Abstract

Bottled water serves an increasingly large percentage of urban poor populations in lower-income countries, yet receives little attention within international development research and policy. This study investigates the impact of packaged drinking water (refill water) on affordability and equity of drinking water access by the urban poor under SDG 6.1, comparing refill water cost and consumption across socioeconomic quintiles drawn from two sub-districts of Jakarta, Indonesia. Analysis of a customer survey (n = 80) and in-depth interviews with 12 small-scale refill water providers reveals the significance of water quality, convenience, and reliability of water in defining affordable water access. Lower-income households perceive refill water to be the most affordable, safe drinking water source available to them, despite representing the second highest per unit cost source. Piped water is considered more expensive despite its low per unit volume cost, because of total costs associated with guaranteeing its reliability and quality. We suggest that the combined costs of securing domestic and drinking water for poorer households need to be considered for future approaches targeting the provision of inclusive water access under SDG 6.1 in Indonesia. Packaged water needs to be taken into account in the strategies designed to increase access, and measurements of affordability and equity of access.

INTRODUCTION

On September 27, 2015, the 2030 Agenda for Sustainable Development was adopted, initiating the era of the Sustainable Development Goals (SDGs). The SDG drinking water goal 6.1 aims to ‘achieve universal and equitable access to safe and affordable drinking water for all’ (United Nations 2016), reflecting the complexity of water access and responding to the criticisms of the indicator used to measure water access under the preceding Millennium Development Goals (MDGs). The conflation of ‘improved’ sources with ‘sustainable access' during the MDG-era has been questioned repeatedly, as it captures types of sources, rather than water quality (Onda et al. 2012), or availability and accessibility (Satterthwaite 2016). A study from 2012 estimates that ‘1 billion […] people using piped or other-improved water sources receive unsafe water, meaning 1.8 billion people did not have access to safe water in 2010’ (Onda et al. 2012, p. 892). This brings into focus the reality of many urban residents who have access to improved water supply that is neither potable nor reliable (Burt & Ray 2014; Satterthwaite 2016).

The lack of access to safe drinking water, even if ‘improved’ access is provided, is reflected in the growth of the bottled – or packaged – water sector in many cities of middle- and lower-income countries, involving branded bottled water, packaged ‘sachet’ water in West Africa (Stoler 2017) or refill water in Indonesia (Bayer 2013) and in the Philippines (Francisco 2014). However, attention to the bottled water sector within international development research has largely been limited to studies on water quality and public health. Research in West Africa reports that packaged ‘sachet’ water substantially increased the proportion of people able to access (reasonably) safe water and gaining the associated health benefits (Stoler 2012). Similarly, in Indonesia, use of non-branded bottled water is associated with the reduction of childhood diarrhoea compared with available tap water (Sima et al. 2012). While this focus on water quality is justified and relevant, analyses of other dimensions of access to bottled water represent a gap in current research.

This study addresses this gap by investigating the affordability and equity of access to bottled water in Jakarta, Indonesia. The growth of the bottled-water sector in Indonesia has been especially rapid, growing at a pace of more than 12% per year between 2009 and 2014 (Prasetiawan et al. 2017), thereby making Indonesia the second largest market for bottled water in Asia after China (IBWA 2015). The capital city of Jakarta provides a particularly relevant context, given that bottled water is the most universally accessed form of water: 71% of Jakarta's residents rely on some form of bottled water, whereas only 14% consume piped water and 15% consume groundwater for drinking (BPS 2015). The dominance of bottled water reflects the poor water quality status of other sources in the city, where neither shallow groundwater nor piped water supply is safe for drinking (Budiyono et al. 2016).

We focus specifically on consumption of refill water (locally called air isi ulang) which makes up at least 52% of total packaged water consumption in Jakarta (BPS 2015), consumed by rich and poor alike (World Bank 2015). Refill water production uses membrane-based filtration systems to treat raw water sources, typically spring water from surrounding peri-urban areas brought in by trucks. Water is both treated and distributed through decentralized, small-scale refill stations and individual vendors. Given the pervasiveness of refill water in Jakarta it does attract the attention of the urban water development sector (World Bank 2015), but development actors assume it is unaffordable for poorer residents given its higher per unit cost in comparison with domestic groundwater and piped water sources (cf. Bakker et al. 2008; World Bank 2015). Likewise, refill water is assumed to have a negative impact on water equity: in the absence of cross-subsidy through a differentiated tariff structure, poorer households pay proportionally more per unit volume consumed than better-off ones.

