Create a book

GDP at regional level

From Statistics Explained

Data from March 2013. Most recent data: Further Eurostat information, Main tables and Database.
Maps can be explored interactively using the Eurostat Statistical Atlas (see User's manual).
Map 1: Gross domestic product (GDP) per inhabitant, in purchasing power standard (PPS), by NUTS 2 regions, 2010 (1)
(% of the EU-27 average, EU-27 = 100) - Source: Eurostat (nama_r_e2gdp)
Map 2: Gross domestic product (GDP) per inhabitant, in purchasing power standard (PPS), by NUTS 3 regions, 2010 (1)
(% of the EU-27 average, EU-27 = 100) - Source: Eurostat (nama_r_e3gdp)
Map 3: Change of gross domestic product (GDP) per inhabitant, in purchasing power standard (PPS), by NUTS 2 regions, 2008–10 (1)
(percentage points difference between 2010 and 2008; in relation to the EU-27 average) - Source: Eurostat (nama_r_e2gdp)
Figure 1: Gross domestic product (GDP) per inhabitant, in purchasing power standard (PPS), selected NUTS 3 regions, 2000–10 (1) - Source: Eurostat (nama_r_e3gdp)
Table 1: Proportion of the resident population, by NUTS 2 regions, 2010
(%) - Source: Eurostat (nama_r_e2gdp) and (demo_r_d3avg)
Figure 2: Gross domestic product (GDP) per inhabitant, in purchasing power standard (PPS), by NUTS 2 regions, 2010 (1)
(% of the EU-27 average, EU-27=100) - Source: Eurostat (nama_r_e2gdp)

Gross domestic product (GDP) is a key measure of economic development and growth. This article presents a regional analysis of European Union (EU) GDP, based upon the level of GDP per inhabitant (often used as an indicator of living standards), as well as how this measure has changed in recent years. Economic accounts provide important information that may be used to make a regional analysis of the economy. These statistics (which are only available in current price terms) are also used for the allocation of expenditure under the EU’s cohesion policy (see Regional policies within the introductory article). Every region in the EU is covered by cohesion policy: however, most structural funds are directed to NUTS level 2 regions whose GDP per inhabitant is less than 75 % of the EU-27 average (on the basis of a three-year average).

Main statistical findings

GDP is initially calculated in national currencies, and then converted by purchasing power parities (PPPs) which take account of different price levels between EU Member States, allowing for a more accurate comparison. By using PPPs (rather than market exchange rates) these indicators are converted into an artificial common currency called a purchasing power standard (PPS). The use of a PPS makes it possible to compare purchasing power across the regions of EU Member States that use different currencies and where price levels are different. For more information about the use of PPPs please refer to the data sources and availability section below.

Regional GDP per inhabitant

Map 1 shows GDP per inhabitant in each NUTS level 2 region as a percentage of the EU-27 average, which in absolute terms was 24 500 PPS in 2010, up from 23 500 PPS in 2009 but still slightly below the 2008 pre-financial and economic crisis level of 25 000 PPS. Among the NUTS level 2 regions in the EU, GDP per inhabitant ranged from 6 500 PPS (27 % of the EU-27 average) in Severozapaden in Bulgaria to 80 300 PPS (328 % of the EU-27 average) in the capital city region of Inner London in the United Kingdom; between the two ends of the distribution there was a factor of 12.4 to 1. Luxembourg (266 % of the EU-27 average), the Belgian capital city region of Région de Bruxelles-Capitale/Brussels Hoofdstedelijk Gewest (223 %) and the German region of Hamburg (202 %) occupied positions two to four in terms of a ranking of regions with the highest GDP per inhabitant. These were followed by the French capital city region, the Dutch city of Groningen and the capital city regions of Slovakia, the Czech Republic, Sweden and Austria all with GDP per inhabitant lying in a range equivalent to 164 % to 180 % of the EU-27 average. In general, many of the regions with high GDP per inhabitant (equivalent to 125 % of the EU-27 average or higher) were capital city regions or neighbouring regions — this was the case in Belgium, the Czech Republic, Denmark, Ireland, Spain, France, Luxembourg (which is just one region), the Netherlands, Austria, Slovakia, Finland, Sweden and the United Kingdom. In addition, there were several regions with an average GDP per inhabitant at least 25 % above the EU-27 average in southern Germany, around major cities in western Germany, northern Spain and Italy, western Austria, several regions of the Netherlands, the Belgian region around Antwerpen, the island region of Åland (Finland), the far north of Sweden and North Eastern Scotland (in the United Kingdom). As such, the Slovakian and Czech capital city regions of Praha and Bratislavský kraj were the only regions in the Member States that joined the EU in 2004 or 2007 among the 41 regions where the average GDP per inhabitant was 25 % or more above the EU-27 average. The next most prosperous region (by this measure) in the Member States that joined the EU in 2004 or 2007 was a long way behind, namely Bucuresti - Ilfov in Romania at 111 % of the EU-27 average. The Hungarian, Polish and Slovenian capital city regions were the only other regions in the Member States that joined the EU in 2004 or 2007 that reported GDP per inhabitant (in PPS) equal to or above the EU-27 average.

