Vous êtes sur la page 1sur 80

Urban Slums in Gujarat and Rajasthan

Study of Basic Infrastructure in Seven Cities


[The report examines seven urban slums from Gujarat and Rajasthan describing the current provision of basic services in these areas.]

Benjamin Stanwix

Mahila Housing SEWA Trust


401-402, Akashganga Complex, Brahmkshatriya Co-op. Society Gujarat College Road, Navrangpura, Ahmedabad 380 009, INDIA Ph# 91-79-26560558, 65446208 mahilahsg@gmail.com www.sewahousing.org

2009

LIST OF CONTENTS
` FORWARD ..................................................................................................3 LIST OF ABBREVIATIONS ....................................................................................... 4 A BRIEF NOTE ON SEWA AND MHT ....... ................................................................ 5 EXECUTIVE SUMMARY .......................................................................................... 8 SECTION 1 BACKGROUND OF THE STUDY ..............................................................................14 1.1 INTRODUCTION ....................................................................................................... 14 1.2 OBJECTIVES ............................................................................................................ 14 1.3 STRUCTURE OF THE REPORT ....................................................................................... 15 1.4 METHODOLOGY ...................................................................................................... 15 1.5 REPORT OVERVIEW .................................................................................................. 16 SECTION 2 STATUS OF SLUMS IN SELECTED CITIES .................................................................23 2.1 GUJARAT ............................................................................................................... 23 2.1.1. Ahmedabad ...................................................................................................... 23 2.1.2. Nadiad ........................................................................................................35 2.1.3. Godhra ............................................................................................................. 40 2.1.4. Ankleshwar...................................................................................................... 46 2.2 RAJASTHAN ............................................................................................................ 53 2.2.1. Jaipur ................................................................................................................ 53 2.2.2. Bikaner ............................................................................................................. 59 2.2.3. Jodhpur ............................................................................................................ 65 SECTION 3 INTER-CITY COMPARISON ........................................................................................... 69 APPENDIX ..................................................................................................................... 77 REFERENCES ................................................................................................................. 78 `

FORWARD
For a poor woman in the informal sector, her house is her productive asset, and inputs such as water and electricity play a very important role in her entire production process. Apart from this, improved habitat services also have a direct impact on other human development indicators such as health, education and communal harmony. After initiating work towards addressing these issues, Mahila Housing Sewa Trust understood that the major obstacle for most of the poor families regarding improving their housing is to access basic services and housing finance. The reasons for this are manifolds. Provisioning of basic services is a function of the state. At the state level, multiple agencies are involved in the provisioning of these services, and coordination amongst these agencies is a herculean task. The access to basic services is governed by the authorization or the notification of the poor settlements by the states. Housing finance for the poor is not considered viable. In addition, information on the housing status of poor settlements is scantily available. The situation is worse in the small towns and municipalities where resources are scarce. However the Jawaharlal Nehru National Urban Renewal Mission (JNNURM) announced in 2005 has definitely moved towards addressing some of these issues. Over the years, MHT also understood that multiple approaches are required to fulfill the housing needs of the poor. So, other than constructing low cost houses for the poor, MHT also provide necessary support to access essential basic services, like water, sewage and toilets at their doorstep, which subsequently leads to investment in housing by the poor households. Incremental housing by the poor is a faster and cheaper method to expand the coverage of housing. This study has been based on the primary data collected by MHT on the status of housing and basic services in the urban slums, to enable better outreach through government programmes like JNNURM and the Integrated Housing and Slum Development Programme for the small towns. We hope that the study will be useful to the government and the development practitioners to get a larger picture, and assist them in their programmes and policies. In four of the cities MHT itself has been able to initiate its activities in tandem with the government. We appreciate and acknowledge the work of Mr. Benjamin Stanwix in writing this report and Dr. Surashree Shome in guiding the entire process. Bijal Bhatt Coordinator Mahila Housing SEWA Trust, Ahmedabad

List of Abbreviations
AMC BMC CDF FGT HDI JDA MDG Ahmedabad Municipal Corporation Bikaner Municipal Corporation Cumulative Density Function Foster-Greer-Thorbecke Human Development Index Jaipur Development Authority Millennium Development Goals

MHT NIUA NSS PML ST SY2SP UPRS WB WHO

Mahila Housing SEWA Trust National Institute of Urban Affairs National Sample Survey Private Money Lender Schedule Tribe Sahbhagi Yojana 2 Support Programme Urban Poverty Reduction Strategy World Bank World Health Organization

A Brief Note on.......


Self Employed Womens Association (SEWA) Self Employed Womens Association (www.sewa.org), founded by Elaben Bhatt in 1972 is a trade union of poor women workers in informal economy1. SEWAs goals are to organize women workers in informal economy for full employment and self reliance through its dual strategy of struggle and development. The struggle is against the many constraints and limitations imposed on them by society, trade, state and its policies, while development activities strengthen womens bargaining power and offer them with new alternatives. To pursue these complementary strategies, SEWA organizes its members through the joint action of union and cooperatives; provides services of various kinds; advocates for change in the wider policy environment; and builds institutions to mange and sustain its activities. Over the years, SEWA and its members have distilled Eleven Points based on their objective that serves as a guide posts or standards by which SEWA measures its own progress. The Eleven Questions of SEWA are: Have more members obtained more employment? Has their income increased? Have they and their family members are better nourished? Has their health been safeguarded? Have they obtained child-care? Have they obtained or improved their housing? Have their ownership in assets increased? Have the workers organisational strength increased? Has workers leadership increased? Have they become self-reliant both collectively and individually? Have they become literate? Currently, SEWA has its presence in nine states of India. By the end of 2008, total membership of SEWA was over 1.1 million, of which about 50 percent were from Gujarat and rest from other states of the country. More than 65 percent of the SEWAs members are from rural areas and rest are from the various cities of the country, like Ahmedabad, Delhi, Indore, Lucknow, Thiruvananthapuram etc.

The informal economy is comprised of the self-employed who run small unregistered businesses or small-scale production activities as well as wage workers who do not have secure contracts, workers benefits, or labour protection.

SEWA Bank With an initiative of 4000 self-employed women, SEWA established Shri Mahila Sewa Sahakari Bank Ltd (www.sewabank.com) in 1974 to meet the financial life-cycle needs of its members. The objectives of the Bank are manifold: - inculcating the habit of savings; - providing credit and promoting insurance benefits; - enabling members to get out of the clutches of unscrupulous moneylenders and build their own assets and capital; - equipping members with technical and managerial capabilities; and - to make poor self employed economically strong, safe, sound and self-reliant. Currently SEWA Bank has over 318,594 account holders with a total working capital of Rs. 1,291,888,904 (March, 2009). Mahila Housing SEWA Trust (MHT) After 20 years of SEWAs formation, an internal research indicated that more than onethird of its loans taken from SEWA Bank had been used by members to renovate and repair house and to access infrastructures, like water connections, construction of toilets and drainage, and acquiring electricity. Thus, in 1994, SEWA established Mahila Housing SEWA Trust (www.sewahousing.org.), to expand and consolidate its services to poor. MHT is an autonomous institute registered under Bombay Public Charitable Trust Act, 1950. The major objectives of MHT are to: - Improve the housing and allied infrastructural conditions of the poor. - Establish linkages to provide legal advice for secure housing tenure, finance/loans through formal institutions, technical assistance for the construction/renovation of house and infrastructure, information on the housing market. - Influence housing and its infrastructure related policies and programmes for urban and rural poor, and also disseminate the information among the related beneficiaries. Key Activities of MHT are: - Provision of basic infrastructure in slums and low income areas in partnership with Urban Local Bodies (ULBs) - Developing slum up-gradation plans with ULBs to plan and design basic infrastructure in slums - Electrification for low income households - Training of construction workers, especially women, through Karmika School for Construction workers - Planning and construction of low income housing

Technical assistance for the construction of disaster resistant housing in rural areas Training awareness and capacity building of CBOs and ULBs Facilitating access to housing and infrastructure finance with technical support Capacity building of poor on land tenure issues Conducting Social Audits for related government programmes like NREGA Building city level federation/institutions of the poor Action, research and documentation Policy influence, advocacy and networking

EXECUTIVE SUMMARY
Urban slums in India comprise city areas which lack basic infrastructure facilities, where poor live in squalor, lead an insecure life in overcrowded and dilapidated dwellings. The Mahila Housing SEWA Trust (MHT) works to ensure that the housing and infrastructural needs of these communities are included in the rapid process of urban development. MHT has collected data from slums of various cities to outline the communities need for basic infrastructure and services in the hope that a clear overview and understanding of these areas will provide a foundation that would allow a more inclusive improvement. The present report analyses the data from seven urban slum settlements four from Gujarat: Ahmedabad, Nadiad, Godhra and Ankleshwar; and three from Rajasthan: Jaipur, Bikaner and Jodhpur. The main objectives of this report are to provide a picture of the conditions faced by the average household in these slums, and to investigate the underlying relationships between access to infrastructure with health and income of the households2. 1. General Household Information Household size The average number of people per household ranged from 3.7 to 6.3, (Table 1) with Bikaner and Jodhpur having the largest household size, and Jaipur the smallest. Generally families with lower income had larger household sizes, for example, in slums of Bikaner where 92% of population was poor it had an average household size of over 6 people, whereas Jaipur with a poverty rate of 41% had a household size of below 4 persons. This by no means proves a causal link but it reveals that in India a relationship exists among income, poverty and household size, and is worth noting (Ray, 2000). Table 1
Bikaner Jodhpur Ankleshwar Godhra Ahmedabad Nadiad Jaipur
Ave. Household Size 6.3 6.3 5.3 5.0 4.6 4.6 3.7

The report has made use of the primary data from previous surveys, which had covered the four cities from Gujarat and three from Rajasthan. Apart from these primary data, various other studies have been used to understand the current status of, and the major issues facing urban slums in India. Most of the analysis of data has been done using the statistical analysis package STATA, and in some cases, the data was re-coded for ease of use and analysis.

Residency The average number of years a household had lived in their current neighbourhood ranged between 18.6 to 23.5 years, which indicated that most families had been residents for quite a number of years. Table 2
Ave. Years of Residency Bikaner Jodhpur Ankleshwar Godhra Ahmedabad Nadiad Jaipur 23.5 20.0 18.6 19.6

Type of House There were large differences in type of house the slum dwellers lived (Table 3) across seven cities. For example 73% of the residents in slums in Ahmedabad lived in kuccha houses while in Jodhpur only 7% did. It was not clear from the data which factors accounted for such differences. In some cities house type was linked to income, which could explain part of the story. Table 3
Ahmedabad Nadiad Godhra Ankleshwar Jaipur Bikaner Jodhpur Kuccha 73% 37% 45% 75% 54% 36% 7% Type of House Semi-Pucca Pucca 12% 15% 35% 38% 19% 19% 63% 8% 23% 17% 6% 18% 1% 74%

House and Land Ownership In all the cities, except Jaipur, the majority of residents owned their house. Bikaner topped the list where about 99% lived in their own houses, followed by Ankleshwar, Nadiad, Jodhpur, Godhra, and Ahmedabad (Table 4). Although many residents of these slums owned houses, they did not necessarily own the land on which these houses were built. And again those households, who did own the land, did not have any proof to show their legal ownership of the land.