Our study sets out to examine these dimensions in more detail, focusing our analysis on access to refill water among the urban poor, who are most likely to be affected by exclusion from water access in Jakarta (World Bank 2015). The current monitoring guide for target 6.1 defines drinking water as affordable when the ‘payment of services does not present a barrier to access to or prevent people from meeting basic human needs’ and holds that equity ‘implies progressive reduction and elimination of inequalities among population subgroups’ (UN Water 2016, p. 5). We examine equity between two socioeconomic population groups, the lower-poor and middle-poor residents of two sub-districts in Jakarta. We acknowledge the importance of other dimensions of difference which determine access to water, such as age, gender, or (dis)ability, and likewise acknowledge the relevance of comparison between poor and high income residents. However, we focus on this sub-group given the relative gap in knowledge on refill water consumption by low income residents and its relevance for water and development goals.

METHODOLOGY

Research locations

Data collection was carried out over a period of 4 months from 2015 to 2016 in two of Jakarta's sub-districts: Penjaringan in North Jakarta and Gedong/Tengah in the southern part of the city. These areas were selected because of their classification as low-income and because of their differences in physical availability of water sources. Both settlements are legal, rather than informal, and households can apply for household connections to the city's centralized piped water network. There are however significant differences in the quality of piped water services given their respective locations within the piped distribution network system; groundwater quality is also very different between the two locations (Kooy et al. 2016).

Penjaringan is located in the city of North Jakarta, which stretches along the northern coastline of the city. Because of its coastal location and low altitude, shallow groundwater in Penjaringan is very saline and therefore considered unfit as a source of drinking water and most other consumptive purposes (e.g. cooking). Piped water services from the centralized network are available via household connection, or by informal sale between neighbours, but water pressure and continuity of hours of service are very poor as the sub-district is at the edge of the distribution network. The second research area formed by the adjacent districts of Gedong and Tengah, is located within the city of East Jakarta, but geographically situated in the southern part of the city. Here, groundwater in the shallow sub-surface is fresh rather than saline, and can be accessed at depths of up to 15 meters by manual or electric pumps. Shallow wells are closed, and in the MDG definition are considered ‘improved’, although often vulnerable to contamination, for instance from inadequate on-site sanitation systems (Kosasih et al. 2009). The centralized piped water network in this area provides higher quality services in terms of water pressure and continuity.

The research followed a sequential mixed-methods research design. Interviews were conducted with 12 small-scale, independent water providers selling refill water in both research locations and two refill water experts. A household survey was designed based on the interviews, and distributed among the customers of the previously interviewed vendors. Households were chosen by means of a systematic-sampling procedure from lists of customers of refill water depots; ten households were randomly selected from each vendor. A total of n = 80 questionnaires was collected, 40 in Penjaringan and another 40 in Gedong and Tengah.

Conceptualizations

We take the current SDG monitoring guide's definitions of affordability and equity as a point of departure and assess volumes of refill water consumed and its relative and absolute affordability for different income groups. Thus, we consider refill water access to be equitable when relative affordability and consumption of refill water does not vary between income groups. We assess absolute affordability by determining the share of monthly income spent on refill water, as well as through capturing customer perceptions of the price of refill water compared with other sources of water.