Overall, there were 68 level 2 regions with an average GDP per inhabitant that was more than 25 % below the EU-27 average. A total of 25 regions were concentrated in six of the EU-15 Member States: Greece (seven regions), Italy (five southern regions), France and Portugal (three regions each), the United Kingdom (two regions) and Spain (the region of Extremadura). The remaining 43 regions were in Member States that joined the EU in 2004 or 2007: all of these 12 Member States had at least one region below this level except for Cyprus and Malta. Among these regions there were 22 regions where average GDP per inhabitant was less than 50 % of the EU-27 average, and these regions were found in Bulgaria, Hungary, Poland, Romania and Slovakia. Around 38.4 million people lived in the 22 regions whose GDP per inhabitant in PPS was less than 50 % of the EU-27 average, equivalent to 7.7 % of the EU-27 population.

In the EFTA countries (no regional data for Switzerland and no data for Liechtenstein), GDP per inhabitant was above the EU-27 average, ranging from 102 % of the EU-27 average in Hedmark og Oppland to 192 % in Oslo og Akershus (both Norway). There were two other Norwegian regions with GDP per inhabitant more than 25 % above the EU-27 average (Agder og Rogaland and Vestlandet), while Swiss GDP per inhabitant was equivalent to 154 % of the EU-27 average. Generally low averages for GDP per inhabitant were recorded in the former Yugoslav Republic of Macedonia (36 % of the EU-27 average), Turkey (50 %) and Croatia (59 %).

A more detailed regional analysis

Map 2 presents the same indicator as Map 1 but at the more detailed level 3 of the NUTS classification. Understandably the overall analysis is similar to that for the NUTS level 2 regions, although there are a number of NUTS level 3 regions that are atypical for the higher level (NUTS level 2) regions to which they belong. This phenomenon may often result from commuting inflows into central NUTS level 3 regions from surrounding areas, resulting in a concentration of economic activity in the most built-up areas.

For example, in the Bulgarian capital NUTS level 2 region of Yugozapaden the average GDP per inhabitant (in PPS terms) was 75 % of the EU-27 average, but at the more detailed NUTS level 3, the region Sofia (stolitsa) recorded a value of 105 % for this indicator while the remaining four NUTS level 3 regions had values below 40 %. A similar situation occurred in the Polish capital city NUTS level 2 region of Mazowieckie where the NUTS level 3 regions of Ostrolecko-siedlecki and Radomski recorded average GDP per inhabitant (in PPS) that was less than half that recorded for Mazowieckie, the latter being pulled up by a relatively high level for the NUTS level 3 region of Miasto Warszawa.

