Population statistics at regional level
Data extracted in March 2018.
Planned article update: September 2019.
The three Baltic Member States had the largest gender gaps among EU regions for life expectancy at birth.
Among EU regions, the Greek island region of Lesvos, Limnos had the highest crude rate of net migration (4.6 %).
There are considerable differences in regional demographic patterns across the European Union (EU) from overcrowded, dynamic metropolises which may have relatively youthful populations to more remote, rural regions that are often characterised by declining population numbers, a relatively elderly population structure and poor access to a range of services.
Demographic developments have the potential to influence regional economic performance, resource consumption and other environmental pressures. In recent decades, many of the EU Member States have been characterised by an increasing pattern of population concentration, as people move from rural areas towards large cities (and surrounding areas). There has been considerable policy interest in the blurring of borders between urban centres and their adjacent regions, as low-density suburban developments have social, economic and environmental implications. Most population projections indicate that the EU’s population will continue to age as a result of consistently low levels of fertility and extended longevity. Although migration can play an important role in the population dynamics within many of the EU Member States, it is unlikely that it can reverse the ongoing trend of population ageing. The social and economic consequences associated with population ageing are likely to have profound implications both nationally and regionally, for example, impacting the capacity of governments to raise tax revenue, balance their own finances, or provide adequate pensions and healthcare services.
Statistics on regional demography are one of the few areas where detailed NUTS level 3 data are collected and published for each of the EU Member States, EFTA and candidate countries. The demographic characteristics of a given territory are the cumulative result of a range of demographic events: live births, deaths, emigration and immigration.
Historically, increased longevity may be attributed to a range of factors including improved socioeconomic and environmental conditions, changes in working conditions/occupations, lifestyle changes or better medical treatment and care.
In 2016, the life expectancy of a new born child in the EU-28 was 81.0 years. Across the EU Member States, life expectancy at birth ranged from highs in Spain (83.5 years) and Italy (83.4 years) down to lows of 74.9 years in Bulgaria, Latvia and Lithuania.
Life expectancy in the EU ranged from a high of 85.2 years in the Spanish capital city region down to 73.3 regions in the north-western Bulgarian region of Severozapaden — a difference of 11.9 years
Map 1 presents life expectancy at birth for NUTS level 2 regions, detailing the average (mean) number of years that a new born child could expect to live if subjected throughout his/her life to current mortality conditions. The regions with the highest levels of life expectancy were principally located in Spain and Italy; note there were also three regions in Switzerland — Ticino, Région lémanique and Zentralschweiz — where life expectancy at birth reached a similar level.
Looking in more detail, there were 11 NUTS level 2 regions where life expectancy at birth in 2016 was 84.0 years or more (as shown by the darkest shade of yellow in Map 1). The highest level of life expectancy in the EU was recorded in the Spanish capital city region of Comunidad de Madrid (85.2 years), while the three other Spanish regions with high levels of life expectancy were neighbouring, interior regions located in the northern half of the country, running from west to east — Castilla y León, La Rioja and Comunidad Foral de Navarra.
In Italy, Provincia Autonoma di Trento had the highest level of life expectancy, at 84.3 years, and was joined to by two neighbouring northern regions — Provincia Autonoma di Bolzano/Bozen and Lombardia (which includes the city of Milano) — as well as the central regions of Umbria and Marche, such that there were five Italian regions where life expectancy stood at 84.0 years or more.
Aside from these nine regions in Spain and Italy, the two remaining regions with the highest levels of life expectancy were both capital city regions, namely, Île de France (84.2 years) and Inner London - West (84.1 years). The relatively high level of life expectancy in the capital city regions of Spain, France and the United Kingdom may be attributed, among other reasons, to the close proximity and wide range of healthcare services that are available, alongside relatively high levels of income and living conditions. The situation in the United Kingdom (which has two capital city regions at NUTS level 2) was of particular interest insofar as life expectancy at birth in Inner London - West was, on average 2.1 years higher than in the adjacent region of Inner London - East, which, among other things, is generally considered to be less affluent.