We use two poverty lines to inform our analysis: one is Jakarta's official poverty line as set by the Indonesian Central Bureau of Statistics' (BPS) at IDR 2.08 million (about US$156) per household per month. At an average household size of four (BPS 2016), this translates to just below US$1.40 per person per day and lies therefore even below the internationally used $2 per person per day cutoff. In contrast, according to our interviews with civil servants of the Indonesian Ministry of National Development Planning (BAPPENAS), given the costs of housing rental and basic services in Jakarta, the poverty line would more appropriately be set at IDR 4.4 million ($330) per household per month. We take both numbers as a point of reference for our analysis. The setting of the poverty line in Jakarta is contested as it is used to set official minimum wage standards, and there are large deviations between surveys of the Wage Council (Dewan Pengupahan) and the Labour Union (Serikat Buruh) calculating the costs of basic household needs (defined as food, clothing, housing, education, health, transportation, saving and recreation). The Labour Union recently published a survey showing costs of housing rent, water supply and electricity were IDR 1,320,000/person/month (Deny 2016). We also know from field research that renting a one person dormitory room in Jakarta, without water or sanitation facilities, costs IDR 500,000 per month.

We acknowledge the long-standing debates on how to define urban poverty (cf. Mitlin & Satterthwaite 2013), and the challenges in the accuracy of such assessments across the various methods used as respondents often under-report on income or expenditures for a variety of reasons. We believe our results to be robust in capturing the percentage of household water expenditures spent on refill water, and in capturing the differences in relative cost and volume of consumption of refill water between income groups, despite the methodological limitations of our own assessment based on self-reporting of income sources. While the assessment of absolute affordability of safe water sources may be affected by higher than reported incomes in the study area, the role of refill water in these calculations is not affected. The implications of our results for equitable access remains valid, as any increases in average income due to under-reporting do not affect that assessment of differences in relative cost and consumption between income groups.

Data analysis

A variable for total household income was created through the addition of the incomes of all the members of one household and one-time monthly incomes. The variable ‘total household income’ was then log-transformed to follow a normal distribution, which was used as a basis for all correlational analyses. A quintile analysis was performed by dividing all households into five equal-sized income quintiles, ranging from the lowest (Q1) to the highest socioeconomic group (Q5) in this sample, each holding n = 16 cases. With n = 80 our sample fulfils the minimum standards of what is needed for valid statistical analysis given temporary and financial constraints. Quintile analysis was chosen (rather than other n-tiles) to prioritize a higher level of precision (i.e. little loss of information) at the expense of larger samples per quintile. We acknowledge that this leads to relatively large standard deviations (see Table 1). Nevertheless, we present our data in quintiles to give the reader the possibility to compare refill water consumption patterns in lower and higher-income brackets. Statistical analysis was conducted with the help of SPSS Statistics. Qualitative data were analysed using a grounded, open-coding approach using atlas.ti software for qualitative data analysis.

Table 1

Monthly household refill water cost and consumption

 Monthly household income (million rupiah)
Monthly refill water consumption (litres)
Share of monthly income spent on refill water (%)
Absolute
As percentage of poverty lines
meansd12meansdmeansd
Q1 1.38 0.51 66.34 31.38 142.88 50.73 3.45 2.92 
Q2 2.68 0.27 128.83 60.89 183.16 123.12 3.34 5.75 
Q3 3.63 0.34 174.49 82.47 179.36 69.92 1.23 0.36 
Q4 4.92 0.33 236.50 111.78 215.08 124.64 2.32 2.67 
Q5 7.12 1.47 342.25 161.76 202.16 113.62 0.81 0.58 
Penjaringan 3.69 1.97 177.38 83.83 199.88 114.00 2.88 4.30 
Gedong/Tengah 4.13 2.21 198.52 93.83 167.96 83.60 1.65 1.73 
Total 3.90 2.09 187.43 88.60 184.68 101.08 2.27 3.33 
 Monthly household income (million rupiah)
Monthly refill water consumption (litres)
Share of monthly income spent on refill water (%)
Absolute
As percentage of poverty lines
meansd12meansdmeansd
Q1 1.38 0.51 66.34 31.38 142.88 50.73 3.45 2.92 
Q2 2.68 0.27 128.83 60.89 183.16 123.12 3.34 5.75 
Q3 3.63 0.34 174.49 82.47 179.36 69.92 1.23 0.36 
Q4 4.92 0.33 236.50 111.78 215.08 124.64 2.32 2.67 
Q5 7.12 1.47 342.25 161.76 202.16 113.62 0.81 0.58 
Penjaringan 3.69 1.97 177.38 83.83 199.88 114.00 2.88 4.30 
Gedong/Tengah 4.13 2.21 198.52 93.83 167.96 83.60 1.65 1.73 
Total 3.90 2.09 187.43 88.60 184.68 101.08 2.27 3.33 

1 = BPS (IDR 2.08 million); 2 = BAPPENAS (IDR 4.4 million).