Within the German region of Oberbayern (NUTS level 2) there was a very large range in the values recorded for this indicator between the NUTS level 3 regions: Fürstenfeldbruck recorded average GDP per inhabitant (in PPS) that was 76 % of the EU-27 average whereas München Landkreis recorded a ratio of 317 %. In a similar manner in Rheinhessen-Pfalz (NUTS level 2) the NUTS level 3 region Südwestpfalz recorded average GDP per inhabitant (in PPS) that was 52 % of the EU-27 average whereas Ludwigshafen am Rhein (Kreisfreie Stadt) recorded a value of 251 %. The German NUTS level 3 regions of Regensburg, Coburg, Schweinfurt, Wolfsburg, Koblenz and Ludwigshafen am Rhein (all Kreisfreie Städte) each recorded average GDP per inhabitant that was more than double the average for the NUTS level 2 regions of which they were part, namely Oberpfalz, Oberfranken, Unterfranken, Braunschweig, Koblenz and Rheinhessen-Pfalz respectively. In a similar vein, the NUTS level 3 region of Oost-Groningen in the Netherlands recorded average GDP per inhabitant (in PPS) that was 68 % of the EU-27 average, which was less than half the level (180 %) recorded in Groningen (NUTS level 2) as a whole.

Across the NUTS level 3 regions of the EU in 2010, GDP per inhabitant ranged from 5 000 PPS (20 % of the EU-27 average) in Vaslui in Romania to 143 800 PPS (587 %) in the capital city region of Inner London-West in the United Kingdom; between the two ends of the distribution there was a factor of 28.8 to 1. Along with Inner London-West five other NUTS level 3 regions recorded GDP per inhabitant that was at least three times as high as the EU-27 average, four in Germany and one in France: Wolfsburg, Kreisfreie Stadt; München, Landkreis; Frankfurt am Main, Kreisfreie Stadt; and Schweinfurt, Kreisfreie Stadt in Germany; and Hauts-de-Seine in France. In a further 23 NUTS level 3 regions GDP per inhabitant was at least double the EU-27 average and these regions were mainly in Germany (18 regions), with two in the Netherlands and one each in Belgium, France and Luxembourg. At the other extreme, with GDP per inhabitant below 30 % of the EU-27 average, were 27 regions, including 17 in Bulgaria, eight in Romania, and one each in Latvia and Hungary.

Among the level 3 regions in Norway, the capital city region of Oslo recorded GDP per inhabitant equivalent to 248 % of the EU-27 average, while none of the other Norwegian regions saw their average GDP per inhabitant fall below the EU-27 average. Among the level 3 regions of Croatia and the former Yugoslav Republic of Macedonia, GDP per inhabitant ranged from less than 20 % of the EU-27 average in Severoistocen and Poloski (in the former Yugoslav Republic of Macedonia) to 76 % in Istarska zupanija (Croatia), with the Croatian capital city region of Grad Zagreb well above this range, at 109 %.

Changes over time

Map 3 shows the extent to which GDP per inhabitant changed between 2008 and 2010, compared with the EU-27 average (expressed in percentage points of the EU-27 average). The period studied covers the main years of the financial and economic crisis: GDP per inhabitant within the EU-27 dropped from 25 000 PPS in 2008 to 23 500 PPS in 2009 before partially recovering to 24 500 PPS in 2010. As the analysis is based on a comparison with the EU-27 average, a small positive increase for an individual region may still reflect an actual fall in average GDP per inhabitant, albeit by less than the EU-27 average (-500 PPS per inhabitant) over the two years.

Regions that expanded relatively fast, whose GDP per inhabitant increased by more than 5.0 percentage points compared with the EU-27 average, are shown in the lightest sand shade. By contrast, regions which experienced the highest rates of contraction (those with a fall of 5.0 percentage points or more in GDP per inhabitant compared with the EU-27 average) are shown in the darkest shade of purple.

The highest growth rates relative to the EU-27 average were recorded in the Province/Provincie Brabant Wallon in Belgium (13.5 %) and the Polish capital city region of Mazowieckie (12.5 %). As well as these two regions there were a further 20 regions where the change was more than 5.0 percentage points. Six Polish and six German regions recorded increases of more than 5.0 percentage points, accompanied by two each in Belgium, Sweden and the United Kingdom, as well as one region each in Denmark, France, Malta (one region only at NUTS level 2) and Slovakia.

At the other end of the range, a total of 26 regions recorded a fall of at least 5.0 percentage points relative to the EU-27 average, with the largest reductions (10 percentage points or more) in Groningen (the Netherlands) and Ionia Nisia (Greece). These 26 regions were spread across 10 EU Member States: six regions in Greece, five regions in Spain, three regions each in Italy, the Netherlands and Finland, both Slovenian regions, and one region each in Estonia, Ireland, Romania and the United Kingdom. Among these regions were the capital city regions of Estonia (which is just one NUTS level 2 region), Greece, Spain, Romania and Slovenia.

