«Corrigenda to OECD publications may be found on line at: © OECD 2014 You can copy, download or print OECD content for ...»
Road transport: In many countries, air pollution from motor vehicles has replaced coal smoke as the major cause for concern. The growing use of motor vehicles means that efforts to reduce emissions from individual vehicles may be jeopardised by increases in the volume of traffic. In many developing countries, efforts to control pollution are further threatened by the use of old vehicles that do not meet modern pollution control requirements.
Power generation: Generating power produces more pollution (in particular, sulphur dioxide and nitrogen oxides) than any other single industry. Better dispersion of pollutants emitted by tall chimneys brings better dilution in the air and lowers local concentrations of pollutants. This, however, leads to wider dispersion of pollution and to trans-boundary air pollution. Stricter operating practices and the use of modern abatement techniques have resulted in a sizeable reduction in the amount of pollutants emitted from power stations.
High concentrations still occur in many developing countries, however, particularly from older power stations and from the use of high sulphur lignite or coal.
Waste disposal: Landfill and incineration are the most common methods of waste disposal. If not properly managed, landfill sites can cause a number of problems, such as the production of methane gas, dangerous levels of carbon dioxide, and trace concentrations of a range of organic gases and vapours. Poorly managed incineration can result in the production of poisonous chemicals such as hydrochloric acid, dioxins, furans and heavy metals. Hydrochloric acid produced by the burning of plastics contributes locally to acid rain. The burning at low temperature of organic matter and plastics can also lead to emissions of dioxins.
Source: Adapted from EEA (2010), Urban Environment – SOER 2010 Thematic Assessment, EEA, Copenhagen, Denmark
1.6. The geography of inequalities
The various dimensions of inclusion and exclusion are unevenly distributed across space – and often mutually reinforcing. Inequalities in income, opportunity and access to essential services all vary substantially within and across OECD countries. Spatial concentration of these forms of disadvantage vary strongly at various territorial scales and these different sources of inequality can reinforce one another, locking households and communities into circumstances that make it particularly hard for them to improve their life chances.
Wealth and poverty are concentrated spatially. Regional inequalities in GDP per capita have increased in many OECD countries since 1995, especially in Eastern European countries, and are particularly high in emerging market economies. In developing countries, income differentials between urban and rural areas are particularly pronounced, and rising inequality in big cities is sometimes the result of increasing inflows of wealthy people rather than growing concentrations of poverty (Glaeser et al., 2008). However, this is far from universal: and for most countries, poverty is higher in regions with lower average levels of income: poverty, like wealth, tends to be spatially concentrated. Relative poverty rates of regions (incomes less than 50% of the national median) provide a picture often masked by country averages. In Mexico, Turkey and Italy, relative poverty rates can vary from 50% (32% in Italy) to less than 7% (Figure 1.31) (OECD, 2014a).
Relative poverty in many countries is highly concentrated in space Regional relative poverty rates, 2010 Note: Poverty headcounts with poverty line defined at 50% of the national median income. Elaborations from the OECD Income Distribution Data at regional level.
Source: OECD (2014a), How’s life in your region? Measuring regional and local well-being for policy making, OECD Publishing, Paris.
Income inequality is higher within urban areas than elsewhere. In many countries, urban income inequality has been rising faster than overall income inequality. For example, in 2009, 17 of the top 25 US metropolitan areas had estimated Gini coefficients above the US national average (American Community Survey, 2010). In France, the median household income of the top 10% in France was 3.4 times that of the bottom decile, but this ratio varies widely across space, reaching 8.4 in the Aix-Marseille metropolitan region (OECD, 2013l).
Large cities tend to be more unequal than smaller ones, and the relationship between city size and intra-urban inequality seems to be strengthening. This is disconcerting when seen against the backdrop of demographic trends towards increasing concentration of populations and activity in larger cities, especially in emerging economies and developing countries. Evidence suggests that it reflects changes in the nature of cities’ economic specialisations. Very large cities increasingly concentrate specific managerial and administrative functions and the high-value services that support them (corporate headquarters, R&D, finance, etc). However, as Sassen (2006) points out, lower-skilled jobs are also increasingly in demand in such cities (i.e. their skill profiles are more polarised). This can be problematic when such low skilled jobs are not able to pay a ‘living wage’.18 As a result, even the most dynamic metro regions have experienced rising inter-personal inequality (OECD, 2006a).
