URBAN POVERTY IN ULAANBAATAR Understanding the Dimensions and Addressing the Challenges CONTENTS 1.0 ACKNOWLEDGEMENTS 07 6.0 INCOME POVERTY IN ULAANBAATAR 20-23 6.1 Levels and Trends 20 2.0 EXECUTIVE SUMMARY 08-13 6.2 Spatial Distribution of Income Poverty 20 2.1 Introduction 08 2.2 Background and Context 08 7.0 KEY CHARACTERISTICS OF URBAN POOR IN ULAANBAATAR 24-41 2.3 Income Poverty in Ulaanbaatar 10 7.1 Typology of Ger Areas and Relationship to Income Poverty 24 2.4 Multidimensional Poverty 11 7.2 Trends of Growing Unemployment 27 2.5 Vulnerable Groups 12 7.3 Mechanisms that Perpetuate Unemployment 28 2.6 Conclusion 13 8.0 MULTIDIMENSIONAL POVERTY 42-49 3.0 INTRODUCTION: BACKGROUND AND CONTEXT 14-15 IN ULAANBAATAR 8.1 Overview of the Methodology 42 3.1 Poverty Decline in Mongolia: Strong Start but Faltering Progress 14 8.2 Level and Distribution of Multidimensional Poverty in Ulaanbaatar 45 3.2 Uniquely Urban Nature of Poverty in Mongolia 14 8.3 Understanding the Dimensions of Deprivation 46 4.0 AIM OF THE REPORT AND METHODOLOGY 16-17 9.0 THE IMPORTANCE OF NON-MONETARY DIMENSIONS 50-67 4.1 Aim of the Report 16 16 9.1 Urban Poor’s Access to Municipal Services 52 4.2 Methodology 9.2 Urban Poor are Underserviced in terms of Water Delivery 53 5.0 ULAANBAATAR: UNPRECEDENTED GROWTH OF A PRIMATE CITY 18-19 9.3 Burden of Lack of Access to Sanitation is Disproportionate for Urban Poor 55 9.4 Lack of Effective Solid Waste Management is an Important Dimension of 56 5.1 Rapid Urbanization has Coincided with Urban Sprawl 18 Multidimensional Deprivation for Urban Poor and Non-Poor LIST OF TABLES AND FIGURES 9.5 Urban Poor Bear Disproportionate Burden of 57 Dysfunctional Health Care System TABLES Table 1: Urban-Rural national poverty trends: 2010 – 14 9.6 Lower Internet Access among Urban Poor is Likely to 60 Table 2. Regional poverty trends: 2010 - 14 Exclude Them from UB’s e-Governance Initiatives Table 3. District Level poverty Headcount Table 4. Poverty by Household Head's Characteristics, Ulaanbaatar 9.7 Neighborhood Relations in Ger Neighborhoods 61 Table 5. Income by Locality Type Table 6. Labor Force Participation Rate in Ulaanbaatar: 2010-14 9.8 Salience of Alcoholism in Ulaanbaatar’s Neighborhoods 62 Table 7. Distribution of Ulaanbaatar's Population by Household Head's Gender and Linkages with Poverty Table 8. Characteristics of Urban Poor by Quintile (percent) Table 9. Deprivation Line Table 10. Key Hotspots of multidimensional poverty 10.0 MARGINALIZED GROUPS 68-77 Table 11. Poverty Level & Dimensional Contribution to Overall Poverty by Ger & Non-Ger Areas Table 12. Absolute dimensional contribution to multidimensional poverty Table 13. Transitional Probabilities comparing Income versus Multidimensional Poor Individuals based on their 10.1 Rural-Urban Migrants: Not Poor but Vulnerable 68 Income Status – Ulaanbaatar, Mongolia: 2012 Table 14: Access to water by quintile 10.2 Challenges Related to Obtaining a Plot in the Fringe 71 Table 15. Access to Improved Sanitation Facilities Table 16. Frequency of garbage collection by Income Quintiles and Location in the City Table 17. Distribution of migrants and non-migrants across residential locations 11.0 COMMUNITY LIVING IN THE DALAN DAVKHAR CEMETERY 78-79 Table 18. Determinants of (Log) Income in Ulaanbaatar FIGURES 12.0 OTHER VULNERABLE GROUPS 80-81 Figure 1. Distribution of the poor by location: 2010-14 Figure 2. Average Poverty Headcount per Khoroo Figure 3. Relative Contribution of Each Dimension to Multidimensional Poverty 12.1 Disabled 80 Figure 4. Poverty headcount and centralized water access Figure 5: Ulaanchuluut landfill site – in the territory of 26th khoroo, Songinokhairkhan district 12.2 Street Children 80 Figure 6: Administrative procedures for obtaining urban residency status 12.3 Apartment Entrance Hall Watchmen 80 MAPS Map 1. Poverty Headcount Map 2. Kindergartens within a 30 minute walking time: 2014 13.0 CONCLUSION 82-85 Map 3. School Capacity by Khoroo: 2014 Map 4. Kindergartens within a 40 minute walking time: 2014 Map 5. Multidimensional Poverty Clustering Map 13.1 Extend Infrastructure in Line with Urban Population Growth 83 Map 6. Clustering of Income and Multi-deprivation BI-LISA 13.2 Migrant Registration Reform 83 Map 7. Ulaanbaatar Poverty Headcount and School Accessibility Map 8. Access to Sanitation Services 13.3 Develop Mechanisms for Improving Regulatory Oversight and 84 Map 9. Access to Garbage Pickup Map 10. Internet Access in Ulaanbaatar Enforcement Around Corruption Map 11. Perceived Alcoholism within communities Map 12: Dalan Davkhar cemetery – surrounded by khoroos 7, 10, 11 and 12, Chingeltei district 14.0 REFERENCES 86 15.0 ANNEX 87-97 7 1.0 ACKNOWLEDGEMENTS © 2017 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington, DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, The report was authored by Gayatri Singh (Urban Development Specialist, GPSURR) under the AAA on Making denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Cities Inclusive (TTL Judy Baker), with contributions from Gilvan Guedes on multi-dimensional poverty analysis Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World (Professor, Federal University of Minas Gerais) Joseph Mulhausen (Consultant), Battuya Dash (Consultant) and Gauri Bank, all of which are specifically reserved. Gadgil (Consultant). Statistics based on National Household Survey 2014 data was provided by Obert Pimhidzai Rights and Permissions (Economist, GPVDR). Background information on social networks in ger areas was also provided by Sabine Willach (Consultant). The report also benefitted from the overall guidance of Judy Baker, Lead Economist as well as from This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) http://creativecommons.org/ inputs during the initial stages from Meskerem Brhane (Practice Manager, GSU19). The report design and layout licenses/by/3.0/igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: was done by Seetha Raghupathy with support from Lu Pang. Arailym Murat has been instrumental in facilitating the Attribution— Please cite the work as follows: Singh, Gayatri. 2017. Urban Poverty In Ulaanbaatar: Understanding the Dimensions translation and editing of the Mongolian version of this report. and Addressing the Challenges. World Bank. Washington DC. License: Creative Commons Attribution CC BY 3.0 IGO Translations— If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank The team is extremely grateful for the active collaboration from the National Statistical Office (NSO) and the shall not be liable for any content or error in this translation. Master Planning Department (MPD) who provided valuable data inputs. Gayatri Singh also led the survey as Adaptations— If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility well as qualitative data collection. The Mongolia Marketing Consulting Group carried out the household survey of the author or authors of the adaptation and are not endorsed by The World Bank. and focus group discussions.  Ethnographic research on vulnerable groups was carried out by Bulgan Enkhbat Third-party content—The World Bank does not necessarily own each component of the content contained within the work. (Consultant). Additional interviews with service providers as well as survey supervision was carried out by Battuya The World Bank therefore does not warrant that the use of any third-party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with Dash, who also provided invaluable assistance in analyzing FGDs and ethnographic interview recordings. you. If you wish to re-use a component of the work, it is your responsibility to determine whether permission is needed for that re-use and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. The work was carried out under the guidance of Abhas Jha (Practice Manager), Coralie Gevers (former Country All queries on rights and licenses should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Manager for Mongolia), Mark Lundell (former Sector Manager for Sustainable Development for Mongolia and Document design and layout: Seetha Raghupathy China) and James Anderson (Country Manager, Mongolia). Cover design: Seetha Raghupathy Cover photo: © Gayatri Singh/ World Bank The in-depth insights regarding the mechanisms perpetuating urban poverty, which are critical for understanding Pictures. Further permission required for reuse. and addressing the challenges facing Ulaanbaatar, would not have been possible without the information that was p.9 © Gayatri Singh/ World Bank p. 65 © Gayatri Singh/ World Bank generously shared by residents of the city’s ger areas. The team is deeply indebted to all research participants who p.17 © Chandan Desukar/ World Bank p. 67 © Gayatri Singh/ World Bank p. 19 © Ying Yu/ World Bank p. 72 © Bulgan Enkhbat/ World Bank have made this report possible. p. 21 © Chandan Desukar/ World Bank p. 74 © Bulgan Enkhbat/ World Bank p. 23 © Gayatri Singh/ World Bank p. 76 (Box 12) © Bulgan Enkhbat/ World Bank p. 31 © Khasar Sandag/ World Bank p. 76 © Gayatri Singh/ World Bank p.33 © Gayatri Singh/ World Bank p. 77 © Gayatri Singh/ World Bank p. 36 © Khasar Sandag/ World Bank p. 81 © Gayatri Singh/ World Bank p. 44 © Gayatri Singh/ World Bank p. 82 © Khasar Sandag / World Bank p. 64 © Gayatri Singh/ World Bank p. 85 © Gayatri Singh/ World Bank 9 Urban Poverty in Ulaanbaatar: Executive Summary 2.0 EXECUTIVE SUMMARY 2.1 INTRODUCTION The report takes a mixed methods approach, combining quantitative, qualitative and spatial analyses. Availability Mongolia’s economic growth from 2010 has been of rich spatial data and the use of spatial techniques are unprecedented with GDP peaking at 17.5 percent in 2012 used to create a better understanding of the locational when Mongolia was the fastest growing economy globally. distribution of urban poor and the nature of poor places. This unparalleled growth of the economy was accompanied by an equally impressive decline in poverty by 17 percentage points between 2010 to 2014. However, much of this decline 2.2 BACKGROUND AND was during 2010-2012. From 2012-2014 poverty declined by less than 6 percentage points, partly due to slowing growth CONTEXT and declining foreign direct investment. World Bank (2016) Ulaanbaatar generates 65 percent of the country’s GDP1, estimates raise concerns that poverty reduction is likely to 85 percent of power, and 50 percent of investments. come to a standstill due to the negative welfare impact Ulaanbaatar’s average real GDP growth was around 13.3 of the economic slowdown in 2015, which is expected to percent between 2008 and 2012, compared to the national continue through 2016. During 2015-16, urban poverty was growth rate of 8.8 percent. The expansion of Ulaanbaatar projected to increase, and World Bank estimates suggest has been phenomenal, both in terms of population growth as that approximately 6.5 percent of the urban population may well as its urban extent. Ulaanbaatar’s population (housing have already fallen below the poverty line. 42 percent of the nation’s population) rose from about half a million in 2001 to about 1.2 million in 2011, and is projected to rise to 1.7 million by 2025. Much of this population increase The risks of an impending stall in poverty reduction are has been through rapid rural-urban migration in the last two particularly strong in urban areas that are home to almost decades. In 1989, 26.8 percent of Mongolia’s population two-thirds of Mongolia’s population and account for 55.6 lived in Ulaanbaatar; by 2006 that number had risen to 38.1 percent of the poor. Given the trends of rapid urbanization percent; and by the 2010 census, 45 percent of Mongolia’s and growing centrality of cities in Mongolia, this report delves population lived in the capital. deep to examine multiple dimensions of urban poverty and how social groups might differ in their experiences of urban poverty. While urbanization presents an enormous opportunity for promoting prosperity, in Ulaanbaatar, urbanization has coincided with low-density sprawl. Instead of being able to The objective of this report is: leverage the benefits of agglomeration and densification, ++ to carry out an in-depth analysis of the extent, nature Ulaanbaatar is struggling with basic service delivery, and spatial distribution of poverty in the capital city especially for urban poor populations. There are three of Ulaanbaatar, which holds the largest proportion of dynamics responsible for the low-density urban development urban population in the country and serves as the most of Ulaanbaatar’s urban form: important destination for rural-urban migrants, ++ identify sources of vulnerability for poor and excluded populations using a variety of data sources that go ++ The city’s soviet style planning legacy and constitutionally beyond income metrics and highlight the deeply mandated land ownership have led to large-sized multidimensional nature of urban poverty in Ulaanbaatar. urban blocks in the city center with low-rise structures. 1 Asia Foundation 2015 estimate: https://asiafoundation.org/resources/pdfs/EconomicDevelopmentMongolia.pdf 10 11 Urban Poverty in Ulaanbaatar: Executive Summary Mongolian citizens registered who live in Ulaanbaatar household heads, or those dependent on social transfers In Ulaanbaatar, the labor force participation rate steadily employed small-scale traders and vendors also identified are entitled to a free plot of land of up to 700m2 in the and private sector wages. increased for both men and women between 2010 and corruption when attempting to obtain permits to carry fringes and between 400-550m2 in the city center. 2014, However, between 2012 and 2014 the unemployment out their businesses in authorized market spaces. ++ The city has not adequately planned for growth in TYPOLOGY OF GER AREAS AND RELATIONSHIP rate for women almost doubled, signaling that there are ++ Education: Among poor households 68 percent are TO INCOME POVERTY not sufficient jobs to absorb the increasing number of its population, with considerable gaps in availability headed by someone with upper secondary education of affordable housing stock closer to the city center The city classifies Ger areas by three zones, central, mid- women entering the labor force. Unemployment for men in or lower. Only 8 percent are headed by someone with leading to considerable urban sprawl. tier (middle) and fringe based on their location, connectivity Ulaanbaatar also increased slightly. While unemployment higher education. Access to quality education is a ++ The rapid pace of rural-urban migration since the early to engineering networks and housing types. Within the declined overall from 2010 to 2014, some of the gains made key input into the development of human capability 2000s has interacted with the first two conditions and official classification, Central ger areas, where connection between 2010 and 2012 are being reversed. Mechanisms and determines the extent of inter-generational spurred outward growth of the city in its peripheries. for centralized engineering networks is feasible, will be perpetuating unemployment include: disadvantage transferred. An examination of the spatial redeveloped with high-rise and mid-rise buildings. Mid- distribution of schools in Ulaanbaatar highlights the tier ger areas are planned for redevelopment with low-rise inequality of physical access to schools for children from ++ Age and gender bias: Both men and women over The outcome is a patchwork of plots occupied with detached and mid-rise buildings and will be connected to partial urban poor households living in ger areas. 40 report that strict age requirements constrain their single-unit homes leading to low residential density. Such engineering networks. Redevelopment for Fringe ger opportunities for employment. Men report this bias residential neighborhoods have come to be known as ‘ger areas is planned in phases via land readjustment schemes more in manual labor jobs and link it to perceptions areas’, which are a unique feature of Ulaanbaatar’s urban with onsite networks. “Non-ger areas” typically contain of preference for Chinese workers. For women, 2.4 MULTIDIMENSIONAL landscape. Ger areas, particularly at the periphery, have apartment buildings with some single-home plots and ger become home to many urban poor, whose income poverty structures mixed in. the age bias has a gendered dimension as well. POVERTY Prospective employers demonstrate negative attitudes is worsened by the severe lack of basic services and Conditions in ger areas have meant that the urban poor towards women with children, who are perceived as infrastructure provision in these neighborhoods. not only suffer from income poverty but are also adversely unemployable due to their child rearing roles. There is a clear spatial dimension to the distribution of affected by multiple deprivations due to lack of basic urban poor in Ulaanbaatar across its districts and khoroos. ++ Constraints posed by childcare: Analysis of khoroo services and infrastructure. Focusing solely on income More than half of the urban poor report living in gers (57.5 household registry data shows that public kindergarten 2.3 INCOME POVERTY IN percent), compared to less than 1 percent who report living facilities are able to accommodate only 30 percent of poverty rather than multidimensional poverty provides an incomplete measure of the living conditions faced by the children eligible for enrollment. Lack of affordable child ULAANBAATAR in apartments. However, only 38 percent of people living in urban poor. Identifying additional deprivations is relevant ger areas are poor and poor households seem to be evenly care options are an important constraint on women’s Ulaanbaatar is home to approximately 33 percent of from a policy perspective to enable the design of meaningful spread out across the three classifications of ger areas. access to the job market. In the absence of strong family Mongolia’s poor population, with approximately 16 percent social and economic policies. Poverty is not concentrated solely in the fringe ger areas, support systems, particularly in migrant households, of the city’s households below the poverty line. Given the women must stay at home to take care of children. though the poverty headcount is positively correlated with greater shares of population living in Ulaanbaatar, the city ++ Labor exploitation: Between 2010 and 2014, the distance from the city center. In order to understand multidimensional poverty in has a higher proportion of poor than any other sub-national construction sector contributed to more than half Ulaanbaatar, the World Bank Survey 2014 data was analyzed administrative location type, highlighting its centrality in of labor income growth for the poorest 20 percent. using the well-established Alkire Foster methodology. interventions designed to target urban poverty. Levels However, exploitation of informal workers in the form TRENDS OF GROWING UNEMPLOYMENT Results of the analysis show that overall 23.4 percent of of inequality in Ulaanbaatar have shown no reduction, as of non-payment of full wage, fraudulent contracts and the population of Ulaanbaatar are multidimensional poor. demonstrated by the Gini coefficient that has remained Despite the unprecedented growth of the past decade, the unfair dismissals is prevalent particularly within the Multidimensional poverty is consistently higher and more steadfast at 0.33 between 2010-2014. expansion of non-labor intensive industries meant that job construction sector. These practices pose significant intense among ger residents. 39.5 percent of ger residents creation in Mongolia only increased by 11 percent. According hurdles in accessing jobs and discourage job seekers are poor as compared to only 17.8 percent of non-ger to the Institute for National Strategy (INS) of Mongolia, the from staying in the job market. An analysis of National Statistical Office’s Household Socio- residents, meaning a person living in ger areas is twice as number of unemployed seeking jobs has increased from ++ Corruption & Cronyism: Corruption and cronyism are likely to be multidimensional poor as compared to someone Economic Surveys (HSES) data from 2010, 2012 and 2014 39,000 in 2012 to 55,000 in 2014, while the number of new jobs seen as barriers to accessing employment opportunities, living in non-ger areas. for Ulaanbaatar highlights that the incidence of poverty being created has halved. Youth unemployment continues to and job seekers must have influential connections or in Ulaanbaatar is highest among households living in Ger be a big concern, with 69 percent of unemployed falling in pay bribes to access jobs or even be short listed. Self- areas, with unemployed household heads, less educated the age group of 15-34 years. 12 13 Urban Poverty in Ulaanbaatar: Executive Summary Deprivations in the dimensions of ‘Assets & Money’, ‘Water & DISABLED are often single mothers with children under the age of Sanitation’ and ‘Solid Waste’, are the largest contributers to five who have migrated from rural areas and do not have a There are disability allowances within the social welfare multidimensional poverty with notable differences between permanent city address. system that aim to reduce the monetary vulnerability of ger and non-ger areas. In both ger and non-ger areas, disabled populations within Ulaanbaatar. Nevertheless, alcoholism is also seen as not only an important indicator there is little provision for dealing with the everyday affecting the quality of urban livability but also makes a large contribution to multidimensional poverty. challenges faced by those who are physically disabled. 