At the other end of the range, there were 38 NUTS level 2 regions where average life expectancy was less than 78.0 years (as shown by the lightest shade of yellow in Map 1). In 2016, the lowest life expectancy at birth was recorded in the north-western Bulgarian region of Severozapaden, at 73.3 years. This reinforces the link between life expectancy and income and living conditions, insofar as Severozapaden also recorded the lowest level of economic activity in the EU, as its gross domestic product (GDP) per inhabitant was less than 30 % of the EU-28 average.The majority of the regions with relatively low levels of life expectancy were predominantly located in the easternmost regions of the EU, including: all six regions from Bulgaria, all seven regions from Hungary and all eight regions from Romania, as well as three out of four regions from Slovakia, 8 out of 16 regions from Poland, one of the two Croatian regions, and one of the eight Czech regions. There were four more regions in the EU where life expectancy at birth was less than 78.0 years: two of the three Baltic Member States — Latvia and Lithuania (both single regions at this level of detail) — as well as the outermost regions of Mayotte (France) and Região Autónoma dos Açores (Portugal).
The largest gender gap for life expectancy at birth was recorded in Lithuania, where women could expect to live 10.6 years more than men
Figure 1 provides more detailed information on the highest and lowest levels of life expectancy, with a specific analysis by sex. In 2016, life expectancy at birth for women (83.6 years) was some 5.4 years higher than that for men (78.2 years).
Among the top 10 regions in the EU with the highest levels of life expectancy for women, there were six regions from Spain and three from France, the remaining region being Provincia Autonoma di Trento (Italy). A similar analysis for men reveals that the top 10 regions was composed of seven regions from Italy, two from the United Kingdom and one from Spain. The highest levels of male and female life expectancy at birth were recorded in the Spanish capital city region, Comunidad de Madrid, where a new born child could on average be expected to live 87.8 years if female and 82.2 years if male.The largest gender gaps for NUTS level 2 regions were recorded in the three Baltic Member States (each a single region at this level of detail), as women could expect to live 10.6 years more than men in Lithuania, 9.8 years more in Latvia and 8.9 years more in Estonia, and in several Polish regions — Lódzkie, Lubuskie, Warminsko-Mazurskie, Podlaskie and Lubelskie — where the gender gap ranged from 8.9 years to 9.4 years. There were no regions in the EU where men could expect to live longer than women. However, the gap in life expectancy between the sexes was generally quite small in the Netherlands, Sweden and the United Kingdom. The difference was no more than 3.0 years in Devon (the United Kingdom) and three Dutch regions — Drenthe, Overijssel and Utrecht — the latter recording the lowest gender gap, at 2.8 years.
Historically low fertility rates have led to a gradual ageing of the EU’s population structure, which has been driven, among others, by a growing proportion of women choosing to delay/postpone childbirth. Women in the EU-28 are having fewer children, which may, at least in part, be attributed to a growing share of women participating in further education and/or trying to establish a professional career (before starting a family).
It was commonplace to find that childbirth was delayed in capital city regions, while women living in former industrial heartlands tended to give birth at a much younger age
In 2016, the average (mean) age for childbirth in the EU-28 was 30.6 years. Looking in more detail by NUTS level 3 region, the mean age of women at childbirth was generally quite high across Ireland, Spain and Italy and quite low in much of eastern Europe (see Map 2).
