Improving access to clean water by low-income urban communities in Jakarta, Indonesia

(1)

1

By Amber Noordegraaf 10971904

29-2-2016 - Amsterdam University of Amsterdam

Graduate School of Social Sciences Master International Development Studies Master Thesis

Supervisor Netherlands: Michelle Kooy (UNESCO -IHE) Supervisor Jakarta: Indrawan Prabaharyaka

Second reader: Michaela Hordijk (UvA)

Improving access to clean

water by low-income urban

communities in Jakarta,

(2)

2

Abstract

The international development Master Meter (MM) project in Jakarta aims to improve the water security of low-income households by providing access to drinking water through a centralised piped network water supply system. The MM project is implemented by USAID-IUWASH (Indonesia Urban Water, Sanitation and Hygiene), the Indonesian government, the PDAM (public water utility), and private sector water operators (Palyja, Aetra) in collaboration with local communities. The project implements a semi-centralised, community operated water supply system with a reduced water tariff in areas with illegal settlements and low-income households, and teaches the community-based organisation (CBO) members how to operate it. The project is premised on the assumption that access to piped water improves the water security of low-income households, and that households will be able to consume more water, at less costs, than before. These assumptions are not tailored to the water supply context of Jakarta, where piped water supply is intermittent, irregular, and supplemented with other water sources by the majority of urban households. This study queries these assumptions by asking what is the impact of access to piped water on the water security of low-income households in Jakarta and how does access to piped water change their household water strategy in terms of what sources are used, volume used, and per unit costs paid. To answer this question 45 household surveys and 4 interviews with water utilities and NGO’s have been done. Based on an analysis of three MM projects in Jakarta over time (2009 and 2015) this study found that access to piped water for low-income household does increase their access to water as it primarily made access to water easier. On average, households consumed more water, at a lower cost, than before the project. Also, the price, quantity and quality of this piped water source are better compared to their non-piped water sources. However, access to piped water does not change households practices of buying refill or bottled water for drinking, which are much more expensive than their piped water. In addition, the sustainability of the project is questionable as MM model presents a temporary rather than a permanent solution. Also, the piped water is being used for more activities than just drinking. The piped water is being used for domestic chores and even for home industries. Therefore, the study concludes that while access to piped water for low-income households in Jakarta increases their level of water security, the outcomes are more ambivalent than predicted. The positive effect is not a permanent one as the sustainability of the project is at stake, and does not reduce the amount of highest cost, highest quality water purchased by poor households.

(3)

3 This study shows that a project set up to help low-income households to improve their water access on the short-term should consider long-term planning and impacts in order to make a permanent positive impact and should asses for what purposes their water is actually being used.

Keywords: Piped water, water security, water access, low-income households, Jakarta,

(4)

4

Acknowledgements

Firstly, I would like to express my gratitude for my supervisor in the Netherlands, Michelle Kooy, for her continuous support, knowledge and patience. Her guidance and insightful comments always helped me finding my way in setting up and implementing the research, and helped me with writing this thesis. Additionally, I would like to thank her for connecting me to relevant persons and organisations in Jakarta, which have proven to be essential to my research.

Also, I would like to express my gratitude for my supervisor in Jakarta, Indrawan Prabaharyaka. I would like to thank him for his continuous support and knowledge during my fieldwork, for arranging a translator, assisting me to important meetings and helping me find my way in the city.

I also want to thank Michaela Hordijk, from the University of Amsterdam, for being my second reader and for taking the time to read and evaluate my thesis.

My sincere thanks also goes to Foort Bustraan, Tofikurochman Achmad and Zulfa Hananiawati from IUWASH and Haryanti Koostanto, Mas Catur and Pak Aidi from SPEAK, who provided me with the opportunity to visit research locations and introduced me to my key respondents. Without their support this research would have not been possible.

Additionally, my sincere thanks goes to Irma Damayanti and her colleagues at Palyja, and Astriena Veracia and her colleagues at Aetra. I appreciate them for making time to help me with my research, and answering my questions during interviews.

I want to thank my translator, Melinda Martinus, who made time to help me with my research and played an essential role in my research. Without her translation I would not have been able to conduct the surveys in the research locations.

Finally, I would like to thank my family. I want to thank my parents for their support, and especially my father who made time to proofread my chapters. Also, I want to thank my brothers who supported me the whole time, and I want to express additional gratitude to my brother Tim who made time to design my cover page.

(5)

5

List of images, figure and tables

Images

- Conceptual Scheme ... 19

- Division of working area water operators in Jakarta ... 28

- Raw water sources Jakarta ... 34

- Positioning of Citarum River versus West Tarum Canal and Jakarta ... 35

- Polluted Citarum River ... 35

- Lay-out division responsibility Master Meter ... 38

- CBO management model ... 46

- Map with indication of the research locations Penjaringang, Jembatan Besi and Cilincing in Jakarta. ... 47

Figures - Satisfaction with quality of piped water ... 59

- Satisfaction with monthly costs MM ... 61

- Satisfaction with quantity piped water MM ... 65

Tables - Comparison of costs water sources in Jakarta ... 29

- Before and after MM ... 39

- Water sources used before the MM connection divided by subdistrict ... 54

- Water sources used since the MM connection divided by subdistrict ... 55

- Using same amount of water since MM connection? ... 57

- Before and after MM ... 58

- Expenditure on MM water per location ... 58

- Which water source is used for drinking? ... 59

- Reasons for HH to use water source for drinking ... 60

- Sources used for drinking per location ... 63

- Costs for a m3 of piped, refill and bottled water per location ... 64

(9)

9

List of abbreviations

ADB – Asian Development Bank CBO – Community-based organisation DAI – Development Alternatives Inc. ESP – Environmental Services Program GOI – Government of Indonesia

HH - Household

HWS – Household Water Security

IPC-UNDP – International Policy Center for Inclusive Growth – United Nations Development Programme

JMP - Joint Monitoring Programme for Water Supply and Sanitation MDGs – Millennium Development Goals

MM – Master Meter

NGO – Non-governmental organisation NRW – Non-revenue water

UN – United Nations

UNDP - United Nations Development Programme

USAID – United States Agency for International Development IUWASH – Indonesia Urban Water, Sanitation and Hygiene

Indonesian words

Aetra - Aetra Air Jakarta (private water supply operator in Eastern Jakarta, Indonesia) BPS – Badan Pusat Statistik – Statistics Indonesia

DKI Jakarta - Daerah Khusus Ibu Kota Jakarta (Special Capital Region of Jakarta) Girik – Land title based on tax receipts

Garapan – Land title based on land transfer agreements between individuals Kampung – Spontaneous informal settlement

(10)

10 KK - Kartu Keluarga (Indonesian Family Card which states the head and members of a household/family which is kept by the head of the family.)

