Planning the Liveable If Not the Ideal: Frontline Planners’ Discretionary Actions and Inequalities in Everyday Intermittent Water Supply Planning in Tiruppur, India

Abstract

Millions of urbanites across the Global South receive intermittent water supply (IWS). This paper examines how discretionary action by street-level bureaucrats (SLBs) in everyday water supply planning in IWS systems redistributes water quantities and service quality to alleviate unequal scarcity burdens. Ethnographic findings from the small city of Tiruppur, India, reveal that SLBs’ supply plans aim to distribute water volumes equitably within the network. However, SLBs’ communicative strategies to manage socio-material uncertainties and improve supply predictability, a critical dimension of service quality, reinforce existing socio-spatial inequalities. Rethinking SLBs’ contradictory discretionary actions in everyday water supply plans is crucial for managing water scarcity equitably in the IWS systems of Global South cities.

Keywords:

• Street-level bureaucrats
• discretion
• water scarcity
• uncertainty
• small cities
• water access inequalities

Introduction

Cities of the Global South experience chronic piped water scarcity. Even in cities with high coverage of piped water networks, water flows through the pipes intermittently or for a few hours each day or week, creating a state of scarcity. Intermittent water supply (hereafter IWS) is the dominant mode for nearly a billion people in low- and middle-income countries (Kumpel & Nelson, 2016). The causes of intermittency and the characteristics of IWS systems vary considerably across regions and, often, within cities, with implications for unequal experiences of water scarcity and security (Totsuka et al., 2004).

Planning scholarship has investigated water access patterns in IWS or hybrid systems (Beard & Mitlin, 2021; Carolini, 2012) and the outcomes of pilot efforts to convert intermittent systems into continuous systems (Burt et al., 2018). It has also examined user (dis)satisfaction with IWS and their turn to alternative community-managed sources (Das & Takahashi, 2014; González Rivas et al., 2014; Spencer, 2008), willingness-to-pay for improved water services (Pattanayak et al., 2005), demand improved services from utilities (Acey, 2019; Pierce, 2017), and ensuing impacts of these efforts on water access and affordability (Beard & Mitlin, 2021). Despite the ubiquity of piped networks across the Global South, the possibilities for everyday water supply planning within piped IWS systems remain to be critically examined by planning scholars.

Everyday water supply planning by frontline utility staff or street-level bureaucrats (hereafter SLBs) (Lipsky, 1980) in IWS systems is a critical component of water infrastructure planning and everyday service delivery in cities of the Global South, even if scholarly and policy attention is overwhelmingly focused on episodic plans for producing large infrastructures that take years (and sometimes decades) to come to fruition (Anand, 2020). Across these cities, SLBs prepare water supply schedules to deliver water for a carefully calculated duration to each valve area (the smallest geographical unit for water supply) in their city’s piped network. SLBs, thus, decide water quantity – how much water people living in a valve area will receive and aspects of service quality in IWS systems (Kumar et al., 2022) – when users will receive water, for how long, and with what frequency, reliability, or predictability. Supply quantity and service quality tend to vary unequally in IWS systems as land uses, localities, and communities deemed important tend to be privileged in supply schedules, by way of supply timings, duration, and/or supply frequency (Alda-Vidal et al., 2018; Anand, 2017, p. 101). SLBs’ responses to technical constraints like the size and condition of the pipe network, material agency, and uncertainties about water flows (Anand, 2017; Tiwale, 2019), labour shortages (Zérah, 2020), and socio-political pressures (Anand, 2017; Coelho, 2006) also introduce quantity and quality variations. Their everyday water supply schedules, practices, and distribution plans, thus, have material implications for people’s experiences of water scarcity in Southern cities where piped water supply networks are inadequate, under-sized, or still being built. Since SLBs have some latitude in planning schedules and distributing water in IWS systems, examining everyday supply planning also provides a window into the exercise of discretion in everyday planning (Catney & Henneberry, 2012; Forester et al., 2023; Lipsky, 1980) and its relationship to the persistence (or contestation) of socio-spatial inequalities in urban water access.