There were eight regions in the EU where the average age at childbirth was above 33.0 years, these were principally urban regions in capitals, including Voreios Tomeas Athinon (in the north of the Greek capital), Paris (defined here as the centre of the French capital within the confines of the périphérique) and the four London boroughs of Westminster; Kensington and Chelsea & Hammersmith and Fulham; Camden & City of London; and Wandsworth (the United Kingdom). The other two regions were somewhat atypical: Heidelberg Stadtkreis (a university town in Germany) and Medio Campidano (a relatively rural region in Sardegna, Italy). With the exception of the latter, these figures tend to support the view that some women delay having children in order to study or pursue a career.At the other end of the range, there were 40 NUTS level 3 regions located across Bulgaria and Romania where the mean age of women at childbirth was less than 27.5 years in 2016. Among these, there were four city regions in Bulgaria where the average age at childbirth was below 26.0 years: Montana, Pazardzhik, Yambol and Sliven; the last of these recorded the lowest mean age of women at childbirth across the EU, at 25.1 years. The two regions with the lowest mean ages in Romania were also city regions and were located in the south of the country: Giurgiu (26.2 years) and Calarasi (26.3 years). Apart from these bottom 40 regions that were exclusively located in Bulgaria and Romania, the lowest mean ages of women at childbirth were recorded in the neighbouring, industrial regions of Kosický kraj (eastern Slovakia) and Borsod-Abaúj-Zemplén (northern Hungary). These figures tend to support the view that some of the lowest mean ages for women at childbirth were recorded in regions characterised as former industrial heartlands (often with relatively low levels of economic development). This is supported when extending the analysis to other EU Member States, as the mean age of women at childbirth was also relatively low — less than 29.0 years — in regions such as Bremerhaven Kreisfreie Stadt and several parts of Saxony-Anhalt in Germany, Ardennes or Pas-de-Calais in France, and Central Valleys or Hartlepool and Stockton-on-Tees in the United Kingdom.
With life expectancy at birth rising for successive generations and historically low fertility rates, it is not surprising to find that the median age of the EU-28 population has increased. During the most recent decade for which data are available it has risen by 2.7 years, reaching 42.8 years on 1 January 2017. The median age of the population also rose in each of the EU Member States during the last decade, with particularly rapid changes in the age structures of Greece, Portugal and Romania. On 1 January 2017, the lowest median ages among the Member States were recorded in Ireland (36.9 years) and Cyprus (37.4 years), while the highest median ages were recorded in Germany and Italy (both 45.9 years).
Although childbirth was often postponed in capital city regions, some of the lowest median ages were often recorded in these regions
Map 3 shows the median age of NUTS level 3 regions at the start of 2017. It is interesting to note that several of the EU Member States displayed a considerable range of ages: these intra-regional differences were most apparent in France, the United Kingdom, Denmark and Belgium, where the uppermost region had a median age that was at least 1.5 times as high as the lowermost region. Such differences often reflect relatively low median ages in those regions characterised as thriving, urban/metropolitan regions (with a wide variety of educational/employment opportunities and relatively high numbers of migrants), whereas the highest median ages are often recorded in regions characterised as sparsely populated areas and/or regions that are popular retirement destinations.
Across the whole of the EU, the lowest median ages among NUTS level 3 regions were recorded in two overseas French regions, Mayotte and Guyane (17.8 years and 25.6 years); these atypical regions both had relatively high fertility rates and lower than average levels of life expectancy. There followed 14 urban regions from the United Kingdom, where the median age of the population was lower than anywhere else in the EU; this group of regions included Manchester (30.1 years), Nottingham (30.3 years) and Tower Hamlets (eastern London; 30.7 years), as well as five additional boroughs in London. There were several other EU Member States where the lowest median ages were recorded in capital city regions, for example, in Bruxelles-Capitale/Brussel-Hoofdstad (Belgium), Sofia stolitsa (Bulgaria), Byen København (Denmark), Dublin (Ireland), Groot-Amsterdam (the Netherlands), Wien (Austria) or Stockholms län (Sweden). In those cases where the capital city region did not have the lowest median age, it was often the case that the lowest median age was recorded in a suburban region close to the capital, such as Stredocesky kraj (which surrounds the Czech capital of Hlavní mesto Praha), Dytiki Attiki (the western agglomeration of Athens), or Seine-Saint-Denis, Val-d'Oise or Val-de-Marne (each of which is a suburban area close to Paris). Another pattern observed in several Member States was for the lowest median age to be recorded in a university city, for example, Heidelberg Stadtkreis (Germany), Gdanski (Poland) or Manchester (the United Kingdom).