KTP - Kartu Tanda Penduduk (Indonesian Identification Card, issued upon reaching age of 17 or marriage.)

MBR - Masyarakat yang Berpenghasilan Rendah (Low-income communities)

Palyja - PAM Lyonnaise Jaya (private water supply operator in Western Jakarta, Indonesia) PAM Jaya - Perusahaan Daerah Air Minum Jakarta Raya

PDAM - Perusahaan Daerah Air Minum

SPEAK - Strategi Pengkajian Edukasi Alternatif untuk Komunikasi (is a local NGO which works on communication and advocacy in the field of water and sanitation.)

(11)

11

Chapter 1 – Introduction

1.1 Introduction

This chapter is a short introduction to the research on the impact of an international development project on access to clean water by low-income households in Jakarta. First, a brief introduction to the research topic and its context will be given (1.2), which will be elaborated on in following chapters. Additionally, it will briefly introduce the Master Meter project, the project which was used to study the impact of access to (piped) water on water supply strategies of low-income households in Jakarta (1.3). Then, it will also discuss how this research addresses a gap in literature and how this research will provide new insights by discussing the research aim and its relevance (1.4). Finally, a summary of the chapter will be given (1.5).

1.2 Brief background research

Water is a basic need of every human being as it is vital for life. In 2010 the UN declared water (and sanitation) to be a human right and being essential to all other human rights. Nevertheless, the demand for water rises and water distribution falls short. Indonesia is one of the most densely populated countries of the world and within Indonesia a large number of the millions inhabitants are living without safe drinking water and/or (proper) sanitation. These people are often the rural- and urban poor. Poor – or in this case more specifically low income – households deal with the problems of the availability of water on a daily basis. However, the lack of water is not the only problem low-income households have to face with regard to water. Water access and the quality of water can be just as problematic. In the context of Jakarta (officially known as the Special Capital Region of Jakarta or DKI Jakarta) many low-income households live in informal settlements without any land ownership documentation and are not legally entitled to be connected to the formal water supply system. This means they have to get their water from alternative sources via alternative providers. These sources include piped water delivered to communal kiosks; purchasing of piped water from neighbours who have legal connections; shallow groundwater, contaminated with e-coli and saline/brackish in the northern parts of the city; illegal connections to the centralized piped water supply system and bottled or packaged water for drinking. Some of these sources are of very poor water quality, and most of these supply strategies mean that water supply is much more expensive than the per unit charges in the water tariff for piped water supply.

(12)

12 In the past, the Indonesian government committed itself to work on the goal set by the Millennium Development Goals (MDGs) of 2000 on ensuring environmental sustainability; in this case MDG number 7.C on water and sanitation. One of the targets (number 10) was to

‘halve the proportion of the population without sustainable access to safe drinking water and basic sanitation by 2015’ (UN 2013). Indonesia has met this target on water with an increase

of access to drinking water of 39% over its entire population since 1990 (JMP 2015). In total the urban coverage to improved water sources in 2015 was 94%, of which 33% piped water and 62% (rounded up) other improved water sources, and 6% unimproved sources (JMP 2015). These numbers seems to be rather high, but it is important to point out that an improved source is defined as protected from outside contamination by nature or through active intervention and as such this includes protected wells.

But according to Bain et al. (2012), this definition is based on the likelihood of a source providing safe drinking water and so the water quality of water sources can actually differ. Also, most of them have access less than 12 hours a day. So the degree of accessibility is very questionable.

1.3 Master Meter: access to piped water for low-income households

To help achieve the MDGs there have been several initiatives in Indonesia to work on these goals, and also specifically for the goal on improving access to clean water. In Jakarta, there have been several projects to include low-income communities in a water supply system, often not being in the ‘formal system’ but in an alternative one. The Master Meter project founded by USAID-IUWASH (Indonesia Urban Water, Sanitation and Hygiene), the Indonesian government, PDAM (water utility) and the private water operators (Palyja and Aetra) in collaboration with local communities, is one of these initiatives. As will be discussed in detail in Chapter 5, the aim of the Master Meter (MM) program is to improve access to clean water for low-income households living in informal settlements by connecting them to a piped water source. Water operators do not want to connect users to a private connection that have troubles with payment, who do not consume a lot of water and especially are not able to work with users that do not have the legal requirements in relation to land titles (Prabaharyaka & Pooroe 2010). This is specifically a problem for the low-income households in Jakarta, which are deprived of their basic needs by the formal system and need to find alternative water supply strategies. Via the MM low-income households obtain access to a piped water source in or close to their homes for a reduced tariff compared to the normal price of a private connection.

(13)

13 The incentive for the private sector water operators is that the provision of a legal communal connection reduces the volume of water theft from these areas. The local community based organization (CBO) is taught how to manage and maintain this system and gets the responsibility and say over the management within their community.

1.4 Research aim and relevance

The project design is premised on the assumption that access to piped water improves the water security of low-income households. This increase in access to water is identified by the project as the consumption of more water, of higher quality, at lower per unit cost. Presumably, water sources of higher cost, or poorer quality, will be discarded in favour of the use of piped water. However, this global model of pro-poor urban water supply provision (MM), is not specifically adapted for the context of Jakarta. Therefore, it does not take into account the historical and widespread practices of households across all income classes in combining different water sources, of different qualities, according to need (groundwater, bottled/packaged water). Nor does the project adequately account for the lack of reliability (intermittent quantity, poor quality) of piped water supply across the entire networked system. This study addresses the gap, questions the project design assumptions, by asking what is the impact of access to piped water on the water supply strategies of low-income households in Jakarta, and subsequently – how is access to clean water improved or increased via the project. Looking specifically at how access to piped water from the MM changes the types of water sources used, the volumes consumed, the prices paid. Second – examining access to piped water across four dimensions of water access (ease, safety, affordability, continuity) to evaluate if water access has improved/increased.