Research on discretionary action is torn between celebrating frontline planners’ creative improvisations to realise policy goals and critiquing these actors for yielding to political pressures and elite interests (Forsyth, 1999). Particularly in India – where this study is based – relying on discretionary actions by frontline bureaucrats (or state officials) allows a large section of the population that is unfairly and arbitrarily excluded from formal policies to access citizenship entitlements (Chatterjee, 2004). How frontline water supply planners reinforce or address inequalities through discretionary action to correct policy failures or mitigate uncertainties (Beauregard, 2021) awaits further elaboration. To illuminate relationships between discretionary planning and unequal experiences of water scarcity and to analyse the potential of incremental, everyday discretionary actions to improve service delivery, this paper answers the following research questions: In the absence of policy prescriptions for IWS, how do frontline water utility workers (or SLBs) exercise discretion in everyday water supply planning? In allocating water quantity and service quality through everyday supply plans (schedules), how do they (re)distribute the burdens of water scarcity? How do the distributive practices and communicative strategies informing discretionary actions address socio-spatial inequalities?

This paper’s analysis of ethnographic evidence on SLBs’ discretionary actions from Tiruppur, India, finds that frontline planners seek to create everyday supply plans (or schedules) that aim to distribute water quantity as equally as possible to all qualifying connections (if not residents) while providing minimum water volumes to low-income residents and operating within infrastructural capacity constraints. These everyday supply plans and discretionary actions to manage uncertainties rely on communicative strategies that reflect and reinforce socio-spatial differences among users. They improve supply predictability for selected local users while excluding inter-state migrants, inadvertently creating uneven service quality across the city’s piped network. This study thus reveals the potential and limits of everyday discretionary action for addressing unequal water access in IWS systems in contexts without policy safeguards for equitable water quantity and service quality provision for vulnerable populations.

The next section weaves together the planning literature on discretion and uncertainty, interdisciplinary scholarship on everyday water supply by SLBs, and literature on state-society relations in urban India to articulate relationships among discretionary actions, (un)equal water access in IWS, and citizens’ access to the state. Thereafter, I describe my research methods and case context, noting the many causes of piped water scarcity in Tiruppur. The empirical section of the paper describes discretionary actions in everyday water supply, analysing the communicative strategies informing them and how they reinforce or challenge urban inequalities. After discussing the case findings and contributions, the paper concludes with reflections on the possibilities for relying on discretionary, bottom-up actions to improve water infrastructure planning and address unequal water access in urban contexts with IWS.

Discretionary Planning’s Possibilities and Pitfalls

Policy implementation often involves interpretation and the exercise of discretion by frontline planners or SLBs (Lipsky, 1980). Discretionary action is especially common in delivering services like water supply where the required information cannot be easily accessed ex-ante and/or involves informal context-specific knowledge (Pritchett & Woolcock, 2004, p. 194). When planners exercise discretion, they “make practical judgments and choices about how to act: either how to apply rules or how to operate when rules do not exist” (Forsyth, 1999, p. 6). Planning scholars are ambivalent about discretion, celebrating discretionary actions that involve creative policy improvisation through experimentation, collaboration, and negotiations to devise the most appropriate solutions to local problems (Forester et al., 2023; Pritchett & Woolcock, 2004) and criticising discretionary transgressions that pander to elite interests, succumb to political influences or social biases, and produce arbitrary and inconsistent policy application (Biggar & Siemiatycki, 2023; Catney & Henneberry, 2012; Forsyth, 1999).