2.6 CONCLUSION Accessible physical infrastructure is limited, constraining the The analysis reveals important findings about Ulanbaataar’s mobility of the disabled and their ability to be self-sufficient. urban poor, namely that they are not exclusively migrants living in fringe ger areas. Rather, the urban poor are spread 2.5 VULNERABLE GROUPS throughout the three ger areas and are more likely to be Certain social groups face disadvantages and stigma based STREET CHILDREN non-migrants. Employment and education are certainly on their identity or unique social circumstances creating important factors in addressing poverty, though they are According to the Centre for Child Protection in the Police social exclusion and vulnerability. compounded by non-income deprivations including the lack Department of Ulaanbaatar, the number of children living of adequate infrastructure and service provision. on streets of UB has declined from 889 in 2011 to 690 in 2012 and to further 260 children in 2013. In 2014, Ulaanbaatar RURAL-URBAN MIGRANTS Metropolitan Police Department launched a campaign to The analysis highlights primary areas where government can Migrants are often stigmatized by local residents and seen identify homeless and street children in collaboration with initiate policy measures, namely employment, education, as the source of Ulaanbaatar’s high levels of poverty and the Child and Family Development Centre. They identified service delivery, and provisions for widespread alcoholism. urban sprawl, though these beliefs are not substantiated by and registed 74 street children and went on to provided The city government recognizes the importance of providing the quantitative and ethnographic evidence collected for them with health check-ups and attempted to reunite them education, and has already taken steps to address gaps this report. The World Bank 2014 survey finds that migrants with their families or place them in children’s centers in in both access and quality. Alcoholism remains largely on the whole are not urban poor in terms of income. A linear the city. NGOs have raised concerns that the problem of unaddressed, and Mongolia would do well to seek examples regression model for determinants of income poverty shows children’s vulnerability in Ulaanbaatar is shifting from streets from abroad in how to address this growing affliction. Finally, no significant association between migrant status and to children’s homes, which have limited capacity. Ulaanbaatar already has the appropriate policy tools and income when controlling for age, gender, education status, mandate over urban planning and land use management, employment status and residential location in Ulaanbaatar. land taxes, and zoning regulations, all of which can be used to APARTMENT WATCHMEN promote inclusive development through ensuring affordable Since the privatization of Soviet style apartment blocks and efficient service delivery. Better urban planning and land While migrants are not the urban poor in terms of income, started during 1990’s, Unions of Apartment Owners (UAO) management aimed at promoting density and with a view to 60 percent of migrant households are multidimensional poor have been set up and a new work position of “apartment increasing access to services will not only be important for as compared to 48 percent non-migrant households owing entrance hall guard” created. The roles and duties of this the efficiency and cost-effectiveness of service delivery but in large part to their concentration in ger areas lacking work position include cleaning the apportioned public area, is also critical to enhancing the inclusion of urban poor and adequate infrastructure and services. As highlighted in safeguarding the entrance hallways for 24/7, and other creating a livable capital city for all citizens. the previous section, limited water provision and sanitation chores ordered by the apartment council. Typically, the services are significant contributers to multidimensional guard lives under the staircase in the entrance ina space poverty. A majority of migrants moving to Ulaanbaatar live around 1,8х2m (3,6-4м2) in size, with a very low ceiling and in ger districts, though they are distributed across all ger the open side closed with cheap materials. It is a narrow locations rather than concentrated in fringe ger areas as is space with limited air circulation and no natural light that commonly assumed. gets cold in winter, is often noisy, and in many cases has no sanitation and toilet facilities. The residents of this space 15 Urban Poverty in Ulaanbaatar: Introduction: Background & Context 3.0 INTRODUCTION: BACKGROUND & CONTEXT 3.1 POVERTY DECLINE IN 3.2 UNIQUELY URBAN Table 1 Urban-Rural National Poverty Trends : 2010 - 2014 MONGOLIA: STRONG NATURE OF POVERTY IN POVERTY HEADCOUNT RATE (%) CHANGE DISTRIBUTION OF THE POOR START BUT FALTERING MONGOLIA 2010 2012 2014 2012-2014 2010 2012 2014 PROGRESS Mongolia 38.8 27.4 21.6 -5.8 100.0 100.0 100.0 Urban 33.2 23.3 18.8 -4.4 55.2 56.2 55.6 Mongolia’s economic growth from 2010 has been The risks of an impending stall in poverty reduction are Rural 49.0 35.5 26.4 -9.1 44.8 43.8 44.4 unprecedented with its GDP peaking at 17.5 percent in 2012 particularly strong in urban areas that are home to almost Source: World Bank authors’ calculations from HSES 2010, 2012, 2014 in World Bank (2016) when it was the fastest growing economy globally2. This two-thirds of Mongolia’s population and account for 55.6 unparalleled growth of the economy was accompanied by percent of the poor. This is unlike most other developing an equally impressive decline in poverty by 17 percentage countries where poverty is concentrated in rural areas. by severe winter weather), which has historically led to rural areas10. Since 1990, the living standard decreased points from 38.7 percent in 2010 to 21.6 percent in 2014. Poverty declined at a much slower pace in urban areas as significant losses of crops, livestock and rural livelihoods dramatically all over Mongolia, resulting in classic push-and However, much of this decline was during 2010-2012, and compared to rural areas during 2010-2014. Rural areas have and spurred distress migration to urban areas. According pull factors that led to high levels of rural-urban migration, from 2012-2014 poverty declined by less than 6 percentage contributed disproportionately more to the overall poverty to the Government of Mongolia estimates, at the end of but without a strategy to manage rapid urbanization and points. reduction in the same period while constituting a smaller 2015, 50 districts in 16 provinces were affected by dzud, and plan for it. While a temporary de-urbanization was evident population base. World Bank estimates show that poverty in another 120 districts in 20 provinces were facing near-dzud between 1990 and 1995, during which the rural population rural areas declined by 13.5 percentage points during 2010- conditions (ReliefWeb/UN OCHA) 7. grew significantly, migration to urban areas increased after A reason for this slowdown is the sharp downward turn in 12 and 9.1 percentage points during 2012-14, while there were 1995 and even more after 200011. Economic hardship in rural Mongolia’s growth since 2012, with the growth slowing much smaller declines of 9.9 and 4.4 percentage points in areas due to dzud has been a primary push factor for distress down to a single-digit due to plummeting foreign direct urban areas in the corresponding periods respectively (Table Conditions are expected to impact close to a million migration. But another key reason is social aspiration of investment and chronic delays with mining projects3. As 1). Slowdown in the pace of urban poverty has been further people in the affected rural districts (ReliefWeb/IFRC)8 rural families for intergenerational mobility, particularly with much as 11 percent of the country’s population is within 10 accompanied by a lack of decline in the level of urban and may signal a reversal in Mongolia’s poverty reduction regard to better access to education and health services in percent of the poverty line, making them highly vulnerable inequality. The Gini coefficient for urban areas has remained achievements. the case of migrants to Ulaanbaatar, which has become the to these economic shocks4. World Bank (2016) estimates at 0.33 since 2010 whereas inequality declined in rural areas main migrant12 destination and a rapidly expanding primate raise concerns that poverty reduction is likely to come from 0.32 in 2010 to 0.28 in 2014. Levels of inequality in city. to a standstill due to the negative welfare impact of the urban areas are not expected to decline in the short term. During the socialist period, migration was centrally economic slowdown in 2015, which is expected to continue controlled and urbanization for the recruitment of work through 2016. The challenge of achieving long term inclusive forces in factories was promoted. By 1990, approximately 55 Given the uniquely urban nature of poverty in Mongolia and growth of Mongolia is a very real one, especially with the The greatest impact of Mongolia’s economic slowdown to 60 percent of all Mongolian citizens were living in urban the likelihood of an increase in urban poor in the short term, prevailing high levels of inflation5 and dipping growth rates. is being felt in urban areas. In the 2015-16 period, poverty areas9. For the remaining populations, the living standard in it is especially pertinent to better understand the nature of was projected to have increased in urban areas, and the soum -centers was, compared to today, reasonably high challenges faced by urban poor, the mechanisms that are approximately 6.5 percent of the population may have with schools, boarding schools, theatres, public transport, likely to create durable forms of inequality and find ways to already fallen below the poverty line6. Mongolia is currently newspapers and electricity made readily available. This tackle these dynamics. experiencing conditions of dzud (summer drought followed contributed to a relatively high standard of living even in 7 http://reliefweb.int/report/world/asia-pacific-region-el-ni-o-snapshot-january-2016 8 http://reliefweb.int/report/mongolia/mongolia-extreme-winter-condition-emergency-plan-action-epoa-n-mdrmn005 2 The declining trend started after high rates of growth in 2011 (at 17.5 percent), dropping to 12.3 percent in 2012 and 11.7 percent in 2013. World 9 Gilberg, Rolf & Svantesson, Jan-Olof (1999) The Mongols, Their Land and History. In: Ole Bruun & Ole Odgaard (ed.) Mongolia in Transition Old Bank 2016a Mongolia Poverty Report Patterns, New Challenges. Richmond: Curzon 3 ADB East Asia Outlook 2014: https://www.adb.org/sites/default/files/publication/31241/ado-2014_1.pdf 10 Ibid. p. 180. 4 World Bank 2016a. Poverty reduction and shared prosperity in Mongolia: Well begun, but can it continue? Washington DC. 11 Taraschewski, Thomas. (2008). Stadtentwicklung in Ulaanbaatar im Zeitalter fragmentierender Entwicklung. Aktuelle Migrations- und sozioökonomische 5 Differenzierungsprozesse unter besonderer Berücksichtigung der Jurtenviertel. Berlin: Wissenschaftlicher Verlag. Consumer price inflation soared to 12.9% year on year in July 2014 fueled by last year’s highly expansionary fiscal and monetary policies. 6 12 World Bank, 2016 Migrant is a household head who was not born in Ulaanbaatar 17 Urban Poverty in Ulaanbaatar: Aim of the Report and Methodology 4.0 AIM OF THE REPORT AND METHODOLOGY the locational distribution of urban poor and the nature of on the perceived quality, accessibility and outcomes ++ Spatial data and analytics: Extensive spatial data 4.1 AIM OF THE REPORT poor places. Several sources of data utilized in this report of health and education as key locally-provided social was secured from the Ulaanbaatar City Master Planning This report explores various dimensions of urban poverty (as described below) have also been used to produce the services. These focus group discussions were also carried Department (MPD) in July 2014, which forms the basis and how social groups might differ in their experience of analyses for the World Bank Ulaanbaatar Service Delivery out in 2014. Henceforth, World Bank FGDs 2014. The for majority of the spatial analyses examining access urban poverty. The main thrust of this report is to carry out Report (2016) given that the two reports are outcomes of analysis of focus group discussions allowed for a more to services. This includes GIS layers of the distribution an in-depth analysis of the nature and spatial distribution the same broader analytical and technical work. As such, nuanced understanding of how poverty is experienced of public transit nodes and networks, schools, clinics, of poverty in the capital city of Ulaanbaatar as well as the some of the analyses presented in this report overlap with and the mechanisms that serve to perpetuate poverty, streetlights, building footprints, residential classifications sources for vulnerability for poor and excluded populations those presented in the Service Delivery Report. However, any particularly when interpreting quantitative data. etc. In addition, the National Statistical Office (NSO) using a variety of data sources. These challenges are overlapping analyses are not always explicitly mentioned of Mongolia provided access to a poverty map that particularly central for Ulaanbaatar, where population, in the interest of the report’s flow and readability, given ++ Ethnographic Study: The report also utilizes insights estimated poverty at sub-district level (i.e. khoroo poverty and economic growth have become increasingly that the narrative of this report approaches the analyses from an ethnographic study commissioned under the level, or an urban ward). The poverty mapping was concentrated over the last decade. This report goes beyond presented from an urban poverty lens. project, which delves deeper into the understanding carried out using the decennial census and household solely money metric measurements and highlights the deeply of vulnerability in the urban setting of Ulaanbaatar. It survey data through the application of a robust small- multidimensional nature of urban poverty in Ulaanbaatar. captures the experiences of two vulnerable groups, area-estimation methodology 15. The findings allow for The primary data sources used for this study include the urban poverty comparison across districts and khoroos. namely, recent migrants living near a landfill site and following: Henceforth, World Bank Spatial Analysis 2014. ger area community that was established partially on a The multidimensional view taken by this report does not ++ Household Survey on Service Delivery in cemetery. The cemetery has subsequently been fenced diminish the importance of traditional poverty measurements Ulaanbaatar: a geo-referenced, citywide random off and exists alongside the community. Both groups that focus on economic deprivation and form the backbone sample of 3000 households using a two stage random face stigma from local residents in Ulaanbaatar and the of established poverty analysis. Rather, it augments the sampling design, stratified by ger and non-ger areas13. in-depth study allows us insights into their lived urban economic dimension by highlighting aspects of urban It collected information on socio-economic indicators, experiences and coping strategies. Henceforth, World livability, whose absence can exacerbate the impact of migration status, access to water, sanitation, solid waste Bank Ethnography 2014 income poverty on urban poor. More importantly, such collection, functionality of streetlights, access to health dimensions of urban livability (particularly access to and clinics and schools, social capital and neighborhood quality of municipal services, basic infrastructure, education, conditions. Data collection was carried out in 201414 health etc.) are central to supporting local governments in (henceforth, World Bank Survey 2014). creating inclusive cities. ++ Analysis of 2010, 2012 and 2014 Household Socio- Economic Survey Data collected by Mongolia’s 4.2 METHODOLOGY National Statistical Agency, with a focus on results for Ulaanbaatar (henceforth, HSES) The report takes a mixed methods approach, combining quantitative, qualitative and spatial analyses. Availability of rich spatial data and the use of spatial techniques ++ Focus group discussions: 18 focus group discussions further is further used to create a better understanding of with residents from different districts of Ulaanbaatar 13 The two stage sampling design first randomly identified specific khesegs to be targeted from which ten households were selected using simple random sampling. Ulaanbaatar city is divided into nine districts, which are subdivided into 152 khoroos. Khoroos are the lowest sub-district administrative units of the city Khoroos were divided into two strata- ger strata, if more than 50% percent of the households in that khoroo lived in gers, eslse the apartment strata, Khoroos were further subdivided roughly into smaller units of approximately 200 households each, resulting in a total of 1168 units that formed the primary sampling units (PSUs). 200 PSUs were randomly selected from the ‘ger starta’ and 100 from ‘apartment strata’. For the second stage sampling, a listing exercise was undertaken in 300 PSUs, of which ten were randomly selected from each PSU, resulting in a total of 3000 households. 15Small Area Estimation Maps: http://econ.worldbank.org/external/default/ 14 The results of this quantitative survey instrument will be used throughout this report to illustrate the analysis. main?theSitePK=477894&contentMDK=20292228&menuPK=545573&pagePK=64168182&piPK=64168060 19 Urban Poverty in Ulaanbaatar: Ulaanbaatar: Unprecedented 5.0 ULAANBAATAR: Growth of a Primate City UNPRECEDENTED GROWTH OF A PRIMATE CITY Ulaanbaatar generates 65 percent of the country’s GDP, why Ulaanbaatar has received migrants from all parts of the registered who live in Ulaanbaatar are entitled to a free built houses, with low residential density. Such residential 85 percent of power, and 50 percent of investments. country is due to the increased intensity of a series of severe plot of land of up to 700m2 in the fringes and between 400- neighborhoods have come to be known as ‘ger areas’, which Ulaanbaatar’s average real GDP growth was around 13.3 winter storms (called dzud) that have destroyed considerable 550m2 in the city center19. Two, the city has consistently are a unique feature of Ulaanbaatar’s urban landscape. percent between 2008 and 2012, compared to the national rural livestock and resulted in the large out migration of low- suffered from gaps in the availability of affordable housing As will be demonstrated in the subsequent sections, growth rate of 8.8 percent. Up to 99 percent of local output income, low-skilled rural families into Ulaanbaatar16. stock closer to the city center and has not planned for ger areas (especially in the peripheries of Ulaanbaatar) is comprised of construction, mining, transportation services, growth in its population. Three, the rapid pace of rural- have become home to several urban poor, whose income While migrants have contributed to Ulaanbaatar’s population retail, light manufacturing and processing. Construction urban migration since early 2000s has interacted with the poverty is worsened by the severe lack of basic services and growth, they should not be held responsible for the rise in and maintenance works, retail and wholesale industries in first two conditions and spurred an outward the growth infrastructure provision in these neighborhoods (World Bank urban poverty. As discussed in the section titled Vulnerable UB accounted for 75 percent of the total national output in of the city in its peripheries. The 700m2 plot provision has 2016b). Groups later in this report, migrants are not the income poor. 2011. The real estate sector has grown significantly, driven meant that incoming households have the incentive to find The calculations based on World Bank 2014 Survey data mainly by increasing demand (rapid in-migration, rising the closest possible plot to the city, resulting in newcomers find no significant association between migrant status and disposable income, and interest rate subsidies), as well as continuously being forced to settle in the fringes given the income when controlling for age, gender, education status, boosted supply resulting from direct government funding to lack of affordable renting options in the city. employment status and residential location in Ulaanbaatar construction companies, producers of construction materials (See Table 18). However, rural-urban migrants face The outcome is patchwork of plots divided by fences, and developers to support new construction, as part of the considerable discrimination stemming from their migration and occupied with detached single-unit homes ranging government’s 2012 Monetary Stimulus Programs. As these status and also face multiple dimensions of deprivation, as from traditional gers, to more free-standing single level sectors continue to drive growth and generate employment discussed in more detail in latter parts of this report. in UB, further diversification in the service sector, such as financial services, hotel management and restaurants, will expand employment opportunities for UB’s population (World Bank 2014). 5.1 RAPID URBANIZATION HAS COINCIDED WITH URBAN The expansion of Ulaanbaatar has been phenomenal, both SPRAWL in terms of population growth as well as its urban extent. While urbanization presents an enormous opportunity for Ulaanbaatar’s population rose from about half a million in promoting prosperity, its benefits have not been realized 2001 to about 1.2 million in 2011, and is projected to rise to in the case of Ulaanbaatar, which is facing considerable 1.7 million by 2025. Much of this population increase has challenges of service delivery 17. A main reason is that been through rapid rural-urban migration in the last two the urbanization has coincided with low-density urban decades. In 1989, 26.8 percent of Mongolia’s population sprawl of the city instead of being able to leveraging the lived in Ulaanbaatar; by 2006 that number had risen to 38.1 benefits of agglomeration and densification. There are percent; and by the 2010 census, 45 percent of Mongolia’s three dynamics responsible for the low-density urban population lived in the capital. Population growth in the development of Ulaanbaatar’s urban form. One, the city’s capital city is expected to continue at a high rate (current soviet style planning legacy and constitutionally mandated population growth rate being approximately 5%), signaling land ownership have led to large-sized urban blocks in the an urgent need to accelerate the provision of services and city center with low-rise structures (4-5 stories on average, infrastructure. Besides economic growth, another reason 10-12 stories for the highest buildings18. Mongolian citizens 18 Nowadays both spaces commonly taken up by parking. 19 16 According to Article 3 of the Law on Procedures for Observance of the Law on Land Allocation for Mongolian Citizens for Ownership. According to Kamata, T. (2010). Managing Urban Expansion in Mongolia: Best Practices in Scenario-based Urban Planning. World Bank Publications. UB Mayoral Decree no. A/726 of 2013 the entitlement is currently applicable to UB residents who had registered their residency before 30 May, 2013 17 World Bank 2016b. Ulaanbaatar: Inequality in Access to Service Delivery. 21 Urban Poverty in Ulaanbaatar: Income Poverty in Ulaanbaatar 6.0 INCOME POVERTY IN ULAANBAATAR 6.1 LEVELS AND TRENDS 6.2 SPATIAL DISTRIBUTION OF Table 2 Regional Poverty Trends: 2010 - 14 INCOME POVERTY Similar to the national trend, poverty declined quickly   POVERTY HEADCOUNT RATE (%) CHANGE DISTRIBUTION OF THE POOR during 2010-2012 in Ulaanbaatar, and was reduced by 11.4 There is a clear spatial dimension to the distribution of REGION  2010 2012 2014 2012-14 2010 2012 2014 percentage points.. In the following two years (2012-2014), urban poor in Ulaanbaatar across its districts and khoroos, West 52.7 32.3 26.0 -6.3 19.0 15.9 18.2 the pace of poverty reduction declined to a third of previous with district level poverty headcounts ranging from 10.4% in Highlands 51.9 38.5 25.3 -13.2 26.0 27.0 22.2 levels, averaging only 3.4 percentage points in this latter Bayongol to 30% in Singinokhairkhan (Table 3). Khoroos with Central 29.9 28.3 22.3 -6.0 12.3 15.7 16.3 period. From 2012-2014, the pace of poverty reduction in the the highest poverty rates tend to be located in the peripheral East 42.3 33.4 31.4 -2.0 7.7 9.0 10.8 capital city of Ulaanbaatar was one of the slowest, second areas of the city, along the edges of Songinokhairkhan Ulaanbaatar 31.2 19.8 16.4 -3.4 35.0 32.4 32.5 only to the eastern part of Mongolia (Table 2) and also more and Khan-Uul, as well as in the rural districts of Nalaikh, sluggish than the national urban average decline (Table Bagakhangai, and Baganuur. All residential khoroos of Source: World Bank (2016a): calculations from HSES 2010, 2012, 2014 1). Although having the lowest poverty headcount (16.4%) Songinokhairkhan have a poverty headcount over 20 across all regions, the capital region of Ulaanbaatar is home percent. Particularly high headcounts of poor populations to approximately 33% of Mongolia’s poor population. Given (over 34% of the population) are seen in Khoroos 3 and 11 of the greater shares of population living in Ulaanbaatar, the Songinokhairkhan district and in Khoroos 14, 7, and 6 of Khan- city has a higher proportion of poor than any other sub- Uul district (see Figure 2). national administrative location type (Figure 1), highlighting its centrality in interventions designed to target urban poverty. Even among the well off districts such as Chingletei, the levels of poverty range from 4.7 to 30.5. Of the rural districts, Baganuur and Nalaikh have become satellite cities Approximately 16 percent of UB households were below of Ulaanbaatar following a referendum in 2015. There is a Figure 1 Distribution of the poor by location: 2010-14 poverty line in 2014 and a substantial percentage clustered high level of sub-district variation within Ulaanbaatar with around the poverty line, and hence vulnerable to economic khoroos in the city center having less than 6% urban poor as shocks20. World Bank (2016a) estimates suggest that poverty compared to some of the peripheral districts where khoroo rate of Ulaanbaatar (16.4 percent in 2014) is likely to be poverty rates range from 35-45% of the khoroo population. higher in 2015, reaching 22.1 percent of the population. The proportion of poor increases with the distance to UB Projections suggest that the situation may improve in 2016 city center, with central khoroos of Chingeltei (03), Bayongol but the poverty rate is estimated to remain above the 2014 (15), and Sukhbaatar (04 and 06) being the only khoroo’s estimates at 19.4 in Ulaanbaatar. The depth of poverty is also with a poverty headcount below 5 percent. Map 1 shows the likely to increase, particularly among the unemployed and distribution of households below poverty line by khoroos. low-skill wage employees. Levels of inequality in Ulaanbaatar have shown no reduction, as demonstrated by the Gini coefficient that has remained steadfast at 0.33 between 2010-2014. Reduction in inequality will be as important as declines in urban poverty in the coming years to ensure that growth is shared equitably in order to create an inclusive, world-class city. 20 Source: World Bank (2016a) based on HSES 2010, 2012, 2014 Value in 2014 prices 22 23 Urban Poverty in Ulaanbaatar: Income Poverty in Ulaanbaatar Table 3 District Level poverty Headcount MAP 1. POVERTY HEADCOUNT KHOROO POVERTY MINIMUM KHOROO POVERTY MAXIMUM KHOROO POVERTY DISTRICT HEADCOUNT (%) HEADCOUNT RATIO HEADCOUNT RATIO BG 10.4 5 25.9 BK 24.2 20.4 27.9 BZ 21.2 5.6 34.2 CH 21.2 4.7 30.5 KH 27.2 6.3 38.3 SB 17.0 4.2 30.5 SH 30.0 10.8 40.4 Figure 2 Average Poverty Headcount per Khoroo Source: Poverty Mapping carried out by the National Statistical Office (NSO), Mongolia using small area estimation techniques on Census 2010 and Household Socio-Economic Survey Data 201121. 