Table 4
Ownership of House and Land and Legal Proof of their Land Ownership Legal Land Ownership* House Owned Land Owned Ahmedabad 63% 33% 25% Nadiad 85% 38% N/A Godhra 72% 6% 22% Ankleshwar 86% N/A N/A Jaipur 25% 10% 30% Bikaner 99% 38% 47% Jodhpur 83% 3% N/A *Percentage of households with legal documents of land ownership within the percentage who said that they own land

2. Income, Savings and Loans The analysis of per capita monthly income (Table 5) and also the percentage of the population in each city that was earning below the national poverty line showed that percentage of households below the poverty line was lower in large cities like Ahmedabad, and Jaipur. It was above 70% in Jodhpur and above 80% in the remaining four cites, i.e., Ankleshwar, Nadiad, Godhra and Bikaner, which were comparatively much smaller in size and population than Ahmedabad and Jaipur. Table 5
Cities Ahmedabad Nadiad Godhra Ankleshwar Jaipur Bikaner Population Below the Poverty Line Per Capita Monthly Income* Rs.702 Rs.374 Rs.366 Rs.364 Rs.672 Rs.382 % Poor** 32% 85% 87% 82% 41% 92%

Jodhpur Rs.513 71% *The average monthly income earned per person **The percentage of the population below the National Poverty line (Rs.538.60/person/ month)

Not all the surveys provided data on savings, but the ones that were available (Table 6) showed that not many households saved money. The highest saving rates were seen in Ahmedabad, which had the least number of poor households and also has SEWA Bank. Table 6
Savings Cities Ahmedabad Ankleshwar Do you Save? Yes No Rs. 23% 10% 77% 90% 451 460

10

Godhra Nadiad Jaipur Jodhpur Bikaner

3% 19% 2%

97% 81% 98%

452 608 -

Data on loans (Table 7) was available for only three cities, viz. Ahmedabad, Ankleshwar and Godhra. However, the available data showed that Ahmedabad had the highest percentage of households that took loan, followed by Ankleshwar and Godhra. Table 7
Loans Ahmedabad Ankleshwar Godhra Yes 74% 33% 20% No 26% 67% 80%

3. Infrastructure Water Access to water (Table 8) varied across cities; Ahmedabad, Ankleshwar and Bikaner had over 60% of households with access to their own tap. However, Godhra and Jaipur have a relatively small percentage of households with similar access. Table 8
Water Facility In-House Tap Public Tap Neighbour Private Seller Hand Pump Ahmedabad 66% 14% 8% 4% 0% Nadiad 40% 17% 43% 0% 0% Godhra 27% 21% 36% 0% 15% Ankleshwar 65% 35% Jaipur 26% 37% 17% 8% 0% Bikaner 66% 34% Jodhpur 57% 43%

Toilet Access to toilet facility (Table 9) was quite bad in certain cities with half of the population in Nadiad, Godhra, Jaipur and Bikaner slums used open areas for their toilet need. However, in Ankleshwar, most of the households had in-house toilets or they used their neighbours toilet. Table 9
Toilet Facility In-House Neighbour Public Open Area Ahmedabad 72% 3% 11% 14% Nadiad 43% 0% 4% 51% Godhra 37% 0% 5% 57% Ankleshwar 76% 24% 0% 0% Jaipur 40% 0% 2% 49% Bikaner 43% 0% 1% 56% Jodhpur 75% 25%

11

Drainage Data on drainage facility (Table 10) showed mixed results. In most households particularly in Bikaner, Godhra and Jodhpur had extremely poor drainage facility. It was best in Ahmedabad, where 85% of the households had drainage facility. Table 10
Sewerage/ Drainage Yes No Ahmedabad 85% 15% Nadiad 31% 42% Godhra 7% 93% Ankleshwar 77% 23% Jaipur 40% 60% Bikaner 2% 98% Jodhpur 32% 68%

Electricity Compared to the other facilities, access to electricity (Table 11) for most households in all the cities was good, particularly Ahmedabad, where almost all households had electricity. This could be attributed to the pro poor policies of Ahmedabad Municipal Corporation, which resulted in a high percentage of electrification of slums. Table 11
Electricity Yes No Ahmedabad 92% 8% Nadiad 70% 25% Godhra 71% 29% Ankleshwar 65% 35% Jaipur 82% 10% Bikaner 65% 35% Jodhpur 68% 32%

4. Health Preferred Medical Facility The data on usage of health facilities shows that poor families in Rajasthan tend to use the public health facilities whereas those in Gujarat use seem to prefer private health care. This may be due to a better health system in Rajasthan coupled with a higher availability of private care in Gujarat. Table 12
Medical Facility Ahmedabad Ankleshwar Godhra* Public 28% 56% Private 72% 42% *Data is not available for the rest of the households Nadiad 50% 50% Jaipur* 45% 30% Jodhpur Bikaner 85% 15%

Average Monthly Health Expenditure According to the available data (Table 13), the annual expenditure on health ranged between Rs.190 to Rs.270, with Godhra showing the minimum expenditure, and Jaipur the maximum.

12

Table 13
Health Expenditure (Rs.) Ahmedabad Ankleshwar Godhra Nadiad Jaipur Jodhpur Bikaner 220 190 270 228

5. Correlations between Infrastructure, Health and Incomes Investigation into correlation between health and infrastructure showed that the household health was correlated with access to basic amenities; those households who do not have access to water, toilet, and drainage (sewerage and storm water) were spending monthly more money on healthcare than those who had access to these services. The latter group also reported higher rate of illness. It was also clear that a relationship did exist between income and access to infrastructure. The poorer households lacked access to water, toilets, drainage, etc. 6. Correlation between house ownership, land tenure, and expenditure on infrastructure and upgrading The regression analysis also showed that spending on household upgrading and acquiring basic amenities was much higher if the house and the land were owned by the family. Spending was also positively correlated with income.

13

URBAN SLUMS IN GUJARAT AND RAJASTHAN STUDY OF BASIC INFRASTRUCTURE IN SEVEN CITIES __________________________________________
SECTION 1 BACKGROUND OF THE STUDY
1.1 Introduction
Rapid increase in urbanisation together with expanding number of urban slums3 pose considerable challenges to the process of socio-economic development in India, especially with regards to construction of suitable and sustainable human settlements. Vast areas in Indian cities have unsafe dwellings, overcrowded, condition, and lack basic infrastructure.4 In these areas residents live in poor and squalid conditions, and lead economically insecure life. Mahila Housing SEWA Trust (MHT)5 is a non profit organisation that is committed to ensure that housing and infrastructural needs of these communities do not get compromised in the overall development of the cities. They began their work in Ahmedabad and gradually moved to neighbouring states. MHT partners with the slum community, the government, and the private sector to successfully bring about upgrading of slums. This report examines only a small part of that process. It analyses data from selected urban slums in Gujarat and Rajasthan, by outlining the communities need for basic infrastructure and services. This has been done in the hope that a clear overview and understanding of these areas will provide a foundation which allows the necessary improvements to take place more easily.

1.2

Objectives

The main objectives of this report are twofold: firstly, to provide a basic picture of the conditions facing the average household in these slums by presenting and examining data on household characteristics; and secondly, to investigate some underlying relationships in the data which would help to tell the story of these households more accurately. To achieve the latter objective a few relationships which were found to be significant in other studies were tested. Specifically, the two key premises examined were:
3

A UN Expert Group has created an operational definition of a slum as: an area that combines to various extents the following characteristics: inadequate access to safe water; inadequate access to sanitation and other infrastructure; poor structural quality of housing; overcrowding; and insecure residential status (UN-Habitat, 2007). The description serves as an accurate working definition for use in this report. 4 Throughout the report infrastructure is taken to include basic facilities such as access to water, toilets, drainage, electricity, etc. 5 The Mahila Housing SEWA Trust (MHT) was registered in 1994 with the objectives of improving the housing and infrastructural conditions of urban poor women.

14

i)

Households without access to basic amenities, such as piped water, adequate toilet, drainage and waste disposal facilities were more prone to fall ill, and thus spend more money on healthcare than households with access to adequate amenities. This premise was based on the supposition that households that lacked access to basic infrastructure lived in unsanitary conditions, and thus their members often fells prey to number of illnesses. Lack of basic infrastructure was correlated with income level. It was generally the poorer households (in terms of income) who usually lacked access to basic infrastructure. This premise was based on the ground that families with lower incomes would not be able to afford basic infrastructure, also those without basic infrastructure usually have less time to undertake productive activities and would perhaps have to spend more money on expenses such as healthcare. This, in turn, contributes to keeping them poor. So the link ran in both directions.

i)

In addition to these two hypotheses, the report also investigates the link between savings and income, and between property ownership and investment in household upgrading. Though regression analysis was the primary tool used to explore these links and test the two hypotheses mentioned above, the report has also provided a number of tables to support insights.

1.3

Structure of the Report

The report is divided into three sections. Section 1 provides the background of the study wherein section 1.1 provides a brief introduction, section 1.2 states the main objectives, section 1.3 explains the basic structure of the report, section 1.4 explains the methodology used and limitations of the study, and finally, section 1.5 provides a report overview that states how the analysis of each city was conducted and sets out the structure which was used throughout. The overview goes on to introduce and discuss the main topics which are explored in the analysis of each city. Section 2 contains a separate report for each of the seven cities where the survey data is presented and investigated under the headings laid out in the report overview. Section 3.1 presents a comparative analysis of the data on all seven cities. Finally, in section 3.2 some conclusions are drawn.

1.4

Methodology

The report used relevant primary data from previous surveys conducted in the seven cities focussed in this report. The data was taken from the surveys which MHT carried out for various other studies. Apart from these primary data, various other studies were used to understand the current status of, and major issues facing urban slums in India6. Most of the analysis was done using the statistical analysis package, STATA, and in some cases the data was re-coded for ease of use and analysis. Each survey was done
6

SEWAs UPRS and SYS2P studies were used for this purpose

15

independently, so they differed in their design and the questions they asked; some being more comprehensive than the others depending on their initial purpose. This perhaps affected the comparative accuracy of the estimates presented. In all cases the data was adjusted for sample size and the population in question. Despite these differences, all surveys included enough basic questions of interest to make a study of this nature possible and useful. However, the major constraint that restricted the scope of this report was: the data used here had been collected for other purposes, and thus were broad based and lacked the depth needed for the purpose of this study. Some of the specific limitations with regard to the data were; in many surveys information on household health was inadequate only the NIUA survey of Ahmedabad had substantial data on the frequency and type of illness , for some other areas there was data only on monthly health expenditure, while some had no information on health at all. Given this, situation most of the investigation into health levels had to be based on health expenditure which is a poor proxy at the best especially given the fact that some of the households used public healthcare, and others private, and these were not always specified. However, since this was the only available option for investigating health care in some of the cities. The results suggested that despite these limitations it did provide some worthwhile insight. The data on household demographics was generally limited to family size, caste, and religion. Data on occupations or employment was scarce. Information on savings and loans was either varied, or for some cities it was completely absent. Some of the surveys, such as that of Jodhpur, had a very small sample size which affected the accuracy of analysis. Several surveys lacked data on household expenditure. More generally, when compared to studies such as the UPRS or SY2SP (where the questions were prepared specifically) these questionnaires did not ask very detailed questions. Thus while the information provided was adequate to create a profile of the basic conditions in each area, the scope for any detailed analysis was quite limited and must be noted.

1.5

Report Overview

The analysis of each city was done under following six divisions. All the divisions provide a discussion of different problems facing slum dwellers which other studies have noted. This section also explains how the data was analysed and presents a few salient points from the data. i. General Household Information A general description of the households and their housing conditions were provided. Household Size

16

Average household size gave an indication of how many people made up a typical family, and how many children the parents were providing for. It was generally found that larger households had lower incomes, for example in this study, Bikaner had an average household size of over 6 people and 92% of the population was poor, while Jaipur had a household size below 4 and a poverty rate of 41%. This by no means proved a causal link, but it was worth noting that in India a relationship between income, poverty and household size does exist. (Ray, 2000) Caste Data on different castes is shown, where available, as this can influence the functioning of a community in certain cases, especially where caste divisions still translate into class divisions. The development worker would be better off at least having the knowledge of these divisions. It was also found that in certain cities such as Jaipur and Bikaner incomes differed significantly depending on caste. Type of Housing The type of housing used by the slum dweller was a good initial indicator of the living conditions of the residents. The data showed that it was generally the poorer residents who resided in the worst housing. The house is possibly the most important asset for the poor, and can be used to access other funds and opportunities. In many cases the home was a living area, a place of work, and an asset which could be used for access to credit (SEWA Academy 2002, Chen 2000). A pucca house is a standard dwelling made with bricks and cement, and has a solid roof. It is the most desirable possession for the poor. A kuccha house is built with various makeshift materials. Such houses are less durable and more likely to leak in the rainy season. Furthermore, these houses can be difficult to service as they are often constructed haphazardly and are usually not built with drainage and sewerage facilities, or electrical installations. There are also houses which are partly kuccha and partly pucca; they are referred to as semi-pucca or mixed in this report. Ankleshwar had the highest percentage of kuccha houses while Jodhpur had the highest percentage of pucca. Tenure Issues of tenure are important because they deal with housing security and affect the decision a family makes about spending on their home. If tenure is insecure and families could be forced out unexpectedly, they are unlikely to spend money on renovations to make the home a more liveable place, or on acquiring basic services and making it an acceptable living-space. The data for Ahmedabad supported this. It showed that when property was owned, households had spent over Rs 10,000 and more on upgrading their house and in acquiring new facilities. A further concern related to tenure was that without ownership the home could not be used as collateral for loans.

17

Residency Finally, the length of time residents had stayed in the area is important, and can influence the type of project to be initiated in a community. For example, it is plausible that long term residents may be more organised as a community and this would make any development work easier, but then they may also be unwilling to relocate, if the need arose. ii. Income, savings and loans Income In this study income was found to be the most reliable form of welfare measure available for every household, and thus it was used to examine the general wealth of the households. Specifically, per capita monthly income was used to construct a poverty profile for each city. Two techniques of examining income poverty were employed: Cumulative Density Function (CDF) and the FosterGreer-Thorbecke (FGT) measures. An example of a CDF, shown below, illustrates the spread of per capita income across the surveyed population. On the y-axis is the population and on the x-axis is monthly per capita income. This way of illustrating income allows one to escape the constraints of a fixed poverty line, and get a sense of both the extent and the depth of poverty.
Income CDF
1

.8 % of population

.6

.4

.2

0 0 250 IndiaWB 1000 1500 per capita monthly income 2000

The two red lines respectively indicate Indias national poverty line, and the new the World Bank (WB) poverty line, chosen as an international measure. Indias national poverty line for urban areas as per 61st National Sample Survey is Rs. 538.60 per month (2006). The international poverty line is calculated at $1.25 per day (2005 PPP) which is approximately Rs.650 per month (Chen & Ravallion, 2008). From the CDF above one can see that around 30% of the population are below the national poverty line, and approximately half are below the international line.