Whereas Western scholarship and practice approach discretionary actions cautiously, empirical analyses of state-society relations from India, particularly those focused on relations between the urban poor and lower-level state functionaries or SLBs, present a more complicated picture. They observe that the poor, including those inhabiting informal settlements or occupations, routinely engage in corrupt practices such as paying bribes (Anjaria, 2011; Gupta, 2012), leveraging social connections, fixers, or brokers within the bureaucracy (Benjamin & Bhuvaneswari, 2001; Björkman, 2015), and exerting pressure through their political representatives (Anand, 2017; Chatterjee, 2004) to tap into SLBs’ discretionary powers and access substantive citizenship entitlements that are unfairly denied to them and/or remain inaccessible when bureaucratic rules and procedures are strictly followed (Gupta, 2012). These bonds between the poor and SLBs tend to be fragile and contingent (Chatterjee, 2004; Ranganathan, 2014); they reflect the citizen’s social position (class, caste, religious or ethnic identity, gender) but also the urban institutional setting (bureaucracy’s size, complexity, capacity, and distance) that varies across mega- and small cities in India (Heller et al., 2023). Nonetheless, the various bonds that the poor seek to forge and the ‘illicit’ mechanisms through which they do so open up ‘ordinary spaces of negotiation’ (Anjaria, 2011), allowing the poor to influence policy implementation, i.e., lower-level state functionaries’ interpretation and application of rules in practice in the absence of clear guidelines. On their part, SLBs might sometimes deviate from official rules to satisfy locally situated notions of fairness and responsibility (Rankin et al., 2024). These studies suggest that discretionary actions could unexpectedly produce progressive, locally situated forms of pro-poor outcomes, even if they are contingent, reinforce and build on illegal practices, involve the abuse of power by SLBs, and exclude groups without the contextual knowledge or wherewithal to access the state through extra-legal arrangements.

Urban geographers have investigated how and why planning a city’s piped network or the “modern infrastructural ideal” (Graham & Marvin, 2001) reflects or produces socio-spatial inequalities (e.g., Kooy & Bakker, 2008). However, our understanding of how everyday supply planning or the discretionary action of SLBs involved in service delivery reinforces and layers additional inequalities onto this ‘splintered ideal’ is limited. In the absence of clear policy prescriptions for the distribution of water quantity and service quality to various supply areas in IWS systems, SLBs routinely make practical judgements about how much piped water people in a certain locality will receive, when, and for how long (Anand, 2017). Since water allocation is a zero-sum game – more water for some invariably creates scarcity for others, SLBs encounter and resolve ethical dilemmas surrounding equity in daily operations. The water utility’s and SLBs’ notions of ‘deservingness’ of particular localities and user groups therein no doubt influence their actions (Tiwale, 2019), as do local social norms on fairness and political pressures from users and their elected representatives (Anand, 2017; Coelho, 2006; Hossain, 2011; Zérah, 2020). SLBs, thus, have the discretionary agency to challenge (or actively maintain) prevailing socio-spatial inequalities in the piped network and the city more broadly (Coelho, 2006), either by supplying water at convenient times, for longer durations or with greater frequency and predictability in chosen localities. How they do so and to what outcomes in Tiruppur remains to be seen.

Material Agency, Uncertainties, and Discretionary Action

SLBs’ discretionary actions in water supply are distinct from their practical judgements in delivering other services; they are also circumscribed by water availability, network characteristics, infrastructural capacity, and material agency that together introduce many everyday uncertainties for SLBs and consequently for users (Anand, 2017). Seasonal or climate-induced changes in river or lake water levels, burst or leaky pipes, electricity shutoffs, pressure variations, roadside construction, extreme weather events, water pilferages, or a vehicle that is wrongfully parked on a supply valve can all disrupt the best-laid plans (or supply schedules). Uncertainties are complicated by the fact that SLBs across the Global South operate in informal contexts without detailed, updated maps about the piped network or real-time data on water flows across the city, instead relying on their (and users’) experiential knowledge (De Coss-Corzo, 2021). Furlong (2014) notes that water supply schedules help utilities create predictability and reduce inconvenience for users in crisis-prone systems. However, the details of these schedules – how they are created, communicated, or implemented and for whom they alleviate inconveniences and uncertainties (if at all) remain to be studied.

Planning scholars have advocated collaborative planning strategies such as dialogues and negotiations to address socio-material uncertainties (Beauregard, 2021, building on Innes & Booher, 2004) while acknowledging unevenness in the capacity to cope with unavoidable disruptions and unpredictability (Marris, 1996). How SLBs engage in dialogue, negotiations, or other communicative practices to address uncertainties impacting water supply planning and the (uneven) impacts of these strategies on water access or experiences of scarcity await further examination.

In interrogating how SLBs exercise discretionary agency to (re)distribute water scarcity burdens in Tiruppur, I focus on analysing how they conceive, operationalise, and justify notions of fair water quantity and service quality allocations while simultaneously working within (or challenging) the bounds of policy guidelines, material constraints, and available infrastructural capacity. More importantly, by examining the ‘spaces of dialogue and negotiation’ influencing discretionary actions, including in moments of uncertainty, and questioning who influences and benefits from these, this study recognizes the potential of SLBs’ discretionary planning to produce pro-poor service delivery outcomes while critically analysing the associated inequalities and limits to such discretionary action.