21 The map represents the proportion of poor people to the total khoroo population. The khoroo boundaries are the 2010 boundaries to keep the corresponding spatial distribution. 25 Urban Poverty in Ulaanbaatar: Key Characteristics of Urban Poor 7.0 KEY CHARACTERISTICS in Ulaanbaatar OF URBAN POOR IN ULAANBAATAR An analysis of National Statistical Office’s Household (i.e. 54.9 percent of built area in UB). Ger areas refer to land Table 4 Poverty by Household Head's Characteristics, Ulaanbaatar Socio-Economic Surveys (HSES) data from 2010, 2012 occupied by detached houses and by gers. “Non-ger areas” POVERTY DISTRIBUTION OF THE DISTRIBUTION OF and 2014 for Ulaanbaatar highlights that the incidence of typically contain apartment buildings with scatterings of   HEADCOUNT RATE POOR POPULATION poverty in Ulaanbaatar is highest among households living plots with single houses and ger structures. The city classifies in Gers, with unemployed household heads, less educated Ger areas by three zones, central, mid-tier (middle) and   2010 2012 2014 2010 2012 2014 2010 2012 2014 household heads, those dependent on social transfers fringe for future developments in terms of their locations, and private sector wages, and those dependent on social connectivity to engineering networks and housing types. Gender transfers and households living in gers (see Table 4). For Within the official classification, Central ger areas are those Male 29.1 19.0 14.3 70.2 76.7 67.9 75.4 80.4 77.7 example, about 24 percent of people in households headed where connection for centralized engineering networks is Female 37.8 23.7 23.6 29.8 23.3 32.1 24.6 19.6 22.3 by someone with upper secondary education lived in poverty feasible and will be redeveloped with high-rise and mid-rise in 2014 compared to only 3.7 percent amount those with a buildings. Mid-tier ger areas are planned for redeveloping Education Less than primary 66.8 50.0 64.0 1.5 2.3 2.5 0.7 0.9 0.6 degree or diploma. The poverty rate among people whose with low-rise and mid-rise buildings and connected to partial Completed primary 50.8 32.3 23.8 5.6 4.8 2.9 3.4 3.0 2.0 major source of income is social transfers and private wages engineering networks. Redevelopment for Fringe ger areas Lower secondary 50.9 36.9 36.1 17.2 22.1 18.5 10.6 11.9 8.4 was 26 and 17.5 percent. With respect to housing type, is planned in phases via land readjustment schemes with Upper secondary 36.8 24.4 23.9 49.5 50.6 44.2 42.2 41.3 30.3 approximately 38 percent of people living in ger residence onsite networks. Since the maps obtained from the city were Vocational training 21.2 17.9 17.1 19.0 11.8 24.0 28.2 13.1 23.1 are poor as compared to only less than 1 percent of those dated prior to the time of the World Bank Household Survey Degree or Diploma 15.4 5.6 3.7 7.3 8.3 7.9 14.8 29.8 35.5 living in apartments. Of those who are poor, more than half 2014, some of the newer ger developments in the outskirts of live report living in gers (57.5 percent), which is a larger the city had not been captured by the maps from the Master Housing Type proportion as compared to 2010. Planning Department (MPD). Additional analysis was carried Gers 58.5 40.8 37.7 44.2 53.4 57.5 23.6 26.0 25.0 out to exhaustively classify the surveyed households across Apartments 10.3 3.4 0.6 10.6 5.5 1.5 32.1 32.3 38.4 the ger zones by overlaying the locations of the households Detached House or Flat 32.8 19.8 17.7 43.9 37.6 36.7 41.9 37.9 34.0 In the following sections, the report focusses on key with a combination of GIS layers from MPD and World Bank’s Other 17.5 17.6 27.2 1.4 3.4 4.3 2.4 3.9 2.6 characteristics of poor household presented here. Where PUMA 2013 ‘Artificial Area’ layer for Ulaanbaatar22. possible, these descriptive results are augmented with the Employment Status findings from qualitative research commissioned for this Wage - public sector 24.7 11.6 10.8 10.9 7.0 8.5 13.9 12.0 12.9 study highlighting mechanisms that perpetuate poverty. The findings of the HSES Survey (2010, 2012, 2014) in Table Wage - private sector 30.7 21.1 16.4 35.4 44.3 44.0 36.3 41.8 44.0 4 as well as the World Bank 2014 survey (Table 5) confirm Family business - non-agric 17.5 12.4 11.6 6.4 8.6 9.7 11.6 13.8 13.7 that ger areas are home to a larger proportion of the city’s Herders and farmers 43.4 21.9 2.6 0.8 1.4 0.1 0.6 1.3 0.9 urban poor as compared to the apartment areas. The total Unemployed 41.1 39.6 33.6 5.6 4.5 6.5 4.3 2.3 3.2 7.1 TYPOLOGY OF GER AREAS percentage of the extremely poor and near-poor residing Inactive 38.5 23.6 20.3 40.8 34.2 31.1 33.4 28.8 25.2 AND RELATIONSHIP TO in a ger dwelling is 65 percent. Among the extremely poor, 39 percent live in a ger dwelling, 39 percent live in non-ger Income Source INCOME POVERTY detached housing, and 22 percent live in an apartment. In Agriculture 64.2 16.9 2.9 0.9 0.5 0.1 0.4 0.6 0.8 Ger areas are a unique feature of Ulaanbaatar’s urban built contrast, for the non-poor group, 22 percent live in a ger, Public sector wages 24.2 11.3 12.8 14.2 8.5 12.5 18.3 15.0 16.0 environment, house approximately 60% of Ulaanbaatar’s 33 percent in non-ger detached dwelling, and 45 percent Private sector wages 35.2 21.7 17.5 56.8 54.4 56.1 50.3 49.9 52.7 population and cover more than 9701.03 ha area of the city live in an apartment. However, poor households seem to Household business 24.1 12.7 10.5 9.2 7.2 8.4 11.9 11.2 13.2 Social transfers 37.3 28.3 26.0 16.2 27.8 20.2 13.6 19.5 12.7 Remittances 17.8 12.6 10.7 2.7 1.4 1.8 4.7 2.3 2.8 22 While, MPD’s central and middle Ger GIS layers were predominantly used to represent central and middle ger, the fringe ger layer was enhanced Capital income 3.7 2.2 7.2 0.1 0.2 0.8 0.8 1.4 1.8 using a combination of the MPD fringe GIS layer and the PUMA slums GIS layer for surveyed households whose coordinated were found to be outside of the SPPD fringe boundaries. Additionally, survey points located on the outskirts of the city but within 15 meters of a Ger area were assigned the Ger category when they fell outside both SPPD and PUMA layers to accommodate of the margin of error in the measurement of GPS coordinates. This Total 31.2 19.9 16.4 100.0 100.0 100.0 100.0 100.0 100.0 assumption was further confirmed through visual verification of this subset of points. 26 27 Urban Poverty in Ulaanbaatar: Key Characteristics of Urban Poor in Ulaanbaatar be evenly spread out across the three classifications of Ninety percent of ger area residents own their khashaa 7.2 TRENDS OF GROWING insecurity in employment and creating barriers to labor force ger areas rather than being concentrated in the fringe plot23, and an additional 6.6 percent have legal papers participation, especially among the urban poor. Two focus areas, contrary to the accepted narrative in Ulaanbaatar. to prove possession or were in the process of formalizing UNEMPLOYMENT groups were organized with informal sector workers and One reason for the belief that fringe ger areas are home their tenure title. However, as discussed in the next section, Despite the unprecedented growth of the past decade, job two with unemployed participants to understand in greater to the poorest residents of Ulaanbaatar is based on the ger areas are more likely than non-ger areas to suffer from creation in Mongolia only increased by 11 percent mainly depth the constraints to gainful employment as well as assumption that recent migrants who have settled in the multiple dimensions of deprivation with respect to access because the mining industry is not labor intensive. According challenges of earning a living as an informal sector worker. city’s fringes are the urban poor. As discussed later in this to basic services. Such experience of multiple dimensions to the Institute for National Strategy (INS) of Mongolia, the These discussions revealed considerable barriers faced by report, this belief is not supported by the empirical analysis. exacerbates the experience of poverty. number of unemployed seeking jobs has increased from urban poor in accessing stable employment opportunities, It is also important to note that ger residents enjoy a high 39,000 in 2012 to 55,000 in 2014, while the number of including discriminatory practices (particularly for women level of tenure security overall despite high levels of poverty. new jobs being created has halved. Youth unemployment and middle-aged job seekers) and shared experiences continues to be a big concern, with 69% of unemployed of labor exploitation by employers. Interestingly, most Table 5 Income by Locality Type falling in the age group of 15-34 years. In Ulaanbaatar, participants did not perceive a shortage of jobs in the labor force participation rate (i.e. individuals who are Ulaanbaatar but both groups agreed that employers’ bias employed or seeking jobs) steadily increased for both men towards younger and foreign workforce, exploitative labor LOCATION/DWELLING TYPE* and women between 2010 and 2014 (Table 6). However, it practices, rampant cronyism and corruption were the main PERI-URBAN GER MID-TIER GER INCOME CHARACTERISTICS CENTRAL GER APARTMENT AREA is especially worrisome that between 2012 and 2014, the reasons for unemployment and barriers in accessing jobs. AREA AREAS unemployment rate for women almost doubled, signaling For those in the informal sector, additional barriers were Mean HH income (MNT)  191,818  179,904  220,910  337,795 that there are not sufficient jobs to absorb the increasing created by high rents for space to carry out business, lack Share of quintile  number of women entering the labor force. Unemployment of transparency in permit procedures, and lack of access 1 (lowest) 26 25 22 13 for men in Ulaanbaatar also increased but only slightly. to information regarding job opportunities available. 2 21 23 20 17 Overall, while unemployment has declined from 2010 to Participants who were unemployed additionally highlighted 3 19 22 22 17 2014, some of the gains made between 2010 and 2012 are the following reasons for unemployment in Ulaanbaatar: 4 18 18 19 23 being reversed. declining quality of higher education in public schools, high 5 (highest) 16 11 17 29 rates of alcoholism, underdeveloped small and medium Sum of bottom 3 quintiles 66 70 64 47 sized enterprises, economic downturn, low levels of salaries According to the World Bank Survey (2014) data, offered, and tendency to expect government welfare approximately 73% of male respondents and 53% of female stemming from widespread social welfare availability that is Table 6 Labor Force Participation Rate in Ulaanbaatar: 2010-14 respondents work either full time or part time. At the time of a legacy of previous regimes. the survey, 65% of male and 48% of female respondents were employed full time.. Men working in part-time employment   LABOR FORCE PARTICIPATION RATE UNEMPLOYMENT RATE were more likely to be self-employed as compared to those 2010 2012 2014 2010 2012 2014 in full-time employment. These figures reflect high levels of insecurity in labor force participation with a large proportion Females 50.72 52.13 55.77 11.20 4.99 8.12 either unemployed or working part-time. When we look across household income quintiles, only 39% of respondents Males 63.26 68.21 73.45 13.61 6.58 7.79 in the lowest quintile were employed fulltime as compared to 70% of those in the highest quintile, implying that the urban Total 56.6 59.57 63.97 12.46 5.83 7.94 poor have more precarious employment. Source: World Bank (2016a): calculations from HSES 2010, 2012, 2014 Analysis of the focus group interviews highlight some of 23 A fenced plot will typically have one of more gers among open space, distinctly separate from the next plot. the mechanisms that are perpetuating a high degree of 28 29 Urban Poverty in Ulaanbaatar: Key Characteristics of Urban Poor in Ulaanbaatar 7.3 MECHANISMS In Ulaanbaatar, the female population makes up a significant share of labor force, and there are many families whose single BOX 1: BIASES IN EMPLOYMENT THAT PERPETUATE breadwinners are women, particularly due to high levels of UNEMPLOYMENT alcoholism in Ulaanbaatar’s male population. According to “I recently I went to a store at Sunday Market for a clerk post … interview was going well….but WHO Global Alcohol report, men consume over five times when I mentioned that I have 2 children, I was not hired. Some places have already set age range, AGE AND GENDER BIAS more alcohol than women in Mongolia. As a result, women’s like 20-25” Both men and women over 40 report that strict age unemployment during maternity leave or in circumstance Female, Unemployment FGD. when there are no alternative caregiving provisions, may requirements constrain their opportunities for employment. While men relate this bias more to manual labor jobs and result in economic vulnerability due to loss of income for the household. Given the increase in the level of female “I went to one place to get a job. I was asked whether I have husband. Then I was told that I should link it to perceptions of preference for Chinese workers headed households in Ulaanbaatar between 2012 and 2104 not have a baby for some years in near future. So I thought that a child was a barrier. I replied-I (particularly in the construction sector), for women the age (Table 7), the gender discrimination in access to employment don’t have husband and am unlikely to have a baby in next 3 years. That’s difference between bias has a gendered dimension as well. All women in the focus opportunities is likely to increase the vulnerability of such men and women. [Employers] want continuous employment and don’t want to give maternity groups agreed that prospective employers demonstrate households, particularly with respect to increase in poverty. leave money.” negative attitudes towards women with children, who are perceived as unemployable due to their child rearing Female, 26 years old, unemployed. roles. Others recounted instances of employers explicitly asking about their plans to have children or outright asking women to delay pregnancies. Older women highlighted an “I called after an ad. When I say I’m over 50, then they say-okay okay and hang up. There are jobs employer preference for younger women who were seen as available but the people are discriminated for their heights, appearances etc. That’s what I think. more attractive by employers, particularly with respect to Aged people like us are not even considered as humans. For us, no use in looking for jobs. I’m 50 employment as hotel and restaurant wait staff (see Box 1). years old but I’m not that old, I can at least work as a janitress.” Female, 55 years old, unemployed. “There are jobs available for young people. But no jobs are available for people of our age. There are always age limits set. I searched a lot at first trying to even work as a janitress but now I’m discouraged. [Employers] say-will call you or talk to you later and disappear. People below 40 are hired. But law says women can work till 60 but there are age limits so I don’t know what to do.” Table 7 Distribution of Ulaanbaatar's Population by Household Head's Gender Female, 46 years old, unemployed.   2010 2012 2014 GENDER OF THE HOUSEHOLD HEAD “Unemployment is very high. Around middle ages of 40 to 55, [people] are usually unemployed. When I go to get a job, [they] say, you are old. I registered to a district labor exchange. Then I went Male 75.4 80.4 77.7 to a job offered. I got the letter and went to see an auto mechanic. But I was told that I was old. So I called back to the labor exchange. I told them from the beginning that [employers] don’t hire Female 24.6 19.6 22.3 saying old, what’s use, when I go there, I would be called old again When I want to retire, they say that you haven’t reached the age but when you try to get job, they say you’re old.” Total 100.0 100.0 100.0 Male, unemployed. Source: World Bank Poverty Team calculations from HSES 2010, 2012, 2014 30 31 Urban Poverty in Ulaanbaatar: Key Characteristics of Urban Poor in Ulaanbaatar MAP 2. KINDERGARTENS WITHIN A 30 MINUTE WALKING TIME: 2014 CHILDCARE SERVICE LAW: KILLING THREE BIRDS can be provided on his/her owned, rented or leased space WITH ONE STONE 24 including apartment units, detached houses and gers that meet the specified requirements. Like the kindergartens, Policy-makers passed the Childcare Service Law in July 2015 childcare service provision starts in September and ends to address the high demand for available kindergarten seats on June every year. The government transfer for the costs of in UB city and aimag centers stemming from the urban-rural childcare service provision is set at 116’700 tugrik per child, migration. According the official estimates at that time, with payments by parents not exceeding 60 percent. The about fifty-six thousand children were left out of pre-school permits to run a childcare service are issued by district and education, with over thirty thousand parents unable to work. aimag governors based on the premise assessments by the The new law aims to expand childcare services for preschool- pertinent khoroo and bag governors. age children who were unable to enroll in kindergartens due to overcrowding and unmet deman. Additionally, the law supports the employability of parents and creates jobs for The law took effect in January 2016 and the contracting of the women as well as men. services started in April 2016. With a planned budget of MNT 10.8 billion for implementation in FY2016, over ten thousand children are expected to benefit from new childcare service Policy makers designed a flexible scheme for childcare provision nationwide with about 60 percent residing in UB service provision. where any 18-65 year old person with city. secondary education or above is eligible to provide babysitting services for 2-5 year old children. The childcare provider must complete a 5 day special training, and services SHORTAGE OF AFFORDABLE CHILD CARE AS A facilities are able to accommodate only 30% of children CONSTRAINT ON EMPLOYMENT eligible for enrollment, indicating severe shortages. Lack of affordable child care options are an important constraint on women’s access to the job market. In the These shortages affect urban poor disproportionately. As absence of strong family support systems, particularly in shown by Map 2, the majority of kindergartens are located migrant households, women must stay at home to take care in the city center, which is also home to a larger proportion of children. As of March 2016, Ministry of Finance and Ministry of higher income residents and apartment dwellers. There of Population Development and Social Protection (MPDSP) are a number of settlements in the far north and west of estimated that about forty-six thousand children were left Songinokhairkhan that either have no kindergarten or out of the preschool education and over half of the children which are far outside of a typical walking distance of 30 (23,670) resided in UB city. An analysis of the level of access minutes (see Map 2). Such inequality in spatial distribution 24 Sources: compiled from relevant legal acts and media coverage: Childcare Service Law at http://www.legalinfo.mn/law/details/11223; to public kindergartens in Ulaanbaatar revealed tremendous of affordable child care has an especially negative impact regulations on childcare service provision, financial assistances, and beneficiary ceiling, respectively at webpages of MPDSP at http://www.khun. gov.mn/index.php/mn/service/khunamsermenu/247-huuhedharahcat/2447-saidintushal.html, http://www.khun.gov.mn/index.php/mn/service/ gaps in coverage of eligible populations across the whole on women’s labor force participation, especially in lower- khunamsermenu/247-huuhedharahcat/2444-juram.html, and of National Center for Children at https://www.facebook.com/302578806509179/ city, particularly in poorer khoroos. Analysis of khoroo income households that cannot afford private facilities , photos/pcb.772104542889934/772098076223914/?type=3; consultation on proposed childcare service at Ministry of Education and Science at http://www.meds.gov.mn/Хүүхэд-харах-үйлчилгээний-талаар-нээлттэй-хэлэлцүүлэг-хийлээ; media coverage on initiative implementation at http://medee. household registry data shows that public kindergarten nannies or domestic help. mn/main.php?eid=76219 32 33 Urban Poverty in Ulaanbaatar: Key Characteristics of Urban Poor in Ulaanbaatar LABOR EXPLOITATION BY EMPLOYERS than half of labor income growth for the poorest 20 percent BOX 2: EXPLOITATION OF INFORMAL WORKERS CONTRIBUTES TO DISCOURAGED JOB SEEKERS (World Bank 2016a). Exploitation of informal workers seems to be rampant in “It’s everywhere. They hire a worker and fire out him saying that he didn't do the work after private sector firms, particularly within the construction CORRUPTION AS A BARRIER TO EMPLOYMENT employing the person for several days. The wage goes to their pocket” sector that is a key source of manual labor jobs for men. Corruption and cronyism are seen as the most important Male, Unemployment FGD. As discussed in Box 2, exploitation often takes the form barriers to accessing employment opportunities and of non-payment of full wage, fraudulent contracts, unfair carrying out other income generating activities, unanimously dismissals etc. “Slicing-off” or “tsavchaa ” in Mongolian is “I worked in many construction companies. I worked 2 days and 20 days. Whenever I go for by all participants of the four focus groups. These two a common term for labor exploitation practice, referring job based on ads, I get hired. They say that they employ you for one week as a probation. dynamics often coexist and work together to create negative to the illicit appropriation of a certain portion of wage by After working 6 days, I asked if I could become a permanent employee but was told to work on consequences for job seekers. All focus group participants a middle-person or entity, especially in the construction probation for 3 more months with daily rate of MNT 20,000. After working a month, I requested my agreed on the need to use influential connections or pay sector. Such exploitative practices were highlighted by all payment and eventually got only MNT 50,000 for the whole month.” bribes to access jobs and at times, even for getting short men in both focus groups and seen as significant hurdles in Male, Unemployment FGD. listed. They spoke about cash and in-kind incentives and accessing jobs and discouraging job seekers from staying in termed it as the practice of “looking at hands”, which the job market. Even in cases where written contracts exist, essentially refers to a manner of asking for a bribe in the form grievance redressal mechanisms do not seem to exist. The “Now these private companies say that [they] will see how you will work for 3 days when you go of an ‘advance’ from the potential salary that the job seeker FGD participants overall displayed little knowledge of legal to get a job. After 3 days, [they] say you are not suitable for us and no payment or salaries are would earn after getting the job. The FGD participants recourse options available and a low level of confidence in provided. They do real fraud activities.” spoke about networks of corruption within organizations, the judicial system in resolving such conflicts. Urban poor are particularly government offices, where there is an incentive Male, Informal Workers’ FGD II. particularly at risk of such practices that are likely to create to maintain the status quo that could be disrupted by adding poverty traps and increase the number of discouraged new employees who may not cooperate with the existing job seekers, given that the construction sector is the major “This has become profitable for these organizations. Hire and employ a person for 15 days and dynamics. (See Box 3) employer of the poorer households in urban areas. Between then fire without any wage. Another person is then hired and fired after 15 days. Basically, the 2010 and 2014, the construction sector contributed to more private entities aim to employ people without a pay.” Male, Unemployment FGD. “They throw the contracts at you saying- “take that contract and let’s see where you can go and what you can do.” Male, Unemployment FGD. “Let’s say I’m the director of the contracted company. I’ll assign the work to the engineer. I’ll specify wage per square meter; for example, emulsion paintwork wage would be 9,000 tugriks per square meter. Then engineer gathers several people. In other words, he’s hiring a brigade who will agree to his terms. Then, the wage becomes half 4,000-5,000 tugriks. Sometimes non- construction company which has no engineer gets the state tender through the back door and that company, takes a cut and then allocates the work to other company and eventually wage to the workers drops to 5,000-6,000 tugrik.” Male, Unemployment FGD. 34 35 Urban Poverty in Ulaanbaatar: Key Characteristics of Urban Poor in Ulaanbaatar BOX 3: CORRUPTION AS A BARRIER TO EMPLOYMENT BOX 4: LIMITED EMPLOYMENT SUPPORT SERVICES “… Even when you can start from the beginning, [they] don’t call you to an interview. So [you] need “I sell small quantity of flowers in summer time. [I] set up a tent in Naadam field for 3 consecutive a person to pull you in from the top.” years. The process of getting a permit is a really bureaucratic and permits difficult without Male, Unemployment FGD. knowing someone. So this year I obtained it through my connections by begging them to help me.” Female, Informal Workers’ FGD. “Generally, there are jobs. When we ask people, [they] often look at their hands. They say if you can find 2 million or 3 million tugrik, there are such and such jobs. But where could we find such “I solicited to rent a small square to sell stationary for September 1st. If you don’t have a connection, amount of money? [They say] that give an advance of this and that million tugriks. There were you would rent that small space for 400,000 tugriks. My friend has a stand and with his letter, I several instances… [they] say - if I let you in, give me an advance of 2.8 million or 3 million tugriks. rented for 60,000 tugriks. This kind of slicing-offs is everywhere.” There are many such jobs.” Male, Informal Workers’ FGD II. Female Unemployment FGD. “One pays money to get a job at a Customs Office, people say. In other words, the bribe amount “… there was an exhibition for the opening of SME support project. As of now, none of the people is 10 million tugriks. … Groups of people work as a team. This means that they are not keen on who submitted the proposals received the money. Heard that the friends of khoroo staffs, who welcoming a stranger to their team. They’d like to have someone who wouldn’t disrupt the team submitted the proposals, received the money. 