The exploration of this topic will be in the interest of all stakeholders involved and could lead to new insights in (pro-poor) water management options, research and water supply related policies and projects as it looks at the specific context of Jakarta water supply strategies practised by low-income households.

1.5 Summary

This chapter provided a brief introduction to the research background (1.2) and the background of the Master Meter (1.3).

(14)

14 The MM was set up as a program to help achieve the MDG goal to increase access to safe drinking water and is a cooperation between GOI, USAID-IUWASH, PDAM, private water supply operators (Palyja, Aetra) and local communities. The research aim and the relevance of the research is discussed as well (1.4) and shows that the aim is to research the impact of the Master Meter on water supply strategies of low-income households, and evaluate whether or not piped water has improved water access.

(15)

15

Chapter 2 – Theoretical Framework

This chapter discusses the main concept on which this research is based: water access. Water access is not a singularly defined term as there are multiple different views on what water access actually entails, and usually it is used as a component of water security and not as a stand-alone concept. Section 2.2 discusses the multidimensionality of water access and how it will be assessed with regard to this research. To conclude, in section 2.3 a summary is given.

2.2 Water Access

2.2.1 Definitions of water access and it shortcomings

Access is often mentioned as a part of water security, but by putting access as a variable under water security it loses a lot of important dimensions of which water access can be composed of. When looking at water access as a separate concept, it makes looking at a water situation of a household more complete and context-based. There are multiple dimensions to accessibility and to show the importance of the multidimensionality of water access these different dimensions will be discussed.

Nganyanyuka et al. (2014) argue that statistics on water access (coverage rates on access to improved water sources) are only partially relevant as they often leave out important problems of access (such as quantity, quality and reliability) and ignore the informal private water selling activities that make up a big part of access strategies. Up to now most definitions on water access are aimed at looking at global comparison (comparison between countries for MDGs i.e.), this focus makes that small/local context could be lost in the bigger picture. They argue that water access does not work with a one-size-fits-all definition, as this will not apply in diverse context and conditions. There should be more attention to the improvements necessary to increase water access, as perceived by the citizen/users, rather than looking at coverage rates.

Smiley (2013) says that water access is generally measured by the distance to a source of clean water. She argues that this does not take other important factors into account such as location of a source, reliability, costs, quantity and equity. Sorenson et al. (2011) agree that not just access - as defined as linear distance - should be measured and increased, but also safe access and reliable access to water. They also argue that there should be more attention to gender with regard to water access, as women usually carry the burdens of water fetching with regard to time and energy and related health implications.

(16)

16 Obeng-Odoom (2012) additionally argues that in development literature it is important to look beyond coverage of the piped water network as the definition or indicator of water access, as water access in development theory can be different from development in practice. He argues water access should look at more than the ‘wide’ meaning, it should have a ‘deep’ understanding which includes looking at quality, reliability, cost of water, and equitable distribution of piped water (improved source). He also argues the MDG 7C was too focused on access and should have included these other dimensions as well.

Overall, a good approach to water access is given by IPC-UNDP (2009) which describes water access in slums to be related to the appreciation of ‘the multidimensional nature of the

problem, including income poverty, infrastructural limitations, asset ownership and housing quality’ (2009, p 1). Access to water should be ‘safe, affordable, continuous and easy’

(IPC-UNDP 2009, p 1). For this research water access is also approached by looking at the dimensions of safety, affordability, continuity and ease as mentioned by IPC-UNDP (2009) and with regard for social issues. Mason explains that there is more and more recognition for the importance of social programs and policies to increase water access, which address problems as social inequalities and injustices.

2.2.2 Assessing water access

Hadley & Wutich (2009) discuss some aspects which needs to be assessed to examine water access - which have been mentioned in other literature - : distance to water sources, seasonal variation, time- and economical expenditures and storage capacity (2009, p. 452). But there are some more elaborate ways to examine water access as well. Komnenic et al. (2009) discuss the Water Poverty Index (WPI) that should initially be able to explain society’s inability to provide adequate access to water. Just like water access and water security, water poverty has several different perspectives. Most of them relate water poverty to the availability of water for economic development, communities and countries. Therefore the WPI has its limitations. Generally, the WPI is defined with 5 dimension: resources, access, capacity, use and environment. With the WPI a country can be water poor with high levels of access to water, which seems contradictory. That is why Komnenic et al. recommend to focus on the indicators and not on the composite indices. This approach however, does not take social causes of unavailability into account.

Yang et al. (2012) focus on the aspect of safe water access by discussing the implications of using the categorization of water sources into improved and unimproved.

(17)

17 Improved sources are sources that: ‘(..) distinguishes between sources that, by nature of their

construction or through active intervention, are protected from outside contamination (including piped water, tubewells or boreholes, protected dug wells and springs, and rainwater) and those that are unprotected (such as rivers, streams, and unprotected wells).’

(Yang et al. 2012, p. 1222). They argue that through this binary approach to water sources, individual inequalities may be masked. Bain et al. (2012) add that the improved/unimproved approach does not assess water quality, as the decision on whether a source is improved or not is based on the likelihood of the source providing safe drinking water. They argue that taking water quality into consideration has the consequence that the percentage of people with access to safe drinking water is substantially reduced in their countries of study.

This research will consider the multiple dimensions of access – safety, affordability, ease and continuity – to evaluate how water access has improved – or not – as a result of the Master Meter project.

2.3 Summary

In section 2.2 definitions and limitations of the concept of water access are examined, which often start with the distance to a water source, and continue to the social, environmental, and other dimensions of access. It discusses how water access is often seen as a part of water security, but when we look at it as its own concept it makes it possible to look at multiple dimensions relevant to water access. The approach to water access is shared after discussing the ways to assess water access. The dimensions safety, affordability, continuity and ease will be used to assess the impact of the Master Meter project on the access to water by low-income households in Jakarta.