Case Context and Methodology

This paper draws on ethnographic findings from Tiruppur, a city in the southern state of Tamil Nadu, India. Tiruppur houses over a million residents, most of whom are employed in the city’s export-oriented textile economy. Despite its global economic orientation, Tiruppur is a secondary and small city; it does not dominate popular imaginations of urban India. Although Tiruppur is a single, non-representative case, its vantage point as a secondary, ‘off-the-map’ city (cf. Robinson, 2002) offers some conceptual and empirical advantages for theorizing discretionary action in everyday supply schedule planning. Tiruppur’s relatively small size greatly reduced the diversity of state actors and the complexity of governance arrangements in the municipal piped water network, making it easier to follow SLBs and question their community engagement practices and relationships across the city. Focusing on Tiruppur also allows me to respond to calls for illuminating how everyday planning processes intersect with urban inequalities in smaller, non-metropolitan places that comprise the global urban majority (Kudva, 2015; Randolph & Deuskar, 2024).

Between 2017 and 2020, I conducted multiple rounds of ethnographic fieldwork in Tiruppur, where my data collection methods included a thorough review of locally available water infrastructure plans and water supply data, participant observation of everyday water supply, and in-depth, open-ended interviews in Tamil with 33 informants involved in planning piped water supply. These informants – identified through non-random snowball sampling – included four city administrators, six engineers who knew or planned bulk water supply and the distribution network, and 23 SLBs in everyday water supply operations, sampled from diverse localities across Tiruppur. The SLBs comprised five ward-level assistant engineers, eight tap inspectors, and ten valve operators (locally called ‘watermen’). Figure 1 depicts the organisational structure of Tiruppur’s water department. Not all SLBs worked at the street-level (the assistant engineers did not), though they all played distinct roles in exercising discretion in planning, implementing, or overseeing water supply schedules (as elaborated later) and interacted with service users (clients) and their elected representatives to varying degrees.

Figure 1. Schematic diagram of Tiruppur water department’s organisational structure.

Localities were purposively sampled to include supply areas in the urban core and peripheral villages recently brought under city administration to capture variations across diverse network conditions, population densities, and socio-political pressures. I conducted repeat interviews with four informants and interviewed four new persons in the summer of 2022. In interviews, I probed my informants to elaborate on the technical, procedural, policy, and political considerations that shaped everyday supply and explain their approaches to fair water quantity and service quality distribution. The theme of unpredictable disruptions emerged through the interviews. I triangulated the SLBs’ accounts with current or former ward councillor interviews (n = 14) and surveys of water users (n = 110) across the city, discussed at length in other publications. I translated and transcribed all the interviews and used a combination of iterative descriptive and interpretive analysis to arrive at the conclusions. Informants’ names have been anonymized for privacy reasons.

The Un-Planned Origins of Water Scarcity and Intermittent Water Supply in Tiruppur

Many factors contribute to Tiruppur’s water scarcity, including the city’s location in an arid, water-scarce region. In the last century, rapid and unplanned urbanization, which followed on the heels of industrial restructuring in the global textile economy that was localized in city-specific ways, also generated additional water demand, stressing local, non-perennial water infrastructures meant to serve smaller populations. Tiruppur was a small town (and later a small city), so it was not always prioritized in state-level funding allocations for constructing large urban water supply schemes. A historical analysis of Tiruppur’s water supply schemes constructed in the last century reveals that the long durée between the design of piped water infrastructures, the allocation of state (or donor) funds to build it, and rapid interim growth in the local economy meant that the installed infrastructure lagged true water demand, creating potable water scarcity (Subramanyam, 2021, chapter 2). Despite its shortcomings, this approach to managing scarcity – rooted in episodic plans for new water supply schemes – persists.