While urbanisation can help lift people out of poverty, large numbers of urban residents remain trapped in it. Drawn to the cities by the opportunities they offer, many migrants from rural areas struggle with the high costs of living in cities, giving rise to an “urbanisation of poverty” for those who lack the skills required to compete within city labour markets. In most OECD countries, exclusion and poverty have become urban phenomena. These issues are prominent not only in lessadvanced metro-regions like Mexico City (about 50% of the population are in relative poverty), partly due to rural migration, but also in cities that have faced strong industrial restructuring (Rotterdam, Lille, Detroit) as well as in the suburbs of some of the richest metro-regions (Paris, London, New York).
Unequal access to employment contributes to inter-regional inequalities. In the past decade, employment growth in many OECD countries was highly concentrated in specific regions (OECD, 2013f). On average, 40% of overall employment creation in OECD economies during 1999-2012 was generated in just 10% of their regions. With the economic crisis, employment destruction has likewise been highly concentrated. In fact, in Ireland, New Zealand, France, Estonia, the Netherlands, Canada, and the Slovak Republic, half or more of the gap between current and pre-crisis employment levels could be filled if just one region returned its employment rate to its pre-crisis level (OECD, 2013f). In many countries, regional disparities in youth unemployment have grown wider since the crisis.
Southern European countries and Mexico are of particular concern, because in some regions the youth unemployment rate now exceeds 40%. Furthermore, while large cities drive national employment in many countries, the economic crisis has affected urban labour market conditions. The unemployment rate in metropolitan areas rose more in the period 2008-2012 than it did in the previous 8 years in 26 of the 28 OECD countries. In 2012, 45% of OECD metropolitan areas had an unemployment rate above the national rate (OECD, 2013f).
During 2007-11, the total number of unemployed people in 207 large OECD metros, for which data are available, rose by over 56% (with a peak at over 60% in 2010). Over the same period, total unemployment in the 25 countries where those metros are situated rose by just under 49% (with a 2010 peak 53% above the pre-crisis low of 2007). In 2012, the situation began to reverse, as metropolitan unemployment across the OECD fell, while aggregate unemployment in the countries concerned rose slightly. Nevertheless, the proportion of large OECD metros with unemployment rates above their national averages rose from about 40% in 2007 to 48% in 2011-12. The metros’ share of aggregate unemployment rose OECD-wide and in 18 OECD countries, including six of the G7 (France was the exception).
The concentration of unemployment in large cities reflected not only the shocks that hit them but also, in some instances, migration of unemployed workers from elsewhere to the cities in search of jobs. Either way, the fall-out from the crisis was increasingly felt in the big cities. For national economies stagnant metropolitan labour markets are a serious problem, given that large metro areas accounted for over half of all net employment growth in the OECD area since 2000 (OECD, 2013f).19 At smaller territorial scales, access to employment may be an issue where public transport networks and other infrastructure put households in poorer neighbourhoods at a disadvantage.
Extreme examples of such problems may be found in, for example, South Africa, where Apartheid-era patterns of spatial segregation persist and mean that black households are often located far from employment opportunities (OECD, 2011h). Yet even in far wealthier OECD countries, it may be difficult for those from poorer neighbourhoods to commute to where the job opportunities are, owing to expensive and/or fragmented public transport networks (OECD, 2012g and OECD, 2012h).
Countries have large inter-regional differences in educational attainment. In 2012, one quarter of the OECD population had only a basic education (i.e. no more than lower-secondary educational attainment). Such people tend to be concentrated in particular places. For example, in most of the regions in Turkey, Portugal and Mexico, and in some regions in Australia and Spain, the proportion of the population with only a basic education was as high as 50% (Figure 1.32). Territorial disparities in the share of workers with tertiary education are also important. The United States, Spain, the Czech Republic and Turkey show the largest regional variation in tertiary educational attainments.
For countries with less regional dispersion, the greatest issue is often the concentration of the skilled labour force, particularly in and around the capital.
Regional factors strongly affect access to, and returns from, education and quality of learning. Even when taking into account the socio-economic backgrounds of students, the location of schools matters greatly in determining the quality of education. In the OECD area, 15-year-old students in urban schools outperform those in rural areas on the PISA test by more than 20 points on average, which is the equivalent of almost one year of education (OECD, 2013i). Moreover, the evidence suggests that the returns to education in urban areas are higher than they are in less dense places. This differential can be a major incentive for highly educated individuals to migrate to cities.