(18)

18

Chapter 3 – Methodology

This chapter discusses the research methodology and data collection approach used to implement the research. In order to be able to assess the water access by low-income households in Jakarta appropriate sub questions are needed, which collectively help answer the main research question. First, these research questions are described (3.2). The conceptual scheme (3.3) helps to put things into perspective and shows the approach taken to the major research concept of water access. Following, the units of analysis are discussed and the sampling process (3.4) is explained. Then the methods (3.5) used to collect data during the fieldwork will be discussed. Afterwards, the limitations and ethics of the research (3.6) will be elaborated on. To conclude a summary is given (3.7)

3.2 Research questions

The main research questions is formulated as follows:

‘What is the impact of access to piped water from the Master Meter project on the water supply strategies of low-income households in Jakarta, when considering access as ease, safety, affordability, and continuity?’

Sub questions:

a. What is the goal of the MM project in relation to water access for low-income households?

b. What are the current water supply strategies – and type of access to clean water - of low-income households in Jakarta?

c. How does the Master Meter project change water supply strategies of low-income households in terms of sources used, volumes uses and prices paid?

d. How does the Master Meter project improve – or not – access to clean water by low-income households in the project areas?

Answering these questions is done based on my collected data in the field and personal observations. My primary data collection in low-income communities answered the sub questions B, C, and D. My interviews covered the remaining sub question. Altogether, my sub-questions supported the main research question.

(19)

19

3.3 Conceptual scheme

Image 1 - Conceptual Scheme

3.4 Units of analysis and sampling

3.4.1 Households

The main method of data collection was through surveys with households, which were conducted in three research locations. Two older MM locations and one new location were used. The selection of old versus new MM locations made it possible to compare current experiences with the MM in the older locations versus the expectations for the future in the new location. In total 45 surveys have been conducted in the locations altogether.

It was divided over the following location: 12 surveys in Jembatan Besi, 13 in Penjaringan and 20 in Cilincing (including 3 pilot versions in both Jembatan Besi and Penjaringan). The 12 in Jembatan Besi were from a total of 25 MM households, which accounts for almost 50% of the MM population in that area. The 13 in Penjaringan were from a total of 58 MM households, which accounts for 22% of the MM population in that area.

(20)

20 The 20 in Cilincing were from a total of 25 MM households, which was 80 % of the MM population. It has to be noted that, only 25 households were connected at that moment, but shortly after the surveys more household would be connected. This would end up to be around 100 to 150 households.

The amount of surveys was chosen in relation to the time limitations of the research and the availability of the households. Initially 40 surveys would be done over 4 locations, but due to the unavailability of a fourth location, 45 surveys were done over three locations (pilot included). More details on the research locations of the surveys is elaborated upon in Chapter 5.

3.4.2 Employees SPEAK, IUWASH and private water operators

Besides the households, three employees of each of the two water operators in Jakarta have been interviewed and two employees of each of the two most actively involved organisations (USAID-IUWASH and NGO SPEAK) have been interviewed as well.

Employees of SPEAK were relevant to interview because they work closely with the MM users as they facilitate trainings for the CBO for the implementation of the MM in their communities. These employees all know the CBO’s very well and often know a lot of the users too and that is why their perspectives on low-income households and water access are important. The two employees interviewed were selected because they had accommodated me to all of the research locations, they work closely with MM users, they were up to date with my research aim and were available for interviewing.

Employees of IUWASH were relevant to interview as they were one of the initiators of the MM program and work as stimulators/mediators between civil society, the government and the water operators. Being one of the initiators of the program it is important to know what their perspectives are on low-income household and the goal of the Master Meter. The two employees of IUWASH interviewed work specifically with the MM program, were up to date with my research aim and were available for interviewing.

Employees of Aetra and Palyja, specifically working with the MM, were relevant to interview as they are the private water suppliers (via Pam Jaya) in West- and East Jakarta. IUWASH arranged for me to speak with three employees of Aetra that work with the MM and three employees of Palyja that work with the MM, and were available for interviews.

(21)

21

3.5 Methods

3.5.1 Mixed methods approach

This research used a mixed methods design as the research used both qualitative and quantitative methods to obtain data. This approach was chosen in order to be able to look at the topic on water access by low-income households in Jakarta and to get a comprehensive and in depth understanding of the situation. The mixed methods approach also makes triangulation possible as quantitative and qualitative data can be compared for corroboration (Bryman, 2006). Additionally, quantitative methods can obtain a lot of usable and interesting data, but adding qualitative methods can also help explain, or at least allow the researcher to elaborate on, the findings of the quantitative data collected. The other way round can quantitative data show to what extent qualitative data is a trend and where interesting correlations occur. Overall semi-structured interviews have been used as qualitative method and the household surveys mainly collected quantitative data, but some qualitative data as well.

3.5.2 Households surveys

Household surveys were conducted as questionnaires for an individual of a household or for multiple individuals of a household to provide samples on these households. The surveys were the primary source of data collection and collected qualitative, but mainly quantitative data. The survey has a set of predetermined questions which starts with questions about the composition and details of the household, followed by adding questions on their water sources, their water consumption and ranking questions to make the respondents show how important different water related topics are to them and how satisfied they are with MM. Additionally some questions focused on the differences of water sources/consumption etc. before and after the MM. The household survey was chosen as method because it is a good way to collect data on the same topic for a larger amount of people, so as to make an overview for each community. Essentially, it helps to collect data on household characteristics within each research location. The quantitative data from the survey can be used to put the qualitative data of the other method into perspective. In an example it can show to what extent certain themes from the qualitative data are actually visible within and representative for a community.

(22)

22 The respondents were visited in their homes within their community. Households living in the research locations were asked to (verbally) fill in a household survey under the guidance of the researcher and with help of the translator.