Looking in more detail at those regions with the highest median ages, perhaps the most striking feature of Map 3 is the high concentration of regions in eastern Germany. Of the 31 regions in the EU-28 where the median age of the population on 1 January 2017 was 51.5 years or more, 24 were located in eastern Germany and they were joined by Osterode am Harz in north-west Germany. In the aftermath of German reunification, many of these regions were characterised by a lack of varied employment opportunities and comparatively low living standards, which may have stimulated particularly young people to move in search of more varied and better paid work. This pattern was more generally repeated in most of the other EU Member States: for example, the remaining six regions where the median age was 51.5 years or more were composed of the central Greek region of Evrytania (54.3 years; the highest value across the EU), the central French region of Creuse (51.9 years), the Danish island of Bornholm (51.8 years), Zamora in north-west Spain, Alto Tâmega in northern Portugal and Etelä-Savo in southern Finland (all 51.5 years). Another characteristic apparent for several regions with relatively high median ages was that they were sometimes popular retirement destinations, with their median ages being pushed up by an influx of retirees. For example, this pattern could be seen on the Danish island of Bornholm, in the northern Italian coastal regions of Savona, Genova (both Liguria) and Trieste (Friuli Venezia Giulia), or several coastal regions in the United Kingdom (Dorset CC, North & West Norfolk, or the Isle of Wight).
The high share of young people living in many urban regions of the EU may be linked to lifestyle choices that are linked to education and labour force opportunities
The information shown in Map 4 is based on the old-age dependency ratio, defined here as the number of elderly people (aged 65 years and over) compared with the number of people of working age (aged 15-64 years). On 1 January 2017, this ratio stood at 29.9 % across the whole of the EU-28; in other words, there were just over three people of working-age for every elderly person. According to Eurostat’s population projections, continued increases in longevity, coupled with low fertility rates can be expected to result in the old-age dependency ratio continuing to rise in most EU Member States during the next 20-30 years, raising concerns for the sustainability of public finances, pension and healthcare systems. These concerns may be mitigated, to some degree, by overall improvements in health status coupled with changes to social security systems that may result in a greater share of elderly people remaining active within the labour force.
The information presented in Map 4 mirrors, to some degree, that shown in the previous map, as old-age dependency ratios are usually closely related to the median age of the population. That said, other demographic events — births, deaths and net migration — may also impact on these ratios. The lowest old-age dependency ratios in the EU-28 were generally recorded in urban centres — often capital cities and their surrounding areas — or in the outermost regions of the EU which are generally characterised by high fertility rates and low levels of life expectancy. The lowest rate among NUTS level 3 regions on 1 January 2017 was recorded in the French region of Mayotte (4.9 %), where there were, on average, slightly more than 20 persons of working-age for each elderly person; another French overseas region, Guyane, had the third lowest ratio in the EU (8.3 %). Tower Hamlets (7.9 %) and Hackney & Newham (9.9 %) — neighbouring boroughs in the east of London — were only other regions in the EU to record old-age dependency ratios below 10.0 %.At the other end of the ranking, there were 10 regions in the EU which recorded an old-age dependency ratio that was higher than 50.0 % on 1 January 2017; in other words, where there were fewer than two people of working-age for every elderly person. Several of these were already mentioned when analysing the regions with the highest median ages — Dorset CC (the United Kingdom), Dessau-Roßlau Kreisfreie Stadt, Suhl Kreisfreie Stadt (both Germany), Creuse (France), Arr. Veurne (Belgium) and Evrytania (Greece). The last of these had by far the highest old-age dependency ratio, at 65.0 %; in other words, for every two elderly persons there were approximately three people of working-age. The other four regions with old-age dependency ratios above 50.0 % were: Arta, Preveza in north-west Greece; Lot and Nièvre (largely rural regions in south-western and central France); Ourense in north-west Spain.
