Factors shaping regional productivity disparities in Europe

Productivity is the ultimate driver of sustainable economic growth and long-term well-being. However, as we have seen in the first article of this Dossier («European productivity from a regional perspective»), neither its level nor its evolution over time are uniform across different territories, as they depend on multiple structural factors. In this article, we review a broad set of variables covering institutional, geographical and technological aspects, as well as others linked to the economy’s productive structure, in order to distinguish the different groups of European regions according to their productivity level. This framework serves as a prelude to the third article,1 in which we quantify their explanatory capacity relative to the dynamics observed over the last 20 years, seeking to understand why some regions have seen an acceleration in their productivity while others have stagnated.

This section provides a brief overview of the aspects most frequently cited in the economic literature to explain territorial productivity differences and the transmission channels.

Firstly, institutional quality plays a crucial role. Regions with better governance tend to exhibit higher productivity and even enhance the returns of other factors such as training and innovation through regulatory efficiency, protection of property rights and the confidence of economic agents.2 Conversely, weak institutions constrain the development of human capital and R&D expenditure, as well as for their translation into efficiency gains. Institutional reforms can be slow, but they are crucial for development.

Secondly, geographical aspects have a significant impact. Densely populated and urbanised regions are conducive to agglomeration economies that boost productivity.3 The concentration of firms and workers facilitates specialisation, mutual learning, and more efficient services, while a high proportion of the population living in metropolitan areas tends to correlate with higher GDP per worker due to better access to markets and knowledge. Furthermore, neighbouring high-productivity regions increase the likelihood of a territory improving its relative position compared to others with a similar level of productivity.4

Thirdly, the structure of the regional productive fabric is a determining factor. A greater relative weight of the manufacturing sector tends to be associated with higher productivity and long-term growth, as it is in their industries – especially those with high technological complexity – where most innovation and efficiency gains are generated. Recent studies indicate that the relative decline of the manufacturing sector in European regions has been accompanied by a slowdown in productivity growth.5 Similarly, business size plays an important role. Regions where a significant portion of employment is in medium-sized and large firms – with greater capital, technology, and economies of scale – tend to be more productive than those dominated by microenterprises.6

Finally, technological factors are decisive in the regional productivity gap. A higher share of jobs in high-tech sectors (both in industry and in services) is associated with higher levels of productivity, as activities such as computing or electronics tend to provide high value added per worker. Similarly, R&D intensity has a positive impact by boosting efficiency and generating spillover effects that benefit the entire productive fabric of the economy. Several analyses have indicated that part of Europe’s low productivity growth in recent decades is due to a technological deficit compared to other advanced economies, including lower private investment in R&D, a lower dissemination of cutting-edge technologies and slower adoption of digitalisation.7

It is worth noting that these factors do not act in isolation but interact with each other. For example, good institutions enhance the positive effect of urban agglomeration or technological innovation. Similarly, skilled human capital is less likely to emigrate if the region offers a dynamic environment with attractive cities, cutting-edge sectors and good governance. The most prosperous European regions typically combine these ingredients virtuously, which explains much of the dispersion in productivity observed between territories.

On the basis of the aspects identified in the previous section as relevant for explaining differences in productivity levels, we will now group Europe’s regions into productivity quintiles, differentiating them according to the value of the variables that represent institutional, geographical and technological aspects and those linked to the productive fabric (see the table for a description of the variables used and their sources).8

In the institutional sphere, we use the European Quality of Government Index (EQI) developed by the University of Gothenburg, which has been published every three years since 20109 and includes aspects related to the quality of public services and the perception of corruption. We observe that the most productive regions tend to exhibit significantly superior institutional quality, with good governance and effective public services (see first chart). This advantage has remained relatively stable over time, while the less productive regions show very limited improvements.

For the geographical dimension, we use three variables: population density, measured as the number of inhabitants per square kilometre published by Eurostat; the share of the region's population living in metropolitan areas, defined as functional urban areas;10 and the productivity of neighbouring regions, which we construct as a population-weighted average. The most productive regions coincide with large metropolitan centres, and this trend is reinforced over time. In less productive regions, urban growth is more limited, which hinders the generation of agglomeration effects. Something similar is observed in the case of density: it is higher in the regions that make up the most productive quintile. Finally, neighbouring regions can influence the productivity of each region through proximity to other markets, the possibility of cross-border cooperation, technological diffusion and access to shared infrastructure. The most productive European regions are also surrounded by highly productive regions (see second chart). In contrast, in less productive regions, the productivity of their bordering regions is also low. Throughout the three periods, a progressive improvement is observed in the upper quintiles, especially in those with the highest productivity (quintile 5), where the productivity of the bordering regions intensifies. This could reflect better economic integration, the utilisation of European networks and greater business dynamism. In the middle quintiles, the progress is more moderate, while in the lower quintiles there are hardly any advances, indicating persistent structural barriers.

If we focus on the dimensions related to the business structure, the results are also noteworthy. Regarding the share of employment in industry, it is observed that this is higher for regions in the lowest quintile and then shows no clear pattern as the regions become more productive. This characterisation reflects the fact that Eastern Europe – with a good number of its regions at the lower end of the distribution – plays a significant role in Central European industrial value chains. On the other hand, the sector's role in the economy has steadily decreased over time, reflecting the progressive shift towards a service-based economy consistent with countries’ more advanced economic development. Also, the regions with higher productivity have a business structure that is made up of larger firms, specifically with a higher share of employment in firms of more than 10 workers; this suggests that more scalable firms have higher productivity, as has been empirically documented in the economic literature (see third chart). This difference persists over time, although the intermediate quintiles show some improvement. In less productive regions, employment in microenterprises predominates, which limits the ability to scale.