Most of the respondents did not speak English so the researcher conducted the survey in English and the translator translated from English to Bahasa Indonesia and back. The household survey has been conducted in three different versions. The first version was the pilot version, this one has been used for 6 households. Afterwards alterations were made to make the survey more complete for research purposes and more understandable for the respondents. The new version also included a visual aid for ranking questions, with the aim to make the questions more accessible for the respondents. Then 19 surveys were done with this second version in Penjaringan and Jembatan Besi. After a short introductory meeting with the CBO of Cilincing another version of the survey was made for this specific location. This version was made because the questions on average piped water consumption and satisfaction with the MM were – in most cases - a bit too premature, due to the fact the water had been running for just two weeks and the first bill was yet to come. A total of 20 households in this area got this third version in order to still get interesting and usable data for the research.

3.5.3 Semi-structured interviews

Semi-structured interviews were primarily chosen as a method to obtain qualitative data on the goal of the Master Meter. As IUWASH and SPEAK work in the field of water provisioning and work on the target of ‘water for all’ an interview is a good way to ask the respondents what the Master Meter does according to them.

The same goes for the concept of low-income households, for which the respondents of Aetra and Palyja were also asked to share their view on this. By using semi-structured interviews respondents can freely discuss their view on the topics discussed. Using semi-structured interviews makes it possible to ask follow-up questions and look for explanations.

The semi-structured interviews had a number of predetermined questions which were asked in a certain sequence in order to get the necessary data, but also left some room for relevant follow-up questions which were thought of during the interview itself.

(23)

23 Employees working at SPEAK (an Indonesian NGO), and IUWASH and the water operators Palyja and Aetra have been interviewed in order to get an idea of the context of the current and former situation of water access at the research locations. These organisations have been working in these research locations for quite a while and are up to date with the current situations, its residents, its challenges and opportunities. During all the interviews the respondents were asked about their work and day-to-day activities, their connection to MM, the procedure for selecting communities for MM and asked to give a definition of low-income households. Water operators got additional questions about non-revenue water (NRW) and service coverage rates and SPEAK and IUWASH got additional questions about the mission of their organisation, the mission of MM and what the ideal water situation of a household looks like.

For the coding and analysis of the interviews thematic/axial coding was mainly used, because most of the themes (codes) were known in advance due to the specific research topics and research questions. Some sub-questions were set out to be answered primarily through the interview data in order to collect (different) views on the research concept and linking this to existing theory on these concepts. During the process of coding not only thematic coding has been done but open coding as well to make sure that everything that may be important is taken into account and can be taken into consideration where appropriate.

3.6 Limitations and ethics

3.6.1 Limitations

In every research there are always limitations, besides the limitation of time due to fieldwork of 2,5 months there were some other limitations that are essential to elaborate on. One of the most important limitations for this research was the language barrier. The low-income households overall, some exceptions, did not speak any English but only Bahasa Indonesia. This made it necessary to arrange a translator as I do not speak the local language. Via a local university (Tarumanagara University) a student of urban and regional planning was asked to fill in this function. Although her English proficiency was good, using a translator always creates several layers of interpretation. Also, as researcher you are depended on your translator’s time as you cannot conduct the household surveys without him/her.

(24)

24 Another important limitation was the fact that I arrived on the very day the Ramadan started. As Indonesia is the biggest Muslim country in the world this meant the daily life for all Indonesian citizens changed greatly as they are not allowed to eat or drink from sunrise until sundown. Ramadan lasts about a month so for the first month I was advised by IUWASH to conduct surveys after sunset or in the weekends. As the traffic in Jakarta is really bad, especially in the evening (people going home after work) initial meetings with CBO’s were done in first few weekend (Penjaringan and Jembatan Besi). After the Ramadan ended, there were two weeks of Idul Fitri holidays to celebrate the ending of Ramadan and most Indonesians travel back to their families (outside of Jakarta). This meant a lot of households and all companies that could be available for surveys or interviews in Jakarta were off for vacation.

Moreover, while conducting the household surveys occasionally I needed some help to find new respondents in the community that were connected to the MM. This mainly occurred in Jembatan Besi and Penjaringan. At such moments the help of CBO-member in the research location was asked in order to find a new household to conduct the survey with. Often this would lead to a situation where the CBO-member would then stick around and be present during the survey. It seemed ungrateful and inappropriate to send them away/ask them to leave. As a consequence this meant that, as the CBO is the managing body of the MM in the community, their presence could lead to influencing the answers given by the respondents. There were some questions about the management of the MM by the CBO as well, so it is not clear to what extent their answers were affected by this.

During the pilot surveys there always was a SPEAK-employee present and the presence of a IUWASH-member during a survey occurred once or twice. The presence of these people could have affected the answers given by the respondents.

Another important limitation in relation to the surveys was that most of the new MM locations were not running yet due to delays. This excluded some potential locations which were initially selected to do surveys in. This also made it impossible to research a total two new locations versus the two older ones, as only one new location was operational. That is why in Cilincing 20 survey were done instead of 10, with the aim to get an even number of surveys in old and new MM locations.

(25)

25 Unfortunately, from the total of 45 surveys not all of the data collected could be used in the analysis. This is due to the unreliable nature of some of the data obtained. One survey is excluded completely as the survey was answered by someone who was not actually a member of the household, but a close friend/neighbour. Additionally, the data on m3 usage of piped water of two households is excluded from analysis. One of the households mentioned their meter actually measured the water consumption of two households and not just theirs and so gave false information of their personal m3 usage. The other household explained their m3 usage of that month was much lower than usual, because they were away for most of the month. Because the data on their m3 piped water usage was not representative, this is not used in the analysis either. The data deriving from the other survey questions of these two households can still be used. Ultimately this means a total of 44 surveys are used for analysis, except for the m3 consumption of piped water where only 42 can be used.

Additionally, as a researcher I have influence on the respondents and the research as well. Being a young western white woman coming to their community via IUWASH could have led to the respondents to develop some level of expectations.

Although the respondents were informed about the purpose of the research, it cannot be excluded that they would expect that, because they were asked about their experience and satisfaction with MM, their answers could make (positive or negative) changes to their MM possible. Although the most important outcomes were shared with IUWASH, this does not mean that they would do anything with this data for better or for worse.