18

To get a more precise statistics on poverty the FGT measures was found to be more helpful. This is a generalised way of measuring poverty and gives exact figures for a chosen poverty line7. The FGT measures provide information on the extent of poverty and the intensity/depth of poverty. The results shown below were calculated for the national poverty line (Rs. 538.60), using the same population as the CDF above.
Foster-Greer-Thorbecke poverty indices, FGT(a) a=0 a=1 a=2 0.3376 0.0833 0.0317 FGT(0): headcount ratio (proportion poor) FGT(1): average normalised poverty gap FGT(2): average squared normalised poverty gap

The output shows that 33.76% of the population live below the national poverty line [this is the headcount ratio (a=0)], while the poverty gap (a=1) tells us that, the average earnings of the poor who fall below the poverty line are 8.33% below the line of Rs. 583.60. This indicates the depth of poverty. The normalised poverty gap (a=2) has no easy interpretation as it is essentially the squared distance below the poverty line, thus giving more weightage to those at the bottom-end of the distribution. It is therefore a very t useful tool for comparison, but it will not be mentioned in this report. Among the cities covered in this report, Bikaner had the highest percentage of poor households at 92%, while Ahmedabad (which has been used as an example here) had the lowest at 34%. Savings The practice of saving is important, especially for the poorer families as it serves as a cushion in times of financial shock and helps to avoid the unnecessary taking on of debt (Rakodi, 2000). It is obvious from the data that most families did not save; the highest percentage of household who saved was 23% in Ahmedabad, and the lowest was in Bikaner with only 2%. Intuitively one can understand that poorer households are less likely to save simply because they have less money. Loans Data on loans was minimal but whatever could be gleaned from the data has been presented in the report. The data on loan sources was available only for Ahmedabad, and here it is clear that the majority of households took loans from informal sources such as private money lenders (PML) and relatives. There could be two main reasons for this: firstly, most households perhaps did not own the assets necessary for acquiring a loan from formal institutions; and secondly, in
7

The general formula for calculating FGT measures id given in the Appendix

19

the case of a financial emergency such as a family death when money was needed urgently, loans from private lenders and family were easily accessible and required no paperwork. Furthermore, borrowing from relatives generally meant that no interest was charged. In Ahmedabad 74% of the households borrowed money but in other cities this figure fell to less than 50%. The recent success of micro-lending schemes in many poor areas of the world suggests that more thorough data on loan activities could provide valuable insights for development. iii. Infrastructure water, toilet, drainage, waste disposal, electricity There are number of strong arguments for improving and upgrading the infrastructure of slum settlements, many of which are interrelated: improving living conditions can bring gains to the quality of life, health, and productivity of slum residents, it can help to break the cycle of poverty, ease the burden on women, and can also be a public good with positive spill-over effects on the wider economy and society. These arguments have been discussed in more detail below. The positive impact of infrastructure improvement on the quality of life of residents is itself a valid reason for undertaking such an activity. Life in the absence of adequate access to basic services such as water and ablutions can be precarious; it is detrimental to health, safety and the dignity of communities (Karn et al., 2003). Many commentators have highlighted the link between poor living conditions which characterise slums, and ill health. This relationship has been explored by Hancock and Devas in their report (Hancock 2000, Devas et al. 2001). A UN-Habitat study shows that lack of safe drinking water and poor sanitation can lead to a range of diseases, while factors such as overcrowding and pollution can also contribute to health problems (UN-Habitat 2008). Other reports reveals that providing good access to water and sanitation can reduce the amount of stagnant water and thus the risk of malaria and a variety of waterborne diseases (WHO 2000, UN-Habitat 2008). With regards to water and sanitation, it was found that in Jaipur only 26% of households had their own tap, while in Godhra 57% of the households used open areas for defecation which alone posed a serious health risk. Illnesses also force households to pay extra medical expenditure which can push them deeper into poverty and indebtedness (UN-Habitat, 2003). Furthermore, the number of working days for self-employed women and others are reduced when family members become ill, thus lowering their income further. Amis (2001) notes how illness leads to asset depletion and debt in order to fund private healthcare, and thus is one of the most powerful forces pushing households into poverty (Amis, 2001:106). Good sanitation and sewage facilities ensure a safer and healthier lifestyle, which in turn lead to a healthy workforce with higher productive capacity. Less money would be spent on

20

illnesses, and in the long-term, these health gains would benefit not just the individuals concerned, but the wider economy and society. One can expect that public expenditure on healthcare will be lower and a more productive and healthy workforce will be created. The WHO estimates that simply meeting the Millennium Development Goals (MDG) target on water and sanitation could result in an annual cost saving of $7 billion in the health sector, and an annual global saving to the value of $750 million from working days gained as a result of improved health (WHO, 2003). Life in slums is particularly difficult for women as the burden of household work in the absence of infrastructure usually falls on them, for example fetching and carrying water (Chen, 2000). They also have to struggle to maintain household hygiene in the context of poor drainage and sanitation. It is women who are most vulnerable to harassment or assault when using open areas in the absence of toilet facilities, or due to insufficient street lighting at night (UN Habitat, 2003). Thus women may derive the greatest benefit from improvements in infrastructure (Amis 2001, Rakodi 2002). Improvements in water supply, access to toilets, and drainage reduce this burden on women, and increase their time for other activities. Access to water, for example, is a vital part of everyday life. Without reliable and easy access to this resource much time and effort have to be dedicated to acquiring it. In the slums of Ahmedabad about 3200 personhours were spent each morning in just collecting water. This time could be productively spent elsewhere if households had access to their own reliable water source. This would be especially benefit the housewives and home-based workers as it would result in increased space for household and income-earning activities (WHO, 2003). Provision of street lights would increase womens personal security, allow them the use of more outside space and lengthen the working day (Amis 2001, Devas et al. 2001). Benefits such as these are also in line with the MDG of promoting gender equality, thus empowering women. Improvements of slum areas would also improve the overall image of the slum, and can therefore stimulate investment and business/service growth in the area. Amis (2001) found that the provision of paved roads provided a powerful confirmation of the legality of the slum settlements, and hence led to subsequent investment. It has also been noted that one effect of slum improvement is integration of slum settlements into the wider life of the city (Amis, 2001). Provision of affordable infrastructure and services in the urban slums thus has numerous benefits, and it makes a key contribution to urban poverty reduction. Projects such as MHTs Parivartan, where the results of installing basic services and infrastructure in urban slums have been documented, provide a compelling example of the difference that these measures make to a community (SEWA Academy 2002, UPRS). Such local evidence is well-supported in international

21

literature; particularly research done by the United Nations (UN-Habitat, 2003, 2008). Not only does the provision of basic infrastructure increases their standard of living but it also creates more time for family members to engage in productive work and contribute to family earnings. Importantly, access to water, proper toilet and drainage facilities have significant health benefits. iv. Health Health is obviously a key aspect of human well-being, and its relation to living conditions in urban slums has been mentioned in the discussion above. Unfortunately the data on health for many of the cities was limited and thus relationships between health and infrastructure were analysed only where it was possible to do so (this was done in the regression analysis in section v). Regression Analysis The last section on each city examines some of the relationships in the data using basic linear regression analysis. The choice of variables was generally dictated by the availability of data. For every city the connection between income and infrastructure was examined. It was generally found that higher income is related to better access to infrastructure, which is what one would expect. Where data on health was available the links between health and access to infrastructure (water, toilet and drainage facility, waste disposal) were explored. For Ahmedabad the relationship between property ownership and investment on infrastructure was investigated. Regression output has been presented after checking for multicollinearity, heteroscedasticity, and autocorrelation which were not problematic in these regressions. There was surprisingly little correlation between infrastructural variables such as water, toilet, and drainage facilities, suggesting that households which have access to one facility do not necessarily have access to other facilities. The regression output tables give the coefficients, standard errors and statistical significance of each variable; additionally it also shows the Adjusted R - which gives the explanatory power of the regression and the sample size (number of observations). The city-by-city analysis is then given where the data is presented for each of the sections discussed above. vi. Summary The findings and key features of the data were summarised at the end of each section.

v.

22

SECTION 2 STATUS OF SLUMS IN SELECTED CITIES


2.1 Gujarat 2.1.1. Ahmedabad
Ahmedabad is the financial and industrial capital of Gujarat. In 2006 the city had approximately 1,000 slums (AMC, 2006). Moreover, the slum population in Ahmedabad has been growing faster than that of its overall population. It almost doubled in the two decades since 1976 to over 41% of the total population (Dutta and Batley, 1999). Therefore, it is likely that the number of slums too must have grown in the last three years. Data for this analysis was taken from the 2007 NIUA dataset, which surveyed 12,773 households in 39 urban slums of the city. i. Household Information The average household size in the urban slums of Ahmedabad was 4.6, and approximately 40% of households had more than three people living in one room. With regards to residency, the average household was found to have lived in the same area for almost 19 years. From Figure 1, below, it is clear that by far the majority of residents lived in kuccha houses (72%). Figure 1
House Type

80% 70% 60% 50% 40% 30% 20% 10% 0%

72%

15%

12%

Kuccha

Pucca

Semi-Pucca

It is useful to examine whether the type of house that a family lived in had any link with their level of income. Table 1, below, separates the monthly per capita income of all households into five groups (or quantiles), and tabulates this alongside the type of house which the family occupies. The data shows that, out of those who lived in kuccha and semi-pucca houses, the highest number belonged to the poorest (1st) quantile. Of those who lived in pucca houses, 35% were in the richest (5th) quantile. Overall the data revealed a relationship which one would expect to see those with higher incomes live in better houses and vice versa.

23

Table 1
Income quantiles 1 2 3 4 5 Total Type of House Kuccha 29% 12% 22% 19% 18% 100% Semi-Pucca 44% 11% 18% 14% 13% 100% Pucca 17% 10% 17% 21% 35% 100%

Another aspect of the housing situation is tenure. In these slums 60% families owned their houses and 40% paid rent. The average rent paid was Rs. 495 per month. Only 33% of families owned the land on which their houses had been built. Moreover, many families who owned property or a house did not have any legal proof of this ownership. The Table2 below suggests that type of house seemed to differ slightly based on whether the household owned the house or lived in rented ones. A higher percentage of residents who owned houses lived in pucca houses, while a higher percentage of those who paid rent lived in kuccha houses. Intuitively this would suggest that when a house is owned by the family they are more likely to spend money to upgrade it. This hypothesis is investigated later in more detail. Table 2
Ownership Own Rent Pucca 17% 10% Type of House Semi-Pucca Kuccha 12% 70% 12% 78% Total 100% 100%

Table 3 shows that families who owned land invested on their property more than those who did not own land. On average, those who owned land spent Rs. 10,444 for upgrading and acquiring basic facilities for their house. This too has been explored further in the regression analysis. Table 3
Summary of Investment After Purchase Ave. Amount (Rs.) Own Land 50,450.14 Do Not Own Land 40,005.36

ii. Income Average per capita monthly income of the slum dwellers was Rs. 702. Figure 2 given below illustrates the spread of per capita income across the survey population using the CDF measure; the national and international poverty lines are shown in red.

24

Figure 2 % population 1 .8 .6 .4 .2

Ahmedabad Income CDF

0 0 250 India WB 1000 1500 per capita monthly income 2000

It can be seen that around 30% of the population were below the national poverty line. If we apply the international poverty line, which has a cut off point of income Rs.650 per month to assess poverty, then we find that approximately half of the slum population could be regarded as poor. To get an exact measure of poverty FGT measures were applied (Table 4). These measures used the Indian poverty line of Rs. 538.60/month to calculate poverty. Table 4
Foster-Greer-Thorbecke poverty indices, FGT(a) a=0 a=1 a=2 0.3376 0.0833 0.0317 FGT(0): headcount ratio (proportion poor) FGT(1): average normalised poverty gap FGT(2): average squared normalised poverty gap

The headcount ratio suggested that 33.76% of the population lived below the national poverty line, while the poverty gap stated that on an average the poor were 8.33% were below the poverty line. The study also revealed that 70% of households had ration cards. This is important given the fact that ration card is the only available form of identity card for many poor households.

