24499 ,TONAL 8,< ,A WOL BANK v | May 2002 No.2 A regular series of notes highlighting recent lessons emerging from the operational and analytical program of the World Bank's Latin America and Caribbean Region OOC0UL rEKCLU$UH OM UMESAS UNMUOUA Judy L. Baker In many developing countries the characteristics of the In Uruguay, the process of social exclusion is a rela- urban poor go beyond the traditional definition of pov- tively new phenomenon and is particularly visible erty. While the latter generally refers to the lack of ac- given that the country has a relatively low level of cess to material resources, there are other factors which poverty, negligible indigent poverty, and has tradition- are linked to poverty. In the case of Uruguay one of ally been characterized as a homogeneous society with these factors is an increase in the perception among the the most equal distribution of income in Latin poor of exclusion from the rest of society. Although America. There is only one small ethnic group (Afro- this group only represents a small proportion of Uruguayans) representing about 5 percent of the popu- Uruguay's total population, the phenomenon is of rel- lation, and generous welfare policies over the past de- evance to policy makers given demographic factors cades have ensured that most Uruguayans receive ac- (the poor tend to have more children), and the cyclical cess to basic health, education and other social ser- nature of poverty. To assess the different aspects of so- vices. Historically, city neighborhoods were quite het- cial exclusion both quantitative and qualitative meth- erogeneous, with households of different income lev- ods were used. els living side by side and sharing the same public space. This integration provided a social cohesion be- [ExcWsison: Defining the 'lNlew Poor' tween individuals of different cultural and socio-eco- ___ f i ,nomic backgrounds. It also provided wider social net- The concept of social exclusion is defined in the litera- works (for job search, etc.), the presence of adult role ture as a multidimensional process which weakens the models for youth coming from more disadvantaged links between individuals and the rest of society (ILO, families (those without stable employment or educa- 1996). These links can take on an economic, political, tional attainment), and better opportunities for social socio-cultural, and a spatial perspective. The more di- mobility. l mensions by which a person is excluded, the more vul- l nerable this person becomes. The emergence of a 'new poor' resulting from job losses during the recessions in the 1990s and other 'TOODdI ncnnU tss5stance to Uruguay The World Bank's support for the reform programme initiated by the current administration has been focused on several key areas, adapting to the policy and institutional stance of the government. The Bank has supported the deregulation of the transport sector, competition in generation of electricity, and improvements to the transparency and regulation of the financial sector, as well as creation of the private pension system. These efforts are now being extended to the water and sanitation sector building on earlier support. In terms of investment, financing has been directed to modernizing the transport system, particularly to meet Mercosur standards and the needs of the emerging commercial forestry industry; diversifying crops, and upgrading technology and environmental considerations in the agriculture and irrigation sectors; introducing new management methods in public hospitals, and importantly in expanding the capacity and quality of the primary and pre- school system. structural changes in the labor market, has likely con- Spa. t Dimensions of Exdusion tributed to an increasing segmentation in Uruguayan A key aspect of the social exclusion process in the society. Many of these individuals have left their resi- Uruguayan context appears to be a geographic polar- dences in middle class neighborhoods in the center of ization among the population, or more specifically, the Montevideo where rents and utility bills are high, and clustering of the poor in marginal areas. Analysis fo- have moved into one of the growing marginal neigh- cused on understanding the spatial polarization process borhoods surrounding Montevideo and other urban ar- was carried out using household survey data from the eas. They have also lost their access to the extensive Continuous Household Survey for the period 1989- social benefits linked with formal employment. The 1996. The income data are decomposed into "within" marginalization may be made worse