WIIW The Vienna Institute for International Economic Studies - Wiener Institut für Internationale Wirtschaftsvergleiche

image credit: istock.com/BalkansCat

By Ambre Maucorps

Introduction

After the 1957 Treaty of Rome established the European Economic Community, the EU grew steadily until, by the mid-2010s, it included 28 member states (though Brexit put a stop to its continuous expansion as a thriving market that ensured the free movement of people, goods, services and capital). In particular, the fall of the Iron Curtain and the progressive opening of Central and Eastern European economies in the 1990s raised people’s hopes and expectations – on both sides of the then divided Europe – of better socio-economic prospects for communities and businesses.

Enshrined in the Treaty of the European Union (TEU), the EU is bound to strive for ‘an ever closer union among the peoples of Europe’ (Article 1) and should promote ‘economic, social and territorial cohesion, and solidarity among Member States’ (Article 3). In particular, the EU Cohesion Policy was designed specifically to address structural disparities and to foster socio-economic convergence within the EU. Starting out with a strong focus on the poorest and most backward regions in the 1980s and 1990s, from 2000 onwards – and particularly with the Lisbon Treaty – the Cohesion Policy’s priorities shifted towards growth, jobs and innovation,[1] thereby providing tailored support for all EU regions.

In its Communication on growth and cohesion in EU border regions, the European Commission (2017) recalls that ‘border regions generally perform less well economically than other regions within a Member State’. Yet borders do not necessarily hinder growth and cohesion equally: their impact depends on topography and accessibility, political, administrative and socio-economic systems, currency and language, among other factors. Diverging historical development paths, in particular between Western and Eastern European regions, could play a major role in this respect. Likewise, intra-EU border regions are more likely to experience different socio-economic conditions than regions located on the EU’s external borders.

Evidently, crises exacerbate the impact of borders. For instance, the COVID-19 pandemic caused unparalleled economic disruption both between and within countries. In Europe, mitigation measures taken by regional and national governments slowed the progress of the disease, but also undermined economic activity – especially where borders were closed, often in an uncoordinated manner. A study carried out for the European Commission reveals that the impact of border-related measures on EU cross-border communities and local economies differed from one border region to the next, depending on pre-existing degrees of interdependence and the toughness of the restrictions (Peyrony et al., 2021).

More recently, Russia’s invasion of Ukraine has put border regions in Poland, Slovakia, Hungary and Romania in the spotlight, as they became the first places to welcome and host refugees. At the same time, the borders between the EU countries (i.e. Finland, Estonia, Latvia, Lithuania and Poland) and both Russia and Belarus have been – sometimes literally – fortified, in particular to deal with the risk that those two countries could instrumentalise migration flows. This poses fresh challenges for regions located on the EU’s external borders, the majority of which are less developed or transition regions.[2]

With the war in Ukraine unlikely to end any time soon and with the enlargement debate now revived, further challenging the EU’s economic recovery and convergence processes, the concepts of integration, cohesion and solidarity, as phrased in the TEU, deserve even greater attention. This article touches on these key issues by focusing on the economic situation of border regions in Europe and how they have developed over time. In doing so, it provides novel insights into two important issues for EU policy makers: first, what is now commonly referred to as ‘left-behind places’, and second, what is generally called ‘the costs of non-Europe’.[3]

Europe: a continent full of borders

As of 2023, the EU includes close to 1,170 regions at the NUTS 3 level of territorial disaggregation,[4] including 384 land border regions.[5] This means that one third of the EU regions are border regions. It is also noteworthy that this proportion is as high as 58% in EU-CEE (in other words, more than half of the EU-CEE regions are actually border regions).

In fact, many European countries are made up largely (if not exclusively) of border regions. Indeed, all regions of North Macedonia and Slovakia are border regions, while Luxembourg and Montenegro are composed of just one NUTS 3 region – a border region by definition. Likewise, and focusing on EU-CEE only, more than half of the regions in Czechia, Croatia, Slovenia, Hungary, Bulgaria, Lithuania, Latvia and Estonia are actually border regions.

This territorial feature throws into sharp relief the importance of cross-border European integration (in the form of e.g. cross-border labour markets or cross-border transport infrastructure), especially for the EU-CEE countries as they transition to open-market economies. In CESEE more generally, the different stages of economic opening-up and of the deepening of cross-border cooperation – and conversely the absence of these features – have logically shaped the economic development of the regions located there.