25

Savings It was found that 23% households residing in slums saved on average Rs. 451 per month. The rest had no saving of any kind. Table 5 shows that the majority of the saving was done through formal institutions, like SEWA and nationalised banks. Table 5
Saving Options Nationalised Bank SEWA Bank SHG's Relatives V.C At Home Others 38.32% 27.89% 12.37% 8.81% 5.78% 0.36% 6.47%

Table 6 examines the relationship between household income and saving. It is clear that the majority of the poorest households did not save. Almost 50% who did not save were from the poorest two quantiles, whereas almost 70% of those who saved were in the highest two income brackets. Table 6
Relation between Household Income and Saving Income Quantile Not Saving Saving 1 36% 9% 2 12% 10% 3 23% 13% 4 16% 26% 5 13% 43% Total 100% 100%

Loans Around 74% of slum households borrowed money. Table 7 lists the various sources from where they borrowed. It is clear that the majority of the households took loans from informal sources like private money lenders and relatives. Table 7
Loans Private Lender Nationalised Bank Relatives SEWA Bank Institutional Employer Others Not Loaning 46.16% 11.09% 9.25% 6.43% 1.53% 0.14% 2.32% 23.08%

26

Other Assets About 78% of the slum households did not have insurance, of which only 14% expressed desire to get insurance. Information on other household assets was minimal; however, average total value of household assets which included land, house and cattle came to Rs. 29,754 for this community. Infrastructure Water The majority of households in these slums had an in-house tap (66%) but there were still a large number of households who had to fetch water. Figure 3
Main Water Source

70% 60% 50% 40% 30% 20% 10% 0% In-House Tap Public Tap Neighbour Private Seller

66%

14%

7%

3%

Fifty-eight per cent of households spent more than 30 minutes per trip to fill water, and 27% of the families had to travel more than one kilometre to obtain water. In addition, for those who used public taps installed by the municipality, water supply was available on average for only 2.5 hours per day. Table 8 examines the time spent in fetching water by comparing the time spent collecting water by households who had a tap against those who did not. It shows that 99% of households who had their own water facility spent little time collecting water, while a large number who did not have their own tap spent substantial time on collecting water. These households spent about 3200 personhours every morning to collect water, which is a basic necessity. Table 8
Time Spent 0-1hr 1-2hrs 2+hrs Own Water No Yes 42.1% 99.3% 46.5% 0.1% 11.2% 0.0%

This highlights the time which could be saved and spent more productively if more families had access to their own water facility.

27

When asked to rate the importance of water on a scale of 10 (1 being the lowest), the slum residents gave an average rating of 8.6. When asked to compare different amenities which residents would like to have them installed or used to upgrade their communities, water was notably the most important. This is shown in Table 9. Table 9
Most important facilities Water 69.66% Toilet 14.17% Gutter 11.55% Garbage disposal 2.11% Street light 1.65% Public transport 0.87%

Toilet/Sanitation The majority of the slum families had in-house toilets (74%), but there were still a large number of families who had to defecate in the open (14% or 1827 households). Figure 5
Toilet

80% 70% 60% 50% 40% 30% 20% 10% 0% In-House Toilet Open Area Public Toilet 14% 9% 3% Neighbour 2% Pay and Use 72%

Of those who did not have their own toilet facility, 2700 households stated that they urgently required a toilet in their house, and of these families 80% showed willingness to pay to acquire a toilet. The average payment offered was Rs. 338. Drainage and Sewerage It was found that 17% households did not have any storm water drainage, and only 10% had no gutter for sewerage. Of those that did have some kind of drainage, 20% reported problems of water clogging. One-third of all households had to pay for repairs on their houses in the rainy season because of lack of drainage system or due to dysfunction of drains. The regression analysis done later in the report examines the impact of drainage on health. Waste Disposal

28

In many slums solid waste was not properly collected. The Ahmedabad Municipal Corporation had provided collection bins only in some slums. From Figure 6 we find that most residents had dustbins at home. Around 19% of households made use of the common dustbin to dispose off their waste, 10% threw it on the street, and only a small number of households burnt their waste. Figure 6
Waste Disposal

70.00% 60.00% 50.00% 40.00% 30.00% 20.00% 10.00% 0.00% Dustbin at Home Dustbin in Locality Throw in Street 19.43% 10.47% 0.53% Burn it 69.57%

In the case of recyclable waste, over 95% families said that they recycled their waste. Electricity Ninety-two per cent of the households in these areas had electricity in their homes. Of these a vast majority had an in-house meter. It is important to mention here that Ahmedabad Municipal Corporation has played a pivotal role in large scale electrification of slums. On average, each household paid Rs. 235 per month for electricity. iii. Health In the year preceding this survey (2006-2007), many households reported that family members had suffered from a number of diseases, and many of these were related to poor sanitation. The data is presented in Table 10. It must be noted that this data had been gathered only from reportage of the respondents and so it is not clear whether these illnesses had been professionally diagnosed. Table 10
Types of Illness Annual Illnesses Observations Diarrhoea 695 Malaria 464 Typhoid 352 Chikun Gunia 5805 Cholera 317 Other 2411

Illness not only affects the patient physically but it also means loss of valuable work days for him. Table 11 presents the number of households who reported members missing

29

work due to illness (recall that the total number of households is 12,773). A large portion of the population missed more than 15 days of work due to illness. Table 11
Workdays Lost due to Illness (per year) < 10 2164 10 15 817 15+ 1084

Illness would also require spending on medical care, and on average, these households incurred medical expense of Rs.220 per month. The Figure 7 below gives the breakup of this spending, with the bulk of households spending between Rs.50 to Rs.200. Figure 7
Monthly Medical Expenses (Rs.)

Rs.0 < Rs.50 Rs.50 - Rs.100 Rs.100 - Rs.200 >Rs. 200 0% 5% 10%

13% 17% 26% 25% 18% 15% 20% 25% 30%

With regards to medical facilities, most families in these slums reported that they preferred private medical facility over government hospitals mainly because service provided by public hospitals was inferior. Figure 8
Preferred Medical Facility

27.44% Private Municipality

71.99%

iv. Regression Analysis In this section some multiple regression equations have been formulated in order to better analyse the data. Three main relationships were explored. The first regressions (1a and 1b) looked at the relationship between infrastructure and health; the second

30

one examined the effect of ownership on household investments; and the third explored the link between income and infrastructure. Equation 1(a) This first equation tries to estimate the relationship between health expenditure and infrastructure (access to basic services). Monthly health expenditure is the dependent variable and is regressed against a series of categorical variables: Water, toilet, sewerage and drainage (storm water). Numerical 1 denotes that the service is present and 0 denotes it is not present. The equation is: Health Exp. = C + 1(Water) + 2(Toilet) + 3(Sewerage) + 4(Drainage) + Regression Output 1
Health Expenditure Coefficient Std. Error P-Value Water* Toilet* Sewerage* Drain C (constant) Observations Adjusted R *10% Significance -13.16 -9.31 -81.53 -15.91 283.48 8,386 0.1 4.62 7.79 5.9 4.57 76.14 0.07 0.09 0.08 0.7 0

The above results support the common beliefs. Analysing the output in the order it has been presented, we see that: o Average monthly spending on healthcare of a household with access to water was Rs. 13.16 lower than a household that did not have access to water. o On an average a household that had access to toilet facility spent Rs.9.31 less than the household that had no toilet facility o A family that did not have sewerage facility spent Rs.81.53 more than a family that had this facility. The magnitude of this point was very high. o The mean monthly spending on healthcare of a household that had drainage facility was Rs.15.91 lower than the one that did not have drainage, although this was found to be statistically insignificant. o The interpretation of the constant term is that, when a household has none of these facilities they spend an average of Rs.283.48 per month on healthcare. The spending would decrease with addition of these facilities. The adjusted R-Squared tells us that 10% of the variation in health expenditure can be explained by these four variables. Equation 1(b) No. of Illnesses= C + 1(Water) + 2(Toilet) + 3(Sewerage) + 4(Drainage) +

31

The regression here is essentially presenting the same hypothesis as the one above that basic services have an effect on health. The difference being that in this case health is measured by, the number of illnesses in the household over one year as opposed to monthly expenditure on healthcare in the previous case. However, similar results are evident which is encouraging for the reliability of analysis which uses health expenditure as a proxy for family health. Regression Output 2
No. Illnesses water* toilet* sewerage drain* _cons Observations Adjusted R *10% Significance Coef. -0.01 -0.02 -0.67 -0.08 0.83 7,118 0.11 Std. Err. 0.01 0.01 0.02 0.02 0.03 P>t 0.06 0.1 0.24 0.05 0

Plainly, a literal interpretation of the coefficients in this case does not make sense (one cannot imagine 0.01 of an illness, for example) instead these results have been included as they support the intuition of the findings in the first regression. From the results one can see that access to water, toilet facilities, sewerage and drainage all decreased the number of illnesses households experienced the magnitudes being relatively unimportant here. The Adjusted R-Squared is 0.11 which suggests that while some of the movement in illnesses is explained here, there are also other factors that influence the dependent variable. Equation 2 This second regression tries to find out which variables might influence how much a family would spend on upgrading their house and in installing basic amenities. In this equation the amount which families invested is the dependant variable and the independent variables are; whether the family owns the land, whether they own the house, and their per capita monthly income. Inv. on House = C + 1(own-land) + 2(rent-house) + 3(pc income) +

32

Regression Output 3
invonhouse ownland** renthouse* pcinc* _cons Observations Adjusted R *5% Significance **10% Significance Coef. 683.22 -11537.32 22.71 32863.76 2,410 0.32 Std. Err. 200.7 975.16 3.5 3151.67 P>t 0.09 0.02 0 0

These results revealed an important link which has been alluded to earlier. Firstly, they showed that owning land was linked to larger investment (Rs.683) on upgrading and acquiring infrastructure relative to households who did not own their land. Secondly, if the family rented out their house they were likely to spend considerably less on investment (Rs.11, 537) compared to those who owned their house. Thirdly, as expected, higher per capita income was related to increased investment, this simply suggests that those with more money invest more. The Adjusted R value was 0.32, and again the variables were statistically significant. These results support the claim that tenure and ownership matter; when the land and house are owned this increases security, and generally leads to greater investment, which ultimately has positive effects for the household. Equation 3 This final equation examines the relationship between income and infrastructure. Monthly Income = C + 1(sewerage) + 2(toilet) + 3(water) + 4(drain) + The regression is used to summarise what has been reported in some of the data tables above; that poorer families generally have fewer amenities available to them. Regression Output 4
Monthly Income sewerage* toilet* water* drain** _cons
Observations Adjusted R *1% Significance **10% Significance

Coef. 44.83 99.11 28.67 25.32 675.37 8,792 0.14

Std. Error 15.17 23.27 10.16 13.76 26.11

P-Value 0.00 0.00 0.01 0.07 0.00

33

Examination of the output revealed that, on average: o Households with a facility to deal with sewerage earned Rs.44 more than those who did not have this facility. o Households with access to toilet facility earned Rs.99 more than those without toilet facility. o Households with their own tap earned Rs.28 more than those who had no tap. o Households with storm water drainage earned Rs.25 more than households with no drainage. o The regression explains 14% of the variation in monthly income according to the Adjusted R. v. Summary and Conclusion Overall, the data showed that about one-third of the families who lived in slums were poor. Many lived in inadequate houses and had insufficient access to basic infrastructure such as water, toilet and drainage facilities. When asked to rate which resources were most urgently needed, the first three were water, toilet and drainage, respectively. Regarding savings, only few households said that they saved, and those who saved were almost exclusively households with higher incomes. Nearly two-thirds of households took e loans, and the majority of them took loan from informal sources. The regression analysis suggests, in agreement with other studies, that the basic amenities have relationship with the health of the households, and therefore, also with the monthly amount spent on healthcare. Furthermore, it appeared that ownership of land and house both played a role in the amount of investment that a household would make on their estate. Finally in general, households who were poorer (earning less) were also those who do not have access to basic facilities.

34

2.1.2 Nadiad
Nadiad is a small city in Gujarat, and it is the administrative centre of the Kheda District. It has a population of about 196,000 and the information presented here was gathered from 6,566 households in 97 different slums (Census, 2001). i. General Household Information The average household size in these selected slums was 4.59, and most of the residents were from the Scheduled Tribe (ST), as shown in Figure 9, below. Nearly all families had at least three children, and the gender ratio here was balanced as the number of women and men who resided in these slums was equal. Figure 9
Caste

Baxi

3%

Other

9%

S.C

17%

S.T

64%

0%

10%

20%

30%

40%

50%

60%

70%

The usage of different types of housing for residing has been shown in Figure 10. The split among kuccha, pucca and semi-pucca was fairly equal with a higher number of household living in kuccha houses in comparison to pucca and semi-pucca houses. However, as we shall see later, most families were extremely poor both in terms of income and their access to basic services. Figure 10
House Type

40% 35% 30% 25% 20% 15% 10% 5% 0%

36%

34%

23%

Kuccha

Semi-Pucca

Pucca

Table 12 examines the relationship between income and house type, and reveals that most of the families who lived in kuccha houses were in the bottom two income

35

quartiles, so again we see that there is a connection between income and the type of house a family resides in; the poorer members of the community live in worst houses. Table 12
Income Quartiles 1 2 3 4 Total Kuccha 34% 43% 15% 7% 100% Type of House Semi-Pucca 27% 39% 21% 13% 100% Pucca 16% 30% 22% 31% 100%

With regards to ownership, it was found that 85% of families owned their houses while 15% paid rent. Thirty-eight per cent of residents owned the land on which their houses were built, the rest of the land was owned by the government and the municipality. With regards to taxes, it was found that 66% of the households were supposed to pay tax to the municipality. Of those who had to pay tax, 6% admitted of not paying it, 30% did not answer, and 63% said they paid their tax. Almost 82% of respondents said that they did not have a ration card. ii. Income Figure 11
% of population 1

Nadiad Income CDF

.8

.6

.4

.2

0 0 250 India WB 1000 per capita monthly income 1500 2000

Mean monthly per capita income of these households was Rs. 373.51, which clearly showed that this was a poor community in terms of income. The Income CDF, (Figure

36

11) shows that if the national poverty line was applied to measure poverty then approximately 85% of the population would be considered poor, and under the international line 90% of the population would be considered poor. The FGT measures, presented in the Table 13, have been calculated on the Indian poverty line of Rs.538.60 per month, and revealed that approximately 87% of the families earned less than this. The poverty gap (a=1) which gives a sense of the depth of poverty tells us that, on average, the poor in this sample were 40% below the poverty line. Table 13
Foster-Greer-Thorbecke poverty indices, FGT(a) a=0 a=1 a=2 0.87019 0.40359 0.2255 FGT(0): headcount ratio (proportion poor) FGT(1): average normalised poverty gap FGT(2): average squared normalised poverty gap

*No data on savings or loans is available iii. Infrastructure Water Seventy-one per cent of households had access to some kind of water facility, while the rest stated that they had no regular access to water. Of those who had access to water, the various sources from where they accessed water have been illustrated below in Figure 12. Figure 12
Water Source

60% 50% 40% 30% 20% 10% 0%

55%

22% 12% 6% 5% Tanker

Municipal Connection

Stand Post

Hand Pump

Kuvo

The division between those who had access to water and those who did not was related to income levels, as the regression analysis in Section 5 will also show. Table 14 presents data on income and access to water, and indicates that the majority of the households who had access were from the top two income groups while most of those who did not have access to water facility were from the bottom two groups.