Ethical considerations were adhered to strictly and followed Diener & Crandall's (1978) principles of ethical social science research: written informed consent was acquired from all respondents preceding their interview/survey participation and the authors are not aware of any deception having taken place vis-à-vis the research participants. Questions were formulated non-intrusively and participants were given the option of not answering the questions.

RESULTS

Affordability

In this study, the average household had a monthly income of IDR 3.9 million ($293), which translates to just above $2.50 per person per day (see Table 1). Thereby, these households – on average – fall below the IDR 4.4 million/month poverty line identified by BAPPENAS officials, while the lowest-income quintile Q1 even falls below the BPS's already low cut-off. This supports the general classification of this sample as low-income and middle-low income, with some households affected by extreme financial deprivations.

Considering water expenditures, as shown in Table 2, refill water has the second highest per unit price of household water supply sources in Jakarta after branded bottled water. However, refill water is almost exclusively used for drinking (21.25%), cooking (1.25%), or a combination of the two (55%). In this sample, respondents also used refill water for ‘home enterprise’ purposes (22.5%) – referring to the use of refill water for commercial purposes from home. Refill water is therefore used at much smaller quantities than bulk water sources, such as piped water and groundwater. At an average consumption of two to three gallons per week (refill water is sold in 19-litre reusable containers, locally referred to as one gallon), households using refill water for drinking purposes (n = 71) (for this calculation, households using refill water for other purposes, such as home enterprise, but not drinking were excluded) pay between IDR 16,000 and 420,000 ($1.19–31.47) per month for refill water consumption. This range stems from households using from 1 gallon (19 litres) up to 112 gallons (2,128 litres) of refill water monthly and can be explained by variations in household income, household size, and the number of purposes for which refill water is used. Overall, almost all the surveyed households, namely 94.4%, spend less than 5% of their income on refill water.

Table 2

Average price per unit and customer perceptions of price and quality of various water sources

Water sourcePrice (USD/m3)Perception priceaPerception qualitya
Branded bottled water 56.16 3.80 4.20 
Refill water 18.33 1.55 3.65 
Nyelang 2.13 3.50 3.07 
Piped waterb 0.38 3.35 3.25 
Groundwater 0.06 1.08 3.39 
Vended water – 3.00 3.00 
Water sourcePrice (USD/m3)Perception priceaPerception qualitya
Branded bottled water 56.16 3.80 4.20 
Refill water 18.33 1.55 3.65 
Nyelang 2.13 3.50 3.07 
Piped waterb 0.38 3.35 3.25 
Groundwater 0.06 1.08 3.39 
Vended water – 3.00 3.00 

Sources: BRPAM, Authors.

a1 = cheap/poor, 5 = expensive/very good.

bexcl. installation costs and monthly maintenance costs and based on the lowest tariff class for low-income households.

Given that drinking water represents only one component of total household costs on water supply, we also recorded the domestic, non-potable water sources accessed by households in each location. In both research locations, only a quarter of households are connected to the piped network via a formal, direct connection. However, when considering the number of households accessing the piped network indirectly via nyelang and/or vended water (nyelang, from the Dutch word slang, meaning hose, refers to the practice of transferring piped water between neighbours with a water hose; vended water is piped water sold via small-scale vendors), a much larger proportion of all households in the northern area (72.5%) access the piped network than in the south (27.5%). Not surprisingly, a much larger share of households in the south report access to groundwater (72.5%) versus the north (42.5%).