In Denmark, Austria and Poland every region achieved a change in GDP per inhabitant (in PPS) between 2008 and 2010 that was at least equal to the EU-27 average if not higher; this was also the case in Luxembourg and Malta which are both just one region at the NUTS level 2. The former Yugoslav Republic of Macedonia (one region at level 2) and Turkey (no regional data available) also both recorded an increase in GDP per inhabitant (in PPS) relative to the EU-27 average.

By contrast, every region in Ireland, Greece, Spain and Slovenia recorded a fall in GDP per inhabitant (in PPS) that was greater than the EU-27 average, as was the case for Estonia, Cyprus, Latvia and Lithuania (which are all just one NUTS level 2 region). Both Croatian regions and Iceland (a single region at level 2) also recorded falls between 2008 and 2010 relative to the EU-27 average.

Focus on longer-term changes in selected regions

The three parts of Figure 1 show GDP per inhabitant (in PPS) as a percentage relative to the EU-27 average (set at 100 %) for 15 selected regions. The first part shows the five regions with the highest GDP per inhabitant (in PPS) in 2010 and shows how their GDP per inhabitant developed over the previous 10 years, always with respect to the EU-27 average in each of those years (note that the first part of the figure has been rebased to 2000 = 100 to allow for the information to be interpreted more easily). Overall, four of these regions displayed relatively stable developments, the exception being Wolfsburg (Kreisfreie Stadt) in Germany which was more volatile; it is likely that shorter working hours — Kurzarbeit — at a major car plant in this region led to the significant reduction in GDP per inhabitant in 2009, while the subsequent rebound in 2010 may be associated with a return to longer working hours.

Comparing these five regions with the top five regions with the highest GDP per inhabitant (in PPS) in 2000, four regions were common to the ranking: the main change was that Paris (France) dropped out of the top five from 2000 and was replaced by the neighbouring region of Hauts-de-Seine in the 2010 ranking.

The second and third parts of Figure 1 show the regions with the strongest expansions and contractions of GDP per inhabitant (in PPS) over the 10 year period to 2010. The regions where this indicator increased most were all in Romania and Bulgaria, the Romanian region of Giurgiu increasing from 14 % of the EU-27 average in 2000 to 36 % in 2010. Among these five regions was also the Bulgarian capital city region, where average GDP per inhabitant (in PPS) increased from less than half (47 %) of the EU-27 average in 2000 to just over the EU-27 average by 2010 (105 %). Three of the five regions where GDP per inhabitant (in PPS) fell most strongly already had a lower GDP per inhabitant (in PPS) than the EU-27 average in 2000 and developments over the period 2000–10 saw these regions move further away from the EU-27 average; two of these regions were Greek and one was Belgian. The two other regions with the largest falls for this indicator were also Greek: in Korinthia GDP per inhabitant fell from 23 % above the EU-27 average in 2000 to 16 % below it by 2010, while in Voiotia it remained above the EU-27 average but fell from 77 % above the average in 2000 to just 18 % above by 2010.

Around a quarter of the EU’s population lived in regions where GDP was less than 75 % of the EU-27 average

Table 1 presents an analysis of the proportion of the population in 2010 living in regions that had an average GDP per inhabitant (in PPS) below 75 % of the EU-27 average and the proportion equal to or above 125 % of the average. The proportion of the population living in regions where the average GDP per inhabitant was less than 75 % of the EU-27 average was 24.2 %, while the proportion living in regions where this value was 125 % or more of the EU-27 average was 18.4 %; the proportion of the population in the mid-range (GDP per inhabitant ranging from 75 % to less than 125 %) was 57.4 %.