37

Table 14
Income Quantiles 1 2 3 4 Total Water Facility Yes No 24% 35% 18% 41% 20% 17% 38% 7% 100% 100%

Toilet Out of the total slum population only 43% families had adequate toilet facility. Of the families without access to proper toilet facility, 58% used open areas to defecate, and only a small percentage made use of public toilets. Figure 13
Toilet

60% 50% 40% 30% 20% 10% 4% 0% Open Area In-House Public Toilet 2% Other 58% 36%

Waste A total of 58% households had drainage in their homes, while the other 42% had no such facility. Around 65% households did not have a dustbin close by, but they admitted that in most cases there was a facility of a community dustbin. Roads Only 63% of the roads in these slums were kuccha; the rest were pucca. Electricity It was found that 70% houses had electricity meters, and their average monthly bill was Rs.306. Three quarters of the community had street lights. Drainage Around 90% houses had no storm water drainage.

38

iv. Health Unfortunately no data on health expenditure or frequency of illnesses was available for Nadiad. By examining other indicators, we can surmise that an equal number of families chose to use the public facilities over private facilities. v. Regression Analysis As there was a lack of data on health and other variables of interest, the only regression presented here is one which investigates the link between per capita monthly income and having access to basic infrastructure. It was found that there was an inverse relationship between infrastructure and income, this summarises some of the earlier findings. The equation used to examine this link was: Per Capita Monthly Income = C + 1(rent) + 2(water) + 3(drainage) + 4(sewerage) + 5(toilet) + Regression Output 5
pcminc water* drainage sewerage* toilet* _cons Observations Adjusted R *1% Significance Coef. 43.32561 -9.713703 33.79729 87.58142 289.0916 6,086 0.11 Std. Err. 8.036331 9.674141 6.649395 6.491663 7.465252 P-Value 0 0.3 0 0 0

The output showed that monthly income was higher for households who had the amenities stated earlier. One exception was the drainage facility, which showed a negative sign, suggesting that if a household had drainage then their income was lower; however the variable was not statistically significantly different from zero. The output for the other three variables, water, sewerage, and toilet, tells us that a household which had access to these facilities earned respectively Rs.43, Rs.33 and Rs.87 more than the households who did not have access to such amenities. Intuitively this is what we would expect to see: it is predominantly the poorer households who do not have adequate access to basic amenities. vi. Summary The general access to basic services in these slums was bad, one-third of the respondents did not have adequate access to water, and over half defecated in open areas for lack of toilet facilities. However, in the case of housing, the situation was not as severe; only 36% of the population lived in kuccha houses. However, using income as the measure of poverty it is clear that a large proportion of families earnings were below the poverty line.

39

2.1.3 Godhra
Godhra, is a municipal town in Gujarat with a population of around 130 000 (2001 census). It is also the administrative capital of the Panchmahal District. The population comprise 52% males and 48% females, and the literacy rate is 73% (Census, 2001). A total of 4,249 households were surveyed in Godhra for this study, across 47 different urban slums. The report is presented below. i. General Household Information The average household size in these slums was 4.95, and most residents were from Bhakshi caste (Figure 15). A large proportion of the women in this community were married (85%) and in terms of occupation most were housewives. Figure 15 Caste
5%

Other

S.T

13%

Bhakshi

58%

S.C 0% 10% 20%

25%

30%

40%

50%

60%

From Figure 16 we find that majority of the residents, 45%, lived in kuccha houses, about 38% families lived in mixed houses, and only a few in full pucca houses. Figure 16
House Type

50% 45% 40% 35% 30% 25% 20% 15% 10% 5% 0%

45%

21% 17% 17%

Kuccha

Pucca Wall, Kuccha Roof

Full Pucca

Kuccha-Pucca Wall

40

An analysis of the relationship between the types of house a resident used and his income level, it can be seen that those residing in kuccha houses were predominantly from the lower income groups. In general the higher income households lived in better houses . Table 15: Relationship between Type of Houses and Income Level
Quartiles Income 1 2 3 4 Total of Pucca 11% 34% 16% 38% 100% Pucca Wall, Kuccha Roof 14% 32% 29% 24% 100% Kuccha-Pucca Roof 21% 33% 25% 20% 100% Kuccha 35% 41% 15% 9% 100%

As far as ownership and tenure are concerned, 76% of all households possessed some kind of property, the majority of which were houses. Figure 17
Property Ownership

3%

6%

House Land House & Land

72%

Almost all families lived on land that they did not own. It was owned either by the Nagar Palika or individual persons. Of those who owned property only 22% had legal papers supporting their claim. The households who did not own their dwellings paid an average Rs 320 per month as rent. Around 82% households did not possess a ration card. ii. Income The average per capita monthly income was as low as Rs.366.48 in the slums in Godhra. The average expenditure of the household was Rs. 354 per month. From the CDF below (Figure 18) we can see that a large group of households earned incomes that placed them below both the national and the international poverty lines.

41

Figure 18
% of population 1

Godhra Income CDF

.8

.6

.4

.2

0 0 250 India 1000 WB per capita monthly income 1500 2000

The FGT measures gave a more accurate estimation of the extent of this poverty. Table 16
Foster-Greer-Thorbecke poverty indices, FGT(a) a=0 a=1 a=2 0.86886 0.39807 0.21019 FGT(0): headcount ratio (proportion poor) FGT(1): average normalised poverty gap FGT(2): average squared normalised poverty gap

Table 16 shows that the headcount index (a=0) was 0.868, which means that approximately 87% of the population would be considered poor under the National poverty line. The World Bank poverty line measured at Rs. 650 gave an even higher estimate of 93%. The second red line in Figure 18 indicates the World Bank poverty line. The poverty gap in this population is 39%. Saving Almost 97% of households did not save. The handful of families who did save, were able to save a mean amount of Rs. 452 per month. Only 20% of households stated that they made use of loans, and the data on the sources of these loans was unreliable as most households did not answer the questions. iii. Infrastructure Water

42

The sources of water for families in these slums were quite diverse, and from the Figure 19 given below, we can see that the number of households with their own tap was smaller than those found in other cities. It is not clear from the survey what their other sources of getting water were. For households who relied on conventional water sources, approximately 81% stated that they had running water for only 1-2hrs per day, and half of the residents said that the water pressure was inadequate to meet their needs during that time. Figure 19
Water Source

40% 35% 30% 25% 20% 15% 10% 5% 0%

36% 27% 21% 15%

Other

House Tap

Public

Hand Pump

Toilet The residents faced serious lack of adequate toilet facility, with more than half of the families having to defecate in open areas (Figure 20). The health risks and other problems associated with this have already been mentioned. Figure 20
Toilet Type
57% 60% 50% 37% 40% 30% 20% 10% 0% Open Space Own Public 5%

About 44% of households had bathroom facility in their house, around 26% of the population used public bathrooms to wash and bathe, and 30% used other facilities. Figure 21

43

Bathroom Facility

Other

30%

Public

26%

House

44%

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

Waste The study showed that 40% of the residents had no garbage disposal facility and this led to accumulation of waste in the slums. Drainage Only 7% of the residents had drainage in their homes and the rest had facility.

no drainage

Electricity Around 71% of the families had electricity, and all of them used an in house meter. iv. Health Despite the extent of poverty in these slums and lack of adequate infrastructure not many cases of illnesses were reported. In fact 93% of the residents said that there were no frequent illnesses in their families. However, this could be because many households did not answer the question on illness. About 56% of the residents said that they used government medical facilities, and the rest went to private practitioners. The average expenditure on healthcare was Rs. 190 per month. v. Regression Analysis Initial analysis to assess whether there existed any relationship between health (expenditure on health) and infrastructure in the slum population of Godhra revealed that there was no a significant correlation or insight into the relation between these variables. However, this could be due to the fact that not many had responded to the questions on illness. Another reason for the lack of correlation could be because these households were extremely poor and simply could not afford to spend money on healthcare (calling the use of health expenditure into question). Perhaps people were falling ill due to poor sanitation and lack of infrastructure but they were simply too poor to spend money on healthcare, and thus no relationship between health expenditure and infrastructure showed up.

44

However, regression analysis showed that income was correlated with insufficient access to water, toilet, drainage, garbage facilities, and that these were also the reasons why the household did not save. The equation and output are presented below.
Per capita monthly income = C + 1(water) + 2(toilet) + 3(drainage) + 4(save) + Regression Output 6
pcminc water* toilet* drainage* save* _cons Observations Adjusted R *1% Significance Coef. 30.159 90.663 57.834 210.359 305.026 4,215 0.26 Std. Err. 7.644 7.736 15.891 25.098 6.643 P-Value 0.000 0.000 0.000 0.000 0.000

Here we can see that some of the independent variables have a strong relationship with income. Particularly it is obvious from the output that the group of households who saved money were earning far more than those who did not save. On average, those who saved earned Rs.210 per month more than those who did not. The other independent variables also showed a strong link with income, especially access to toilet facility. All were significant at the 1% level. The result corroborated what we would intuitively expect the ones who did not have access to the services were the ones who earned lower incomes. Overall, the regression showed relatively good explanatory power, and standard errors for the regressors were low. vi. Summary The data on slums in Godhra revealed that most families there faced a fairly bleak situation. Most of them were below the poverty line, had little access to basic amenities (over half of them did not have access to proper toilet facility), and very few had adequate housing. The regression analysis failed to pick up any link between health and access to basic infrastructure, but did show that there was a correlation between being poor and having inadequate access to basic infrastructure.

45

2.1.4 Ankleshwar
Ankleshwar is an industrial township in Baruch District of Gujarat. It has a population of 112 643 (Census, 2001), and has a literacy rate of 76%. A survey of 293 households in 40 urban slums of Ankleshwar was carried out for this study. i. General Household Information The average household size here was 5.32 people per house, and the majority of the people earned daily wages. Table 17 shows that 81% women were daily wagers, 16% were house-wives and a small percentage of 3% had their own business. Table 17: Occupation of the Respondents (%)
Daily Job Own Business Daily Wager House Wife Women 3% 81% 16% Others 9% 91% 0%

The survey showed that housing conditions were poor; 75% of the families lived in kuccha houses. Figure 22
Type of House

75% 80% 70% 60% 50% 40% 30% 20% 10% 0% Kuccha Mixed Pucca 19% 6%

Table 18 tabulates the earnings of households against the houses they lived in. This was found to be helpful in describing the community. The majority of the poor families lived in kuccha dwellings, while most of the families living in the pucca or semi-pucca (Mixed) houses were in the upper three income quantiles. Noticeably there were no families in the poorest quantile who owned a Pucca house.

46

Table 18
Income Quantile 1 2 3 4 5 Type of House Kuccha Mixed 30% 13% 29% 15% 22% 24% 7% 9% 13% 39% 100% 100% Pucca 0% 15% 31% 31% 23% 100%

As far as tenure was concerned, it was found that the majority of households owned the houses in which they lived (86%) and a comparatively small number, 14%, lived in rented houses. Figure 23
House Ownership

14%

86%

Own

Rent

ii. Income The monthly mean per capita income among these households was Rs. 364.25, while their average per capita expenditure was Rs. 359.10. The CDF in figure 24 illustrates how income was spread across population. One can see that over 80% of the households were earning below the national poverty line, and around 90% were below the international line. This showed that the residents of these slums were very poor.