When asked about their perceptions of the price of refill water as a drinking water source, a large majority of customers considers it to be cheap (1), based on a scale ranging from 1 (‘cheap’) to 5 (‘expensive’). Although groundwater is perceived as the least expensive drinking water source, it is in fact rarely used for drinking, and involves additional water treatment expenses (cf. Nastiti et al. 2017). Refill water is identified as the second cheapest water source, after groundwater, and is seen as significantly cheaper than piped water (see Table 2) despite the much higher per-unit price. In fact, more than two-thirds of respondents consider refill water to be cheap and all but one respondent rank the price level of refill water between 1 and 3 out of 5. At the same time, consumer perceptions of the quality of water options does not vary as much and stays between 3 (‘average’) and 4 (‘good’) for most water sources. These results challenge a simplistic price-per-unit definition of what can be considered affordable drinking water for an urban low-income household. We expand on this observation below.

Equity: affordability across quintiles

All households in this study were sampled from settlements with concentrations of low-income households, hence our calculation of the average income at just below or above the poverty line. However, there are differences in monthly income among the surveyed households, ranging from IDR 400,000 ($30) to IDR 10,700,000 ($800). This variation allows for an exploration of the differences in refill water access between low-income and middle-low income households through a quintile analysis.

This analysis suggests that higher-income households consume more refill water than poorer ones: there is a moderately positive, significant (p = 0.05) correlation between the level of income and the amount of refill water consumed for drinking and cooking. Further, when comparing mean refill water consumption between Q1 and Q5, it becomes apparent that households in the highest income category consume almost 60 litres more refill water per month than those in the lowest-income quintile (see Table 1). However, the likelihood of getting refill water delivered for an extra IDR 1,000 (less than $0.08) delivery charge and perceptions of price and quality of refill water do not correlate with the respondents' income level. The analysis does show that the poorest households in our sample spend a relatively larger share of their monthly income on refill water while consuming less refill water. This confirms the assumptions found in development reports on urban water supply and we return to this point in the discussion.

Refill water vendors' perspectives on affordability and equity

To complement the customer survey, refill water vendors were asked to indicate how affordable refill water was to the people in their neighbourhood. Five of the 12 believed there were no people in their operating area unable to afford refill water. Others indicate that there might be ‘one or two’ households that cannot afford refill water (Interview RWV-7) and ‘none, but the poor might prefer boiling’ (Interview RWV-10). This is in line with the impression of the secretary general of a refill water organization, who believes that ‘the poor can afford it, but the very poor cannot. About 25% of the poor people cannot afford it – the ultra-poor’ (Interview D-1).

From a business perspective, this is not particularly surprising. The manager of a large refill water franchise company explains that while his company does cater to people of all incomes, when selecting locations for new refill stations, they need to consider prospective neighbourhoods' socioeconomic characteristics:

We business people also … have to know where the profit comes from and whether we can survive in the long run. If the calculation cannot meet the ends, we cannot do it, because the investment is very high too. And we are business people, although our business is doing a social project, there has to be some logic behind it’ (Interview O-1).

Nevertheless, many refill water vendors indicate awareness of the social aspect as part of the business logics they engage in. For almost all of them, identifying a need for clean drinking water among their neighbours sparked their interest in starting a business. They express that they consider theirs a ‘noble’ form of business, one that can solve a problem (i.e. lack of clean water) and add value to the neighbourhood. Moreover, some refill water vendors engage in what could be considered a social form of entrepreneurship. For instance, almost all of them have some form of informal credit system. This allows poorer people, who, for instance receive daily or weekly salaries to refill their water and pay when they can: ‘There are many workers who work during the day and then at night they pay. So there are low income households who are in debt [with the vendor], but within two to three days the household will pay’ (Interview RWV-4). In addition, many of them have ‘charity cases’, as illustrated by the following quote from an interview with the manager of a refill water station in Gedong: ‘There are at the moment 32 gallons which have never been paid by customers. But my boss says ‘it's okay, just give it to them’. It's a kind of charity. So households who totally cannot afford to buy water, he just gives it for free, two gallons per week’ (Interview RWV-5). These examples together with the insight into refill water access and the (extremely) poor illustrate both the challenges for water equity created by refill water and its potential for providing a decentralized and flexible option of safe drinking water for lower-income households which is afforded by the specific sociality in a community, as in the charity cases.