At present, Tiruppur Municipal Corporation (the city government) receives bulk (source) water from two water supply schemes located about 50 kilometres away from the city: (i) a city-owned water supply scheme operated by the parastatal, Tamil Nadu Water Supply and Drainage Board (TWAD Board); and (ii) a water supply scheme owned and operated by the New Tiruppur Area Development Corporation Limited (or NTADCL, a public-private partnership company). The TWAD Board scheme was designed to supply about 30 MLD (Million Litres per Day) (or 25%) of Tiruppur’s total daily potable water requirement of 120 MLD in 2011. However, the city receives only 11 MLD because of increasing water withdrawals from several fast-growing towns and villages along the supply pipeline. Consequently, Tiruppur relies on the NTADCL scheme to satisfy its remaining daily water demand of 110 MLD. Since NTADCL is a for-profit operator, these additional withdrawals have increased the city’s water expenses, which remain uncovered from user tariffs.¹ The additional expenses notwithstanding, these supplementary water purchases have not fulfilled the city’s daily water requirement.² In 2020, engineers and administrators projected that the city needed an additional 32 MLD to satisfy a population (and water demand) growing at 6% annually. Hence, the city started constructing a new water supply scheme that is yet to begin operations.

Although Tiruppur Municipal Corporation has taken steps to address its bulk water scarcity, other reasons contribute to piped water scarcity and intermittency. The city’s distribution network (or ‘d-system’ in engineers’ speak) consisting of overhead reservoirs and distribution pipes was designed to distribute just about 30 MLD water from the TWAD Board Scheme to a maximum population of 700,000 residents in 2011. Tiruppur’s population exploded in the early 1990s in response to an unforeseen boom in textile exports, exceeding 850,000 in 2011. Consequently, the d-system proved inadequate for the city’s growing water needs. This d-system was not upgraded when the city started purchasing additional bulk water from NTADCL; it was expanded incrementally in a piecemeal fashion. IWS is also labour-intensive compared to continuous 24x7 supply, as multiple supply valves must be switched on and off at different times (Totsuka et al., 2004). Unfortunately, Tiruppur’s human infrastructure required to operate its d-system stagnated at 1990 levels, thanks to the institution of neoliberal state-wide hiring freezes. The infrastructural labour force did not expand with the construction of the new water supply scheme and an overhaul of the physical d-system.

Tiruppur’s water supply system is under-designed and intermittent because scheme designs underestimate true water demand; they ignore the city’s circular migrant population and widespread informality in their calculations. Although the city’s circular migrant population has not been systematically enumerated, surveys conducted during COVID-19 lockdowns indicate the presence of 240,000 interstate migrant workers who comprise over 25% of the city’s population (Press Trust of India, 2021). Most of these migrant workers live in informal rental housing on subdivided farmland, sharing a single water connection between four and fifty households. These water scarcities – created by oversights in scheme design – are hardly unique to Tiruppur; they occur in other Indian cities, too (cf. Anand, 2017).

Everyday water supply in Tiruppur involves rectifying these policy and planning oversights that produce water scarcity. SLBs exercise discretion to plan supply schedules that distribute scarce piped water judiciously. The next section details how SLBs in Tiruppur’s water department distribute the burdens of water scarcity over space and time through everyday supply planning.

Everyday Water Supply Planning in the Face of Scarcity

Tiruppur’s piped water network serves 185,349 connections, covering approximately 71% of the population (Census of India, 2011). Network coverage is reasonable and higher than the average in similarly sized cities (Subramanyam, 2020, p. 477), even if uneven. Most Tiruppurians experience water scarcity because of intermittent piped water supply, which unevenly distributes water quantity and service quality across the city.

When measured along multiple dimensions, service quality is uneven across Tiruppur’s piped network. Take, for instance, supply frequency. SLBs strive to supply water at least once weekly to each supply area. However, some areas receive water once every six days, and others, especially the peripheral villages, receive piped water only once every 8–10 days (see Figure 2). SLBs ascribe these differences to variations in the distribution network capacity and smaller pipe sizes in the peripheral villages originally designed to supply fewer connections. This has reduced water pressure and the number of connections that can be supplied concurrently, causing supply delays.

Figure 2. Map showing piped water supply frequency at the ward-level in Tiruppur in April 2022 (cartography by Talia Fan; water supply data obtained from tap inspectors in summer 2022).