by continuing mi- and "between" area components based on neighbor- gration from the poorer rural areas of Uruguay to the hood clusters using the Theil Index (Table 1). Polariza- major centers, with these people also settling in peri- tion is occurring if the between area variance in mean urban locations. Together, these movements appear to per capita income is increasing, or the within area vari- have contributed to a fragmentation within Uruguayan ance is decreasing society between the poor and non-poor as these neigh- borhoods are often geographically isolated, have higher The data show that indeed there have been changes in levels of crime and violence, lower levels of services, the spatial distribution of the urban population along little public space, and generate an area stigma which income lines. For example, the poorest four neighbor- affects access to labor markets and social networks. hood areas (see Map 1) all show signs of polarization. Table 1: Spatial Changes in Montevideo by Neighborhood Clusters Census Section, Percentage of Theil Decompositions (neighborhood), Iinter-census Population in Change in Change in Monitevideo, by Mean Icoe Population Irregular Within Area Between Area M veo,b M Growth (1986- Settlements Income Income income(89-96) 1996) (1996 Census) Inequality, Inequalty, . 1989-1991 1989-1992 Lowest-income 13 65.0 21.0 45.0 - __ + 17 67.6 6.6 13.3 + 16 70.8 13.9 15.9 11 72.0 6.8 18.7 - + 25 76.3 9.7 16.7 0 + 2 83.9 -30.15 0.0 0 0 9 84.8 17.7 16.3 - 21 85.5 6.5 12.4 + + Mid-income 20 93.3 -2.1 3.3 0 22 98.0 -2.9 2.2 - 19 102.8 -9.7 0.0 - 3 113.3 -17.2 0.0 + 12 116.6 -5.3 0.0 + 14 125.6 -5.5 1.7 8 129.0 -10.8 0.0 10 143.3 2.1 3.1 + + Highest-income 5 146.6 -10.4 0.0 + + 15 147.7 0.0 X + 7 148.8 -10.7 0.0 + + 23 150.3 -7.2 0.0 - 24 152.3 -0.1 0.0 + 0 4 166.5 -14.0 0.0 + + 6 188.1 -11.0 0 0 + 18 193.8 -1.7 0 + + 'A negative change in the within area Theil decomposition indicates a decrease in inequality, (or polarization by income group) as demonstrated by the shading, a positive change indicates an increase in income inequality within area. A zero refers to no change 2Apositive change in the between area Theil decomposition indicates an increase in inequality. (or polarization by income group) as demonstrated by the shading, a negative change indicates a decrease in income inequality within area. A zero refers to no change. 2 Map 1 - Montevideo: Income Distribution marginal areas than in many other residential areas which impacts on the population both in the short term, and on longer term human capital development. Social networks such as public organizations, churches, and non-governmental organizations can provide an impor- tant link to society for individuals and groups, and serve as a coping mechanism against vulnerability. These networks do exist to some extent in the marginal X Nfid E 3 -income neighborhoods, but participation tends low, again put- ting vulnerable groups at a disadvantage. Highest-income Policy Conclusions Data: 1996 The findings of this research point to several policy The results of the analysis also show that polarization conclusions related to urban poverty and development is not symmetric - poor individuals are clustering in Uruguay and potentially the rest of Latin America. more, but the higher income areas still have a higher First, policies which prioritize improvements in access income variance and more inequality. to quality basic services, particularly education, health, transportation, social assistance, more flexible land use This inter-neighborhood income inequality has con- policies, as well as public information for those in mar- tributed to a fragmentation between the poor and non- ginal areas could help to provide an important link to poor which is not captured in the aggregate data on jobs and human capital development, and reduce some income distribution. The clustering of the poor in of the facets of social exclusion. New community marginal areas can also result in a 'neighborhood ef- based programs designed and implemented by neigh- fect' which may impact negatively on individual be- borhood members based on specific needs, or expan- havior and outcomes, such as children's life chances. sion of existing community programs, could also help to reverse the marginalization process. Characteristics of social exclusion Second are policies related to housing. Policies that Beyond the spatial dimension of social exclusion, foster the maintenance of traditional integrated neigh- several other characteristics were analyzed based on a borhoods, or minimize the concentration of the poor