The three Baltic Member States, each with just one NUTS level 2 region, had all of their population living in regions with an average GDP per inhabitant below 75 % of the EU-27 average in 2010; this was also the case in Croatia (just two regions). In Romania, Slovakia, the Czech Republic, Poland, Bulgaria, Hungary, Portugal and Slovenia, more than half of the population lived in NUTS level 2 regions with an average GDP per inhabitant lower than 75 % of the EU-27 average. By contrast, Denmark, Germany, Ireland, Cyprus (one NUTS level 2 region), Luxembourg (one NUTS level 2 region), Malta (one NUTS level 2 region), the Netherlands, Austria, Finland and Sweden reported that none of their population lived in a NUTS level 2 region with an average GDP per inhabitant that fell below 75 % of the EU-27 average. Indeed, the entire population of Luxembourg (one NUTS level 2 region) lived in a region with an average GDP per inhabitant (in PPS) of 125 % or more of the EU-27 average; in Ireland, the Netherlands and Austria, more than half of the population lived in such regions, as was the case in Norway.

On the islands of Cyprus and Malta (each just one NUTS level 2 region) the entire population lived in regions with a mid-range average GDP per inhabitant (from 75 % to less than 125 % of the EU-27 average), as did the majority of the population in the United Kingdom (86.8 %), France (79.5 %), Spain (77.9 %), Sweden (72.8 %), Finland (71.0 %), Germany (70.4 %), Denmark (69.5 %), Belgium (60.3 %), Greece (54.1 %) and Italy (53.6 %).

On the basis of this analysis, Italy was the EU Member State where there was the highest disparity in living standards between different regions — as 29.0 % of the Italian population lived in regions (principally in the south of the country) where average GDP per inhabitant (in PPS) was less than 75 % of the EU-27 average, 53.6 % of the population lived in regions where average GDP per inhabitant was in the mid-range, and 17.4 % of the population lived in regions (principally in the north of the country) where average GDP per inhabitant was 125 % or more of the EU-27 average.

In the Czech Republic, the capital city region of Praha (home to 11.9 % of the Czech population) had an average GDP per inhabitant (in PPS) that was 72 % higher than the EU-27 average in 2010, while the seven remaining NUTS level 2 regions in the Czech Republic (home to the remaining 88.1 % of the population) each reported average GDP per inhabitant that was below 75 % of the EU-27 average. The same pattern was observed in neighbouring Slovakia, where GDP per inhabitant in the capital city region of Bratislavský kraj (with 11.5 % of the population) was 77 % higher than the EU-27 average, while the remaining three NUTS level 2 regions (with 88.5 % of the population) each recorded GDP per inhabitant that was below 75 % of the EU-27 average.

Major regional differences within countries

There were large differences in GDP per inhabitant between NUTS level 2 regions within the same Member State; Figure 2 provides an analysis of these substantial regional differences within countries. Note that Estonia, Cyprus, Latvia, Lithuania, Luxembourg and Malta consist of only one region at NUTS level 2.

In 2010, the highest level of regional average GDP per inhabitant was at least three times as high as the lowest level in the United Kingdom, Romania, Slovakia and France, whereas it was more than twice as high in Bulgaria, Belgium, the Czech Republic, Hungary, Germany, Poland and Italy. The difference was least in Slovenia (a factor of 1.4 to 1) and highest in the United Kingdom (reaching a factor of 4.7 to 1).

In many of the EU Member States, the capital city region (at the NUTS level 2) had the highest GDP per inhabitant (in PPS): this situation occurred in all of the Member States except for Germany, Spain, Italy and the Netherlands; and this was also the case in Croatia. Germany was the only Member State where the capital city region at NUTS level 2 had average GDP per inhabitant below the national average. Aside from Germany, Italy and the Netherlands were the only other Member States where the capital city region did not record the highest level of GDP per inhabitant (in PPS).

Data sources and availability

The European system of national and regional accounts (ESA) provides the methodology for regional accounts in the EU. ESA95 is fully consistent with worldwide guidelines for national accounts, the 1993 system of national accounts (1993 SNA). Following international agreement on an updated version of the SNA in 2008, the ESA is also being revised.

GDP is the central measure of national accounts, summarising the economic position of a country or region. It can be calculated using different approaches: the output approach; the expenditure approach; and the income approach. However, at the regional level the expenditure approach cannot be used, because it would require the measurement of regional exports and imports; this is not possible in the EU Member States.