3.6.2 Ethics

Ethics are always an important part of any kind of research. Research should be conducted in a manner that respects the respondents and does no harm in any way. In order to do so, several ethical considerations have been made in advance and were implemented in the field. First of all, before a household survey was conducted the respondents were provided with sufficient information on the research purpose, that the participation is voluntary, that they can refuse to answer any question and that they can stop with the interview at any time. Secondly, it was explained that any answers they give will be treated with respect and will be kept confidential.

(26)

26

3.7 Summary

This chapter discussed the methodology of the research. Section 3.2 shared the research questions and explained how some of the sub questions were altered due to new research opportunities, which presented itself during the research. The main research question ‘‘What is the impact of access to piped water from the Master Meter project on the water supply strategies of low-income households in Jakarta, when considering access as ease, safety, affordability, and continuity?’ shows the main focus of this research. Section 3.3 displayed the approach to the main concept of water access with a conceptual scheme. Section 3.4 elaborated on the selection of the units of analysis for the surveys and the interviews which were households for the surveys, and employees of SPEAK, IUWASH and the private water operators Aetra and Palyja for the interviews. It explains the sampling of 45 households within the three chosen research locations Penjaringan, Jembatan Besi and Cilincing and additionally discussed the sampling of the two SPEAK, two IUWASH, three Aetra and three Palyja respondents. Section 3.5 explained the methods used were household surveys and semis-structured interviews and elaborates on why a mixed method approach was chosen. Section 3.6 discussed the limitations and ethics of the research and for example explained which steps were taken in order to conduct the research in a manner that respects and protects the respondents.

(27)

27

Chapter 4 – Context Jakarta

This chapter discusses the context of the research as it took place in Jakarta. In order to understand the impact of access to piped water via the Master Meter project (which will be elaborated on in chapter 5), it is important to understand the context of Jakarta which influences and shapes this project. First, the water supply system of Jakarta is discussed, as this partly explains the presence of the Master Meter in the first place, and the limitations it is faced with in implementation (4.2). Then, the existence of informal settlements and its relation to the Master Meter is explained (4.3). Also, the seasonality of Jakarta’s/Indonesia’s climate is discussed and the specific weather conditions during the research, as this could have influenced the answers given in the household surveys (4.4). To conclude the chapter provides a short summary (4.5).

4.2 The water supply system of Jakarta

4.2.1 PSP and the two-division in water provisioning

The current urban water supply system of Jakarta is based on Private Sector Participation (PSP). The involvement of the private sector in the 1990s was a reaction to the low coverage rates realised by the government, and how water was increasingly being viewed as a commodity to sell instead of a service that needed to be provided (Robbins 2003). At that time governments were gradually seen as unable, and therefore unfit, to supply ‘for all’ and private investment should be able to make sure that the urban poor would be supplied as well. PSP would be able to increase the amount of water connections to the poor by improving performance and increasing access (Bakker 2006). The PSP concession contract ‘(…)

promised to improve water quality, mobilize international finance for network expansion and thereby improve and increase access to water supply for Jakarta residents – particularly the poor.’ (Bakker, 2006, p. 115). With PSP in mind Jakarta set a new target for their national

service coverage rate: ‘(…) from 41% in 1996 to 100% in 2022 (…).’ (Lanti, 2006, p. 255). In 1997 the PSP concession contract was signed by a French- and a British water operator and has officially been effective since 1998 (Lanti 2006). The PSP contract was the beginning of a two-division in ownership of Jakarta’s water supply system, which is still effective today.

(28)

28

Image 2 - Division of working area water operators in Jakarta (source: Dr. Riant Nugroho via slideshare.net)

The international water operators British Thames Water and the French Ondeo (Suez-Lyonnaise des Eaux) each signed a 25-year contract with PAM Jaya (government-owned provincial water supplier, in Jakarta), through which the ownership of water assets was transferred to them (Bakker 2006). The international water operators subsequently appointed local water operators to manage the system for them. The British operator contracted Aetra for East Jakarta and the French operator contracted Palyja for West Jakarta. The division of service areas between Aetra and Palyja is marked by the Ciliwung River which runs through Jakarta from North to South. Within these two areas each operator has three service zones. See Image 2 for a map of the current division of service areas, which has not really changed since the initial division.

This division between the two water operators, that have to work together with the government-related water supplier PAM Jaya, makes the water provisioning more complicated. Jakarta is split in two parts, working completely separate from each other. The Master Meter requires good collaboration in order to succeed as International Development project for providing in access to clean water, these separate working areas make this cooperation difficult.

Also, PSP is mainly a for profit model and the Master Meter is not founded to be a project to make profit, but to help the poor. This focus on profit is visible in some of the requirements of the MM as communities have to do a down payment of two months (of assumed monthly average of MM costs) and the tariff that is being used (higher than the tariff for low-income households with a private connection).

(29)

29 Overall, these private operators are more concerned with their own interests such as decreasing NRW (non-revenue water) and increasing (water) sales than with the social mandate.

4.2.2 Current state of the water supply system and its water sources

Nowadays, Jakarta’s formal water supply system is still far from inclusive and the coverage rate of access to drinking water resources are low. ‘(…) in 2002, there were 63% of Jakarta

residents who had access to reliable drinking water sources but that figure fell to 23% in 2012 due the increasing consumption of bottled water and refill water.’ (Prabaharyaka 2014,

p. 61). This occurred because bottled water and refill water were not seen as reliable drinking water sources and therefore the numbers dropped. The use of bottled and refill water is a way of alternative water supply for low-income households. As will be elaborated upon in 4.3, slum dwellers are generally ignored in the water supply system in Jakarta, which has the consequence that service coverage for people living here is low or non-existent.

This problem is really visible in Jakarta, and the rest of Indonesia, through the enormous presence of water vendors (McIntosh 2014). This lack of coverage means that Jakarta’s low-income households have to rely on alternative water supply strategies such as bottled water, refill water, well water, hydrant water and groundwater. These alternative water supply strategies are more expensive in comparison with the costs for piped water, as can be seen in Table 1, which adds to keeping the poor residents poor.