2 of my friends also submitted the proposal but rules and expose their work. Newcomers often tend to work sincerely so their interest is either to didn’t receive the money, neither.” find their own person or keep it vacant.” Female, Informal Workers’ FGD. Male, Unemployment FGD. “…indeed, the project is a matter that works out through connections. [They] say SME support project but [they] say that your proposal wasn’t selected when you submit one. And [they] never say who was selected when [we] ask.” Self-employed small-scale traders and vendors also On one hand, the self-employed FGD participants identified corruption within government as a barrier to highlighted the benefits of flexibility within informal work Female, Informal Workers’ FGD. accessing opportunities, specifically when attempting to (particularly women with children) like the ability to receive obtain permits to carry out their businesses in authorized wages directly without paying cuts to any intermediary. market spaces. Rent-seeking and clinetelism practices On the other hand, these participants also reported being “…One senior man advised me not to submit my project proposal to the Khoroo office.[He] told me were also reported within government initiatives promoting exploited by police and public service providers (e.g. staff that they would register my proposal under other’s name and I would be just giving the proposal I small and medium enterprises, creating barriers for the from the employment offices) who see them as easy targets developed to other person. This is how khoroo people works.” beneficiaries. Non-transparency in selection outcomes to extract unofficial ‘fees’ for allowing them to carry out their Female, Informal Workers’ FGD. within such initiatives and limited dissemination of daily business. (See Box 5) Another drawback of informal work information seem to support such corrupt practices. Not highlighted by the participants was the inability to access surprisingly, employment support services offered by local formal banking institutions and social insurance programs “…if it’s a project [support], it works out if you have a connection. The project money is transferred administrations are seen as ineffective,.Overall, there are due to the absence of payment records. . The majority of for us but the money goes to completely different people, not us. If that project had been reaching low levels of awareness of available welfare services (See participants found interest rates of non-banking institutions its beneficiary, we wouldn’t have such credit-related problems. [They] tell us that khoroos receive Ulaanbaatar Service Delivery Report for more details on the and informal lenders unaffordable, which constrained their funding of 100 million tugrik each but it’s not clear where [it] goes to.” type of welfare provisions available). (See Box 4). ability to expand their businesses. Female, Informal Workers’ FGD. 36 37 Urban Poverty in Ulaanbaatar: Key Characteristics of Urban Poor in Ulaanbaatar EDUCATION: ACCESS AND QUALITY bottom two quintile who have primary or less than primary education, with a majority of urban poor having at least high As shown in Table 4 urban poor are less likely than their school diplomas. (See Table 8). better-off counterparts to have high levels of educational attainment. Of those who are poor, 68% live in households headed by someone with upper secondary education or While it is important to understand the educational less and only 8% in households with heads having higher attainment of urban poor, which is often a key determinant education. This is confirmed by the finding of World Bank of urban poverty, it is equally important to understand the 2014 Survey (Table 8), which finds that three times more quality and access to education available to the children of respondents in the highest income quintile have attained urban poor. Access to quality education is a key input into graduate education than those in the lowest income quintile. the development of human capability and determines the That said, there is overall a very high level of literacy among extent of inter-generational disadvantage transferred. urban poor in Ulaanbaatar with only 3-4% respondents in the Table 8 Characteristics of Urban Poor by Quintile (percent) BOX 5: EXPLOITATION BY POLICE AND PUBLIC OFFICIALS   1 (LOWEST) 2 3 4 5 (HIGHEST) “Recently, I rented a stand for 4 days was 270,000 tugriks at Misheel Expo. We were brought there by the District Employment Department. We sat there thinking it was free of charge but the staffs TENURE STATUS of Employment Department collected money. We were not previously informed and money was Possession or owner 55 59 62 63 64 collected without our signature. We are often in contact with the Employment Department so we just gave the money to avoid trouble.” Renter 10 8 7 12 9 Female, Informal Workers’ FGD. No Certificate 35 33 31 25 27 EDUCATION “They come from khoroo offices and say okay lady, go away, take the things you are selling with Primary or below 3 4 3 2 1 you. And if you give some tugriks, they leave at once.” Middle school 18 12 8 8 5 Female, Informal Workers’ FGD II. High school 48 47 38 32 22 Technical and vocational 10 9 13 10 8 “Now the police have such mentality. In lunar new year when the market activities become regular, Undergrad or above 21 27 37 48 63 the police say that you can stand and 3,000 tugriks per person and tell one person to go and count the number of people and collect the money.” EMPLOYMENT Male, Informal Workers’ FGD II. Unemployed 29 17 15 13 7 Pensioner 16 16 20 18 16 Student 6 5 2 3 1 “[They] just take money and leave. [They] say-you, lady, give 5,000 tugrik or go away.” Female, Informal Workers’ FGD II. Self-Employed 16 18 17 18 23 Employed 32 44 46 48 53 Source: Data from World Bank Survey (2014) 38 39 Urban Poverty in Ulaanbaatar: Key Characteristics of Urban Poor in Ulaanbaatar ACCESS TO SCHOOLS not optimize the use of existing educational facilities. Map 3 “IT’S A GER AREA SO IT’S DIFFICULT. EARLY IN THE MORNING WHEN shows the total estimated school capacity in each khoroo25. As demonstrated by the Ulaanbaatar Service Delivery Dark green-shaded khoroos represent an excess of demand CHILDREN LEAVE [FOR SCHOOL], IT’S DARK AND FAMILIES HAVE MANY Report 2016 (World Bank 2016b), most educational facilities for the available amount of school space, and light green DOGS. CARS DON’T EVEN SIREN WHEN THEY ENTER THE ALLEYS. OUR in the city are publicly provided and, in principle, free of cost to ensure affordability for urban poor. The success of free shows khoroos with adequate school space. These areas NEIGHBORHOOD GENERALLY LACKS STREET LIGHTING. [CHILDREN] are found just outside of the central city area, for example public education provision is noted in the fact that a majority GO IN THE DARK, THEN THE DOGS OFTEN ATTACK... MANY DRUNK in Chingelei, Sukhbaatar and Bayanzurkh. Orange and of households in apartment and ger areas (approximately 90 red khoroos show that the supply of space far exceeds DRIVERS IN THE EVENING TIME AND THEY HIT SMALL CHILDREN.” percent of respondents with school-age children) reported the demand. The map also shows that school capacity is sending their children to a public school (World Bank Survey, 2014). However, the number of schools only meets 86 percent sufficient mainly around the central city area, but becomes “WHEN CROSSING THE ROAD, THE CARS DRIVE LIKE IT’S ABOUT increasingly overcrowded in the core areas. By contrast, of existing demand based on data from Household Registry TO HIT. WHEN IT’S EVENING, CHILDREN ARE FOLLOWED AND THEN khooros in Bayanzurkh, Songinokhairkhan and Nalaikh tend (2014). A close examination of the spatial distribution of to have an oversupply of space. These are also the khoroos ROBBED BY HOLDING THEIR HANDS AND FEET. MY CHILD’S FOOT schools in Ulaanbaatar highlights the problem of inequality with the largest proportions of their population residing in WAS RUN OVER BY A CAR WHILE CROSSING A ROAD.” in the ease of physical access to schools for children from gers. urban poor households living in ger areas. As shown in World Bank (2016b), the distribution of schools across the city does While there are fewer schools serving the fringe ger areas In addition to distance from kindergartens, majority of of the city compared to the central city areas due to parents from both fringe and central ger areas raised MAP 3. SCHOOL CAPACITY BY KHOROO: 2014 population density, schools in the fringe areas are also significant safety concerns for their children during the located far from the neighborhoods that they serve. There focus group discussions26. The difficulties highlighted in are a number of settlements in the far north and west of the discussions were linked to the conditions and quality Songinokhairkhan that either have no school or which are far of the ger area neighborhood environments. These include, outside of a typical walking distance of 40 minutes (see Map aggressive stray dogs, road and traffic accidents due to 4). The time taken to walk to schools differs considerably for rash driving in tertiary roads, dark streets without functional children depending on where they live. Almost 40 percent street lights, presence of alcoholics and theft and bullying in of children living in ger areas take 40 minutes or longer to the neighborhoods when children walk back and forth from walk to schools, compared to only 17 percent of apartment schools27. Due to the growing concerns of children’s safety area residents; and on average, ger area children have in both fringe and central ger areas, Municipal Police and to walk twice as long as the apartment-area children (18 UB City Council have jointly commenced a recent initiative minutes versus 9 minutes). For example, more than 50% called “school police” that involves patrolling of streets of the population of Songinokhairkhan lives farther than during school opening and closing hours by concerned 40 minutes’ walking distance from a school. In contrast, parents, which the parents spoke about positively28. all residents in Khan Uul are served by a school within 40 minutes’ walking time. 26 Other less frequent contextual challenges reported were related to the terrain in the ger areas and transportation for the school children. In terms of terrain, at least one participant from both geographical locations spoke of the difficulties created by the absence of footpath and slippery road or pathway in wintertime. Whilst, only two female participants from fringe ger areas spoke of the issues related to the transportation of the school children, namely, charging fee for school buses and their non-functional uses, i.e. transporting school staffs instead of the children. 27 4 female participants from fringe ger areas and 3 male participants from the central ger areas spoke of their concerns with regard to the aggressive dogs of the households in their neighborhoods that frighten and, in some cases, bite their children whereas all female participants from the central ger areas in addition to 3 male participants from the same geographical location shared the same concern. 5 participants (2 female and 3 male participants) from the fringe ger areas and 6 participants (2 female and 4 male participants) from the central ger areas spoke of the traffic and road Source: Data from World Bank Survey (2014) related challenges that their children encounter in commuting to and from schools, respectively. 28 25 This is estimated by comparing the total number of potential students per khoroo to the total available seats in all of the schools within the khoroo. 3 parents from each geographical location spoke positively about the initiative that Municipal Police and UB City Council jointly commenced on 25 October 2013. 40 41 Urban Poverty in Ulaanbaatar: Key Characteristics of Urban Poor in Ulaanbaatar MAP 4. KINDERGARTENS WITHIN A 40 MINUTE WALKING TIME: 2014 BOX 6: QUALITY OF EDUCATION “But, it’s generally difficult. It’s more difficult in wintertime. Snow storms, crowded buses, it seems that good teachers are taken by the central schools or they just don’t stay. So, it’s like a test rabbit.” Male, fringe ger area, SBD 17. “Generally, fringe schools are of no use. I once enrolled my child at the school in Khujirbulan and pulled out. [My child] told me that it’s same, dad. Tends to lag behind the modern requirements. Central schools seem better.” Male, fringe ger area, BZD 20. “I guess the central schools are fine but the teaching skills of teachers at fringe schools are very poor... for the ger area schools, they are quite lower when compared with the levels of the schools closer to the center.” Female, central ger area, CHD 8. “In terms of teaching, indeed, teachers seem like they don’t overcome overcrowding... generally, the children’s education is poor from the early grades .” Male, central ger area, SBD 18. Source: Data from World Bank Survey (2014) “My child’s school is old so the chairs and desks are broken, it’s a poor school. Money is collected QUALITY OF EDUCATION the education quality and availability of proper facilities to get the parents’ support. Something should be done with that money but nothing. We don’t to be superior in apartment areas as compared to ger supervise everything so don’t know very well.” Participants of World Bank FGDs 2014, were asked to rate areas. Particularly in fringe ger areas, parents complained Male, fringe ger area, SBD 17. their children’s school facilities and quality of education on about dilapidated sports facilities, damaged or insufficient a scale of 1 to 10, with 10 being the best and 1 being the number of desks and chairs in the classrooms, condition “For example, i give saving fund money for all of my four children. My four children monthly give worst. On average, parents from both fringe and central ger of sanitation facilities, and lack of playgrounds. Sanitation 6000 tugriks for cleaning… four kids also purchase textbooks for 3500 tugriks. Actually the school areas rated schools at 5.8 and 5.6 on average respectively. facilities were reported to be located outside the schools is free of charge. However, other costs when combined constitute the school fee.” Parents who rated their children’s schools the highest were without proper protection from cold and at times, without those who reported high quality of the facilities and high Male, Central ger area, BGD 16. door locks. Pit latrines were reported to be in desperate caliber of teachers. Inadequate ability of the teachers to need of maintenance, however, only the toilets used by impart quality education was the most important reason for children but not by the teachers. Parents felt that the lack Poor condition of school facilities appears to place a on a monthly basis. Participants also felt that children from low ratings by parents from both geographical locations. of sanitation hygiene affected the educational experience considerable burden on low-income households as schools poorer households were more likely to be discriminated Parents felt that the quality of education and teachers’ of their children, especially of adolescent girls. In other parts routinely carry out unofficial cash collections for improving against in the classroom and paid less attention to by the ability in ger areas suffers from the overcrowding, poor of the world, lack of access to adequate toilets has been and maintaining the classrooms. Five participants from the teachers if the parents are unable to fulfil the unofficial quality of education in teacher training institutions and high associated with higher high school dropout rates in urban fringe ger areas and seven participants from the central requests for cash or in- kind contributions from the schools. turnover rates in fringe schools resulting in a shortage of poor communities29. ger areas reported that teachers collect cash and in-kind Such practices do not allow poorer households to fully experienced teachers. Most FGD participants considered contributions to improve the classrooms (e.g. replacing the benefit from the state-funded free public education system. windows, curtains, printer toners, repairs, purchasing new 29World Bank; International Monetary Fund (2013) Global Monitoring Report 2013 : Rural-Urban Dynamics and the Millennium Development Goals. cabinets and so on) and charge parents for maintenance Washington, DC 43 Urban Poverty in Ulaanbaatar: Multi-Dimensional Poverty 8.0 MULTI-DIMENSIONAL in Ulaanbaatar POVERTY IN ULAANBAATAR The discussion so far has highlighted that ger areas are not within a population. Such deprivations, especially when Quality). The reason for carrying out this analysis is twofold. dimensional indicator. The application of the first cut-off is only home to a larger proportion of urban poor but also pose co-existing, function as collective structures of constraints One, this investigation allows for an understanding of the used to identify those individuals who were deprived in each particular development challenges of service delivery given on individuals preventing them from fully realizing their level of deprivation across subgroups, in this case, ger and of the dimensions to get a dimension specific deprivation their low population density, locational disadvantage, and capabilities, increasing their vulnerability and reducing the non-ger strata households. Two, and more importantly, it for each individual in the data. For example, within the unique urban morphology. Such conditions in ger areas have likelihood of promoting shared prosperity. While income allows us to decompose the contribution of constituent dimension of Water and Sanitation, 58% of households meant that urban poor not only suffer from income poverty poverty captures constraints of individuals’ budget sets, dimensions to overall multidimensional poverty to better are deprived with respect to the indicator on ‘access to but are also adversely affected by multiple deprivations multidimensional poverty highlights the aspects of poverty understand the nature of urban poverty. sanitation’, which measures whether or not a household has due to lack of basic services and infrastructure. As a result, that are actually experienced. These findings are particularly access to a flush toilet. Similarly, deprivation in the dimension a sole focus on income poverty becomes an incomplete relevant for policy makers in the as they highlight important of accommodation is captured by dual indicators, including measure for capturing the lack of wellbeing and deprivation dimensions that require state action to improve the lives of The AF method identifies multidimensional poor using a “dual ownership status (renter versus owner) and residence in a in access to dignified living conditions faced by urban poor urban poor. More importantly, the use of multiple dimensions cut-off” method. First, a cut-off is applied to each indicator ger or not. in Ulaanbaatar. Identifying the profile of deprivations faced of consumption (food and non-food) and living standard (see definitions of Deprivation Line in Table 9), below which Following the first cut-off using the deprivation line, a second by urban poor in addition to income poverty is particularly allows us to identify more vulnerable groups than one would or within which a person is considered deprived within that cut-off is applied to specify the breadth of deprivations, i.e. relevant from a policy perspective to enable the design of on solely do on the income dimension. It is possible that meaningful social and economic policies. This section uses these people might be just above the poverty line and as Table 9 Deprivation Line rich sources of data available for this report to quantify such missed by the income poverty measure but equally DIMENSION INDICATOR DEPRIVATION LINE RAW MEAN the nature and composition of multidimensional poverty vulnerable as those below the poverty line in their experience for residents of Ulaanbaatar by taking into account both of multiple depravations. Deprived in access to close bus stop More than 15 minutes of distance 0.17 Accessibility economic dimensions as well as non-economic dimensions Deprived in fast access to workplace More than 60 minutes of distance 0.14 central to urban livability. Deprived in secure tenure Renter 0.05 Accommodation Deprived in proper accommodation Ger 0.26 8.1 OVERVIEW OF THE Deprived in household assets Scale average 0.43 An operationalization of a view that places importance METHODOLOGY Assets & Money Deprived in per capita household income 118,668 Mongolian Tughriks 0.33 on both economic and non-economic aspects of urban In order to understand multidimensional poverty in Deprived in school attendance of youth No attendance of 6-7 y.o. children 0.01 deprivation can be borrowed from the literature that views Ulaanbaatar, the World Bank Survey 2014 data was analyzed Education Deprived in access to internet No access 0.44 poverty (and wellbeing) as essentially multidimensional using the well-established Alkire Foster methodology (Alkire 2008; Chakravarty, and D’Ambrosio, C. 2006; Anand Deprived in education Up to middle school 0.12 (AF method, hereon) (Alkire and Foster, 201131), which and Sen 1997; Bourguignon and Chakravarty 2003, to name Deprived in employment Average unemployment share 0.38 was applied to 22 indicators from the survey data that a few30). A multidimensional understanding of wellbeing Employment Deprived in secure employment Informal sector 0.21 were grouped across eight dimensions (namely, Assets is especially important when we are concerned that the Deprived in regular employment Part-time (<11 months) 0.28 & Money, 32Accessibility, Accommodation, Education, experience of poverty may be different for social groups Deprived in access to garbage collection Employment, Solid Waste, Water & Sanitation, Community No access 0.28 Solid Waste service Deprived in regular garbage collection service No collected + irregular + doesn’t know 0.47 Deprived in water supply No centralized water supply 0.57 30 Alkire, S. 2008. Choosing dimensions: the capability approach and multidimensional poverty. In: Kakwani, N., Silber, J. (eds.) The Many Dimensions of Poverty, pp. 89–119. Macmillan, New York; Chakravarty, S.R., D’Ambrosio, C. 2006. The measurement of social exclusion. Rev. Income Wealth 53(3), Deprived in close water source Not on plot / in house 0.57 Water/Sanitation No flush toilet with central sewage 377–398; Anand, S., Sen, A. 1997. Concepts of Human Development and Poverty: A Multidimensional Perspective. UNDP, New York; Bourguignon, F., Deprived in access to sanitation 0.58 Chakravarty, S.R. 2003. The measurement of multidimensional poverty. J. Econ. Inequality 1(1), 25–49. discharge 31 Deprived in sole access to toilet Toilet shared by more than 1 household 0.39 Alkire, S and Foster, J. 2011. "Understandings and misunderstandings of multidimensional poverty measurement," Journal of Economic Inequality, Springer, vol. 9(2), pages 289-314, June. Deprived in Community Employment Major problem 0.59 32 Deprived in Absence of Wife Beating in The key economic aspect of wellbeing was