DISCUSSION: AFFORDABILITY AND EQUITY UNDER SDG 6.1

The results of our survey indicate the need to include bottled water in the debate about what is considered equitable water access for the SDG era. First, a more differentiated understanding of what constitutes ‘affordable’ water access for the urban poor is crucial. A frequent argument against both bottled water and small-scale providers is the high price of these water sources and, as a result, the disproportionally high price paid by lower-income households. However, this argument is based on an understanding of ‘affordability’ divorced from assessments of water quality, convenience, or reliability. Our survey indicates that low income households consider all of these variables in their assessment of refill water – the second highest per unit cost source – as ‘cheap’. As highlighted in the debates on water access leading up to the SDGs, these dimensions of water services are relevant for the majority of urban residents globally, where piped water supply does not guarantee drinking water quality, and is not provided on a continuous basis or during the hours it is needed.

In comparison, while the MDGs have placed emphasis on expanding access in urban areas to piped water as the most desirable ‘improved’ water source, it is not considered affordable by Jakarta's low-income households. While per unit volume costs for the lowest tariff of consumers is much lower than refill water, initial connection fees are significant (ranging from about IDR 1 million/$75), and fixed monthly charges for meter maintenance are considered expensive in the months or weeks when no water is actually provided (PAM Jaya 2015). For low-income households who cannot afford an in-house connection, or who cannot legally apply for one in the absence of formal land ownership and residency, the purchase of piped water from neighbours is based on what they can negotiate within their social network and ranges from IDR 13,000 up to IDR 89,000 per cubic metre ($0.97–6.63), compared with IDR 1,050 to 9,800 ($0.08–0.73) per cubic metre from a formal connection (Kooy et al. 2016, p. 6). Additionally, the poor service quality of the piped water network requires that households incur additional costs for water storage, such as building a ground floor cement tank or a rooftop storage, and electricity costs for operating the electric pumps to ‘pull’ water from the pipes in times of low pressure.

There are also additional financial costs for household water treatment of either piped or groundwater sources. While point-of-use (POU) treatments, such as chlorination and ceramic filtration, provide a relatively low-cost option, they are unpopular among Jakarta's lower income population and the Indonesian Government (although recently forms of POU treatment have become more popular in Jakarta's vertical housing, which is mostly occupied by middle to high income population); boiling water is the most common treatment practice across Indonesia (see e.g. Sima et al. 2012), and in our sample. There are indications that the costs resulting from treating water through boiling might be more expensive than refill water: A 2006 USAID study investigating the relative costs of the water sources available to low-income households reported that ‘at a fuel cost of Rp.1,100/L, the cost for boiling 10 L of water (average daily family consumption) is approximately Rp.500’ (Weimer 2006, pp. 3–4). If we accept this calculation, boiling water is then more expensive than buying refill water: 19 litres of boiled water cost a household IDR 8,000 ($0.60) in fuel (March 2017 prices), in addition to the price for the untreated water itself, whereas 19 litres of refill water are available for IDR 5,000 ($0.37).

While these calculations require further investigation, they do call attention to the need for more differentiated calculations of water affordability in current and future development debates. Refill water should not be dismissed out of hand because of its higher per unit costs or assumed to be unaffordable for low-income residents. Moreover, a more detailed assessment of the costs of securing, storing and treating water should inform the monitoring of SDG 6.1. Calculations of affordable water access need to include these costs and/or base calculations of household expenses on water on the different combinations of water used, involving costs for both drinking water and domestic water sources.