47

% of population 1 . 8 . 6 . 4 . 2 0 0 25 0 Indi a

Figure 24 Ankleshwar Income CDF

W 100 B per capita0monthly income

150 0

200 0

The FGT measures tell us that 80.7% of the population lived below the Indian poverty line. Looking at the poverty gap it is clear that the depth of poverty was considerable with 36% below the national poverty line of earning of Rs. 538 per month. Table 19
Foster-Greer-Thorbecke poverty indices, FGT(a) a=0 a=1 a=2 0.80783 0.36151 0.19399 FGT(0): headcount ratio (proportion poor) FGT(1): average normalised poverty gap FGT(2): average squared normalised poverty gap

Savings The survey showed that only 10% of the families saved while the remaining 90% had no saving. Those who saved managed to put away around Rs 460 per month. Also most of the respondents who had saving, they had put their money in bank. Table 20 below illustrates that the majority of those who saved came from the upper three income quantiles, while most of those who did not save were in the bottom two quantiles. Table 20
Income Quantiles 1 2 3 Saving No 26% 28% 22% Yes 11% 11% 22%

48

4 5 Total

6% 18% 100%

22% 33% 100%

Loans With regards to loans, the survey showed that 33% of the families took loans, and the average loan amount was Rs. 11,500, with a median value of Rs. 10,000 being borrowed. It was interesting to note that all income groups irrespective of their income profile took loans; hence no clear pattern emerged. No data on the sources of loans was recorded. iii. Infrastructure Water As Figure 25 shows 65% of the families had individual water connection, and the remaining 35% did not have access to private connection. Figure 25
Water Connection

35% Individual Connection No Connection 65%

Data on spending for water facilities showed that on average a household had spent Rs.1,010 to install water connection in their house and the families that did not at present have water connections were willing to pay Rs.420 to get a connection. Table 21, below, illustrates that it was predominantly the households with lower earnings that did not have access to individual water connection: 41% of those who did not have water connection were from the poorest quantile. Table 21: Households having Individual Water Connection
Income Quantile 1 2 3 4 5 Total Individual Water Connection No Yes 41% 16% 25% 10% 20% 23% 6% 26% 7% 24% 100% 100%

49

Toilet As per Table 22, around 24% did not have their own toilets. Those who had their own toilet had paid an average of Rs.3526 for the facility, and those who did not have were willing to pay approximately Rs.604 (the maximum offer was Rs. 2000). Again, the bulk of those who did not have toilets were from the poorest two quantiles of earners. This is what one would expect, because poorer households generally do not have the financial capacity to afford their own toilets. Table 22
Own Toilet No Toilet Toilet Percentage 76% 24% Paid/Willing to Pay Rs. 3526 Rs. 604

Waste Seventy-seven per cent of families had some kind of sewerage connection, in or nearby their house, while 23% did not. The average amount spent to obtain this connection was Rs. 1222 per family, and those who did not have a connection were willing to pay a mean amount of Rs. 444. Here too it was found that the majority of those without sewerage disposal were households with the lowest income. Drainage Eighty-nine per cent of houses did not have adequate drainage facility to deal with rain water. Around 80% complained that they faced water logging when it rained. Roads Sixty-four per cent of households stated that there were no roads in their area. Around 73% stated that when it rained lack of proper roads caused water to accumulate, and when that happened, no one came to attend to this problem. Electricity The survey showed that 64% of the houses had electricity meter and on average each family spent Rs. 1227 per month on electricity. Those who did not have a meter were willing to spend Rs. 954 to get one. It was also found that 74% of the houses were located on streets that had street-lighting, but they also stated that 20% of the street lights did not work. iv. Health No data was collected on health. v. Regression analysis We did not see the need to use regression analysis in this case as the tables and figures presented clearly delineate the condition. However, use of another regression on

50

income and infrastructure was found to be helpful as an overview, and has thus been presented here. Per capita monthly income = C + 1(save) + 2(toilet) + 3(water) + 4(sewerage) + 5(electricity) + Regression Output 7
pcminc Coef. 176.372 64.924 82.869 43.051 3.720 266.796 254 0.16 Std. Err. 42.406 37.930 30.143 38.209 13.649 21.537 P-Value 0.000 0.088 0.006 0.261 0.785 0.000 save* toilet** water* sewerage elec _cons Observations Adjusted R *1% Significance **10% Significance

The results of this regression indicate that families who had access to basic amenities and had saving had comparatively higher incomes than those without amenities and those who did not save. On an average, households that saved earned Rs.176 per month more than the households who did not save. Higher income was also found to be linked with having access to adequate toilet and water facilities, and sewerage connection. The magnitudes can be judged by the coefficients. The result was similar to what was found in the analysis above. The standard errors were relatively high, which may be due to small sample size. vi. Summary Finally we can conclude that a large number of households in slums of Ankleshwar were very poor, most of them did not have access to basic infrastructure and services, and those who were most in need were also the ones who earned the least. Only a small number of families saved. Overall, the housing conditions were poor, and most residents lived in kuccha dwellings. Regression analysis showed that households with higher income were the ones that saved, had access to water and toilet facilities, and also had a sewerage connection. But the standard errors were large.

51

52

2.2

Rajasthan

2.2.1. Jaipur
Jaipur, commonly known as the pink city, is the capital of Rajasthan, and has a population of around 2.3 million (Census, 2001). The annual economic growth rate of Jaipur is 4.5%, and the city has a large handicrafts industry which employs 22% of the workforce (CDP, 2006). Jaipurs water supply covers about 86% of the population, but is under pressure due to steady increase in its population. Many of citys surface water sources are heavily polluted. Its sewerage network covers only 56% of the population, and they do not cover the slum areas. Only half of the solid waste gets collected in the city. Jaipur also faces rampant unplanned growth and development. The city has a total of 183 slums where 31% of its total population reside (CDP, 2006). For this survey, data was collected from 3,706 households residing in 15 different urban slums. i. General Household Information The average household size in these slums was just 3.72 members per house. One-third of the women residing in slums worked at home, and those who worked outside lived within a 4km radius of their workplace. The survey showed that the general condition of slums in Jaipur was superior to the condition of slums in most other cities studied in this survey. Secondly, the families residing in these slums were better off than those residing in slums of other cities. The residents of slums were largely caste Hindus. This is illustrated below in Figure 26. No identifiable link was found between household income and caste, indicating caste not to be a determinant factor in earnings. Figure 26
Caste

OBC

16% 18% 9% 47%


5% 10% 15% 20% 25% 30% 35% 40% 45% 50%

S.T

S.C

General 0%

As shown in Figure 27 over half of the residents lived in kuccha houses. Closer analysis of the data showed that the type of housing was not as closely linked to income as in the slums of other cities. This suggests that in addition to income there were other factors that influenced the type of house that a family chose to live in, such as length of residency.

53

Figure 27
Type of House

60% 50% 40% 30% 20% 10% 0%

54%

18%

18%

Kuccha

Pucca

Semi-Pucca

The histogram below (Figure 28) shows the number of years residents had lived in the area, where the average length of stay was about 19 years. Figure 28: Length of Stay
25 Percent 0
0

10

15

20

20

40 60 Length of Stay (years)

80

100

By linking the type of house with the length of stay (Table 23) it became clear that on average those who had lived in the area for a long time were more likely to live in better houses. Table 23:Length of Stay in Type of Houses
House Type Kuccha Semi-Pucca Pucca Length of Stay 13.71 18.96 22.44

Land and house ownership were also factors that could influence a familys investment in home. In these slums 65% of land was owned by Jaipur Nagar Nigam, 23% by the Jaipur Development Authority (J.D.A), so only a fraction of the households, about one

54

quarter, owned their land. Of these 70% did not have any legal documents to prove their ownerships. Overall, the data suggested that current housing conditions were relatively insecure, and while this could influence the housing conditions of residents, it did not come out very clearly in the data. ii. Income The average monthly per capita income of each household was Rs. 671.80, which was much higher than most of the other slums which had been examined. The CDF below illustrates that around 40% of the people in these slums were earning below the national poverty line, and around 45% were below the international poverty line. While income was not a complete welfare measure, it did suggest that the families residing here were relatively better off than the ones in the other slums in this study. Figure 29
% of population 1 .8

Jaipur Income CDF

.6

.4

.2

0 0 250 India WB 1000 1500 2000 per capita monthly income

Examining the FGT measures one can see that the headcount index was 40.9% and the poverty gap was only 6% which tells us that the depth of poverty was not severe. Table 24
Foster-Greer-Thorbecke poverty indices, FGT(a) a=0 a=1 a=2 0.40974 0.06419 0.02201 FGT(0): headcount ratio (proportion poor) FGT(1): average normalised poverty gap FGT(2): average squared normalised poverty gap

55

Savings The survey showed that 81% households did not save; of the rest who did, the average monthly amount saved per household was Rs. 608. Again this was substantially more than the average savings of households in other areas. iii. Infrastructure

Water The residents used a variety of water sources; 26% households availed municipal water supply, 37% used public supply, 17% obtained water from neighbours, 8% fetched water from the river, and 2% used a hand-pump. Figure 30
Water Facility
37%

40% 35% 30% 25% 20% 15% 10% 5% 0%

26% 17% 8% 2% Public Municipal Connection Neighbour River Hand Pump

Toilet Approximately 40% households had their own toilets, and almost half of the population in these slums used open areas for their ablutions. Figure 31
Toilet Facility
49% 50% 45% 40% 35% 30% 25% 20% 15% 10% 5% 0% Open Area 40%

9% 2% Own Toilet Other Community

56

Drainage Around 60% households did not have either gutter or sewerage tank, while the rest did. As will be shown later, the lack of sewerage facility was correlated to higher expenditure on healthcare. Roads Fifty per cent of the roads in these slums were pucca, while the other 50% were kuccha. Electricity Only 82% households had electricity. Around 55% of houses reported that there were no street lights near their houses. iv. Health The average household spent Rs. 237 per month on healthcare; this was within a range of Rs.20 to Rs.1500, with the median being Rs.200. Fifty-five per cent of households used private hospitals, and on average the distance to private hospitals (4km) was double the distance to government hospitals (2km). Analysis of the links between healthcare expenditure and infrastructure are presented in Section 5. v. Regression Analysis Data for expenditure on healthcare was available for Jaipur slums, and thus some analysis of the relationships between expenditure and infrastructure was possible. This is presented first followed by the usual regression on income and infrastructure. Equation 1 The first equation was similar to the equation used in Ahmedabad analysis. Health Expenditure = C + 1(toilet) + 2(sewerage) + 3(per capita monthly income) + 4(water) + Regression Output 8
Health exp Co-eff. toilet 6.157 sewerage** -17.47 pcminc* 0.069 water* -34.642 _cons 240.777 Observations 3,382 Adjusted R 0.14 **5% Significance *1% Significance Std. Err. 8.418 8.843 0.01 7.521 13.675 P-Value 0.465 0.048 0 0.003 0

The results revealed that there was a relatively strong correlation between water and health expenditure, and a minor link between sewerage and health expenditure, controlling for the other independent variables. More precisely, having a sewerage

57

connection decreased the average monthly expenditure on health by Rs.17, and having an in-house tap reduced it by Rs.34. These two variables were significant at the 5% level. Having access to a toilet was related to higher spending on health, which was not what one would have expected, but the variable was insignificant and the standard error was high. Per capita income was positively related to expenditure on healthcare but it was very small. Overall the equation was significant and the Adjusted R was 0.14. Equation 2 The second equation analysed the relationship between per capita monthly income, infrastructure, and savings. Per capita monthly income = C+ 1(save)+ 2(sewerage) + 3(toilet)+ 4(water)+ Regression Output 9
pcminc Coef. 507.867 122.173 32.172 34.628 664.842 3,143 0.15 Std. Err. 23.393 14.961 14.097 19.512 20.455 P-Value 0.000 0.000 0.023 0.076 0.000 save* sewerage* toilet** water _cons Observations Adjusted R *1% Significance **5% Significance

The outcome of the survey conducted in Jaipur slums was similar to that of other cities. Two variables, which appeared to correlate with large changes in income, were saving and sewerage. Families who saved were likely to earn Rs.507 per month more than those who did not. Families with sewerage connection earned Rs.122 per month more than those who had no connection. These variables were significant at 1% level. The other two variables (toilet and water) were also related to higher income but had relatively high standard errors. Overall these four variables explained 15% of the variation in income. vi. Summary Urban slums in Jaipur had some interesting characteristics. The per capita average income was relatively high and the family size was small. While these were positive indicators, it was also revealing that most families lived in kuccha houses, half of them had no access to toilet and sewerage facilities, were poor, and their access to water was varied. The analysis in Section 5 highlights the correlation between some of these basic amenities and the money spent every month on healthcare, indicating that families who had adequate access to water and had sewerage facilities generally spent less every month on healthcare; the equation controls for income. Additionally, we found that poorer households were less likely to save and were also less likely to have a sewerage connection, and access to toilet and water facilities.