Development gap between border and inner regions

In many CESEE countries, the capital region – typically a non-border region – performs disproportionately better than the rest of the country, raising concerns for social, economic and territorial cohesion (Maucorps, 2022). That is also the case for many countries in Western Europe, so that the discrepancy between the average level of development of border and inner regions, respectively, remains generally sizeable in Europe as a whole – as well as in the EU, despite decades of economic and political integration efforts.

When looking at individual countries, that gap varies widely – and is, of course, heavily influenced by the number of border vs inner regions within each country, as well as by whether the capital region is a border or an inner region. In Latvia, for example, the only non-border region is the capital region, whose borders dovetail with those of the capital city Riga. As a result, in the country’s border regions average GDP per capita in Purchasing Power Standards (PPS) is not even half that of the (sole) non-border region. On the other hand, Lithuania’s only non-border region is not the capital region (which happens to abut Belarus), but is in fact Kauno apskritis, the county surrounding the country’s second-largest city of Kaunas. Consequently, the country’s average GDP per capita in PPS in the border regions is almost equivalent to that of the (sole) non-border region.[6]

Still, in CESEE countries that encompass more than one border and one non-border region, the prevailing picture is one of border regions lagging significantly behind their inner peers (Figure 1). In Bulgaria, Hungary, Turkey, Estonia and Czechia, border regions achieve a GDP per capita in PPS that is only 50-60% that of their inner peers. In Albania, Poland and Romania, the figure is around 65%. It is worth noting that most of these countries border both EU and non-EU countries – as opposed to most Western European countries that have long enjoyed a full-fledged EU neighbourhood.

Figure 1 - Average GDP per capita (in PPS) of border regions, as a percentage of that of non-border regions, in selected CESEE countries and Austria (2019)

Economic growth path of border vs inner regions

The evolution of the development gap between border and inner regions over time is also a topic of discussion for policy makers and those concerned with territorial cohesion. As a matter of fact, the times series of the gap between border and inner regions’ economic output reveals some strong East-West disparities in Europe that single-year values conceal. Indeed, the backwardness of border regions (vis-à-vis inner regions) in Western Europe has declined overall since the turn of the century (although to a limited extent), whereas by contrast in the border regions of Central and Eastern Europe that backwardness has largely deepened.

The most extreme case is that of Bulgaria, where in 2000 the border regions had (on average) GDP per capita (in PPS) equivalent to 77% of that of inner regions; by 2019 that figure was down to a mere 51% (Figure 2). Nonetheless, it does seem that joining the EU somewhat curbed this divergence trend, as evidenced by the slowing pace of the decline observed since 2007. That year, a large number of Bulgarian border regions effectively became neighbours of other EU regions.

Figure 2 - Average GDP per capita (in PPS) of Bulgaria’s border regions, as a percentage of that of non-border regions, and trend lines (2000-2006 and 2007-2019)

Regional convergence in the EU: a two-tier process

Viewed through the lens of EU-wide convergence patterns, the distinction between border and inner regions allows for some novel insights into the complexity behind the concept of territorial cohesion. Indeed, the Eighth Cohesion Report (European Commission, 2022) states that ‘Since 2001, less developed eastern EU regions have been catching up with the rest of the EU, leading to a substantial reduction of the GDP per capita gap.’ While this general comment signals positive achievements in terms of convergence in the EU, it masks the different speeds of convergence among the different types of region. By way of illustration, Figure 3 shows the evolution of the average GDP per capita (in PPS) of border and inner regions as a percentage of the EU average for two EU countries, Bulgaria and Austria.

Figure 3 - Average GDP per capita (in PPS) of border and inner regions, as a percentage of the EU average, Bulgaria and Austria (2000-2019)

This example allows for some interesting observations, namely:

• The steady increase in Bulgaria’s border and inner regions’ GDP as a percentage of the EU average confirms the East-West convergence process observed over the past two decades.
• The upward inflexion of the curve around 2006-2007 (in particular for Bulgaria’s inner regions) indicates an acceleration of the convergence process just after Bulgaria’s entry into the EU, emphasising the role of EU membership in economic development.
• The widening gap between Bulgaria’s inner and border regions points to a decrease in territorial cohesion within the country.
• The narrowing gap between Austria’s border and inner regions’ economic development signals positive spill-over effects from the EU enlargement process: in 2000, around half of Austrian border regions were located on the EU’s external borders, whereas in 2019, only a few were (those neighbouring Switzerland and Liechtenstein).

Conclusions

There is little literature that addresses border-related regional development issues on an EU-wide or Europe-wide scale. As a matter of fact, cross-border development and integration are mostly examined in specific locations, based on the policies or programmes being implemented there (e.g. cross-border Interreg programmes) and the availability of relevant data. On the other hand, territorial cohesion and convergence processes within the EU are usually tackled at the NUTS 2 level of territorial disaggregation – that being the territorial level at which Cohesion Policy eligibility criteria are defined – therefore overlooking the impact of country borders on local economies.

