period. See the box entitled “Higher profit margins have helped firms hoard labour”, Economic Bulletin, Issue 4, ECB, 2024.

  3. See, for example, the article entitled “The slowdown in euro area productivity in a global context”, Economic Bulletin, Issue 3, ECB, 2017.

  4. The construction sector benefited from a tax support scheme introduced in 2020 to mitigate the economic impact of the pandemic on households and businesses. The scheme allowed homeowners to deduct up to 110% of the cost of renovating their homes from their taxes under certain conditions.

  5. For the sectoral analysis, we use gross value added divided by employment or hours worked, which gives slightly different figures than when GDP is used as a numerator; this can be seen in Charts A and B.

  6. Productivity figures for non-market activities are mainly affected by labour costs. However, the stronger increase in employment creation in relation to gross value added weighed on productivity.

  7. See Arce, O. and Sondermann, D., op. cit.