International comparisons of economies and societies tend to be undertaken at the country level. Statistics on production capacities refer to gross domestic product (GDP) for example, while health and education levels tend similarly to be measured and debated in national terms. However, the differences between countries are often not as great as the disparities within them. In Chile, Poland, Portugal and Turkey, for example, the GDP per worker in the best performing region is more than three times higher than in the worst performing region. In several places, disparities have widened recently, as evidenced by the doubling of the difference in employment growth rate across OECD regions after the 2008-09 recession. Understanding the differences and similarities in regional economic structures is essential for designing effective strategies to resume and maintain aggregate growth. At the same time, statistical evidence
specific to regional contexts can help strengthen the recovery of OECD countries by identifying potential and challenges of places to become engines of sustainable economic and social development.
OECD Regions at a Glance aims to make these differences across "places" apparent, and to unpack what they imply for national performance and individual welfare. It does so by providing region-by-region indicators that help to identify areas that are outperforming or lagging behind in their country, as well as the 34-country OECD area. Patterns of growth and the persistence of inequalities are analysed over time, and insights are provided on the relationships between the characteristics of regions and the outcomes they achieve.
This fourth issue of the OECD Regions at a Glance series is organised in three sections, showcasing the contribution of regions to build stronger, fairer and cleaner economies: I. Regions as drivers for competitiveness; II. Inclusion and equal access to services in regions; III. Environmental sustainability in regions and metropolitan areas. This edition includes a number of novelties compared with the previous issues. The analysis of regional competitiveness has been expanded to include measures of public investment in regions, regional connectivity in innovation activities and trends of skilled migrants. For the first time, regional indicators of carbon emissions, air quality, change in urban land and in forest cover, and access to network infrastructure have been produced to assess the environmental sustainability of regions. Statistics on labour mobility and immigration in OECD
regions enrich the analysis of demographic trends. The territorial dimension of access to services is explored both for OECD countries (education, health and broadband access) and developing countries (water, sewage and electricity). Finally, regional statistics for emerging economies (Brazil, China, India, the Russian Federation and South Africa) are included for the first time.
Local factors matter for national sustained growth. Around 40% of OECD GDP, employment and population growth in the past 15 years are largely due to a small number of regions (the 10% top performing regions). The importance of economic agglomerations for growth is an established fact. However, data show that the pace of growth has slowed in many large metropolitan regions in OECD countries, while predominantly rural regions grew at a faster rate than predominantly urban regions in the past 15 years, narrowing the gap between urban and rural areas in some countries.
Increases in labour productivity have a more significant effect than changes in labour utilisation on the GDP per capita growth rate of OECD regions during 1995-2007. Differences in labour productivity growth among regions are invariably the result of multiple national and local factors, including labour-market policies and institutions as well as innovation and skills. For these reasons, national and regional governments are increasingly putting co-ordinated policies for education, skills and innovation in place.
Differences in the tertiary educational attainment of the labour force are greater among regions in the same country than across countries. The Czech Republic, France, Spain and the United States show the largest variation in tertiary educational attainment. For instance, in 2008, in West Virginia (United States) 16% of the labour force had completed tertiary education, while in the District of Columbia this value was 44%. Highly skilled human capital is increasingly mobile. Regions boosting their human capital endowments are those capable of attracting and retaining highly skilled workers. Regional data on international migration show that regions already hosting a significant number of highly educated foreigners have an advantage in the global competition for talent, due to the decisive role of network effects.
Technology-based innovation is very concentrated in a few places. Around 10% of OECD regions account for one-third of the total OECD expenditure in research and development (R&D) and more than half of patent applications. R&D and patenting are still concentrated in the top regions of knowledge-intensive OECD member countries (Japan, Germany, United States), having different technology specialisations (green technologies; biotechnology; and information and communication technology, for example). However, the last decade has witnessed the emergence of new regional top players. For example, Østjylland and Vestjylland (Denmark), Madrid and Barcelona (Spain) and Seoul and Gyeonggi-do (Korea) have appeared as hot spots in patenting on renewable energy. Improving collaboration to address global challenges is high on national agendas both in OECD countries and emerging economies: Shanghai (China) and Karnataka (India)
have tripled the number of partners around the world with which they patent in green technologies in the last five years. Several regions are catching up with national leaders in high-tech manufacturing employment and knowledge-intensive sectors. Studying these regions can provide lessons on how to replicate a positive environment for business creation and technological upgrading.
The 2008-09 economic recession experienced by the OECD area has had a differentiated impact on the loss of jobs within OECD countries. Three-fourths of OECD regions that showed employment growth between 1999 and 2007 shifted to an employment decline between 2008 and 2009. All typologies of regions experienced on average a decline in employment. There are however important differences in the trajectories and it is useful to compare the pre-crisis growth profile of regions that managed to sustain employment growth during the recession ( "resilient regions" ) with the one of regions that shifted from positive to negative total employment ( "recession-hit regions" ). Resilient regions experienced larger increases in their qualified human capital prior to the crisis, as well as in participation rates and in the productivity of business services, the public sector and
Countries have responded to the economic crisis by taking measures to support business and employment and by improving conditions for long-term growth. Such structural policies emphasise the complementarities of efficiency, equity and environmental protection objectives. The long-term consequences of the crisis on families' living conditions and social mobility, as well as the degree of heterogeneity in policy responses, point to the need for statistics that go beyond the economic efficiency of a region to encompass economic and non-economic measures of the quality of life. Spatial differences in access to key public goods and in opportunities for families to invest in health and education have important, dynamic effects, since "fragile" regions can get trapped in downward spirals, and see their capacity to attract people and businesses shrinking over time.
Regional disparities in households' income are largest in Chile, Greece, Hungary, Israel, Italy and the Slovak Republic. Regions in these countries have also experienced structural difficulties in employing young adults (the youth unemployment rate in some regions was as high as 30% in 2009) and women (less than half of women in working age were employed in 2009) and they display higher than average long-term unemployment (more than one-third of the unemployed in these regions have been out of the labour market for more than one year).
Income differentials only partially explain the quality of life in OECD regions. Disparities among regions in basic education attainment, health or natural resources are signalling the challenges of policies to overcome local disadvantages in many countries. Data show that students in city schools outperform rural areas by 20 points in the OECD PISA reading test, or the equivalent of almost one year of education, even after having taken into account the different socio-economic background of students. The age-adjusted mortality rate in regions of Canada, Portugal and the United States can be double that in other regions of the same country. In Chile, Italy, Korea and Mexico, the regional exposure to air pollution ranges from values below the World Health Organization's recommended levels of 10 micrograms of PM2.5 per cubic meter to more than 25 micrograms. Some provinces in China and states in India surpass a
threshold of 60 micrograms of PM2.5 per cubic meter, a level proven to pose serious health risks.
For the first time, estimates of CO2 carbon emissions at the regional level comparable across countries are presented in this volume. In the Czech Republic, the Russian Federation, Spain, South Africa and the United States, CO2 carbon emissions in some regions are more than three times higher than the country average, highlighting the limited value of national figures to design policies to address climate change. Cities are hot-spots for carbon emissions, and work is currently underway to produce city-level estimates of the carbon efficiency of urban economies.
In sum, OECD Regions at a Glance provides a multidimensional and dynamic snapshot of how life is lived - and can be improved - from region to region in OECD countries and emerging economies. Gaps in data availability and geographical coverage, as well as the need to identify better indicators in certain areas, are highlighted in the publication, suggesting next steps to further improve our knowledge for evidence-based policy making.