Using Eurostat’s typology of border and inner (non-border) regions at the NUTS 3 level of territorial disaggregation and the related dataset of GDP per capita in PPS, this article outlines the size and the evolution of the economic gap between both types of region. It shows that this gap is significant in many European countries, especially EU-CEE countries. There, the gap has been mostly widening since 2000, in contrast to the general trend observed in Western Europe.

Finally, the comparative example of Bulgaria and Austria illustrates the two-tier convergence process at work in the EU: while Austria’s border regions have been catching up with their inner peers, Bulgaria’s border regions have been losing ground to their inner peers, even though both types of region have been converging on the EU average. For the sake of economic, social and territorial cohesion, this latter finding must not be ignored by policy makers at either the national or the EU level.

At the national level, territorial development should be concerned with the possible adverse effects of disparities within the country widening (e.g. popular discontent, political polarisation, distrust of governments). At the EU level, Cohesion Policy should ensure that regional (NUTS 2) convergence is not achieved at the cost of sub-regional (NUTS 3) divergence – whereby economies located closest to country borders face mounting challenges on their path to (sustained) economic growth and cohesion. Most importantly, efforts towards EU integration should be inclusive of the latter, especially at the EU external borders.

 

Footnotes:

[1] In the 2014-2020 Multiannual Financial Framework (MFF) programming period, EUR 352bn were earmarked for Cohesion Policy, representing almost a third of the total EU budget.

[2] According to the categorisation of the 2021-2027 Cohesion Policy’s ‘Investment for Jobs and Growth’ goal, defined as follows: less developed regions are those with a GDP per capita in Purchasing Power Standards (PPS) that is less than 75% of the EU average; transition regions have a GDP per capita (in PPS) of between 75% and 100% of the EU average; and more developed regions have a GDP per capita (in PPS) of over 100% of the EU average.

[3] See, for instance, Rodríguez-Pose (2017) on left-behind places and Camagni et al. (2017) on the costs of non-Europe.

[4] Based on the 2021 NUTS classification. NUTS regions are generally defined according to administrative and governance systems. Regulation (EC) 1059/2003 states that each member state’s NUTS 3 territorial class shall have an average population size in the range of 150,000 to 800,000. Many non-EU countries in Europe (including Montenegro, North Macedonia, Albania, Serbia and Turkey) also use the NUTS classification. For more information, see Eurostat (2022).

[5] Border regions are defined by Eurostat as ‘All NUTS 3 regions along land borders, plus NUTS 3 regions that have at least 50% of their population in areas of 25 km width along a land border’. Henceforth, ‘regions’ refer to regions defined at the NUTS 3 level of territorial disaggregation; ‘border regions’ to border regions as per Eurostat’s definition; and ‘inner regions’ to regions that are not border regions as per Eurostat’s definition.

[6] Source: Eurostat, GDP by other typologies, indicator [URT_10R_3GDP].

 

References:

Camagni, R., Capello, R., Caragliu, A. and Toppeta, A., 2017. Quantification of the Effects of Legal and Administrative Border Obstacles in Land Border Regions. Final Report, Expert contract number 2016CE160AT091. Luxembourg: Publications Office of the European Union

European Commission, 2017. Communication from the Commission to the Council and the European Parliament, Boosting growth and cohesion in EU border regions, Brussels, COM(2017) 534 final, 20.9.2017

European Commission, 2022. Cohesion in Europe towards 2050: Eighth report on economic, social and territorial cohesion. Luxembourg: Publications Office of the European Union

Eurostat, 2022. Statistical Regions in the European Union and Partner Countries: NUTS and statistical regions 2021 (2022 edition). Luxembourg: Publications Office of the European Union

Maucorps, A., 2022. Chart of the month: The role of capital regions in CESEE. In Vasily Astrov (ed.), wiiw Monthly Report, No. 06/2022. Vienna: wiiw, pp. 7-8

Peyrony, J., Rubio, J. and Viaggi, R., 2021. The Effects of COVID-19 Induced Border Closures on Cross-border Regions: An empirical report covering the period March to June 2020. Luxembourg: Publications Office of the European Union. https://data.europa.eu/doi/10.2776/092793

Rodríguez-Pose, A., 2017. The revenge of the places that don’t matter (and what to do about it). Cambridge Journal of Regions, Economy and Society, 11(1), pp. 189-209. https://doi.org/10.1093/cjres/rsx024