58

2.2.2 Bikaner
Bikaner is the fourth largest city in Rajasthan with a fairly large population of around 724,000. It is situated in the North-West of Rajasthan in the district of Bikaner (Census, 2001). The data presented here has been taken from a survey of 266 households across 12 urban slums in Bikaner. i. General Household Information The average household size in Bikaner slums was 6.25, and all of the respondents in the survey were married. Three quarters of the families were Hindus. Figure 32 shows the caste wise breakup of the population. Figure 32
Caste

Other S.C S.T General 0% 10% 2% 23% 20% 30% 40% 50% 15%

60%

60%

From the Figure 33 we find that the type of house occupied by the residents varied across slums; 46% families lived in houses which had mixed walls (both kuccha and pucca), 36% lived in kuccha houses, 18% had pucca walls and but a Kuccha roof, and notably only 1% families lived in proper pucca houses. The type of house appeared to be related to income. Like all other slums studied so far here too most families who lived in kuccha houses came from the lowest two income brackets, the families who lived in mixed-wall dwellings also came from the lower half of the income distribution, and those with pucca walls were predominantly from the higher income brackets. Figure 33
House Type

50% 45% 40% 35% 30% 25% 20% 15% 10% 5% 0%

46% 36%

18%

1% Mix walls Kuccha Pucca wall, Kuccha roof Pucca

59

With regards to tenure, the study revealed that 90% families in these slums owned their houses, but only 7% owned the land on which they had built houses. Most of the land was owned by Bikaner Municipal Corporation (BMC) and other private owners, while a small portion was government land. Table 25 presents the ownership data. Table 25
Land Ownership Private 38% Gov 9% BMC 53%

On average, many families had been living in these slums for 23 years, and some claimed that their families had been living in this area for 100 years. Figure 34
Residency
15 Percent 0 5 10

20

40 years

60

80

100

It was surprising that after living in the same area for so long very few residents had built pucca houses. Furthermore, over half of the residents did not have any legal papers to prove ownership of their houses or land. ii. Income On average, per capita monthly income was Rs.381.63, which was relatively low, and the per capita expenditure per month was Rs.362. One can see from the CDF below that over 90% of the population had income below the national poverty line, and if one used the international line then almost entire population would be classified as poor.

60

% of population 1 .8

Figure 35 Bikaner Income CDF

.6

.4

.2

0 0 250 India WB 1000 1500 2000 per capita monthly income

The FGT indices (Table 26) show that 92% of the population earned less than Rs.538.60 per month. The poverty gap reveals that on average those below poverty line were earning 32% less than the poverty benchmark. Table 26
Foster-Greer-Thorbecke poverty indices, FGT(a) a=0 a=1 a=2 0.92 0.32 0.13 FGT(0): headcount ratio (proportion poor) FGT(1): average normalised poverty gap FGT(2): average squared normalised poverty gap

With regards to savings, it was found that only 2% households saved, and 67% of the families took loans from various sources as shown in Figure 36. We find that most families borrowed from private money lenders (PML). Figure 36
Loan sources

70.00% 60.00% 50.00% 40.00% 30.00% 20.00% 10.00% 0.00% 22.64% 8.30% 3.02% 67.55%

Bank

NGO

PML

Other

61

iii.

Infrastructure Water Figure 37


Water Sources

70% 60% 50% 40% 30% 20% 10% 0%

62%

25% 13%

Own Tap

Stand Post

Tanker

Only 66% households had piped water supply and on average, water was available for 1.5hrs per day. Those who did not have piped water used either public stand posts, or hand pumps, or water tankers. Toilet Many of the residents in this community did not have adequate access to toilet facility. Only 43% residents had their own toilets and the rest used open areas for their toilet needs. The survey also showed that around 56% households had bathrooms inside their house. Figure 38
Toilet

Ow n, 43% Open Space, 56%

Joint/Public, 1%

Roads The slums in Bikaner had 65% kuccha roads. The rest were pucca. Around 42% of the roads did not have any drainage system. Waste Collection The waste disposal system in these areas was very poor as 96% of the residents had no waste disposal or waste collection facility.

62

Electricity Around 65% households in these slums had electricity meters. On average, households were paying Rs 465 per month for electricity, and more than half of the residents stated that their usage of electricity was legal. With regards to street lights, 55% of the residents said that they had street lights in their neighbourhood. iv. Health The average monthly expenditure on healthcare in these slums was Rs. 228, and a vast majority of the households (85%) made use of the public medical facility. Most households lived within 3 km radius of the nearest hospital, but over half of the residents had no childcare facility in their area. The data on health also included some information on whether households reported having frequent illnesses in the family. The data showed that 16% households experienced frequent illnesses. v. Regression Analysis The first analysis again used health expenditure as the dependent variable in order to see whether families who spent more on healthcare were generally also those who had poor access to basic services. To reiterate, the reasoning was that those without basic services were more likely to get sick from living in unhealthy conditions, with poor water quality and bad sanitation. The second equation was the usual regression where income was regressed against a number of infrastructural variables. The specified equation was: Health Expenditure = C + 1(per capita monthly income) + 2(piped-water) +3(waste) + 4(toilet) + Regression Output 10
healthexp pcminc Piped-water waste* toilet* _constant Observations Adjusted R *10% Significance Coef. 0.066 -30.306 -20.924 -43.011 161.297 236 0.18 Std. Err. P-Value 0.078 5.234 6.771 14.265 34.172 0.394 0.131 0.071 0.078 0

The results revealed some expected relationships. Higher levels of income were associated with a very small increase in health expenditure. Looking at the infrastructure variables the results showed that on average, households with access to piped water, waste (sewerage) disposal, and toilet facility spent respectively Rs.30, Rs.21, Rs. 43 less on healthcare, which showed a fairly large impact on expenditure. Both toilet and waste were significant at 10% level.

63

Equation 2 This equation analysed the link between households monthly income, savings and infrastructure. The formula was similar to the one used in the previous analyses. Per capita monthly income = C+ 1(save)+ 2(piped-water) + 3(sewerage)+ 4(toilet)+ 5(drain) + Regression Output 11
pcminc Coef. Std. Err. 66.792 23.156 62.248 22.547 114.287 17.906 P-Value 0.755 0.003 0.102 0.671 0.698 0.000 save 20.906 pipedwater* 68.887 waste** 79.595 toilet -9.597 drain -44.406 _cons 345.116 Observations 245 Adjusted R 0.09 *1% Significance **10% Significance

However, in spite of using the same equation the outcome of the survey of slums in Bikaner did not provide much insight. One could see that if a household had access to piped water they generally earned Rs.68 per month more than those who did not have this access, and that, those who had waste disposal service also earned higher incomes (Rs. 79) than those who did not have this facility. However, the standard errors for waste were very high. None of the other variables were significant, while toilet and drain report signs which are opposite to what we would expect. Thus only conclusion that can be drawn from this study is that the poorer households were less likely to have access to piped water. Furthermore, the adjusted R value was relatively small, so the model lacked explanatory power. It is possible that small sample size had affected the results. vi. Summary Households in the slums surveyed here were very poor by income standards, and had limited access to basic amenities such as water, toilet facility, and sewerage disposal. There was no real saving among the residents, which of course one could understand in such a low-income setting. Most of the loans were availed from private money lenders, and though this was typically not the best loan source for poor households as interest rates were high, they were forced to go to the private money lenders because presumably other more conventional sources were inaccessible to them due to lack of collateral. Closer analysis of some of the relationships within the data revealed that having piped-water, waste disposal and toilet facility were strongly linked to how much a family would spend on healthcare. There was also a link between access to piped water and earning higher monthly income.

64

2.2.3 Jodhpur
Jodhpur is the second largest city in Rajasthan with a population of about 2,860,000 (Census, 2001). It is sometimes called the blue city due to the colour of many houses. It has a big handiworks industry. The city has a score of 0.56 on the Human Development Index (HDI), thus ranking 13 in Rajasthan based on the HDI and the literacy rate is 57% (www.jodhpur.nic.in, 2009). The data for this report came from a survey of 120 households from Jagadamba Colony in Jodhpur. It must be noted that this was a relatively small sample. i. General Household Information The average household size in this area was 6.3, which was higher than any of the other slums in this study. Most residents belonged to the General Caste (78%). Figure 39 illustrates the type of houses that residents used. Clearly the majority of the families lived in pucca houses (74%), which was considerably more than in other cities. On average, the houses had two rooms. Figure 39
House Type

80% 70% 60% 50% 40% 30% 20% 10% 0%

74%

14%

12%

Pucca

Kuccha

Semi Pucca

Again it is useful to illustrate how income was found to be linked to the type of houses families lived in. Even intuitively the link made sense: those with more money could afford better housing. One can see from Table 27, below, that not a single family out of those who lived in kuccha houses, came from the top two income groups, while the majority of those who lived in pucca houses were from the two highest income groups. Table 27
Income Quartiles 1 2 3 4 Kuccha 38% 63% 0% 0% Type of House Semi-Pucca 56% 33% 0% 11% Pucca 16% 18% 24% 43%

65

The survey revealed that 83% families owned their houses, while the rest paid rent. However, in terms of land ownership almost all land was owned by private owners or the municipality. It also revealed that 77% families did not have a ration card. Figure 40 illustrates the length of residency among households. The average length of residency was found to be 20 years. Figure 40
Length of Residency
15 Percent 0
0

10

20

40 60 yrs residency

80

100

ii. Income The average per capita monthly income in this community was Rs.512.50 for the 120 households. The CDF, Figure 41, illustrates the spread of income across this population. It was found that about 70% of the population earned less than the national poverty line, and about 85% earned less than the World Bank specified international standard. Figure 41
% of population 1 .8

Jodhpur Income CDF

.6

.4

.2

0 250 Indian WB 1000 per capita monthly income 1500

66

To gain a clearer sense of poverty the FGT measures were applied. As Table 28 reveals about 71% households were earning incomes which were below the national poverty line, and the poverty gap was 20%. Table 28
Foster-Greer-Thorbecke poverty indices, FGT(a) a=0 a=1 a=2 0.70755 0.20624 0.08763 FGT(0): headcount ratio (proportion poor) FGT(1): average normalised poverty gap FGT(2): average squared normalised poverty gap

iii.

Infrastructure Water It was found that 42% households had municipal water connection in their houses, and while 6% used public stand-post; no data were available for the remaining households. Toilet As shown in Figure 42, 75% of respondents had toilet facility; unfortunately it was not clear from the data what type of toilet facilities they had or what the other 25% of households used. Figure 42
Toilet Acess

Toilet No Toilet

Waste/Drainage A large percentage (68%) of residents had no sewerage facility in their homes, while 83% said that they had no storm-water drainage for their homes. Roads All roads in this community were kuccha. Electricity

67

The survey revealed that around 68% households had electricity, and on average, households paid Rs. 910 per month for it. Three quarters of households said they had streetlights in their neighbourhood. iv. Health No data on health was found. v. Regression Analysis The lack of data for Jodhpur, as well as the small sample size made any statistical analysis difficult, but the regression analysis here attempts to show the link between income and infrastructure As in the other cities studied in this report, here too it was found that a household with higher income generally had access to basic amenities such as water, toilet, sewerage, and storm water drainage. Per capita monthly income = C+ 1(water)+ 2(sewerage) + 3(drainage)+ 4(toilet) + Regression Output 12
pcminc water sewerage drainage toilet _cons Observations Adjusted R Coef. 142.782 12.787 65.120 48.171 376.599 97 0.1 Std. Err. 70.426 74.777 135.231 85.668 63.520 P-Value 0.046 0.865 0.631 0.575 0.000

From the output it is clear that water had a very strong relationship with earnings, i.e. if a household had access to water their income was Rs.142 more that those households who had no access to water. But this is the only significant variable and had a high standard error. All the other variables were insignificant and had very high standard errors, which called their accuracy into question. The main reason for this was the small sample size of only 97 observations. Thus, looking at this data one can only conclude that there was a relationship between income and water. vi. Summary In the slums in Jodhpur the number of people who lived in pucca houses was more than those found in other cities, and the majority had access to water, toilet and sewerage facilities. Given these indicators the community was found to be relatively well-off compared to those in some of the other cities. However, looking at the monthly per capita earnings, approximately 70% of the households were earning below the national poverty line. The regression analysis showed that the households who lacked access to water were poorer than those who had access to water. That is, there was a correlation between being poor and having a lack of access to water, but the small sample size affected the accuracy of this result.