Based on our understanding of equity as defined in the current SDG monitoring guide we conclude that refill water is consumed differently by households of different income groups: poorer households consume less and spend relatively more money, in comparison to their income, to access safe drinking water. However, as no household costs of water treatment are subsidized, this suggests that all forms of drinking water treatment now practised are inequitable: lower income households would pay proportionally more for drinking water than would higher income households. We also suggest that like for monitoring of affordability, assessments of equitable access should look across the range of water sources that households have access to. When standing alone from domestic water consumption, almost all refill water consumers spend less than 5% of monthly income on drinking water. However, in another household survey in areas of Jakarta similar to the ones surveyed in this study, we found that total household expenses for water (drinking plus domestic) made up more than 5% of household income of the poorest households (Kooy et al. 2016). In that survey, we also found that when comparing the numbers for overall water expenditure with drinking water expenditure for refill water, drinking water costs make up a much larger share of total water expenditures for Q1 than for Q5. This suggests that there is a need for development interventions specifically targeting the provision of drinking water and/or inclusive water treatment.

CONCLUSION

Currently, refill water reaches more than 3 million people daily in Jakarta alone: branded bottled water and refill water combined reach more than 7 million people. The continued growth and the scale of the sector suggests that this is unlikely to be a temporary phenomenon that phases out, presuming the eventual expansion of the piped network. Similar assumptions have been made with regard to groundwater use: groundwater abstraction in Jakarta was assumed to decrease and eventually come to a halt with the increase of piped water connections. However, today, groundwater use is widespread even where the piped network reaches, and evidence suggests that piped water and groundwater are used in conjunction rather than exclusively (Kooy et al. 2016).

Accepting bottled drinking water as a permanent feature of Jakarta's water supply, rather than a temporary phenomenon, would help to focus attention on aspects of the sector that need further scrutiny. Water quality, plastic waste management, and sustainability of raw water sources in peri-urban source locations are three key problems which have been raised in reference to both Jakarta's water, and more generally for the sector globally (Stoler 2012; Roekmi 2017). Seeing bottled water as an integral aspect of urban water supply is also supported by the recent re-classification of water sources under SDG Target 6.1, ‘The JMP recognizes that bottled water … can potentially deliver safe water … [and] will treat them as improved and classify them as ‘limited’, ‘basic’ or ‘safely managed’’ (WHO 2017: p. 13) according to their accessibility, availability and quality. This confirms the need to attend to the role of bottled water in providing access to improved water supply, while our research highlights the need to assess criteria such as affordability and equity across the multiple water sources (drinking and domestic) used by urban households in cities like Jakarta.

Recognition of the role of bottled water currently plays in providing access to improved water sources for low income urban populations is particularly urgent in the context of Indonesia, but also, we suggest, for other urban contexts where the majority of ‘improved’ sources under the MDGs are not potable. Several studies show that the importance of packaged water extends beyond Jakarta, playing an increasingly important role in urban water access in cities across Thailand (Hadipuro 2010), the Philippines (Francisco 2014), and West Africa (Stoler 2017). This is a strong indication that future approaches targeting the provision of inclusive access to both domestic-use water and drinking water in cities in low- and middle-income countries need to take packaged water into account in both the strategies designed to increase access, and the measurements of how affordable or equitable access really is.