captured by household income in the survey data. However, the income information collected in the Community Major problem 0.07 survey was provided in a categorical form (income intervals) and had to be converted to a continuous measure, such that each household could be Community assigned a specific value before applying the first cutoff to determine deprivation in this dimension. A simulation method was applied to transform it Quality Deprived in Absence of Alcoholism in Major problem 0.50 into a continuous income variable for each household. Additional details on methodology available from Task Team. Community Deprived in Community Safety Major problem 0.23 44 45 Urban Poverty in Ulaanbaatar: Multi-Dimensional Poverty in Ulaanbaatar how many dimensions should a person be deprived on to be combining survey data with a matrix of responses from focus 8.2 LEVEL AND DISTRIBUTION twice as likely to be multidimensional poor as compared to considered poor. Results presented in this report are based group participants who were asked to rank each relevant someone living in non-ger strata. on a 30% cut-off point, implying that for a household to be indicator by its overall contribution to the experience of OF MULTIDIMENSIONAL considered deprived, it has to be deprived in at least 30% of poverty. A total of 16 focus groups generated a matrix of POVERTY IN the 22 indicators (i.e. 7 or more indicators) used to construct responses from 128 individuals that were used to weight A spatial examination analyzing the clustering of the multidimensional poverty index (MPI)33. the indicators used to constitute the eight dimensions of ULAANBAATAR multidimensional poverty further demonstrates that while poverty34. Given that qualitative weighting is more context- Results of the analysis (utilizing qualitative weighting) show khoroo clusters of highly deprived neighborhoods are located specific, it is likely to produce a more realistic representation that overall 23.4% of population of Ulaanbaatar can in all districts, large hotspots of deprivation are located in A final step in the AF method is assigning of weights to of the depth and dimensionality of deprivation. be categorized as multidimensional poor. However, Bayanzurkh, Chingeltei, and in Songinorkhairkhan, as well dimensions or indicators constituting the dimensions. multidimensional poverty is consistently higher and more as in Bayangol (See Table 10). Most of the high deprivation Often the choice of weights (whether equal or differential) intense among Ger residents and this result is robust clusters seem to be midway from the city center to the edge is highly subjective and solely at the discretion of the regardless of the levels of cut-off applied. Using the of the city i.e. outside the city center. In contrast, the city analyst. In this report, we use an innovative approach by abovementioned methodology and specifications, 39.5% of center is home to clusters of low deprivation or high levels integrating qualitative weighting with the estimation ger residents are poor as compared to only 17.8% of non- of multidimensional wellbeing, particularly in Khan-Uul (KH11, of multidimensional poverty in Ulaanbaatar, Mongolia, ger residents. In other words, a person living in ger strata is KH15), Bayangol (BG03, BG04, BG17, BG02) and Sukhbaatar (SB03, SB02, SB04, SB01). Table 10 Key Hotspots of multidimensional poverty DISTRICT KHOROO NUMBERS Bayanzurkh BZ12, BZ08, BZ16, BZ19, BZ14, BZ04, BZ05, BZ02, BZ15 ,BZ18, BZ25 Chingeltei, CH10, CH11, CH09 and Ch15, CH13, CH12, CH14,CH18, SB14 Songinorkhairkhan SH04, SH20 SH26, SH05, SH06 Bayangol BG20, BG21, BG09, BG10, BG11, BG12, BG14, BG23 33 Further details on the dimensions, rationale for the cut-off point and the steps employed within the AF method can be found with the Task Team. 34 Additional details on methodology available from Task Team. 46 47 Urban Poverty in Ulaanbaatar: Multi-Dimensional Poverty in Ulaanbaatar MAP 5. MULTIDIMENSIONAL POVERTY CLUSTERING MAP Table 11 Poverty Level & Dimensional Contribution to Overall Poverty by Ger & Non-Ger Areas UNCENSORED DEPRIVATION CENSORED DEPRIVATION HEADCOUNT HEADCOUNT DIMENSION INDICATOR TOTAL NON-GER GER TOTAL NON-GER GER Access to close bus stop 16.6% 12.2% 29.2% 12.5% 8.2% 24.8% Accessibility Access to workplace 14.4% 12.4% 20.1% 8.6% 5.5% 17.4% Secure tenure 5.0% 3.2% 10.2% 4.5% 2.8% 9.5% Accommodation Proper accommodation 26.0% 0.0% 100.0% 21.6% 0.0% 83.2% Household assets 42.8% 27.4% 86.6% 33.9% 20.2% 72.8% Assets & Money Per capita household 33.2% 28.9% 45.6% 29.9% 24.3% 45.6% income School attendance of 1.3% 1.1% 1.6% 1.1% 0.9% 1.4% youth Education Access to internet 44.0% 36.2% 66.1% 29.1% 19.7% 55.6% Education 12.2% 7.6% 25.5% 8.9% 4.7% 21.1% Employment share 38.4% 33.4% 52.6% 25.6% 18.4% 45.9% Employment Secure employment 20.8% 20.6% 21.3% 11.7% 9.8% 17.0% Regular employment 28.3% 26.5% 33.3% 16.0% 11.9% 28.0% Access to garbage 8.3 UNDERSTANDING times more ger residents are deprived in the education collection 27.8% 34.6% 8.6% 10.0% 10.7% 8.1% dimension as compared to non-ger residents, which includes Solid Waste THE DIMENSIONS OF primary or less than primary educational attainment. Regular garbage collection 46.5% 39.2% 67.3% 34.6% 25.1% 61.7% DEPRIVATION The second panel provides the percentage of households Water supply 57.3% 42.7% 98.9% 43.6% 30.0% 82.4% The multidimensional poverty estimates can be further that are deprived on particular indicator along with being Close water source 57.4% 42.9% 98.6% 43.7% 30.2% 81.9% decomposed in order to provide estimates of how much each simultaneously deprived in at least other 6 indicators Water and (namely, ‘Censored Deprivation Headcount’). In other words, Sanitation indicator (and dimension) contributes to the overall poverty Access to sanitation 58.2% 43.5% 100.0% 44.2% 30.5% 83.2% level. Table 11 allows us to see three types of results that this panel provides the proportion of multidimensional are important to understand the nature of multidimensional poor who are also deprived on that dimension. For Sole access to toilet 39.3% 28.6% 70.1% 33.3% 22.9% 63.1% poverty in Ulaanbaatar. The first panel of results presents example, overall 29.9% of overall residents are below the Unemployment in 59.2% 58.7% 60.9% 36.3% 29.7% 55.1% the proportion of households deprived with respect to a absolute poverty threshold for household income in addition community specific indicator regardless of whether or not they are also to being deprived on the income indicator. When broken Wife beating in community 7.0% 7.3% 6.0% 5.5% 5.4% 6.0% Community deprived on other indicators i.e. regardless of whether they down by strata, the proportion of households with income Quality deprivation is 45.6 percent in Ger areas and 24.3 percent Alcoholism in community 50.2% 51.5% 46.4% 30.5% 25.9% 43.5% are multidimensional poor (namely, ‘Uncensored Deprivation Headcount’). For instance, only 12.2% of the households are in non-ger areas. Community safety 23.2% 23.1% 23.6% 15.6% 13.3% 22.0% deprived in terms of adequate education levels attained in Ulaanbaatar. But when stratified by location type, three Source: Ulaanbaatar Survey Data (World Bank, 2012): Note: Sample size = 2983 48 49 Urban Poverty in Ulaanbaatar: Multi-Dimensional Poverty in Ulaanbaatar Table 12 Absolute dimensional contribution to multidimensional poverty The relative contribution of each dimension overall can be apartments in itself is less likely to be important for reducing seen in Figure 3, with the key dimensions being economic multidimensional poverty but its importance is likely to ABSOLUTE CONTRIBUTION TO MULTIDIMENSIONAL (Assets and Money), access to basic services (Water and be expressed through lowering of contribution of other POVERTY DIMENSION INDICATOR Sanitation), social problems at the neighborhood level dimensions such as better access to water and sanitation, TOTAL NON-GER GER (Community Quality) and garbage collection (Solid Waste). garbage collection etc. It should also be noted that these Interestingly, the same dimensions are important for ger and results are weighted by the ranking of dimensional indicators Access to close bus stop 0.13% 0.08% 0.26% Accessibility non ger households (see Annex 1). by the qualitative preferences expressed by the focus group Access to workplace 0.09% 0.06% 0.18% respondents. If we give each of the dimensions equal weight, the results are somewhat different but the importance of Secure tenure 0.29% 0.18% 0.63% An additional strength of this methodology is its ability the dimensions of Assets & Money, Water & Sanitation, Solid Accommodation to estimate the absolute contribution of constituent Waste and Community Quality. Proper accommodation 0.32% - - dimensions to overall multidimensional poverty level. Household assets 0.17% 0.10% 0.38% This offers a starting point to policy makers with respect Assets & Money to the decisions about prioritization of resources, while As shown in Table 12, the largest contributions to Per capita household income 5.80% 4.72% 9.14% keeping in mind institutional constraints. For example, it multidimensional poverty are made by deprivations in School attendance of youth 0.09% 0.08% 0.12% may not be possible or sustainable to bring all the income dimensions of ‘Assets & Money’, ‘Water & Sanitation’ and poor households out of poverty by increasing household ‘Solid Waste’, with considerable differences between ger and Education Access to internet 0.15% 0.10% 0.29% incomes through social transfers, especially during a phase non-ger residents. The dimension of neighborhood quality of economic downturn. But it may be possible to develop based on the presence of social problems is also important, Education 0.04% 0.02% 0.11% interventions targeting urban infrastructure, basic service particularly with respect to employment and alcoholism. Employment share 0.13% 0.09% 0.24% provision, social interventions addressing community level Interestingly, in both ger and non-ger areas, alcoholism is social problems, public health provision or improving solid seen as not only an important indicator affecting the quality Employment Secure employment 0.06% 0.05% 0.09% waste management services. Such interventions can reduce of urban livability but also makes a large contribution to multidimensional poverty and improve the quality of life of multidimensional poverty. Regular employment 0.08% 0.06% 0.14% urban poor. Access to garbage collection 0.90% 0.96% 0.75% Solid Waste Regular garbage collection 3.10% 2.25% 5.71% Table 12 shows the absolute contribution of each Figure 3 Relative Contribution of Each Dimension to dimension to the overall multidimensional poverty. This Multidimensional Poverty Water supply 0.22% 0.15% 0.43% table describes the absolute amount of multidimensional Close water source 0.22% 0.15% 0.42% poverty that would be reduced if all resources were Water and Sanitation put into bringing each household deprived in terms of 18.5% Access to sanitation 4.18% 2.88% 8.12% a dimensional indicator above the deprivation line. For example, multidimensional poverty would decline by 5.80% Sole access to toilet 3.15% 2.16% 6.15% 33.1% (i.e. from 23.4% to 17.6%) if we could raise the household Unemployment in community 1.26% 1.03% 1.98% incomes of all households above the poverty line. Most interestingly, addressing the deprivation in the Water & 17.1% Wife beating in community 0.33% 0.32% 0.37% Sanitation dimension, could reduce multidimensional 1.1% 1.2% Community Quality poverty substantially from 23.4% to 15.7%. Even a simpler 25.4% Alcoholism in community 1.82% 1.55% 2.68% intervention such as improving the reach and efficiency 2.6% Community safety 0.93% 0.80% 1.35% of Solid Waste collection could alleviate multidimensional poverty by 4 percentage points. Interestingly, provision of Source: Ulaanbaatar Survey Data (World Bank, 2012); Note: Sample size = 2983 51 Urban Poverty in Ulaanbaatar: The Importance of 9.0 THE IMPORTANCE Non-Monetary Dimensions OF NON-MONETARY DIMENSIONS A transitional matrix approach is further used to estimate 100% of people who are in the bottom 40% of the income Table 13 Transitional Probabilities comparing Income versus Multidimensional Poor Individuals based on their Income Status the probability of being multidimensional poor and non- distribution are also multidimensional poor. In other words, – Ulaanbaatar, Mongolia: 2012 poor, given the income poverty status. Table 13 shows these non-monetray poverty seems to be concentrated among PANEL A - ABSOLUTE POVERTY LINE transitional probabilities for two different cut-offs for the bottom 40% of the income distribution and the richer Multidimensional Poverty (k = 30%) monetary poverty. In Panel A, monetary poverty is shown populations have a better quality of life as compared to the bottom 40%. Differences between ger and non-ger area Income Total Ger Non-Ger based on the poverty line evaluated at 118,668 Mongolian Tugriks of monthly household income in 2012. In Panel B is populations continue to persist. Non-poor Poor Non-poor Poor Non-poor Poor based on monetary poverty evaluated as the bottom 40% Non-Poor 66.2 33.8 30.8 69.2 75.7 24.3 of the income distribution. Two main results stand out from Poor 10.1 89.9 0.0 100.0 15.7 84.3 this table. A spatial representation of the concurrence of income poverty with multidimensional poverty reveals some clusters             with high multidimensional deprivation and low income (i.e. PANEL B – BOTTOM 40% The first important result in Panel A is that one third of the High –Low). Such areas are in need of immediate attention Multidimensional Poverty (k = 30%) income non-poor are also poor in multiple non-monetary from policy makers as they are likely to deteriorate given the Income Total Ger Non-Ger dimensions (33.8%)35 and that non-monetary deprivation is presence of multiple disadvantages and lack of opportunities. mainly concentrated among residents in Ger areas (peaking Examples of such areas are khoroos in Chingeltei (CH13, Non-poor Poor Non-poor Poor Non-poor Poor at 69.2%). Also, in the overall population, 10.1% households Ch12, Ch11, CH10, CH07), Songinokharkhan (SH06, SH27, Non-Poor 91.3 8.7 87.6 12.4 92.3 7.7 who are income poor are not poor enough in sufficient SH04 SH11, SH31), Bayanzurkh (BZ04, BZ02), Khan-Uul (KH10, Poor 0.0 100.0 0.0 100.0 0.0 100.0 dimensions to be considered multidimensional poor. These KH08) and Bayangol (BG21, BG09, BG10,BG22,BG23). Source: Ulaanbaatar Survey Data (World Bank, 2012); Note: Sample size = 2983 poor households likely live in closer to the city center or in apartment areas and as such, benefit from the generally MAP 6. CLUSTERING OF INCOME AND MULTI-DEPRIVATION BI-LISA higher levels of service provision. However, among the Ger In the following sections, the report delves deeper into residents all those who are income poor are also deprived a discussion of some of the key indicators of deprivation in other dimensions i.e. the overlap between monetary and highlighted above and their relationship with urban poverty non-monetary poverty measurements is complete. in particular. Leveraging the geo-referenced nature of the survey data as well as other spatial data sources available for Ulaanbaatar, the report highlights the spatially distribution of The second result worth noting is the higher level of overlap deprivations that constitute multidimensional poverty, with between income and non-monetary dimensions of poverty a particular focus on their relationship with the distribution when measured based on the 40% poorest threshold (i.e. of urban poor within the city. Where possible, the analysis with a monthly household income of 139592 Mongolian is strengthened by qualitative findings to understand why Tugriks). While the proportion of income non-poor individuals some of the patterns emerge or affect the lived experiences who are simultaneously deprived in at least 30% of non- of deprivation for urban poor. monetary dimensions was 33.8% in Panel A, in Panel B this proportion declines to 8.7%. Panel B further shows that 35 In other words, deprived in at least 7 other dimensions beyond income since income is accounted for in the matrix. 52 53 Urban Poverty in Ulaanbaatar: The Importance of Non-Monetary Dimensions ULAANBAATAR MAP 7. ULAN POVERTY BATUR POVERTY HEADCOUNT HEADCOUNT AND AND SCHOOL SCHOOL ACCESSIBILITY ACCESSIBILITY Figure 4 Poverty headcount and centralized water access PERCENTAGE OF POPULATION PER KHOOROO WITH ACCESS TO WATER Source: Census 2010 data 9.1 URBAN POOR’S ACCESS If we look at streetlight as an approximate proxy for economic 9.2 URBAN POOR ARE available by NSO), areas with low access to centralized prosperity, Khoroos with high poverty headcount rate are left water systems (piped access or kiosk linked with centralized TO MUNICIPAL SERVICES out; SH22, SH20, SH01, SH24, KH13, and KH14 (2010 Khoroo UNDERSERVICED IN TERMS water pipeline) are those with high poverty headcounts The World Bank Service Delivery Report (2016) highlights names) are good examples. Following the same rational, rich OF WATER DELIVERY (greater than 30% -e.g. SH22, SH31, SH20, KH14, KH12, SB20, the presence of spatial inequality in the access to services neighborhoods are also the most lit; Bayangol district is the SB19). While overall majority of khoroos with high poverty As shown in the previous sections, access to water supply for populations living in different parts of Ulaanbaatar most lit district and has the lowest poverty rate of all districts headcount ratio lack access to centralized water, there is a key component of multidimensional deprivation. The due to the topography and the continuous urban sprawl in with 10%. also seem to be some khoroos with poverty headcount ratio results in Figure 4 further show that lack of access to water Ulaanbaatar over the last decade, which has made a rapid greater than 20% that have good centralized water access affects urban poor disproportionately. expansion of services difficult. The findings of the present e.g. SH10 or KH04. report suggest that the burden of lack of access to services is more significant for urban poor. In other words, the spatial When compared with poverty headcount based on Poverty inequality in distribution of services and social infrastructure These findings are supported by the World Bank 2014 Survey Mapping using Census 2010 and 2011 HSES 2011 data (made affects urban poor disproportionately, in turn lowering their which finds that only 19 percent households in the lowest access to opportunities and a reasonable quality of life. 54 55 Urban Poverty in Ulaanbaatar: The Importance of Non-Monetary Dimensions income quintile have piped water, as compared to more supply to its residents, this is an area where investments by 9.3 BURDEN OF LACK OF quintile lack access to improved sanitation as compared than half of those in the highest quintile (Table 14). More the city will not only reduce overall multidimensional poverty to 43% of the richest quintile. While these data suggest importantly, households in the lowest quintile are more but can also improve the quality of life for urban poor. ACCESS TO SANITATION that poorer households face significant barriers to basic dependent on water delivery to kiosks via trucks. Kiosks served However, in order to do so, the city urgently needs to address IS DISPROPORTIONATE water and sanitation services, they also signal that some by trucks are known to face severe water shortages during urban sprawl and increase the density of population in areas households who are income non-poor are also underserved winter and rainy months when the heavy vehicles cannot close to the existing infrastructure while at the same time FOR URBAN POOR by the existing networks for service provision. This is not access the areas in the steeper topography, particularly planning future infrastructure investments in a manner that Map 8 represents the probability of access to sanitation an entirely surprising result given that lack of access to in fringe ger areas that lack proper roads. Given that takes into account population distribution across the city, measures by access to flush toilets. Overall only parts services also has a spatial dimension. Almost 100% of Ulaanbaatar City is responsible for the provision of water including interventions to improve access for urban poor. around downtown are more than 70% likely to have access apartment dwellers (majority of whom are close to the to sanitation. The spatial pattern of access to sanitation also city center) have flush toilets, and most ger dwellers utilize Table 14 Access to water by quintile follows the poverty headcount ratio map and emphasizing unimproved sanitation facilities (including open pit latrines that the level of access to sanitation positively correlated and open defecation). Both networked services of water and POVERTY QUINTILES with level of income. Results from the World Bank Service sanitation, spatial location within the city and the dwelling Delivery Survey also support the spatial results. With respect type is highly relevant.   1 (LOWEST) 2 3 4 5 (HIGHEST) to urban poor, 77% of household in the poorest income WATER DELIVERY Piped water 19 28 31 42 54 MAP 8. ACCESS TO SANITATION SERVICES Kiosk connected to central pipeline 24 22 23 20 15 Tubewell & other 19 17 18 13 11 Kiosk with truck delivery 37 33 28 25 20 Table 15 Access to Improved Sanitation Facilities POVERTY QUINTILES   1 (LOWEST) 2 3 4 5 (HIGHEST) ACCESS TO TOILET FACILITY Flush 19 28 31 41 54 Improved latrine 3 3 2 3 3 Unimproved or none 78 69 67 56 43 56 57 Urban Poverty in Ulaanbaatar: The Importance of Non-Monetary Dimensions MAP 9. ACCESS TO GARBAGE PICKUP Table 16 Frequency of garbage collection by Income Quintiles and Location in the City POVERTY QUINTILES LOCATION IN THE CITY PERI- 1 5 MID-TIER NON- URBAN CENTRAL   2 3 4 GER GER (LOWEST) (HIGHEST) GER GER AREAS AREA AREA FREQUENCY OF GARBAGE COLLECTION Multiple times per week 12 20 18 24 29 16 5 10 40 Multiple times per month 32 31 34 30 28 14 29 42 31 Once a month 35 33 26 24 20 32 44 38 8 Irregular or not collected 21 16 22 22 23 38 21 11 21 9.5 URBAN POOR BEAR provide healthcare services free of charge to the residents, temporary and non-residents without discrimination40. There DISPROPORTIONATE are district hospitals in Bayanzurkh, Khan-Uul, Sukhbaatar, BURDEN OF and Songinokhairkhan districts. The Ulaanbaatar Service Delivery Report 2016 highlights spatial inequality in the DYSFUNCTIONAL HEALTH distribution of healthcare facilities, in particular the family CARE SYSTEM clinics, which are concentrated in the central city area. Map 3.5 (see Annex 2) shows the residential areas in which As discussed in detail in the Ulaanbaatar Service Delivery health clinics are located within 30 minute-or-less walking 9.4 LACK OF EFFECTIVE While more non-poor than poor have the highest frequency Report, public healthcare services and facilities are distance. The concentration of family clinics in and around of garbage collection, both rich and poor segments are both managed by the city government. Funded by the state SOLID WASTE likely to report similar levels of infrequent collections (Table budget, public hospitals36 in UB consist of four categories: the central city area has left large parts of fringe and remote MANAGEMENT IS AN 16). The findings resonate with the spatial distribution of the 1) district hospitals and health centers37; 2) maternity residential areas un-serviced, which are also the areas with probability of garbage pickup/access in neighborhoods higher incidence of poverty. More importantly, the report hospitals38; 3) specialized health centers39; and 4) family IMPORTANT DIMENSION that was developed based on the georeferenced survey goes on to highlight deeper problems in the quality of clinics. The family clinics are under the Municipal Health OF MULTIDIMENSIONAL data (Map 9). The spatial pattern does not clearly map Agency, while the other three types of facilities are under the public health services available to Ulaanbaatar’s residents, onto poverty but low levels of garbage pick-up seem to be including overcrowding of health facilities, blatant neglect DEPRIVATION FOR URBAN Ministry of Health. The largest segment of coverage is family of patients by health care workers and rampant corruption correlated with peripheries of the city. In fact, nearly 40 clinics or household health centers; and there are 130 such POOR AND NON-POOR percent of ger residents in peri-urban areas reported lack of facilities throughout the city. These clinics are mandated to within the health system. garbage collection or irregular collection despite payment of the municipal fee, as compared to only 21 percent of 36 apartment area residents and 11 percent of central ger The list of all municipal health service providers are available on www.ubhealth.mn 37 residents (Table 16). The district health centers are outpatient clinics that provide preventive, diagnosis, treatment services, while the district hospitals are in-patient clinics. 