(30)

30 Also, these water sources are often of bad quality, which causes sickness among the consumers. Mainly groundwater, collected from wells, is of bad quality. Overall, the water quality in Jakarta is not good and it is ranked as one of the lower scoring countries on a worldwide scale. ‘This simply means pollution, disposal systems and sanitation are at

appalling rates’ (Wibowo & Mohamed 2010, p. 506).

For a lot of areas in Jakarta groundwater is one of the most important water sources, around 70% of Jakarta’s residents relies on it as main water source (Colbran 2009). The quality of groundwater is generally quite poor (Ali 2010). Just like the river water it is very contaminated, especially by human activities. Open defecation and the absence of proper septic tanks is a big factor. Around 55% of Jakarta’s households possess a underground water reservoir which has a distance to a waste disposal of less than 10 m (Tejalaksana 2012). Shallow groundwater is therefore often contaminated by domestic waste (Ali 2010) and deep groundwater in the Northern parts of Jakarta experiences saltwater intrusion as well (UN-Water 2015). Nevertheless, most households still use shallow groundwater as it is free (excluding electricity costs for a pump), the quality is often perceived as acceptable and they have been relying on this water source for centuries (Colbran 2009). This makes groundwater is easily accessible and affordable, but it is not safe and not really continuous throughout the year.

With regard to the other water sources used; hydrant water is a treated piped water source through the local water utilities and is carried to consumers by means of push carts or carried by the consumers themselves. Because this water is treated the quality is fairly good. Hydrant water seems to be safe and affordable, but not really easy as it needs to be transported by the households themselves (to keep it affordable) and this uses a lot of time and energy. Also, as it is being provided by the water operators it is depended on the water supply which varies throughout the year and per season.

Compared to groundwater, bottled water is water of good quality as it stems from protected springs. This type of water is actually branded and quite expensive compared to other sources. Indonesia is one of the largest consumers of bottled water in Asia (Frost & Sullivan 2010). Compared to refill water, bottled water is over four times more expensive (Sima et al. 2012). At urban scale 60% of the market share is accounted for by bottled water in Jakarta (Darmawan 2013 in Prabaharyaka 2014). Bottled water is easy, safe and continuously accessible, but is not affordable for low-income households.

(31)

31 Refill water is one of the cheaper water sources for low-income households and is very popular in Indonesia.

Refill water kiosks ‘make use of either ultrafiltration followed by UV disinfection or reverse

osmosis treatment to purify and disinfect water.’ (Sima et al. 2012, p. 979 ) It is sold to

consumers which do not have access to piped water sources, or unsafe piped water, and cannot afford bottled water. At the urban scale 40% of the market share is accounted for by refill water (Darmawan 2013 in Prabaharyaka 2014). ‘The percentage of Jakarta households

who consumed bottled water and refill water increased steadily and in 2012 65.5% of Jakarta households sourced their drinking water from bottle water and refill water, which implied that poor households were also part of consumer use.’ (Prabaharyaka 2014). But there are

also many households that do have access to piped water, that also buy refill water as additional water source. Refill water is a relatively easy, affordable, safe and a continuous accessible water source and seems to be the best alternative for low-income households, which explains the popularity of the source.

The number of service coverage by piped water connections has increased to 61% in 2012 from 42% in 1997 (Prabaharyaka 2014). Important to note about piped water is that the water of the piped water system is of varying quality and none of it is potable (Bakker 2006). This means users have to boil it first before they can use it. This mainly because the water that is supplied via PAM Jaya (through Aetra and Palyja) is likely to be contaminated somewhere between the treatment plant and residence of the a household. This is an example due to rusted pipes or other bad maintenance issues. Also, piped water is not always a continuous water source. Households have problems with the accessibility of water, i.e.: only water flowing early in the morning for 2 hours a or no water flow at all). Even wealthy households have these problems with their piped water connections (Colbran 2009). Therefore, a lot of households are quite hesitant about piped water, as it is costly, unreliable and does not provide certain clear benefits that for example groundwater does provide (Colbran 2009).

Overall, due to the low service coverage rates for the low-income households, and the bad/questionable quality of several water sources, the majority of Jakarta’s residents rely on a mix of water sources and types of service providers (Kooy & Bakker 2008). The ease, safety, affordability and continuity of these mixed water sources vary a lot, but they can complement each other to generally have relatively easy, safe, affordable and continuous access to water.

(32)

32

4.3 Informal settlements Jakarta

The target group of the Master Meter’s program are low-income households living in informal settlements that cannot legally apply for a private water connection, as they do not have land titles or identity cards (KTP) that show that they have permission to live and work in Jakarta (ID-card) and permission to live on the land they use (land-certificate) (Prabaharyaka & Pooroe 2010). The reason these people live in communities which are classified as informal settlements, is due to the failure of the formal framework to provide them with affordable housing/social housing (Zhu & Simarmata 2015). Informal self-help housing is then the only option for the urban poor to satisfy their basic need of having a roof over their heads. Informal settlements thus emerge as a way to deal with the lack of housing: if the formal system does not provide or is incapable to provide such housing, the informal system will emerge bottom-up, provide solutions and will become a way of urban life (Zhu & Simarmata 2015).

Winayanti & Lang 2004 mention that these spontaneous informal settlements, which are inhabited by the urban poor, are referred to as ‘kampungs’. Most kampungs are located in disputed land, state land or private unoccupied land. A City Bylaw ‘defines illegal locations to be human settlements along railways, along right of ways, along riverbanks, under bridges, and along green paths and parks.’ Bylaw (City Bylaw Peraturan Daerah no. 11/1988 as mentioned by Winayanti & Lang 2004, p. 42).

One of the main drivers behind informal settlements is rapid urbanization, which is often driven by rural poverty. Jakarta has a known drastic number of increase in population, due to population growth and migration ever since early urbanization. This process of rapid urbanization has resulted in the creation of slums and squatter settlements. However, informal settlements are not a sustainable solution, as urbanization continues to occur in combination with acute land scarcity (Zhu & Simarmata 2015).