The final section in this chapter provides a description of changes in the total number of inhabitants living in NUTS level 3 regions. Historically, population growth in the EU has been largely driven by natural population change (the total number of births minus the total number of deaths), with a relatively minor role being played by migratory patterns. However, following the end of the post-war baby-boom, the rate of natural population growth started to slow from the 1970s onwards. This was followed in the 1990s by a quickening pace to political and economic union, as successive enlargements of the EU took place alongside the development of the European single market; this period was characterised by an increase in the relative importance of net migration (the difference between immigration and emigration) for explaining overall changes in population numbers. Since many of the EU Member States do not have accurate figures on immigration and emigration, net migration often has to be estimated; this is usually done by analysing, each year, the difference between the total population change and the natural change.
There are wide-ranging differences in patterns of demographic change across the EU. Some of the most common medium-term developments may be summarised as follows:
- a capital city region effect, as populations continue to expand in and around many capital cities which exert a ‘pull effect’ on national and international migrants associated with (perceived) education and/or employment opportunities;
- an urban-rural split, with the majority of urban regions continuing to report population growth, while the number of persons resident in many peripheral, rural and post-industrial regions decline;
- a north-south split between EU Member States, with a high proportion of the population in northern Member States being single and living alone, whereas Mediterranean regions are often characterised by lower birth rates but a more important role for family units;
- regional divergences within individual EU Member States which may impact on regional competitiveness and cohesion, for example, between the eastern and the western regions of Germany, or between northern and southern regions of Belgium, Italy and the United Kingdom.
Map 5 presents the crude rate of total population change in 2016: this is composed of two different effects: natural population change and net migration. Between 1 January 2016 and 1 January 2017, the EU-28’s population rose by 1.2 million inhabitants, equivalent to a growth rate of 2.4 per 1 000 inhabitants; this increase was wholly attributable to net migration, as the number of births and deaths was balanced. Among the 1 342 NUTS level 3 regions shown in Map 5, a majority (768) reported an increase in their overall number of inhabitants, while there were 568 regions that recorded a decline in population numbers, leaving six regions where the number of inhabitants remained unchanged.In 2016, the fastest expanding populations were often concentrated in eastern Ireland, western Germany, southern Sweden, and the south-eastern corner of the United Kingdom. The darkest shade of blue in Map 5 shows the 95 NUTS level 3 regions where the population grew, on average, by at least 12 per 1 000 inhabitants during 2016. Among these were 15 regions where population growth was at least 20 per 1 000 inhabitants (or 2.0 %): five of these were located in Germany, while there were two regions from each of Greece (both island regions), Spain (both island regions) and France (both outermost regions) and single regions from Malta, Romania (Ilfov, which is a commuter belt surrounding the Romanian capital city region of Bucuresti), Sweden (Uppsala län, just to the north of the capital city region of Stockholm) and the United Kingdom (Tower Hamlets, an eastern borough of London). The last four of these regions were synonymous with a relatively common pattern, insofar as some of the highest crude rates of population change were recorded in capital cities and/or their surrounding areas: see, for example, the Czech Republic, Germany, Estonia, Ireland, Greece, Spain, France, Hungary, Poland or the United Kingdom in Map 5.