38 There are three maternity hospitals, which provide services related to pregnancy, delivery, post-delivery and infants. Pregnancy control services are provided mostly through district health centers. 39 Includes the emergency center, addiction treatment facility, rehabilitation center, dentistry & maxillofacial center and a clinic serving vulnerable groups. 40 According to Government Resolution No. 364 of 2011 on Comprehensive Regulation on Household Health Centers, item 4.1 & 4.3. 58 59 Urban Poverty in Ulaanbaatar: The Importance of Non-Monetary Dimensions While these issues affect all users of the public healthcare Majority of study participants from central as well as ger BOX 7: IMPACTS OF CORRUPTION WITHIN HEALTH SERVICES ON system, their impact is the highest on urban poor who are areas rated that the quality of care at the family clinics URBAN POOR less able to access private hospitals given their economic below the average. Using the scale of 1 to 10, majority of constraints. Mistreatment of urban poor by hospital staff participants scored family clinics in Ulaanbaatar below looking for bribes is also more severe given their relative average (ranging from group averages of 2.5 to 4.5), inability to pay the bribes or access influential connections with 10 being the best and 1 being the worst. Beyond the “One must to have friends and acquaintances at district hospital. I have 2 friends from our to expedite access. Maternal health services seem to be experience of corruption and its link with negligent care, a province who work there. I call them before I go to the district hospital. I put MNT 5,000 in the particularly notorious for corruption practices, including key reason for low rating was a strong perception of poor pocket of my friend. Then I follow her around in the hospital and able to visit all doctors. This is how giving money to specialists and other staff at each phase ability and qualifications of doctors working in family clinics, I manage everything.” of obstetric care. Tip amounts for medical service providers particularly with regard to diagnostic skills. As a result, some Female, central ger area in Bayanzurkh district. mentioned during the FGDs included: MNT 5,000 for medical urban poor opt to pay higher costs at private hospitals doctors; MNT 50,000-100,000 for head of doctors’s team that instead of accessing the public health system where they helps pregnant women give birth; MNT 20,000 for nurses; and have to deal with long queues, waiting times and perceived MNT 10,000 for service staff. According to both male and low ability of health workers. Delays in accessing health care female respondents from the focus group, the lack of such were emphasized by all focus group participants, who linked “I needed to be admitted to the hospital but had to to wait for about a month in order to stay in “gifts” may result in poor treatment or negligent care (World such chronic delays as having detrimental consequences for hospital for treatment. Hospital which is supposed to provide 10 days of treatment only allows a Bank Focus Groups 2014). Urban poor who are unable to their health outcomes. Box 8 provides some narratives that week of treatment and sends off patient. Poor people are poorly treated like this. Rich people pay adequately are likely to be neglected, often leading to emphasize this point. are treated for 2 weeks and given more than 10 injections. Such people with means and wealth negative health outcomes, as highlighted in Box 7. and can give things to doctors are provided with high quality treatment and treated for 14 days.” Female, central ger area in Bayanzurkh district. BOX 8: DELAYS IN ACCESSING HEALTHCARE “I shirk from the wrong diagnosis of the family clinic and 2 to 3 month queue at the district health center so instead of going there, a sick person goes to private hospitals and pays money. Nowadays, one will live healthy if got money or there’s no other way to live healthy.” “I lost my baby while giving birth for the first time in the Maternity Hospital no1. I heard that Female, central ger area, BZD 12. pregnant women usually gave 100’000 tugrik for bribe. I didn’t give that money directly so i was neglected. I was in pain continuously for 10 hours and was scolded for screaming.” “There’s an old saying: a Mongolian person reaches out to a hospital before dying. What it means is Female, central ger area in Chingeltei district. that the person has been trying to access a hospital from the beginning after getting sick and gets necessary treatment just one day before dying. It’s not that the person didn’t pay attention to his/her health. It happens from queuing.” Male, central ger area, BGD 16. “I spent 10 days to get treatment. There was one guy whose condition worsened and was admitted “To get hospitalized, one waits for at least 20 days. [They] say-wait till the beds become available or you can to intensive care after being sent home twice due to the bed unavailability. Money issues are get hospitalized in the corridor. One puts a mattress in the corridor to get hospitalized. It’s difficult to stay in the explicit… Those with social connections receive treatments while those without social connections corridor. I stayed 7 days at home after getting a queue for hospitalization and then I went to the emergency unit are left behind, like people who are always ready to die.” for treatments for 12 days. I stayed in the hospital for 11 days as my condition was serious and I was released. Male, fringe ger area in Sukhbaatar district. People like me can’t get a complete treatment. There’s a severe overcrowding and this happens only in ger areas.” Male, Central ger area in Bayangol district. 60 61 Urban Poverty in Ulaanbaatar: The Importance of Non-Monetary Dimensions 9.6 LOWER INTERNET likely to have Internet access than households farther away 9.7 NEIGHBORHOOD to social interactions. Men from fringe ger areas spoke of from the downtown area. Songinokhairkhan, and Khan- various meeting points such as street corners, community ACCESS AMONG URBAN Uul households are the least likely to have Internet access RELATIONS IN GER exercise area, water kiosks, bus stop and corner shops as POOR IS LIKELY TO (particularly in SH22, SH25, SH08, SH07, SH23, SH22, NEIGHBORHOODS meeting points. Female residents perceived more limitations SH32 and KH04, KH08, KH12, KH13, KH14). Households on in areas for social interaction and community gatherings EXCLUDE THEM FROM the outskirts of Bayanzurkh and Sukhbaatar and Chingeltei World Bank Survey 2014 shows that majority of respondents in the neighborhoods. In contrast, apartment dwellers in both ger areas and apartment areas report a low level UB’S E-GOVERNANCE are also less likely to have Internet access. It is important spoke more positively about the ability of public spaces for of familiarity/interaction with their neighbors with levels of to note that inequality is not only between districts e.g. socialization within neighborhoods. The disparity in access INITIATIVES familiarity being particularly low within apartment areas. Songinakhairkhan and Bayangol but also between khoroos to public spaces within poor versus wealthier neighborhoods A high proportion of households in Ulaanbaatar are without 68% of households in apartment strata claim that they either within the same district (e.g. KH11 vs KH14). When these is likely to lower the formation of strong social capital and access to internet (44.0%) in the whole city. Within Ger do not know their neighbors at all or do not know them findings are examined in relation to poverty, we observe social cohesion in poorer neighborhoods. areas 66% of households lack internet access (World Bank very well, as compared to 54% of households in ger strata. that khoroos with very high poverty headcount (>30%) are Survey 2014). While the relative contribution of lack of Within ger areas, fringe ger households report slightly higher highly correlated with low Internet access (e.g. SH32, SH22, internet access to overall multidimensional poverty, it is levels of familiarity with neighbors as compared central and Such lack of strong social networks (especially among SH01, SH10, KH12, KH14, KH13). Similarly there is a clear link important to note that urban poor are disproportionately middle ger areas households respectively. However, a large urban poor communities) not only implies low levels of between khoroos with low poverty headcount (<5%) and high affected. Map 10 represents a household’s the probability proportion of respondents across all there strata report low community cohesion and a lack of availability of social internet access, particularly in the downtown area. Given of having internet access based on the World Bank 2014 or negligent familiarity with neighbors at 45% in fringe, 57% support from neighbors, but also represents a low level of that UB is increasingly focusing on e-governance initiatives, Survey data (see Annex for methodology of poverty maps). in middle and 53% in central ger areas. Respondents within collective capacity within communities to organize efforts access to internet will be critical for ensuring the inclusion of The clear spatial pattern emerging in this map suggests focus group discussions for fringe ger residents who tend to on issues affecting their neighborhoods. It is not surprising urban poor in Ulaanbaatar’s drive to increasing efficiency of that households closest to the downtown areas are more be poorer, were more likely to report neighborly interactions, that the World Bank 2014 Survey data shows a conspicuous service delivery. especially among women who engage with other women absence of community groups that enable residents within MAP 10. INTERNET ACCESS IN ULAANBAATAR in the community while strolling with their babies, helping neighborhoods to come together to address issues of each other with household tasks and discussing issues of common concern across Ulaanbaatar, particularly absent in common concern. In contrast, neither men nor women FGD fringe ger area neighborhoods. Less than 10% of households participants from ger areas closer to the city center report surveyed in fringe ger areas have any household member low levels of social interaction, citing high neighborhood participating in a civil society group or neighborhood turnover and busy working hours. Focus group participants association (compared to 42% of apartment area residents). who reported low familiarity or interaction with neighbors Low levels of collective capacity among urban poor in turn in all groups showed low relations of reciprocity within the means that these citizens cannot effectively participate community, including rarely providing or receiving support within the democratic local governance structures and from their neighbors41. These respondents explained the influence municipal decision-making on important issues absence of community support and close ties in terms of such as khoroo -level investment priorities. Even after lack of trust within the context of urban life in Ulaanbaatar. controlling for respondent’s age, gender, education and They also reported that they largely turn to their relatives employment in a logistic regression model, we find a strong followed by friends and colleagues in the absence of close statistically significant difference in the odds of membership community ties. to community organizations between central versus other ger area locations. Odds of fringe and middle ger households having membership to any community organization are 47% Ger area residents, especially from ger areas close to the and 42% lower as compared to central ger residents42. city, emphasized the lack of community spaces as a barrier 41Two or three respondents in both male and female groups in the fringe ger areas and female groups in the mid-tier ger areas spoke of lack of community supports whilst most of or half of the respondents in male groups in mid-tier ger areas and apartment areas reported the same. 62 63 Urban Poverty in Ulaanbaatar: The Importance of Non-Monetary Dimensions However, while there is a lack of organized civil society in ger The beginning of rampant alcoholism in Mongolia can be vulnerable group in the city. It is difficult to say, whether areas of Ulaanbaatar residents in focus groups mentioned traced back to the 1990’s, coinciding with the country’s alcoholism induces poverty and vulnerability, or whether it several instances of households coming together informally abrupt transition from a socialist to a capitalist system. A is desperation that encourages alcoholism. As it is pointed to work together for a common neighborhood concern. This simultaneous shift from a dominant nomadic system in rural out in the UNDP 2007 report, “the causative link between was emphasized more by participants in fringe ger areas, areas to an increasing need to adapt to a market economy unemployment, poverty and alcoholism is complex, and where respondents perceive low availability of support from spurred in new socio-cultural norms. The effects of these these studies cannot be considered proof that alcoholism the city. For example, fringe ger residents reported several transformations on the country’s social fabric have been is causing poverty and unemployment or vice versa.” (UNDP, instances of organizing themselves informal self-help groups profound and the capacity of this historically relatively 2007: 46)44 . to voluntarily help neighbors in assembling gers, putting up isolated population to cope with the changes have been fences, growing vegetables, maintaining alleys (e.g. cleaning limited (Armstrong et al, 2007)43 . These changes, along alleys and illegal waste dump areas in their neighborhood) with the relatively uncontrolled access to potent alcohol and addressing other issues of common concern (e.g. at cheap prices, have placed a large number of people at covering stagnant water with black earth) etc. risk of becoming substance dependent. At the same time, anthropologists argue that social norms and symbols of MAP 11. PERCEIVED ALCOHOLISM WITHIN COMMUNITIES masculinity embedded into the transformation processes, put pressure on the male population to over consume the 9.8 SALIENCE OF substance. Alcoholism is considered a major obstacle ALCOHOLISM IN to employability of many Mongolians, especially poor households for whom alcohol abuse is contributing to a ULAANBAATAR’S vicious poverty trap (Mongolia Human Development Report, NEIGHBORHOODS AND 2007). Unemployment, alcoholism and poverty mutually LINKAGES WITH POVERTY reinforce each other, as shown by the in-depth ethnographic work in the community living in Dalan Davkhar (or the Alcoholism in Mongolia is pervasive and considered Sambalkhundev’s Cemetery) that was commissioned for the a national problem. Factors contributing to the high purposes of this report (See Box 8). levels of alcoholism are manifold, particularly related to unemployment. Growing up in a society with a higher level of alcohol abuse, a number of children, particularly boys, In this location, alcoholism is a feature of the poorest have been reported to start consuming alcohol at an early households. People addicted to alcohol typically survive by rate, with underage drinking on the rise. Findings of the doing menial labor jobs at the market such as cleaning sheep World Bank 2014 survey suggest that alcoholism is a citywide heads or collecting plastic bottles and other recyclable problem in Ulaanbaatar. The map represents the probability garbage materials from the dumpsite or by wandering of perceiving alcoholism as a problem within the community throughout the city in search of such materials with large based on the responses of survey respondents. It highlights sacs on their backs. Their daily income fluctuates around that alcoholism affects communities across the city and from MNT 5-20’000 per day, with most of it is spent on does not present a particular spatial pattern and suggests alcohol. Children born to such parents are often not enrolled that policies addressing alcoholism should be coordinated in schools, exploited for child labor (collecting garbage, among districts rather than being spatially targeted. carrying loads at the market, washing cars etc.) or engaged in begging on the street Street Children form a particularly 42 Calculations by the World Bank team using World Bank 2014 Survey data 44Mongolia, and United Nations Development Programme (Mongolia). 2007. Mongolia human development report. http://www.mn.undp.org/content/ 43 Armstrong, Sean (2007): Alcoholism in Mongolia: Social norms or cultural beliefs? mongolia/en/home/library/National-Human-Development-Reports.html. 64 65 Urban Poverty in Ulaanbaatar: The Importance of Non-Monetary Dimensions BOX 9: ALCOHOLISM IN THE CONTEXT OF POVERTY AND VULNERABILITY Chimgee is an alcohol addict and lives alone. She has been drinking for about a decade. Chimgee is one of the many people who failed to adapt to the structural changes in Mongolia. Chimgee lost her job at the outset of the economic crisis that accompanied the transition to a market economy when the factory she was working in was closed. She is divorced from her husband, and her only child, a son, lives in the city with her husband’s family. Chimgee’s sister’s family tends the few livestock she owns in the countryside and sends her a little bit of meat and dairy for consumption. When asked why her people are becoming alcohol addicts, Chimgee answered, “Alcoholic people are depressed. The main challenge is finding a job. When there are full-time jobs, they often don’t pay the small wages (MNT 200,000 or 300,000 per month) they promised or after working many days on probation, only pay the half so that’s why young people don’t want to be employed by companies. Instead they can earn better money (daily around 20’000 tugriks) by working informally at the market, such as washing and burning sheep heads, deboning meat, or carrying loads by wheelbarrow, but for an old person like me it is impossible to find a job.” Then she added, “I don’t want to blame our government for everything, but, it must be not too hard to provide jobs for only 3 million people. There are countries with more than 50, 60 million people where all their citizens have jobs so it must not be so impossible given that our country is livestock and resource rich”. She also mentioned that neither she nor her neighbors receive(s) any welfare support from the government. There are no extended social institutions or social services the challenges of alcoholism, particularly since the problems specifically targeting alcoholics. During winter, police in associated with alcoholism are felt firsthand at local level. Ulaanbaatar often have to resort to gathering up alcoholics Local governments are well placed to respond to the in sobering-cells to prevent them from freezing or attacking problems related to alcohol as local institutions are already others. Public health professionals have criticized the lack responsible of the delivery of various social services (e.g. of funding for national program to address this issue. The health, education and welfare) to their people. They are alcohol industry has been far more effective in promoting likely to have the most precise and relevant information alcohol consumption than public health campaigns in and data on their respective constituents. The local policy highlighting the “hazards” of alcohol45. and actions on alcoholism can complement the national efforts to address alcoholism and may have a significant With respect to addressing the problem of alcoholism, role between the national level policies (e.g. excise tax) and examples of interventions from other countries can be individual level actions (e.g. treatment). sought. Countries around the world have increasingly emphasized the role of the local governments in addressing 45 Armstrong, Sean & Tsogtbaatar, Byamba (2010): The dual nature of alcohol use and abuse in Mongolia: Reflections through policy. Asia Pacific Journal Public Health, Volume 22, no. 3, pp. 209-215 66 67 Urban Poverty in Ulaanbaatar: The Importance of Non-Monetary Dimensions BOX 10: THE ROLE OF LOCAL GOVERNMENTS IN ADDRESSING ALCOHOLISM: CASE OF AUSTRALIA In Australia, the role of local governments in addressing the challenges of alcoholism is greatly recognized by the national, state and local governments. The National Alcohol Strategy 2006-2011: Towards Safer Drinking Cultures1 reports that Australian local governments recognize the alcohol issues as the most significant drug related issues they have to deal with. The local governments have both leading and supportive roles in developing and implementing alcohol policies and actions. They often initiate and manage various programs on alcohol and facilitate the coordinated responses of local service providers towards the alcohol related problems in their communities. The National Drug Strategy2, which is an overarching framework for the actions on alcohol in Australia, highlights the relationship between socio- economic disadvantages and alcohol abuse and thus, the importance of integrated service approaches. The local actions on addressing the challenges of alcoholism in Australia range from the development and implementation of local alcohol policies and establishment of the local liquor accords to specific and broad programs on alcoholism. Adoptions of local alcohol policies are a common practice across Australia. For example, the cities of Gosnells and Armadale in Western Australia adopted a comprehensive alcohol policy and management plan in 2000 after forging a partnership involving their police and health centers to develop strategies to address alcoholism and related social problems in 1999. Another way of addressing alcoholism involves creation of local alcohol action groups, which include the representatives from concerned organizations such as social welfare, treatment center, health, police and local government authorities. Such action groups or teams set up in New South Wales work towards informed and coordinated alcohol policies and actions. The policies and action groups on alcohol provide a framework of the development and implementation of the programs on alcoholism. One example of alcohol action program is the Neighborhood Renewal Program in Victoria, which addresses the alcohol related problems through increased opportunities for the disadvantaged communities. The program involved the Department of Human Services, Victoria state government, and local governments and focused on the socio-economic conditions in specific neighborhoods or towns to reduce social problems including alcohol abuse. The program promoted the across government approaches to planning and service delivery and provided a range of opportunities for training, job creation, health promotion and community safety. On the other hand, the local governments also work with the local liquor licensees, police, and health centers to promote the responsible service policies, compliance with the national and local legislations and regulations on alcohol sales in order to reduce alcohol related incidents and hospitalizations. One example of such collaboration is the Maitland Liquor Accord in Maitland city in New South Wales. The Accord has a membership of 90 percent of local licensees and implemented public education campaigns on alcohol in addition to the responsible service programs. Source: Trevor King and Jennifer Richards, Australian Local Government: Alcohol Harm Minimization Projects 2004, accessible at http://www.ihra.net/ files/2011/07/21/04.6_King_-_Australian_Local_Government_-_Alcohol_Harm_Min_._Projects_.pdf Note: 1 National Alcohol Strategy 2006-2011: Towards Safer Drinking Cultures, accessible at http://webarchive.nla.gov.au/gov/20140801013918/http:// www.alcohol.gov.au/internet/alcohol/publishing.nsf/Content/nas-06-09 2 National Drug Strategy 2010-2015, accessible at http://www.nationaldrugstrategy.gov.au/internet/drugstrategy/Publishing.nsf/content/nds2015 69 Urban Poverty in Ulaanbaatar: Marginalized Groups 10.0 MARGINALIZED GROUPS The discussion so far has focused on monetary and 10.1 RURAL-URBAN MIGRANTS: effort a family puts into the education of a child by sending Further investigation suggests that migrants on the whole are non-monetary dimensions of poverty and inequality in him/her to the capital to pursue higher education, find a not urban poor in terms of income. A linear regression model Ulaanbaatar, namely access to basic infrastructure and NOT POOR BUT good job and earn enough money to later care for the rest for determinants of income poverty shows no significant services. While these dimensions undoubtedly place VULNERABLE of the family. Main factors for migration today are migration association between migrant status and income when limitations on the ability, opportunity