REFERENCES

Badan Pusat Statistik BPS
2015
Statistik daerah: Provinsi DKI Jakarta
.
http://jakarta.bps.go.id (accessed 9 January 2017)
.
Badan Pusat Statistik BPS
2016
Gini Ratio by Province
. .
Bakker
,
K.
,
Kooy
,
M.
,
Shofiani
,
N. A.
&
Martijn
,
E.
2008
Governance failure: rethinking the institutional dimensions of urban water supply to poor households
.
World Development
36
(
10
),
1891
1915
.
Bayer
,
R.
2013
Drinking Water as a Source of Income: How Microentrepreneurs in Water Business Perceive Drinking Water as a Source of Income
.
Master's Thesis
,
Lund University
,
Lund
,
Sweden
.
Budiyono
,
Y.
,
Aerts
,
J. C. J. H.
,
Tollenaar
,
D.
&
Ward
,
P. J.
2016
River flood risk in Jakarta under scenarios of future change
.
Natural Hazards and Earth Systems Sciences
16
,
757
774
.
DOI: 10.5194/nhess-16-757-2016
.
Burt
,
Z.
&
Ray
,
I.
2014
Storage and non-payment: persistent informalities within the formal water supply of Hubli-Dharwad, India
.
Water Alternatives
7
(
1
),
106
120
.
Deny
,
S.
2016
Tuntut UMP Rp 3,8 Juta di 2017, Begini Hitungan Buruh
. .
Diener
,
E.
&
Crandall
,
R.
1978
Ethics in Social and Behavioural Research
.
University of Chicago Press
,
Chicago, IL
.
Francisco
,
J. P. S.
2014
Why households buy bottled water: a survey of household perceptions in the Philippines
.
International Journal of Consumer Studies
38
(
1
),
98
103
.
DOI: 10.1111/ijcs.12069
.
Hadipuro
,
W.
2010
Indonesian water supply regulatory framework: between commercialisation and public service?
Water Alternatives
3
(
3
),
475
491
.
International Bottled Water Association
2015
Bottled Water 2015: U.S. and International Development and Statistics
. .
Kooy
,
M.
,
Walter
,
C. T.
&
Prabaharyaka
,
I.
2016
Inclusive development of urban water services in Jakarta: the role of groundwater
.
Habitat International
(in press)
.
Kosasih
,
B. R.
,
Samsuhadi
,
S.
&
Astuty
,
N. I.
2009
Kualitas air tanah doe kecamatan tebet Jakarta selatan ditinjau dari pola sebaran Escherichia coli
.
Journal Teknologi Lingkungan
5
(
1
),
12
18
.
Mitlin
,
D.
&
Satterthwaite
,
D.
2013
Urban Poverty in the Global South: Scale and Nature
.
Routledge
,
London and New York
.
Nastiti
,
A.
,
Muntalif
,
B. S.
,
Roosmini
,
D.
,
Sudradjat
,
A.
,
Meijerink
,
S. V.
&
Smits
,
A. J. M.
2017
Coping with poor water supply in peri-urban Bandung, Indonesia: towards a framework for understanding risks and aversion behaviours
.
Environment & Urbanization
29
(
1
),
69
88
.
DOI: 10.1177/0956247816686485
.
Onda
,
K.
,
LoBuglio
,
J.
&
Bartram
,
J.
2012
Global access to safe water: accounting for water quality and the resulting impact on MDG progress
.
International Journal of Environmental Research and Public Health
9
(
3
),
880
894
.
DOI:10.3390/ijerph9030880
.
PAM Jaya
2015
Biaya Sambungan Baru
. .
Prasetiawan
,
T.
,
Nastiti
,
A.
&
Muntalif
,
B. S.
2017
‘Bad’ piped water and other perceptual drivers of bottled water consumption in Indonesia
.
WIREs Water
e1219
.
DOI:10.1002/wat2.1219
.
Roekmi
,
R. A. K.
2017
Safe water at a premium
.
Inside Indonesia
127
,
1
3
.
Satterthwaite
,
D.
2016
Missing the Millennium Development Goal targets for water and sanitation in urban areas
.
Environment and Urbanization
28
(
1
),
99
118
.
DOI: 10.1177/0956247816628435
.
Sima
,
L. C.
,
Desai
,
M. M.
,
McCarty
,
K. M.
&
Elimelech
,
M.
2012
Relationship between use of water from community-scale water treatment refill kiosks and childhood diarrhea in Jakarta
.
The American Journal of Tropical Medicine and Hygiene
87
(
6
),
979
984
.
United Nations
2016
Millennium Development Goals and Beyond 2015
. .
UN Water
2016
Integrated Monitoring Guide for SDG 6: Targets and Global Indicators
. .
Weimer
,
M.
2006
Action Research on Point of Use Drinking Water Treatment Alternatives as Appropriate for Underprivileged Households in Jakarta
.
Report Environmental Services Program
,
USAID
,
Jakarta
,
Indonesia
.
WHO/UNICEF
2017
Safely Managed Drinking Water
,
(accessed 11 June 2017)
.
World Bank
2015
More and Better Spending: Connecting People to Improved Water Supply and Sanitation in Indonesia – Water Supply and Sanitation Public Expenditure Review (WSS-PER)
.
Water and Sanitation Program, World Bank
,
Jakarta
,
Indonesia
.