With regard to land titles Indonesia has two systems: the Western law as implemented by the Dutch in colonial times and the indigenous ‘adat’ law (Zhu & Simarmata 2015). Adat law is regarded as informal as it is means traditional and is customary. Then there are two types of ‘official’ land titles: 1) ‘Girik’ which can be proven by showing land tax receipts which suggests the land holder pays tax over its land and therefore is land owner, and; 2) ‘Garapan’ which is based on land transfer agreements between individuals (Zhu & Simarmata 2015).

(33)

33 Girik is seen as the strongest claim of the two for land entitlement, although girik and garapan titles are viewed as quasi-legal titles, because they are not registered at the National Land Board (Winayanti & Lang 2004). Overall, residents must have a certificate for the land they live on which indicates that the State sanctioned the right for them to live there and use the land (Colbran 2009).

Living in informal settlements is always related to informal economies, as most of the inhabitants of these settlements work as street vendors, maids or becak drivers (Zhu & Simarmata 2015). These groups of people living in informal settlements are often ignored in the city’s water supply system and plans (Prabaharyaka & Pooroe 2010).

The people living in informal settlements, often occupying land illegally, are deprived of access to basic infrastructure and services. However, many of these settlements have alternative ways to gain access to services via the informal system and via the market, as the market plays an important role in filling this gap. Also, NGO’s in Indonesia play an important role in the mobilisation of resources for the urban poor (Winayanti & Lang 2004). The Master Meter is an example of a project that helps to improve access to resources, although more parties than just NGO’s are working together (GOI, ADB, USAID, IUWASH, SPEAK and private water operators).

4.4 Water sources & seasonal variation

4.4.1 Raw water sources of piped water in Jakarta

Even though Jakarta has a lot of rivers, of which the Ciliwung River is the biggest one, most of the water is too contaminated to use due to domestic and industrial waste. Jakarta greatly relies on the water supplied from the Jatiluhur Reservoir/Dam (via the West Tarum Canal) in West Java (62,5%). The second biggest source is from treated water from Tangerang (31,8%) and some water comes from the rivers in Jakarta (5,7%). Image 3 shows where the raw water sources of Jakarta are positioned in West Java. The red dots/squares are water treatment plants.

(34)

34 The water deriving from the Jatiluhur Dam, Tangerang and some river water is mainly used by Aetra and Palyja for their customers with piped water connections. According to Colbran (2009) most of the water flowing through the West Tarum Canal also comes from the Citarum River Basin, which is viewed as Indonesia’s - if not the world’s - most polluted river (see Image 4 and 5). As a result, this raw water needs extensive treatment and this results in piped water with a strong smell of chlorine and other chemicals (Colbran 2009).

(35)

35

Image 4- Positioning of Citarum River versus West Tarum Canal and Jakarta (source: The New York Times 2008)

Image 5 - Polluted Citarum River (Source: austroindonesianartsprogram.org)

4.4.2 Seasonal variation and water sources

Indonesia, and therefore Jakarta, is known to have two seasons: the dry season (usually from June to September) and the rainy season (usually from October to May) in combination with an average of 28 to 30 degrees Celsius during the entire year. During the rainy season flooding is a major problem, especially in the North of Jakarta. Land subsidence is a big problem, and excessive use of groundwater is an influential factor here (Jakarta Post 2015).

‘Overexploitation of groundwater has resulted in some critical problems, including contamination by pollutants entering groundwater, salinization of aquifers and land subsidence.’ (FAO 2012, p. 291). Also, land-use change made it so that rain water flows

directly into the river (without being absorbed by the soil) and then into the ocean. As the rivers cannot contain all the water flooding is the result (Tejalaksana 2012). During the rainy season flooding can contaminate (ground) water sources and during the dry seasons water sources rapidly decrease in supply due to drought. Clearly, the seasonality of the weather affects the water sources of Jakarta depending on the availability of sources per section of Jakarta (see 5.7 for more details on water sources in research locations).

4.4.3 Summer of 2015

In the summer of 2015 - the research period - the spokesperson for the Meteorology, Climatology, and Geophysics Agency (BMKG) on climate said that the dry season would most likely last longer that year (Dewi 2015). Residents were told to save water as the El Niño (semi-annual monsoon) had made the dry season even drier.

(36)

36 The BMKG asked Jakarta’s residents to use water wisely. Several reports on water shortage in Jakarta occurred during the months of June, July and August 2015. Examples from The Jakarta Post are ‘Residents told to save water’ and ‘Tangerang residents get water after days of shortage’. In the case of Tangerang mostly the Northern Part of Jakarta experienced water shortages as the Tangerang supply decreased, and water was not flowing in the pipes for local residents (in Tangerang) for days (Dewi 2015). They had to decrease the supply to all connections. Overall, the summer of 2015 was a quite dry period and this definitely influenced the water availability in the city.

4.5 Summary

This chapter elaborated on the current water supply system of Jakarta that is based on water provisioning by two private water operators Palyja (In Western Jakarta) and Aetra (Eastern Jakarta) since 1997. This division makes collaboration more difficult and good collaboration is needed to succeed with the MM, also PSP is mainly a for profit model and not to help the poor. Although, the piped water coverage of Jakarta was only 61% in 2012. Piped water is not available for everyone as people living in informal settlements without ID-card and/or without land certificates cannot apply for an individual piped water connection. Informal settlements pop-up where the formal system does not provide enough (social) housing for the low-income households. Households living in informal settlements are not legally able to get an individual water connection and therefore have rely on other water sources, and alternative supply systems. In Jakarta groundwater has been one of the main water sources for its residents for centuries, but the supply is decreasing fast and the quality is not very good. The majority of shallow groundwater is contaminated by domestic and industrial waste. Additional water sources are public hydrant water (piped water supply), bottled water (branded) and refill water (very popular in Indonesia). Most household use a combination of several sources as the prices and quality of these sources vary.

Jakarta is known to have two seasons: the dry and rainy season. The seasons affect water resources as the rainy seasons often leads to flooding and contamination of water. On the other hand, the dry season is often characterised by water shortages. During the research period (the summer of 2015) there was period of drought as the El Niño was particularly strong that year. This influenced the water supply in multiple parts of Indonesia, and also specifically the water sources of Jakarta. Residents were asked to save water as much as possible.