Economic growth in the EU has been persistently slower than in the US over the past two decades. Economic growth has been slowing, mainly due to weakening labour productivity growth according to the Draghi Report (Draghi 2024). While much of the debate has focused on investment gaps, regulatory barriers, and labour market dynamics at the national level, less attention has been paid to the role of economic structure at a fine geographical scale in shaping Europe’s diverging growth trajectories.

Our recent analysis (Dijkstra et al. 2025) aims to fill this gap by building on previous studies (Enflo 2010, Le Gallo and Kamarianakis 2011, Gómez-Tello et al. 2020, Kilroy and Gana 2020, Martin et al. 2018) and using novel data from the Annual Regional Database of the European Commission (ARDECO). Armed with these data, we examine for the first time productivity dynamics across metropolitan regions (‘metros’) in Europe over the period 2001–2021 with a ten-sector disaggregation analysis.

The results reveal a nuanced geography of economic growth. In capitals, growth was fuelled by both productivity growth and employment growth, which may explain why they also saw the highest population growth (Table 1). In other metropolitan areas, however, employment, productivity, and populations grew at half the rate of capitals. In the rest of the EU (i.e. non-metros), populations shrank and employment barely grew, but labour productivity grew almost as fast as in capitals.

Table 1 Decomposing the growth of gross value added (GVA) per capita in EU capitals, other metros, and non-metro regions, 2001-2021

Note: A decomposition means that growth is split into its components:  GVA per capita growth = Productivity growth + Employment over population growth (A = B + C); Employment per capita growth = Employment growth – Population growth (C = D – E).

When we look at the drivers of productivity growth, one pattern stands out: productivity growth occurred mostly within economic sectors rather than through shifts to more productive sectors in all three types of regions, although its relative importance varied. Capitals experienced high productivity growth, but employment growth was higher in less productive sectors, suggesting that the concentration of highly productive sectors – such as finance and professional services – generates more demand for employment in other sectors, such as retail, arts, and sports (Moretti 2012). Other metros and non-metros also achieved part of their growth through higher employment growth in more productive sectors, reflecting the fact that structural transformations are still ongoing.

Changes in employment by sector confirm this. Between 2001 and 2021, capital regions expanded their employment shares in services (e.g. information and communication services, professional services), while employment in industry, and trade, transport, and hotels declined (Figure 1). Other metropolitan regions followed a similar but less pronounced trajectory. In contrast, non-metropolitan regions remained more dependent on traditional sectors and experienced limited employment growth. This implies that the shift of employment happened through reductions in employment in industry and agriculture rather than through labour expansion.

Figure 1 Employment per sector by type of region in the EU in 2001, 2011, and 2021

Productivity growth over the period 2001-2021 was fuelled by:

• capital city status (capital metro regions lead growth, concentrating both economic and political power; see Figure 2);
• population density (but not enough to prevent other metro regions from lagging behind);
• lower initial productivity levels, showing signs of convergence;
• patenting activity, as well as the employment share in ICT and finance; and
• good transport infrastructure.

Figure 2 Labour productivity growth in capitals and other metros, 2001-2021

Our findings suggest that productivity at the local level is more nuanced than simply “cities are good and other places are lagging”. The findings contribute to the growing debate on agglomeration economies and labour productivity inequalities. Specifically, our work underscores the need to assess why other metro regions have underperformed over the past two decades, and whether non-metro regions will continue to converge or whether their growth will stall once they have transitioned to more productive sectors.

Innovation can increase regional productivity, as shown in our regression analysis and the literature. This analysis is relevant for regional development policy, especially in the context of the debate on EU cohesion policy in the next programming period.  Our study highlights the different trends in productivity growth and sectoral composition of capitals, other metros and non-metros. This suggests that a tailored approach to address the distinct challenges and opportunities of different regional contexts may be more successful. Furthermore, the findings can help to identify strategies that enhance European competitiveness by embracing regional specificities (Capello and Rodríguez-Pose 2025).

References

Capello, R and A Rodríguez-Pose (2025), “Europe’s quest for global economic relevance: On the productivity paradox and the Draghi report”, Scienze Regionali 24(1): 7-15.

Dijkstra, L, M Kompil and P Proietti (2025), “Are cities the real engines of growth in the EU?”, Geography and Environment Discussion Paper No. 2025, LSE.

Draghi, M (2024), The future of European competitiveness, European Commission.

Enflo, K S (2010), “Productivity and employment—Is there a trade-off? Comparing Western European regions and American states 1950–2000”, The Annals of Regional Science 45(2): 401-421.

Gómez‐Tello, A, M J Murgui‐García and M T Sanchis‐Llopis (2020), “Exploring the recent upsurge in productivity disparities among European regions”, Growth and Change 51(4): 1491-1516.

Kilroy, A and R Ganau (2020), “Economic growth in European Union NUTS-3 regions”, Finance, Competitiveness and Innovation Global Practice, World Bank.

Le Gallo, J and Y Kamarianakis (2011), “The evolution of regional productivity disparities in the European Union from 1975 to 2002: A combination of shift–share and spatial econometrics”, Regional Studies 45(1): 123-139.

Martin, R, P Sunley, B Gardiner, E Evenhuis and P Tyler (2018), “The city dimension of the productivity growth puzzle: the relative role of structural change and within-sector slowdown”, Journal of Economic Geography 18(3): 539-570.

Moretti, E (2012), The New Geography of Jobs, Houghton Mifflin Harcourt.