68

SECTION 3
Inter-City Comparison
This section presents a comparative analysis of the data on general household information, income, infrastructure and health, collected from all the seven cities. The comparison would help readers to get an overview of the condition of the slums in these cities. An initial summary of the cities and the survey data collected is shown in the Table 29 presented below. Table 29: Sample for the Study
City Name Ahmedabad Nadiad Godhra Ankleshwar Jaipur Bikaner Jodhpur No. of Slums 39 97 47 40 15 12 1 No. of Households Surveyed 12,773 6,566 4,249 293 3,706 266 120

1. General Household Information Household size As shown in Table 30 the average number of persons per household ranged from 3.7 to 6.3, with Bikaner and Jodhpur having the largest household size, and Jaipur the smallest. Table 30
Ave. Household Size Bikaner Jodhpur Ankleshwar Godhra Ahmedabad Nadiad Jaipur 6.3 6.3 5.3 5.0 4.6 4.6 3.7

69

Residency The graph presented in Figure 43 shows the average residency period of the families in the current neighbourhood for only four cities, as data for the other three was not available. It is evident from the graph that most families had been residents for quite a number of years. Figure 43
Average Years Residency
25 20 15 10 5 0 Ahmedabad Jaipur Jodhpur Bikaner 18.6 19.6 20 23.5

House Type There were large differences in the type of houses that residents had across the seven cities. For example 73% of the residents in Ahmedabad slums lived in kuccha houses, while in Jodhpur only 7% did. It was not clear from the data which factors accounted for these differences. However, in some cities type of house used was linked to income, which could be partly the reason. Table 31: Type of House
Kuccha Ahmedabad Nadiad Godhra Ankleshwar Jaipur Bikaner Jodhpur 73% 37% 45% 75% 54% 36% 7% Semi-Pucca 12% 35% 38% 19% 19% 63% 8% Pucca 15% 23% 17% 6% 18% 1% 74%

House Ownership The survey showed that except in Jaipur, in all other cities, the majority of the residents owned their houses. Table 32

70

House Ownership Own Ahmedabad 63% Nadiad Godhra Ankleshwar Jaipur Bikaner Jodhpur 85% 72% 86% 25% 99% 83%

Rent 37% 15% 28% 14% 75% 1% 17%

Land Ownership In contrast to house ownership, the survey showed that most of the slum residents across seven cities did not own the land on which their houses were built. Figure 44, below, highlights this difference. Table 33
Land Ownership Own Don't Own Ahmedabad Nadiad Godhra Ankleshwar Jaipur Bikaner Jodhpur 33% 38% 6% N/A 10% 38% 3% 67% 62% 94% N/A 88% 62% 97%

Comparison of House and Land Ownership Figure 44


Ownership: House vs. Land (% )
100% 80% 60% 40% 20% 0% Ahmedabad Nadiad Godhra Ankleshwar Jaipur Bikaner Jodhpur

House Ownership

Land Ownership

71

Legal Land Ownership Another interesting fact concerning land ownership was that the majority of households who did own land did not have any proof of their legal ownership. Table 34
% Households with legal proof of their Land Ownership* Yes No 75% N/A 78% N/A 70% 53%

N/A N/A *Percentage of households with legal documents of land ownership within the percentage who said that they own land (refer to above table)

Ahmedabad Nadiad Godhra Ankleshwar Jaipur Bikaner Jodhpur

25% N/A 22% N/A 30% 47%

2. Income Table 35 presents data on per capita monthly income, and also the percentage of the population in each city that was earning below the national poverty line. As is evident from this table, Bikaner had the largest number of poor households, whereas Ahmedabad had the smallest. The level of poverty is illustrated in Figure 45. Table 35
Cities Per Capita Monthly Income* Rs.702 Rs.374 Rs.366 Rs.364 Rs.672 Rs.382 % Poor** 32% 85% 87% 82% 41% 92%

Rs.513 71% *The average monthly income earned, per person **The percentage of the population below the National Poverty line (Rs.538.60/person/ month)

Ahmedabad Nadiad Godhra Ankleshwar Jaipur Bikaner Jodhpur

72

Figure 45
% Poor (Headcount)

Bikaner Godhra Nadiad Ankleshwar Jodhpur Jaipur Ahmedabad 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Savings Not all surveys provided data on savings, so whatever was available has been presented in Table 36. It is clear from the table that the number of households who were saving was small; the highest saving rates were seen in Ahmedabad (which also had the lowest number of poor households). Table 36
Savings Yes Ahmedabad Ankleshwar Godhra Nadiad Jaipur Jodhpur Bikaner 23% 10% 3% 19% 2% No 77% 90% 97% 81% 98% Rs. 451 460 452 608 -

Loans Data on how many families had taken loans was available for only three cities and that has been presented in Table 37. Here again we find that Ahmedabad had the highest percentage of households who took loans. Table 37
Loans Yes No Ahmedabad 74% 26% Ankleshwar* 33% 67% Godhra 20% 80% *In Ankleshwar the average loan amount was Rs.11500

73

3.

Infrastructure Water Data on access to water showed that there was variation across cities; in Ahmedabad, Ankleshwar and Bikaner over 60% households had their own taps. However, in Godhra and Jaipur only a small percentage of households had their own taps. Table 38 Access to Water

Water Facility In-House Tap Public Tap Neighbour Private Seller Hand Pump

Ahmedabad 66% 14% 8% 4% 0%

Nadiad 40% 17% 43% 0% 0%

Godhra 27% 21% 36% 0% 15%

Ankleshwar 65% 35%

Jaipur 26% 37% 17% 8% 0%

Bikaner 66% 34%

Jodhpur 57% 43%

Toilet Data on access to adequate toilet facility across seven cities is presented in Table 39. It shows that in four out seven cities around half of the population were using open areas for toilet purpose. Table 39 Access to Toilet Facility
Toilet Facility In-House Neighbour Public Open Area Ahmedabad 72% 3% 11% 14% Nadiad 43% 0% 4% 51% Godhra 37% 0% 5% 57% Ankleshwar 76% 24% 0% 0% Jaipur 40% 0% 2% 49% Bikaner 43% 0% 1% 56% 25% Jodhpur 75%

Drainage Table 40 Drainage Facility


Sewerage/ Drainage Yes No Ahmedabad 85% 15% Nadiad 31% 42% Godhra 7% 93% Ankleshwar 77% 23% Jaipur 40% 60% Bikaner 2% 98% Jodhpur 32% 68%

From the Table 40 we find that drainage facility varied across cities. While in Ahmedabad and Ankleshwar 85% and 77% households respectively had drainage system, in Godhra and Bikaner only 7% and 2% households respectively reported to have this facility.

74

Electricity Overall it was found that compared to all other facilities access to electricity was good in all the cities (Table 41), particularly Ahmedabad, where almost all households had electricity. Table 41 Access to Electricity
Electricity Yes No Ahmedabad 92% 8% Nadiad 70% 25% Godhra 71% 29% Ankleshwar 65% 35% Jaipur 82% 10% Bikaner 65% 35% Jodhpur 68% 32%

4. Health The data on the type of medical facilities preferred was again mixed across all cities, (Table 42) with no clear majority showing preference for private care over public care system. One would have expected the poorer populations to prefer public health care system over private one, but this did not emerge clearly from our survey. Preferred Medical Facility Table 42 Type of Medical Facility
Medical Facility Public Private Ahmedabad 28% 72% Ankleshwar Godhra 56% 42% Nadiad 50% 50% Jaipur 45% 30% Jodhpur Bikaner 85% 15%

Average Monthly Health Expenditure Table 43


Health Expenditure (Rs.) Ahmedabad Ankleshwar Godhra Nadiad Jaipur Jodhpur Bikaner 220 190 270 228

3.2 Conclusion The report has analysed data from selective slums in the seven cities of Gujarat and Rajasthan. It has attempted to provide a detailed description and analysis of the problems facing the households in these slums. A profile for each city was then presented, where the important data was investigated and explained. Additionally, within these profiles some regression analysis was used to explore links in the data for

75

the cities with more data this helped to provide deeper insights, while for others, it served as a summary of the relationships within the data. Examining the general trends that emerged from the data we can draw some broad conclusions about these cities. Most of the households were poor, overall they lacked access to basic services (it was extreme in certain slums), and the housing conditions for the most part were not very good. Despite living in an area for many years most of the families did not own land, and the majority of those who did own land did not have any legal proof. Saving rates in the slums were very low, and with the exception of Ahmedabad, loan rates too were low. Besides this broad assessment, which was the first objective of the report, the study also came up with a few specific findings that need to be mentioned. The report attempted to analyse some of the underlying relationships in the data. Firstly, by analysing the link between health and infrastructure it was found that household health was correlated with access to basic amenities; those households who did not have access to water, adequate toilet facility, and drainage (sewerage and storm water) were spending more money on monthly healthcare and reported more frequent illnesses than those who had these amenities. Secondly, it was clear that a relationship existed between income and access to infrastructure, that is, the poorer households were also the ones who lacked access to water, toilet facility, drainage, and so on. Thirdly, in Ahmedabad specifically, the data revealed that spending on upgrading of house and acquiring basic amenities was much higher if the house and the land were owned by the family. Spending was also positively correlated with income. Fourthly, the data showed that the choice of type of house generally differed by income; with families earning higher income living in better housing. Finally, analysis also revealed that saving was much higher among the richer households. The findings provide helpful insights into the condition facing the urban slum dwellers. While many of the findings were intuitive and perhaps even expected the value of empirical evidence cannot be underestimated. They provide strong support to the claim that proper interventions would result in many direct and indirect positive gains.

76

Appendix
This generic formula describes the FGT (also called the P class) of poverty measures:

P
where : z is the poverty line

1 n

q i 1

z yi z

for

yi is the standard of living indicator (e.g. PCI) of the ith household is the aversion to poverty parameter. The higher the value of , the more sensitive the measure is to the well-being of the poorest person. As

approaches infinity, the measure reflects only the poverty of the poorest person n is the population

P0 ( =0) is the head count P1 ( =1) is the poverty gap ratio P2 ( =2) is the squared poverty gap

77

References
Amis, P (2001), Rethinking UK Aid in Urban India: Reflections on An Impact Study of Slum improvement projects, The Journal of Environment and Urbanization Vol. 13, No 1, April 2001 Chen, Martha (2000). Women in the Informal Sector: A Global Picture, the Global Movement. Washington DC: World Bank. Chen S, Ravallion M, (2008), The Developing World Is Poorer Than We Thought, But No Less Successful in the Fight against Poverty, Policy Research Working Paper 4703, The World Bank Development Research Group, August 2008. Census of India (2001), Retrieved on 30th June 2009, As viewed in www.sangamindia.org Slums in India (2009) City Development Plan (CDP), (2006), Jaipur Municipality, www.jnnurm.nic.in/nurmudweb/cdp_apprep/CDP/CDP-Jaipur.pdf available at

Devas, N (2004), Urban Governance and Poverty: Lessons from a Study of Ten Cities University of Birmingham School of Public Policy Dutta, S & Batley, R (1999), Ahmedabad, Urban Governance, Partnership and Poverty Working Paper 16, International Development Department, University of Birmingham, Birmingham Hancock T (2000), Urban Ecosystems and Human Health Institute of Development Studies Research. www.idrc.ca/ecohealth Karn, S.K., Shikura, S, Harada, H. (2003), Living Environment and Health of Urban Poor: a Study in Mumbai Economic and Political Weekly: August 2003 Meenakshi J.V, Ray R (2000), Impact of Household Size and Family Composition on Poverty in Rural India, Centre for Development Economics, Delhi School of Economics, www.ecite.utas.edu.au/25467 MHT (2009), Endline Report on the Sahbhagi Yojana 2 Support Programme (SY2SP) Ray, R (2000), Poverty, Household Size and Child Welfare in India, Economic and Political Weekly, Vol. 35, No. 39 (Sep. 23-29, 2000), pp. 3511-3520 Published by: Economic and Political Weekly, http://www.jstor.org/stable/4409778

78

Rakodi C with Loughhead S (2002), Reducing Urban Poverty in India Lessons from Projects Supported by DFID in Urban Livelihoods: A People-Centred Approach to Reducing Poverty, C Rakodi with T Lloyd-Jones (Eds). London: Earthscan, p. 225-36 SEWA Academy (2002), Parivartan and its Impact: A Partnership Programme of Infrastructure Development in Slums of Ahmedabad City, SEWA Academy: 2002 UPRS and SEWA Report (unpublished) Urban Poverty Reduction Strategy, Ahmedabad UNDP, (2009), India: Urban Poverty Report, Factsheet, (http://www.undp.org.in/content/factsheets/PovertyReduction/INDIA-URBANPOVERTY-REPORT-2009.pdf UN-Habitat (2003), Global Report on Human Settlements, The Challenge of Slums, Earthscan, London; Part IV: Summary of City Case Studies, pp 195-228, As viewed in Bhatt, R (2003) Understanding Slums: Case studies for the Global Report UN Habitat UN-HABITAT (2007), Twenty-first session of the governing council, Nairabi, Kenya, April 2007, www.unhabitat.org UN-Habitat, (2008), The UN-HABITAT Water and Sanitation Trust Fund ANNUAL REPORT, 2008, UK Department for International Development Reducing Urban Poverty in India http://www.dfidindia.org/pub/pdfs/urban1.pdf UN-Habitat, State of the Worlds Cities 2004/5, www.unhabitat.org WHO/UNICEF/WSSCC (2000), Global Water Supply and Sanitation Assessment 2000 Report. WHO, Geneva, Switzerland WHO with the Stockholm International Water Institute (2003), Driving Development by Investing in Water and Sanitation WHO/SIWI: 2003. See www.siwi.org www.egovamc.com Younes. M, Bartram. J (2001), Waterborne Health Risks and the WHO perspective, Department of Protection of the Human Environment, World Health Organization, Geneva, Switzerland, International Journal of Hygiene and Environmental Health , 204, 255 263

79

Acknowledgment
I would like to extend my thanks to all those who made this research possible through their advice, organisation, and facilitation: Dr. Surashree Shome, Ms. Renana Jhabvala, and Ms. Bijal Bhatt. I would also like to express my gratitude to the women at the Mahila Housing SEWA Trust (MHT), Ahmedabad who always made me feel welcome. It is mainly because of them I found my work to be an enjoyable and educational experience. I sincerely hope that this report will serve its purpose and be valuable in assisting the work that is being done by MHT. Benjamin Stanwix 12th October 2009

80