In 2016, the fastest overall population growth in the EU was recorded in the northern Aegean island region of Lesvos, Limnos (Greece); its total number of inhabitants grew by 4.6 %Figure 2 provides a more detailed regional analysis of the NUTS level 3 regions with the highest and lowest rates of overall population change, with similar information for natural population change and net migration. By far the highest crude rates of natural population change were recorded in the outermost French regions of Mayotte and Guyane, while the next highest rates were in the urban centres of Hackney & Newham, Tower Hamlets (both eastern boroughs of London) and Seine-Saint-Denis (a northern suburb of Paris). On the other hand, the five regions with the highest crude rates of net migration included: two island regions in Greece (Lesvos, Limnos and Ikaria, Samos); two German regions (Bamberg Kreisfreie Stadt (which has a federal centre for migrants) and Salzgitter Kreisfreie Stadt (which decided in 2017 to no longer allow migrants to settle on its territory)); and Ilfov in Romania.
Source data for figures and maps
Eurostat collects a wide range of regional demographic statistics: these include data on population numbers and various demographic events which influence the population’s size, structure and specific characteristics. The data may be used for a wide range of planning, monitoring and evaluating actions, for example, to:
- analyse population ageing and its effects on sustainability and welfare;
- evaluate the economic impact of demographic change;
- calculate per inhabitant ratios and indicators — such as regional GDP per capita, which may be used, for example, to allocate structural funds to economically less advantaged regions;
- develop and monitor migration and asylum systems.
Demographic developments drive a range of policy developments, in particular within the fields of employment and social policy, health, free movement, asylum and migration. Indeed, statistics on population change and the structure of populations are increasingly used to support policymaking and to provide the opportunity to monitor demographic behaviour within a political, economic, social or cultural context. The European Parliament passed a resolution on Demographic change and its consequences for the future of the EU’s cohesion policy (2013/C 153 E/02) which underlined that regional demographic developments should be statistically measured and stressed that demographic change should be considered as a cross-cutting objective in future cohesion policy.
In May 2015, the European Commission presented a European agenda on migration outlining measures to respond to the influx of migrants and asylum seekers arriving in the EU through the Balkans and across the Mediterranean. The agenda also provided a range of options for the longer-term management of migration into the EU, setting out four levels of action for migration policy, namely:
- a new policy on legal migration — maintaining the EU as an attractive destination for migrants, notably by reprioritising migrant integration policies, managing migration through dialogue and partnerships with non-member countries, and modernising the blue card scheme for highly educated persons from outside the EU;
- reducing incentives for irregular migration — through a strengthening of the role of Frontex, especially in relation to migrant returns;
- border management — helping to strengthen the capacity of non-member countries to manage their borders;
- a strong common asylum policy — to ensure a full and coherent implementation of the common European asylum system.
The European Commission announced a new assistance instrument for emergency support in March 2016: this plan allocated some EUR 700 million of aid, over the period 2016-2018, to provide humanitarian assistance through the rapid delivery of food, shelter and healthcare. There followed a number of further initiatives during the remainder of 2016 as the migrant crisis remained high on the political agenda, among which: the implementation of the EU-Turkey statement; additional financial support to Bulgaria, Greece and Italy to help cope with specific migration challenges; further provisions for supporting Syrian refugees (those displaced within Syria and those in other host countries); additional support for the protection of unaccompanied minors; renewed efforts to help save lives at sea and to disrupt smuggler networks; as well as the creation of safe and legal routes for asylum-seekers.
- Regional demographic statistics (t_reg_dem)
- Population (t_demo_pop), see:
- Population on 1 January by NUTS 2 region (tgs00096)
- Population change by NUTS 2 region — crude rates of total change, natural change and net migration plus adjustment (tgs00099)
- Population density by NUTS 2 region (tgs00024)
- Regional demographic statistics (reg_dem)
- Population and area (reg_dempoar)
- Fertility (reg_demfer)
- Mortality (reg_demmor)
- Population (demo_pop), see:
- Regional data (demopreg)
- Population (demo_fer), see:
- Regional data (demofreg)
- Population (demo_mor), see:
- Regional data (demomreg)
- Population (ESMS metadata file — demo_pop_esms)
- Population change — demographic balance and crude rates at regional level (NUTS 3) (ESMS metadata file — demo_r_gind3_esms)