and dignity of urban for education and work, migration for medical reasons, controlling for age, gender, education status, employment During socialist times, migration was centrally controlled residents, they do not explicitly highlight the disadvantage migration due to natural catastrophes and the loss of status and residential location in Ulaanbaatar (See Table 18). and urbanization for the recruitment of work forces in that some social groups face based on their identity or unique livestock and livelihoods and the absence of alternatives for The table presents the results of a linear regression model factories was promoted47. The living standard in the sum- social circumstances. Such disadvantage often expresses self-sufficiency and chain migration. Very little information that shows the association between logarithm of income as centers was, as compared to today, relatively high with itself in the form social stigma, which is instrumental in is available on the structures and conditions of the migration the dependent variable with independent variables of age, schools, boarding schools, public transport, newspapers creating social exclusion and vulnerability. World Bank’s 2013 process, the trajectory of migrant adaptation and the gender, migration status, location of residence, employment, and electricity available in every center. In the years after report titled ‘Inclusion Matters’46 emphasizes the importance conditions in which migrants live in Ulaanbaatar. Migrants and education. Of particular note is the absence of significant 1990, living standards and accessibility to services such as of understanding and addressing the dynamics of social are often stigmatized by local residents, seen as the source association between migration status and log income. This is education and health decreased dramatically in the rural exclusion in order to fully address inequality. This section of Ulaanbaatar’s high levels of poverty and held responsible contrary to the popularly held view that the city’s urban poor areas. While a temporary de-urbanization was evident highlights key dynamics in this regard in Ulaanbaatar, with for urban sprawl, though these beliefs are not substantiated populations are comprised of recent migrants. Instead, key between 1990 and 1995, during which the rural population particular emphasis on two groups: (1) rural-urban migrants by the quantitative and ethnographic evidence collected for determinants of income poverty include residential location, grew significantly, migration to urban areas increased after residing adjacent to the Ulaanchuluut landfill site in a this report. employment status and educational attainment. Gender 1995 and even more after 200048. fringe ger area with low levels of service provision; and (2) and age also play a role, with men earning approximately 10 community living near a cemetery called Dalan Davkhar percent higher log income than women. A separate model (also known as the TV District) (Figure 5). The findings in this World Bank 2014 survey finds that a majority of migrants that included migrants only did not show any statistically Contemporary migration streams from rural-urban areas section are based on ethnographic fieldwork commissioned moving to Ulaanbaatar live in ger districts. However, they significant relationship between income and year of arrival. are governed by both economic hardship as well as social under this study, which investigated intersections between are not necessarily concentrated in fringe ger areas as is ambition49. Economic hardship in rural areas had been the concepts of poverty, vulnerability and social exclusion commonly assumed, but distributed across all ger locations exacerbated by the loss of the herds in extremely harsh in two communities. Insights to the experiences of stigma (Table 17). This also holds true for recent migrants who have Underscoring the previous discussion of spatial inequality in winters (dzud50), increased rural unemployment, and faced by these communities highlight the nature of arrived in the last five years. access to services, individuals living in peri-urban and mid- alcoholism. Individual experiences including death of a family exclusionary dynamics that prevent these communities from tier ger areas earn log income approximately 14 percent member, divorce or imprisonment can also push rural families fully integrating into the city’s urban fabric. and 17 percent lower than those living in central ger areas into poverty. Migration due to social ambition describes the while individuals living in non-ger areas have log incomes approximately 12 percent higher relative to those in central Table 17 Distribution of migrants and non-migrants across residential locations gers. This makes sense in the context of spatial distribution RESIDENTIAL LOCATION NON-MIGRANT MIGRANT TOTAL of deprivations, and the unequal provision of services Figure 5 Ulaanchuluut landfill site – in the territory of 26th across residential areas. Unsurprisingly, unemployment has Central ger 22.12 26.03 24.23 khoroo, Songinokhairkhan district the greatest affect on income with individuals who are (Adapted from UB Municipality’s community mapping web Fringe ger 13.2 12.46 12.8 unemployed earning approximately 68 percent lower log page http://manaikhoroo.ub.gov.mn) incomes than those who are employed. Finally, with regards Middle ger 23.57 26.96 25.4 to educational attainment, individuals who attended only middle school or lower, high school, or technical and Apartment area 41.12 34.55 37.57 vocational school see log incomes that are approximately 60 percent, 45 percent and 40 percent lower respectively 46 World Bank (2013), Inclusion Matters: The Foundation forShared Prosperity, Washington,DC. than individuals who attended at least college, signaling 47 Gilberg/Svantesson 1996: p. 20. substantial returns to higher education. 48 Taraschewski 2008: p. 157. 49 Bruun 2006: p. 180. 50 The term “ dzud ” denotes the effect of extreme weather conditions which is the mass death of livestock 70 71 Urban Poverty in Ulaanbaatar: Marginalized Groups Table 18 Determinants of (Log) Income in Ulaanbaatar connected to the central electricity grid until October 2013 need to secure the future of their children and highlight the and thus was called as “харанхуйнхан” or “those in the dark importance of a good education. Other reasons include,   COEF. STD. ERR. area”. Streets still don’t have lighting and some households unavailability of jobs in the countryside, loss of livelihoods Age -0.0254 0.0069 *** interviewed mentioned that households located further to the Zud and hardships in the rural lifestyle. Some still Age squared 0.0003 0.0001 *** still do not have access to electricity. The community also have a small number of livestock being taken care of by their Male (ref: female) 0.1006 0.0343 *** faces problems related to water supply. The nearest water relatives in rural areas. Even though there are schools in Migrant (ref: non-migrant) -0.0442 0.0334 well/kiosk is more than 5 km away; households fetch water every soum (a rural sub-district) in the countryside, parents by cars or ask their neighbors with cars for help. Schools strongly feel that being far from home in rural dormitories RESIDENTIAL LOCATION (REF: CENTRAL GER) and kindergartens are absent in the area and the nearest that are poorly managed is not easy for children. They Peri-urban ger area -0.1367 0.0569 ** school (that is already overcrowded) is 7-8 km away. People further believe that urban schools provide better quality Mid-tier ger area -0.1682 0.0465 *** with cars are able to send their children to schools in other education and have aspirations for their children to go to Non-ger area 0.1161 0.0444 *** khoroos. Given poor road infrastructure and bus service, the universities. If generational aspirations are a key reason EMPLOYMENT TYPE (REF: EMPLOYED) many households own vehicles. The price of a plot of land for undertaking migration, coupled with a persistent lack of is lower compared to other areas of the city, ranging from quality education in rural areas, it is likely that rural-urban Unemployed -0.6802 0.0625 *** MNT 1.5 - 10 million given its peripheral location and it is still migration will continue to increase. Pensioner -0.1552 0.0757 ** possible to occupy a vacant plot of land for few months and Disabled -0.6061 0.0986 *** then apply for the land certificate to obtain the title free Homemaker -0.2996 0.0755 *** of charge. Given the lack of affordable housing available Other -0.6849 0.1012 *** closer to the city, this area becomes an attractive option for 10.2 CHALLENGES RELATED TO EDUCATION LEVEL (REF: COLLEGE AND ABOVE) households to arrive and settle in the community despite the OBTAINING A PLOT IN THE lack of services. Middle school or below -0.6021 0.0566 *** FRINGE High School -0.4469 0.04 *** Technical and Vocational Training -0.3947 0.0594 *** Over time, newer households have come to depend on URBAN REGISTRATION, SOCIAL NETWORKS AND Observations 2983 the landfill site for their livelihoods, including rural-urban CORRUPT PRACTICES IN LAND OFFICES R2 0.1724 migrants. The majority of households in this area are comprised of adults in their 30s or 40s with small children, While Mongolian nationals are free to move and live anywhere ***, **, and * indicate statistical significance at the 1, 5, and 10 percent level within the country52, citizens migrating from Aimags to UB who have arrived from rural areas over the last decade. While migrants are not the urban poor in terms of income, The garbage collectors work in unsanitary conditions of the Most migrant households reported that they stayed in plots are required to transfer their registration status from rural it should be noted that 60% of migrant households are garbage dump every day. They are often harassed by the owned by their relatives or acquaintances before moving to urban residence and register their city address with the multidimensional poor as compared to 48% non-migrant police and subject to high levels of crime in and around to acquire their own khashaa plots. These households municipal authorities. The residential address registry is the households. In order to better understand migrants’ urban the garbage dump. Many of the newcomers, especially were marked by their industriousness- some had full- basis on which basic public services including land titling, experiences and the mechanisms that increase their rural-urban migrants, are not registered in Ulaanbaatar time formal employment (such as seamstress at factory or pre-school, primary and secondary education and health vulnerability, the ethnographic study focused on a migrant and therefore unable to access all urban services such as driver at a private enterprise) while others had informal or services are provided to all residents in UB city. The Civil community living in the fringe ger areas close to city’s largest schools, health care, social welfare etc. seasonal employment (such as taxi driver or construction Registration Law (Article 25 and 26) requires the citizens landfill site. The Ulaanchuluut’s 19-hectare landfill was worker). Some households run garbage businesses: they moving from one administrative unit to another for more established in 1975 in the western end of Ulaanbaatar city collect scrap metals and other materials from the dumpsite, than 180 day or for good to be registered to the concerned (Figure 5). In 2012, 3110 households (with 12051 people) were The fringe location of this areas also translates into load the scraps on their porter cars and deliver them to administrative units by their residential address. However, the registered with 26th khoroo in Songinokhairkhan district insufficient access to infrastructure and services. Many middlemen who export them to China. When asked about process of registering residency for rural migrants moving to which is the territory including Ulaanchuluut’s dump area51. households have built large detached houses with red their reasons for moving into the city, they emphasize the UB city requires them to complete a set of fragmented yet or other colorful roofs on their plots. The area was not complicated procedures and may place burden and cost 51 52 http://acfmongolia.mn/index.php/mn/90-26 Article 16.18, the Constitution of Mongolia. 72 73 Urban Poverty in Ulaanbaatar: Marginalized Groups This office provides an official form called Civil State The availability of social networks plays an important role in BOX 11: PROFILE OF A RURAL-URBAN MIGRANT Registration Form 22. After this, the citizens are required facilitating urban residency and eligibility for land titling in to visit the State Registrar at the local Khoroo Office in the city. Rural-urban migrants often rely on their connections UB city to submit the required documentation56. Lastly, and acquaintances by registering their residency with the migrants submit the form together with other required households that already have urban residential status. With documentations57 to the Civil Registration Office in UB city, the urban residency status, migrants can access land-titling which validates the registration of a new residential address services for available plots, which are mostly located in the and provides a document known as “address change slip” fringe or undesirable land (hilly or mountainous areas) with to the migrants as soon as the deletion of registry is entered limited access to services. However, rural-urban migrants and confirmed in the online Integrated System of Civil may also suffer from the corrupt practices of service Registration. providers (duplicate land allocation, cash and in-kind bribe demands and intentional provision of false information on the availability of the vacant plots and their future planning) The total cost58 required for this process per person is MNT when they attempt to obtain the titles on the vacant plots Bayabaatar is an ex-herder who moved to UB with his family in 2002 to work abroad but lost his money to 12,000. While the process is expected to take 14 days, local in the fringe. Overall, lengthy administrative procedures and a fraudster. His family lived in his uncle’s yard in a nearby district for several years and obtained registered experience is that it can take up to a month or more. Even requirements for urban residency status and land-titling city residency address. He occupied the plot in the community in 2012 when electricity was unavailable when the procedure of changing registration is followed coupled with both a lack of social networks and corrupt and fenced it. Last year, with the arrival of the electricity, the family built an 8x8 m2 house. correctly, a migrant effectively has no access to public practices of service providers leave rural-urban migrants Bayabaatar has primary education and he is one of many informal cab drivers in the city and has aspirations services in their place of origin or destination throughout the in a precarious situation with regards to land registration. for his children to have a bright future. Every day he drives his two children to a school in the city center registration process due to the initial step of deletion of her The consequences vary from denied access to loss of about 15 km away. His wife Uli works for a cashmere factory as a seamstress and earns a regular income Soum registration. This can increase migrants’ vulnerability opportunities or other economic costs, such as the payment that provides stability and allows the family to plan their expenses and build a life in the city. given the absence of any safety net. of demanded bribes. Figure 6 Administrative procedures for obtaining urban residency status on the citizens. The process of transferring residency from an for obtaining urban residency status, involves various administrative unit outside UB city to a khoroo and required public offices including Bag and Soum Governors’ Offices, 1. Obtain circulation sheet at Bag Governor's Office 2. Obtain signatures and stamps from 10 different organizations/authorities procedures are shown in the Figure 6. Khoroo Office and Capital City Civil Registration Office53. 3. After Bag Governor signs circulation sheet a citizen (s) is (are) deleted from Registry and provided First, migrants are required visit Bag Governor’s Office to BAG with an Affirmation Document 4. Visit State Registrar at Soum Governor's Office obtain a circulation sheet54. When the circulation sheet has with required documents As a result, despite the availability and affordability of the the signatures of all relevant organizations/authorities, the 6. Visit State Registrar at concerned Khoroo office plots in the fringe of the city, obtaining and registering a plot Governor of Bag deletes the previous registry of migrants KHOROO and present required documents 5. State Registrar at Soum Governor's Office is challenging for the rural-urban migrants. Access to land- and provides a document affirming the deletion of registry. issues Civil State Registration Form-22 titling services is tied to urban residency status and provided Then, the migrants are required to visit the State Registrar at 7. Visit Registration office of Capital City and submit at the respective khoroo or district facilities. The process the Soum Governor’s Office with the required documents55. required documents to State Registrars 8. Deletion of Registry at sending soum is entered 9. State Registrar at Capital City Office validates a and confirmed in online Integrated System of Civil UB CITY slip with new address Registration SOUM 53 It should be noted that the migration within UB or among Districts is dealt by the District Civil Registration Departments. 54 The document lists eleven organizations or authorities including the Governor of Bag and the citizens are required to obtain signature and stamp 56 from each organization/authority. These include Banks, Financial Department, Taxation Office, Environment Department, Police, Land Office, Social Needed documentations are: Civil State Registration Form 22; ID Card and a birth certificate of a child under age of 16 in case there is an Insurance, Energy, and Military Headquarter present in the Soum. After obtaining the clearances or signatures from all these organizations and accompanying child; and documents of relatives or parents who are the address holders or a document certifying the ownership of the residential authorities, the Governor of a Bag signs and verifies the circulation sheet. property (apartment or khashaa plot etc.) 57 55 Needed documentations are: ID and its copy; the affirmation document from the Bag Governor; Service Fee of MNT 5’000 payable to Civil At this stage, the needed documentations include: migration registration form 22; MNT 6,000 fee payable to a bank account; and a photograph Registration Department account at a commercial bank and and stamp duty of MNT 1’000 payable to the State Bank; one photograph (3X4); and a (size 3X4). birth certificate of a child under age of 16 and its copy in case there is an accompanying child. 58 Please note that this amount doesn’t take into account any other costs such travel cost that a migrant may incur. 74 75 Urban Poverty in Ulaanbaatar: Marginalized Groups BOX 12: NARRATIVES OF VULNERABILITY There are about ten households that survive by collecting garbage in the Ulaanchuluut landfill. The kheseg Another example is Tsatsraltuya, a woman in her 40s, who completed 8th grade and scavenges daily to leader59 referred them as the most vulnerable households but spoke disparagingly about the causes of support her family of alcoholic husband (who also scavenges) and 4 kids. Her family moved from Bornuur their vulnerability as laziness and alcoholism, which was contrary to the perceptions of interviewees and soum, Tuv province, where they were herders, to UB city 8 years ago and finally settled in the area to have the in-depth interview narratives. According to the kheseg leader, the vulnerable households buy alcohol their own plot (Image 2). Her husband does not have ID documents while her children’s ID documents were using the food vouchers they are given. However, the individual cases mentioned below demonstrate erroneous. This has prevented her family from receiving social welfare benefits such as child allowance, that these households suffer from multiple dimensions of vulnerability and are exposed to exclusion food voucher and others targeted for extremely poor households. She spoke of burdensome process for mechanisms while lacking the support of an urban networks or family base in the countryside that are soliciting the food vouchers and child allowance. Her family could not get a food voucher after soliciting vital to the resilience and success of migrants living in the city . it for four months. She also visited the social welfare office (at district) twice to meet the officer in charge of children related issues to solicit the child allowance but the officer was not available at each time, attending some other sessions. She finds frequent visits to the social welfare office burdensome One example is an old widow named Dorj, in his 70’s, an ex-herder who lost his all livestock during dzud and costly (to go the office, she needs 2 bus transfers and 2’000 tugriks for the bus fares, which is great and came to city from Khujirt soum, Uvurkhangai province several years ago. His stepchildren neglect him amount of money for her). Her family uses sacks as a fuel to heat the ger in cold and currently survives while his son, who is a driver and stays with his in-laws, couldn’t transfer his rural residential address to on the income from scavenging. She spoke of her poor health, particularly her leg illness that prevents the city due to the complicated procedures on obtaining clearances, particularly in relation to the military her from formal employment such as working as a janitress and she finds scavenging gainful. She says enlistment record, in Khujirt soum. Dorj occupied an empty plot of land in the area and built a small shelter “about 1’000 people work at dumpsite, daily 200-300 people are there, children come to work there after with scrap materials collected from the dumpsite. As he’s not a registered resident in UB city, he was school; it’s very dangerous for them as they are prone to truck accidents, half of the people working here not eligible to apply for the land title certificate. Despite living on the plot for many years and building a don’t drink”. She mentions that scavengers earn about MNT 5-20’000 per day on average by selling their shelter, he lost the plot as khoroo social worker. Let someone else to reside on the land, planning to take collected refuse materials (plastic cans, bottles and metal) to middlemen who come with porter cars to it for himself later (See Image 1). Now he has no home and wanders from one household to another. “It load them. She wants to build a fence to obtain the land certificate, but she does not have money to buy is all very complicated for me”, the old man says, “I thought I’ll just build a small hut on the land, which fencing materials. She also mentions that her relatives are aloof and it’s easier for her and her family to was not occupied by anybody and live there without bothering anyone but two, three years ago suddenly live in the neighborhood as they are not discriminated and people don’t say “ugh or eww” when they go people started to take lands and build fences around here, I cannot apply for this plot as I don’t have a to stores. All these contribute to her desperate life. city address”. A young wife, Uranbaigali aged 31, has 2 small kids (4 year old and 9 month old). She migrated from Dornod province to UB city when her family lost all their herds during dzud in 2000. She neither has a birth certificate nor an ID document. She had her documents searched from the National Archive but with no success. Perhaps, she was misspelling her name or her name changed after the birth certificate was registered60. She could not receive any welfare assistances such as the pregnancy or child allowances due to the lack of identification documents. She mentioned that her relatives were distant and aloof while she could not enroll her eldest child to a kindergarten due to lack of connection. Her stomach was getting bigger last year so people thought she had liver disease and was going to die but it was discovered later that she was pregnant and she gave birth. Her family does not afford meat, so only buys jijiglen or small piece of organ meats to survive (MNT500 per piece). Her husband, who is a construction worker (finds work Image 1: Old man built the shelter on the right but plot is Image 2: A ger near the landfill site. Neighbors say that the now occupied by another household household members often drink and fight. Two kids, aged 7 and 11 attend the informal Christian school. 