Regions with the lowest and highest percentage of workforce with only basic education, 2012 Note: Countries ranked by average share of population with only basic education.
Source: OECD (2013f), OECD Regions at a Glance, OECD Publishing, Paris.
Considerable disparities in education can be found within metropolitan regions as well. In the Chicago region, for instance, school districts record high school graduation rates that range from 57% in the city of Chicago to over 95% in suburban areas (OECD, 2012h). In Aix-Marseille, the share of the working-age population without a diploma ranges from 39% in neighbourhoods in northern Marseille to 14% in Aix-en-Provence (OECD, 2013l).
Access to education services in developing countries is increasingly becoming harder in urban areas. In fact, children from poor urban neighbourhoods are less likely to attend school than children from other urban areas and rural areas. A survey in Delhi in 2004-2005 shows a primary school attendance rate of 54.5% among children living in slums, compared to a 90% attendance rate for the city as a whole. Similarly, in Bangladesh, 18% of children attended secondary education in 2009, compared to 53% in urban areas as a whole, and 48% in rural areas (UNICEF 2012). In several African countries, enrolment improved in the non-slum urban areas in the late 1990s, but worsened in the urban slums.
Inequality in skills is very much pronounced at the metropolitan level. There are more skilled workers in urban areas than in non-urban areas, but larger cities also have greater skill and wage disparity within skill groups than do smaller cities (Baum-Snow, N. and R. Pavan 2012). Rising inequality also exists within skill groups and job polarisation. One factor in urban disparity is that skill upgrading provides greater returns in larger urban areas than smaller ones. For example, agglomeration economies for skilled and specialised work may increase income inequalities in large cities by favouring highly qualified workers (Belal and Partridge 2006). There are also substantial differences in the returns to skills in urban areas, related to local concentrations in different industries, and these too are strongly correlated with inequality. Skills inequality results from differences in education, from immigration, from the effect of industry concentrations, and from differences in the returns to skills depending on location (Glaeser et al., 2008). OECD research has also identified regional disparities in the intensity with which skills are utilised by employers; some regions and some cities become trapped in a “low skills equilibrium”, in which the returns to skill are depressed, because local demand for skills is low (Froy et al, 2011, OECD 2014c).
Significant regional disparities in health outcomes are in part the product of unequal access to health services. In 2010 in North America, the life expectancy at birth in Texas (USA) was around 75 years, which was 6 years lower than in Minnesota (USA) and the life expectancy in Chihuahua (Mexico) was only 68 (Figure 1.33) (OECD, 2013f). There are also large differences in the ageadjusted mortality rates within countries. In most countries, the richest regions tend to have a higher number of doctors and lower age-adjusted mortality rates.
Source: OECD (2013f), OECD Regions at a Glance, OECD Publishing, Paris.
In general, the unequal distribution of resources and health services mainly affects rural areas, but the urban advantage is increasingly challenged. In OECD countries, rural areas are more affected by population ageing, diseconomies of scale in the provision of healthcare services and problems in accessing healthcare facilities. However, the rural disadvantage in access to health services is more pronounced in developing countries. A World Bank report found that urban per capita consumption exceeded rural per capita consumption by more than 40% in 72 developing countries (World Bank, 2009b). Limited access to paediatric healthcare in rural areas affects the poorest rural families. The risk of mortality is lower for children from wealthier families living in urban areas with better-educated mothers. Inequalities in under-age-5 mortality rates between urban and rural areas are particularly high in Cambodia and Albania. Providing equal access to healthcare and improving health conditions for vulnerable groups are also becoming a major concerns in urban areas, which confront higher income inequality. Although people living in cities have better access to health services than their counterparts in rural areas, the scale of inequality within urban areas can sometimes equal or even exceed that of rural areas (UNICEF, 2012).
Aedo, C. and I. Walker (2012), Skills for the 21st Century in Latin America and the Caribbean, Directions in Development Series, The World Bank, Washington D.C.
African Development Bank (AfDB) (2012), African Economic Outlook 2012: Promoting Youth Employment, OECD Publishing, Paris.
AfDB (2011), “The middle of the pyramid: Dynamics of the middle class in Africa”, AfDB Market Brief, April 20, AfDB Group, Tunis, Tunisia.
Althabe, F. et al. (2007), “Health inequality in Latin America”, The Lancet, Vol. 370, No. 9599, pp. 1599-1600.
American Community Survey (2010), US Bureau of the Census, www.census.gov/acs/www/.