Comparisons between where people work and where they live

A regional comparison of the level of economic activity can be made by comparing regional GDP with the population of the region in question; this is where the distinction between place of work and place of residence becomes significant. GDP measures the economic activity within national or regional boundaries, regardless of whether this was attributable to resident or non-resident employed persons. As a result, regional GDP per inhabitant is based upon a numerator that reflects the place-of-work (the GDP produced in the region) which is divided by a denominator whose value reflects the place-of-residence (the population living in the region). This drawback is particularly relevant when there are significant net commuter flows into or out of a region. Areas that are characterised by a considerable number of inflowing commuters often display regional GDP per inhabitant that is extremely high (when compared with surrounding regions). This is particularly the case for economic centres such as the regions of London (United Kingdom), Wien (Austria), Hamburg (Germany), Praha (Czech Republic) or Luxembourg. Because of this anomaly, high levels of GDP per inhabitant that are recorded for some regions with net commuter inflows do not necessarily translate into correspondingly high levels of income for the people living in the same region.

Purchasing power parities

Regional GDP is calculated in the local currency of the region (and therefore the country) in question. GDP can be converted into a common currency to make it more easily comparable — for example, converting into euros or dollars.

Exchange rates reflect many factors relating to supply and demand in currency markets, such as international trade, inflation forecasts and interest rate differentials. However, exchange rates do not reflect all the differences in price levels between countries. To compensate for this, GDP can be converted using conversion factors known as purchasing power parities (PPPs) to an artificial common currency, called a purchasing power standard (PPS); this makes it possible to compare the purchasing power of different national currencies. Even within a currency union, such as the euro area, a single currency continues to display different purchasing power across countries, depending on national price levels. In broad terms, the use of PPS series rather than the euro-based series tends to have a levelling effect, as those regions with very high GDP per inhabitant in euro terms also tend to have relatively high price levels (for example, the cost of living in central Paris or London is generally higher than the cost of living in rural areas of the EU).

Calculations for GDP per inhabitant that are based on PPS series instead of euro series can result in considerable differences when ranking regions. For example, in 2010, the Belgian region of the Province/Provincie Limburg was recorded as having a GDP per inhabitant of EUR 26 700, ranking above the German region of Schleswig-Holstein, with EUR 25 400. However, in PPS terms, Schleswig-Holstein, at 24 200 PPS per inhabitant, was above the Province/Provincie Limburg, at 23 800 PPS.

Context

Measuring economic development

Economic development is commonly expressed in terms of GDP, which in the regional context may be used to measure macroeconomic activity and growth, as well as providing the basis for comparisons between regions. GDP is also an important indicator from the policy perspective, as it is crucial in determining the extent to which each Member State should contribute to the EU’s budget, while three-year averages of GDP are used to decide which regions should be eligible to receive support from the EU’s structural funds.

GDP has also come to be regarded as a proxy indicator for overall living standards. However, by design and purpose, it cannot be relied upon to inform policy debates on all issues. GDP does not measure, for example, environmental sustainability or social inclusion, and these limitations need to be taken into account when using GDP for analysis. Indeed, it is increasingly recognised that GDP alone should not be used to measure economic, social and environmental priorities.

A number of international initiatives have focused on this issue and in August 2009, the European Commission adopted a communication called ‘GDP and beyond: measuring progress in a changing world’ (COM(2009) 433 final), which outlined a range of actions to improve and complement GDP measures. The European Commission noted that there was a clear case for complementing GDP with statistics covering other economic, social and environmental issues, on which individuals’ well-being critically depends.

Economic policies

The EU’s regional policy seeks to help every region achieve its full potential, through improving competitiveness and raising living standards of the poorest regions towards the EU average. Regional economic policy seeks to stimulate investment in the regions by improving accessibility, providing quality services and preserving the environment, thereby encouraging innovation and entrepreneurship and the creation of jobs, while overcoming inequalities that may be manifest in social deprivation, poor housing, education and healthcare, higher unemployment or inadequate infrastructure provisions.

See also

Further Eurostat information

Data visualisation

Publications

Main tables

Regional economic accounts - ESA95 (t_reg_eco)
Regional economic accounts - ESA95 (t_nama_reg)

Database

Regional economic accounts - ESA95 (reg_eco)
Regional economic accounts - ESA95 (nama_reg)

Dedicated section

Methodology / Metadata

Source data for tables, figures and maps (MS Excel)

External links


Views