in qualitative survey carried out in 6 marginal neighbor- marginal neighborhoods, may mitigate the exclusion hoods. The survey used focus groups, open-ended in- process. Such policies may not be politically popular terviews with key informants, and direct and partici- among some, but would be deserving of further inves- patory observations. The focus groups were carried tigation. out with specific groups identified as vulnerable such as unemployed males aged 40-50, single female Finally, the aggregate numbers on inequality appear to headed households, and inactive teens. mask what is happening at a more disaggregated level such as the neighborhood. While income inequality The characteristics of exclusion are linked to access- for the country as a whole only shifted slightly during ing labor markets, basic services and social networks. the 1989-1996 period, the within and between area Those living in marginal neighborhoods, particularly changes were significant. Analysis of trends in income the vulnerable groups identified in this study, face distribution in other countries may benefit from similar constraints beyond those in the rest of society in each disaggregation as part of standard research on poverty of these areas. These constraints both contribute to, and income inequality. and are a result of the exclusion process. In access- ing labor markets the constraints include a lack of skills, lack of social networks, and stigma attached with living in poor marginal neighborhoods. Many work in the informal sector where salaries are low, and there are few if any benefits. The quality and quantity of basic services such as health, education, nutrition, sanitation, and security is lower for those in __3 3 For Further Reference Uruguay at a glance Uruguay Region 2000 Alonso, Daniel, Zaffaroni, Cecilia, Mieres, Pablo, 1998. Population, mid-year (millions) 3.3 516 "Encuentros y Desencuentros, Familias Pobres y GNI per capita (Atlas method, US$) 6,000 3,680 "Encuenros y Dsencuenros, Failias Pbres Y Average annual growth, 1994-00 Politicas Sociales en el Uruguay" , CLAEH, Population (%) 0.7 1.6 Montevideo Uruguay Labor force (%) 1.2 2.3 Montevideo, Uruguay Most recent estimate (latest year avail., 1994-00) Poverty (% of pop'n below national poverty line) 17 35 Ellen Ingrid Gould, and Margery Austin Turner. 1997. "Does Urban population (% of total population) 91 75 Life expectancy at birth (years) 74 70 Neighborhood Matter? Assessing Recent Evidence." Infant mortality (per 1,000 live births) 15 30 Housing Policy Debate, 8 (4) 833 66. Child malnutrition (% of children under 5) .. 9 Housing Policy Debate, 8 (4) 833 66. Access to an improved water source (% of pop'n) 98 85 Illiteracy (% of population age 15+) 2 12 Gross primary enrollment (% of school-age pop'n) 109 113 Jencks, Christopher and Susan Mayer. 1990. "The Social Consequences of Growing up in a Poor Neighborhood." 1980 1990 1999 2000(est) GDP (US$ billions) 10.1 9.3 20.7 19.7 In Inner-City Poverty in the United States, National (% of GDP) Agriculture 13.5 8.7 5.4 5.7 Academy of Sciences, Washington, D.C. Industry 33.7 32.9 26.3 26.0 Manufacturing 25.9 26.5 16.0 16.1 Services 52.8 58.4 68.3 68.3 Katzman, Ruben. 1996. Marginalidad e Integraci6n Social en Uruguay. CEPAL, Oficina Montevideo, Uruguay. 1980-90 1990-99 1999 2000 GDP (av. ann. growth) 0.5 3.4 -2.8 -1.3 GDP per capita -0.1 2.6 -3.6 -2.0 World Bank, 2001. Uruguay: Maintaining Social Equity in View more data at http://www.worldbank.org/data/ a Changing Economy. Economic Report No. 2162. About the Author Learn more about our work The author is a Senior Economist with the Latin America and Caribbean Region at the World Bank. This work was prepared in the context of a World Bank study entitled, I "Maintaining Social Equity in a Changing Economy" (World Bank Report No. 21262). The field work on social http://wwwwor1dbank.org/1ac exclusion was carried out by a team led by Patricia Reynoso and Claudia Romano. The quantitative work on polarization was carried out collaboratively with Luisa ..and on Poverty Corrado and Melvyn Weeks. Edmundo Murragarra. carried out work on vulnerability, and helpful comments were re- http://www.worldbank.org/poverty/ ceived on the various stages of the paper by Carine Clert, Shelton Davis, Norman Hicks and Quentin Wodon. For comments, please contact ibaker2(P_worldbank.org. To Subscribe to "en breve" please send an "en breve" might--also be of interest to: email to "en_breve@worldbank.org"1 or write to: Name. Editor, En breve Institution_- MSN 16-604 Address* The World Bank 1818 H Street NW Washington D.C. E-mail: 4