59 60 Khoroo is further divided into so-called khesegs or sections. There are about 1168 khesegs in UB city, with each represented by a volunteer activist This happens when parents change the name of their sick child by asking a monk for a more suitable name, assuming that old name was harsh. It who is recognized by the respective khoroo officials as the leader. becomes difficult later to find the birth certificate with original name from the archive when it is forgotten. 76 77 Urban Poverty in Ulaanbaatar: Marginalized Groups only during warm seasons through his connections and when he does, he earns daily MNT25-30’000), drinks and beats her sometimes. She thinks there is nothing to do in the countryside and the city has better opportunities. However, she says that she encounters problems with finding good jobs; only job she could find was dishwashing with a daily wage of MNT 7’000. Image 3: Uranbaigali, aged 31, from Dornod province. This is the ger set up on old man Dorj’s plot. Lastly, a nurse Navchaale aged 45, lives with her husband (who is an auto-technician) and their children. They migrated to UB city from Tuv province in 1996 and settled in the area in 2011 to obtain their own plot after residing on her sister’s plot nearby. She says that her household stayed without electricity for 3 years and that the khoroo activities targeted at vulnerable groups are not felt that much. “There are many people who are not registered to the khoroo but make a living in the dumpsite. They are the people who are fighting for life for real and making their ends meet in the Ulaanchuluut landfill site. ” 79 Urban Poverty in Ulaanbaatar: Community Living in the 11.0 COMMUNITY LIVING Dalan Davkhar Cemetery IN THE DALAN DAVKHAR CEMETERY The Dalan Davkha r cemetery (also known as Sambalkhundev’s The plots usually consist of detached houses and a ger and Unlike the community adjacent to the landfill site, the and the modern apartment residents who often look down Cemetery) was established in the 1930s and is located are shared between 2 households who are often related. households living near the cemetery have been living there and stigmatize them. within 5 km distance from the city center61. The cemetery Most households hold land titles, though a few are renters. since Mongolia’s early years of transition from a socialist The households that are living right next to the cemetery are is surrounded by the ger areas known as Denjiin Myanga Households often have a car. Nearby services include a water to market society. The majority of households in the area not necessarily the poorest. However, they have been widely (hillside thousand households) and Zuragt (TV district - due kiosk within 500 meters as well as a school, kindergarten and include adults in their 30s or 40s with small children and are stigmatized by the society for residing near the cemetery. to its closeness to Mongolian National Radio and Television bus stops all within close walking distance. The price of a engaged in trade or do menial work (cleaning sheep heads Residents from other districts of UB city would describe the Broadcaster) neighborhoods that fall in the territory of plot of land is higher than other areas of the city given the or carrying loads by wheelbarrows at the market)62 at the residence in the area as unbelievably horrific and tell each Khoroos 7, 10, 11 and 12 of Chingeltei district, with each proximity to the city center, and range from MNT 25 - 20 nearby markets. Some households have formal employment other dreadfully as “each khashaa plot has at least one or populated with 8-12,000 residents (see map below). million. Households in the community point out the lack of (e.g. public school teacher, plumber, construction worker) two graves and their kids play with human bones”. Although street lighting and piles of wastes, which vendors from the while others run small stores in the area or stalls at the everybody admits that there are graves underneath the nearby markets dump, as common problems. markets nearby. Most of the households interviewed in the yards, they are not visible as it is impossible to tell where area report that they migrated from a rural area to the the graves are. They usually sell and buy the plots in the city in search of employment after consecutive dzuds in community through their networks of relatives and friends. their home provinces. They also highlight the importance of They mostly report that they have become accustomed to education opportunities for their children in the city. Adult the place now and there is nothing to be scared or ashamed male members of the households typically attained a 4th of. So there are different perspectives about the place MAP 12: DALAN DAVKHAR CEMETERY – SURROUNDED BY KHOROOS 7, 10, 11 AND 12, CHINGELTEI DISTRICT to 8th grade education, with a few having attained some between the insiders and outsiders. (Adapted from UB Municipality’s community mapping web page http://manaikhoroo.ub.gov.mn) vocational education while female members commonly attained an 8th to 10th grade education, some vocational education and even in a few cases, higher education. Most While there are households who managed to settle migrant households reported that they stayed in the plots successfully and living normal lives, there are ones who fail owned by their relatives or acquaintances before moving to to do so. Those households are described as alcoholics the area to acquire their own khashaa plots. who survive on income from the meager jobs at the market, Dalan Davkhar Cemetery according to the khoroo officers and some residents in the neighborhoods. From washing sheep heads, daily income The reason that they chose to live in the area despite the for one person can be around MNT 15,000 per day so many nearby cemetery is related first of all to its closeness to households appear to rely on this income for a living. People, services (such as school) and Shonkhor and Naran Tuul-2 who do menial or casual jobs such as cleaning sheep heads, markets. Now, some households’ livelihoods depend on these working as butchers, or carrying loads on a wheelbarrow at markets as five households’ heads work at the market; some the market are looked down and stereotyped by the urban households rely on cheap meat or variety meats for daily society as the jobs performed by unfortunate, unskilled, food consumption. On the other hand, other vacant areas in homeless, or alcoholics despite the fact that these jobs the city outskirts are too far away from the services and thus provide reasonable incomes. cost high to commute. Many downtown residents as well as other residents from all corners of the city come to the market to purchase their weekly or monthly meat and food items in bulk. So the market serves as the meeting place for the ger-district residents of Denjiin Myanga/cemetery area 61 62 It was closed in 1998. These jobs reportedly allow the households to earn an income of 15’000 tugriks per day. 81 Urban Poverty in Ulaanbaatar: Other Vulnerable Groups 12.0 OTHER VULNERABLE GROUPS 40 percent between the ages of 14 and 18. Since 1996, the Entrance Halls” project, about 80 percent of the guards to-month depending on the amount of maintenance fees 12.1 DISABLED Center has taken in approximately 25,000 children64. In 2014, living in the staircases are single mothers with children. They collected by the UAO for a particular month. While there are disability allowances within the social Ulaanbaatar Metropolitan Police Department launched are not hired with a formal labor contract, their salaries welfare system that aim to reduce the monetary vulnerability a campaign to identify homeless and street children in are often not paid on time and salaries vary from month- of disabled populations within Ulaanbaatar, there is little collaboration with the Child and Family Development Centre. provision for dealing with the every day challenges faced They identified 74 street children in Ulaanbaatar, registered by those who are physically disabled. Disabled accessible them, provided them with health check-ups and attempted toilets and other facilities are non-existent. People with to reunite them with their families or place them in children’s disabilities living in the ger districts are generally confined centers in the city65. However, NGOs have raised concern to their homes and often cannot participate in public life. that the problem of children’s vulnerability in Ulaanbaatar The lack of disability accessible public transport in a key is shifting from streets to children’s homes, which also have reason for this confinement. Unpaved roads, lack of proper limited capacity. While this is a step in a positive direction, sidewalks, and traffic lights in ger areas further constrain it does not address the underlying causes that are forcing mobility, particularly for persons in wheel-chairs. Such children to take to streets in the first place. restrictions in mobility can make mundane tasks challenging for the physically disabled, such as collecting water from distant water kiosks. 12.3 APARTMENT ENTRANCE HALL WATCHMEN 12.2 STREET CHILDREN Since the privatization of Soviet style apartment blocks started during 1990’s, Unions of Apartment Owners (UAO) The number of homeless and street children in Ulaanbaatar have been set up and a new work position of “apartment grew soon after the end of the socialist era63. Their entrance hall guard” created. The roles and duties of this growth was attributed to a number of sources including work position include cleaning the apportioned public abandonment, unaccompanied migration, domestic area, safeguarding the entrance hallways for 24/7, and violence and orphans running away from orphanages where other chores ordered by the apartment council. Typically, they were ill-treated. They live in groups to protect each the guard lives in the staircase in the entrance. Staircase other from other groups and possible raids. They spend the is a space around 1.8X2m (3,6-4m2) in size, with a very low severe Ulaanbaatar winters nestled up to underground hot ceiling and the open side closed with cheap materials. It is a water pipes and scavenge on food to survive. However, over narrow space with limited air circulation and no natural light the recent years, the numbers of street children seems to be that gets cold in winter, is often noisy, and in many cases has declining. According to the Centre for Child Protection in the no sanitation and toilet facilities. The residents of this space Police Department of Ulaanbaatar, the number of children are often single mothers with children below 5, migrated living on streets of UB has declined from 889 in 2011 to 690 in from rural areas and without city addresses. According to 2012 and to further 260 children in 2013. Around 60 percent a survey carried out by the “1х3=5 or Families Living in the of them are between 9 and 14 years old with the remaining 63 The exact number of street children varies according to source. Odgaard refers to 3000 street children in the years from 1994-1995. (Odgaard 1996: p. 119; Bruun, Ole and Ole Odgaard (Eds.). Mongolia in Transition. Old Patterns, New Challenges, Studies in Asian Topics, No.22. Nordic Institute of Asian Studies. Richmond: Curzon Press) UNICEF refers to the same number in its report on street- and unsupervised children of Mongolia 2003. (UNICEF. (2003). “Street and Unsupervised Children of Mongolia”. [http://www.unicef.org/mongolia/street_children_Report_Eng.pdf 20.11.2013].) 64 Statistics produced from the Centre for Child Protection, Police Department of Ulaanbaatar. 65 http://mad-intelligence.com/37-street-children-returned-to-parents/ 83 Urban Poverty in Ulaanbaatar: Conclusion 13.0 CONCLUSION After years of strong economic growth and impressive This study takes a close look at the nature and spatial 13.1 EXTEND INFRASTRUCTURE urban migrants also affect the ability of other vulnerable declines in poverty, 2012-2014 saw Mongolia’s growth rate distribution of poverty in Ulaanbaatar in order to determine groups to access critical social services. slow from a peak of 17.5 percent to single-digit growth. how patterns of urbanization affected both economic IN LINE WITH URBAN Unsurprisingly, there was also a slowdown in the rate of opportunity and quality of life for residents living in various POPULATION GROWTH: poverty reduction, exposing underlying weaknesses in parts of the city. The findings showed a deep spatial Mongolia’s approach to poverty alleviation. Given that dimension associated with both income poverty and 13.2 MIGRANT REGISTRATION Ulaanbaatar serves not only as the country’s economic multidimensional poverty, with people living in peripheral Ulaanbaatar’s unique sprawl pattern creates REFORM engine, but also houses nearly half of Mongolia’s population, areas of the city faring much worse than those closer to the barriers to adequate infrastructure and service addressing urban poverty is critical to any attempts to city center. The current sprawling urban form in Ulaanbaatar provision as houses sprawl further from the Ulaanbaatar’s current registration system alleviate poverty more broadly throughout the country and makes the delivery of services costly and inefficient and ensure that the gains from growth and urbanization benefit contributes to spatial inequality. Spatial inequality in turn is urban core. The city must use planning tools perpetuates inequality by complicating the all. responsible for exclusion from the benefits of urban living based on tier excellent spatal data, as well process for rural-urban migrants to gain for the urban poor, making the experience of poverty more as land and fiscal management instruments severe. urban residency. Without urban registration, (including better land use management, migrants are excluded from local social land value capture, and zoning regulations) services and are often unable to legally to promote densification in order to lower extend infrastructure services to their homes. overall costs of service delivery and provide Ulaanbaatar can take steps to streamline the quality services to more of the population. registration process such as consolidating The city can also bulid upon the survey data registration forms in one government office from the World Bank’s urban poverty and or using a wider range of documents for service delivery reports, to understand city proof of identity. A more efficient process will growth patterns and areas of deprivation not only reduce vulnerability among migrant in order to effectively target investments in populations, but also reduce opportunities infrastructure in areas of density. for corruption within city offices. Similar registration systems in Vietnam and China Analysis of income poverty revealed that contrary to popular have been unable to stem in-migration and opinion, migrants are not the urban poor. However, their experience of living in Ulaanbaatar is marred by social stigma instead created a class of undocumented and subsequent challenges in accessing gainful employment residents without any rights to the city. and basic services (ie. registering for land). Given the large These systems are now undergoing reforms number of migrants already living in the city and likelihood to better accommodate these vulnerable of sustained migration streams, it is important to address the mechanisms underlying migrant exclusion that are creating groups. Ulaanbaatar would be wise to avoid a class of citizens without a city. Other vulnerable groups, these mistakes. including the disabled and street children, experience the various deprivations faced by migrants. Regulations governing residency and land registration that plague rural- 84 85 Urban Poverty in Ulaanbaatar: Conclusion The largest factor to play a role in income poverty is The analysis highlights primary areas where government can unemployment. The urban poor face a number of constraints initiate policy measures, namely employment, education, in finding employment including age and gender biases, service delivery, and provisions for widespread alcoholism. corruption, limited and unaffordable childcare, exploitative The city government recognizes the importance of providing employment practices, and finally accessibility and quality education, and has already taken steps to address gaps in of education. Looking further to factors contributing both access and quality. A recent intervention tackles not to multidimensional poverty, Assets & Money’, ‘Water & only the issue of limited kindergarten capacity but leverages Sanitation’ and ‘Solid Waste’, are shown to be the largest that to enhance labor force participation and job creation. contributors. Multidimensional poverty is experienced This innovative but simple intervention, the Childcare unevenly across ger and non-ger strata. Service Law, is discussed in depth in an earlier section of the report. Alcoholism remains largely unaddressed, and Mongolia would do well to seek examples from abroad in how to address this growing affliction. Finally, Ulaanbaatar 13.3 DEVELOP MECHANISMS already has the appropriate policy tools and mandate over FOR IMPROVING urban planning and land use management, land taxes, and REGULATORY OVERSIGHT zoning regulations, all of which can be used to promote inclusive development through ensuring affordable and AND ENFORCEMENT efficient service delivery. Better urban planning and land AROUND CORRUPTION management aimed at promoting density and with a view to increasing access to services will not only be important for the efficiency and cost-effectiveness of service delivery but Corruption and exploitative labor practices is also critical to enhancing the inclusion of urban poor and are perceived as primary barriers to creating a livable capital city for all citizens. employment. The city can do more to collect data and investigate claims, and introduce mechanisms to facilitate documenting these events. Low-cost mobile platforms utilizing SMS (I Paid a Bribe in India and Lapor in Indonesia) have allowed citizens in other countries to quickly report incidents in order to create a knowledge base that either government, civil society groups or a combination of both can leverage to initiate audits, investigations and mobilize resources and citizens. 86 87 14.0 REFERENCES 15.0 ANNEX 15.1 POVERTY MAPS ADB. 2014. “Asian Development Outlook (ADO) 2014: Fiscal Policy for Inclusive Growth.” Mandaluyong City, Philippines A1. BUS ACCESSIBILITY Alkire, S. 2008. Choosing dimensions: the capability approach and multidimensional poverty. In: Kakwani, N., Silber, J. (eds.) The Many Dimensions of Poverty, pp. 89–119. Alkire, S and Foster, J. 2011. "Understandings and misunderstandings of multidimensional poverty measurement," Journal of Economic Inequality, Springer, vol. 9(2), pages 289-314, June. Anand, S., Sen, A. 1997. Concepts of Human Development and Poverty: A Multidimensional Perspective. Armstrong, Sean & Tsogtbaatar, Byamba (2010): The dual nature of alcohol use and abuse in Mongolia: Reflections through policy. Asia Pacific Journal Public Health, Volume 22, no. 3, pp. 209-215 Asia Foundation. 2015. “Economic Development in Mongolia.” Ulaanbaatar Bruun, O., & Odgaard, O. (1996). Mongolia in transition: Old patterns, new challenges. Nordic Institute of Asian Studies. Studies in Asian Topics, (22). Gilberg, Rolf, and Jan-Olof Svantesson. "The Mongols, their land and history." Mongolia in transition: Old patterns, new challenges (1996): 5-22. Government of Mongolia and United Nations Development Programme (Mongolia). 2007. “Mongolia Human Development Report 2007: Employment and Poverty in Mongolia.” Ulaanbaatar, Mongolia Government of Mongolia and UNICEF (Mongolia). 2014. “Analysis of the Situation of Children in Mongolia 2014.” Ulaanbaatar, Mongolia IFRC. 2016. “Emergency Plan of Action, Mongolia: Extreme Winter Condition.” Kuala Lumpur Kamata, Takuya; Reichert, James; Tsevegmid, Tumentsogt; Kim, Yoonhee; Sedgewick, Brett. 2010. Managing Urban Expansion in Mongolia: Best Practices in Scenario-based Urban Planning. World Bank. Macmillan, New York; Chakravarty, S.R., D’Ambrosio, C. 2006. The measurement of social exclusion. Rev. Income Wealth 53(3), 377–398 M.A.D. Mongolia Newswire. 2014. “37 street children returned to parents.” Ulaanbaatar, Mongolia OHCA. 2015. “ASIA-PACIFIC REGION: El Niño Snapshot.” Suva, Fiji Islands Taraschewski, Thomas. (2008). Stadtentwicklung in Ulaanbaatar im Zeitalter fragmentierender Entwicklung. Aktuelle Migrations- und sozioökonomische Differenzierungsprozesse unter besonderer Berücksichtigung der Jurtenviertel. Berlin: Wissenschaftlicher Verlag UNDP, New York; Bourguignon, F., Chakravarty, S.R. 2003. The measurement of multidimensional poverty. J. Econ. Inequality 1(1), 25–49. World Bank. 2013. Inclusion Matters: The Foundation for Shared Prosperity. New Frontiers of Social Policy; Washington, DC. World Bank; International Monetary Fund (2013) Global Monitoring Report 2013 : Rural-Urban Dynamics and the Millennium Development Goals. Washington, DC World Bank 2016a. Poverty reduction and shared prosperity in Mongolia: Well begun, but can it continue? Washington DC. World Bank, Forthcoming. “Toward Inclusive Urban Service Delivery in Ulaanbaatar, Mongolia.“ Ulaanbaatar 88 89 A2. ULAANBAATAR POVERTY HEADCOUNT AND BUS ACCESIBILITYY A3. BUILDING DENSITY IN RESIDENTIAL AREAS 90 91 A4. NEIGHBORHOOD TRUST A5. MULTIDIMENSIONAL DEPRIVATION 92 93 A6. WATER SUPPLY A7. CENTRALIZED WATER ACCESS 94 95 A8. UNUSED SCHOOL CAPACITY A9. HOUSEHOLD INCOME 96 97 A10. POVERTY HEADCOUNT 15.2 METHODOLOGY 2014 SURVEY MAPS To calculate spatial clustering of similar data and identify hotspots and coldspots we applied a Univariate Local Moran Kriging, a type of spatial interpolation, is the estimation of to the 2014 survey interpolated maps66 . The results display unknown values by using wighted averages of known values whether data are spatially correlated –correlated data are using their spatial covariance. statistically significant at the 95th percentile. Local Moran I Categorical data: Indicator Kriging was used to also informs about the type of spatial correlation: interpolate categorical survey responses collected in Ulan Batur in 2014. Indicator kriging is based on the probability of ++ If identical values are grouped together in space they exceeding a certain threshold. Each threshold was chosen will be described as high-high or low-low. to best focus on deprivation and poverty. Indicator Kriging, a ++ If opposite values are grouped together in space they sub-category of ordinary kriging is based on the assumption will be described as high-low or low high. that the data follow a Gaussian distribution. Statistically significant group of similar values in space are Discrete data: Empirical Bayesian Kriging (EBK) was used often named hotspot and coldspot. to interpolate discrete survey responses collected in Ulan Batur in 2014. The assumption that data follow a Gaussian We also applied a Bivariate Local Moran to two different distribution is not needed for Empirical Bayesian Kriging, datasets; household income and multi-deprivation. The instead EBK uses the estimated semivariogram model to run results display whether the data are spatially co-correlated hundreds of simulation and obtain a more realistic model for and whether co-clustering of high or low values occur. The the data. map interpretation is the same as with the Univariate Local As with every statistical method, Kriging is based on a set of Moran I with one difference, Bivariate Local Moran analyzes assumptions, the main one being that the data are assumed the spatial variation of two dataset instead of one. to be spatially correlated. Some of of the datasets display This leads to a slight difference in how to read the results: higher spatial correlation than others and other factors may affect the spatial variability. ++ High-high or low-low are identical values grouped together of one variable with identical (e.g. high) values HOTSPOT MAPS grouped in space of another variable. ++ Opposite values, high-low or low-high, are identical Spatial autocorrelation: Describes the relationship values of one variable (low or high) grouped together in of objects close to each other in space and their respective space with identical values of another variable grouped values. Objects can be clusterd if they are close to each together in the same spatial area. other and have similar values or objects can be dispersed if In a Bivariate Local Moran if the results are significant the they are close to each other with dissimilar values. implications are that both variables values are linked to each Univariate and Bivariate LISA are statistical test to examine other by their location in space. the spatial correlation of values. 66It is important to note that the maps generate using kriging are by definition based on a spatial pattern defined by their respective exstimated covariance model. The results of LISA test should therefore be interpreted with caution.