Not only is supply frequency variable, but supply timings are not fixed; they are difficult to ascertain. However, since early 2022, Tiruppur’s water department has started posting area-wise, fortnightly supply schedules on the Corporation website. These schedules are prepared by the tap inspectors in consultation with the watermen and then approved by the ward assistant engineers, who forward them to the central office for online posting. The schedules announce the dates on which particular localities will receive water supply but do not specify the hour. All the SLBs assert that exact timings are difficult to specify because of many uncertainties. They note that technical disruptions like power failures, leakages, or burst pipes routinely introduce delays and throw supply off schedule. Hence, SLBs refrain from committing to a specific time (or time window) to avoid complaints in case of schedule deviations. Since supply timings are difficult to predict, many residents (especially those without an underground sump and pump storage facility) skip work, wait expectantly all day on scheduled supply days, and/or turn to alternative sources. Despite the low supply frequency, unclear supply timings, and the overall scarcity of piped water, the city has not descended into chaos. Below, I detail how Tiruppur’s frontline planners exercise discretion in planning and communicating everyday water supply schedules. I demonstrate how their planned schedules and on-the-fly improvisations allow them to distribute water flows and supply quantities in ways that assure minimum water volumes to marginalized residents, while responding to socio-material uncertainties and averting foreseeable conflicts. However, these discretionary improvisations nonetheless result in uneven service quality.

Improvising Water Distribution through Intermittent Supply Schedule Planning

National and state-level water policies in India prescribe norms or service levels for urban service provision. In the municipal water sector, these service levels are tracked using performance indicators like piped water supply coverage, per capita quantum of daily water supply, supply continuity, water quality, and complaint redressal efficiency (Ministry of Urban Development [MoUD], Government of India [GoI], 2010). Whereas the service-level benchmarks and indicators developed by the National Ministry of Urban Development recognize the widespread prevalence of IWS (Ministry of Urban Development (MoUD), Government of India (GoI), 2010, p. 31), they do not prescribe any norms or guidelines for intermittent supply planning. Government manuals for designing and operating urban water supply systems continue to naively encourage the development of continuous (24x7) supply systems rather than prescribe how to improve IWS (Meyer & Subramanyam, 2023). In the absence of policy prescriptions or guidelines, SLBs devise contextually appropriate norms for service quality and attempt to achieve them through their intermittent supply plans.

Tiruppur’s SLBs have developed water supply norms and schedules that are pro-poor, drawing on the watermen and tap inspectors’ embodied knowledge of household-level water connection, collection, and storage practices, the acknowledgement of informality and the widespread practice of sharing water connections, and communications with water users and/or their elected representatives. They aim to distribute a minimum weekly quota of piped water for drinking and cooking needs to low-income, working-class tenants within the network by carefully controlling supply duration. Whereas water supply frequency and supply timings varied across Tiruppur, supply duration was more or less fixed at 2 hours per supply in every valve area across the city. Uniform supply duration across Tiruppur contrasts observations from other cities where supply duration was significantly lower in slums, even after accounting for pressure variations (Alda-Vidal et al., 2018; Anand, 2017). The fixed supply duration also shows that SLBs’ discretionary actions are not entirely ad hoc but have become locally codified over time, reducing confusion for users (cf. Biggar & Siemiatycki, 2023).

How did Tiruppur’s SLBs arrive at the 2-hour supply duration norm? Assistant engineer Ramesh rationalized the supply duration by noting how it complied with policy prescriptions,

The [policy] norms require that we supply water at the rate of 110 lpcd (litres per capita per day). We assume that each household [and connection] has about 4 persons. Since we supply water once every seven days, we must provide 110x4x6 or 2640 litres per supply. In 2 hours, each connection gets about 2500 litres on average at the prevailing pressure. (Interviewee #TIRENG11, personal communication, 11 June 2022)

Although the locally devised supply duration helped the city meet national-level water quantity and adequacy norms, it overlooked water contamination risks common to IWS (Kumpel & Nelson, 2016). Even if each household in the network received the prescribed water quantity, it remained responsible for storing water safely for a week (until the next supply) to prevent further microbial contamination. Since supply duration varies inversely with supply frequency, I pressed tap inspector Dinesh to explain why duration could not be reduced to supply water more frequently, noting that this would help alleviate storage and contamination issues. Unlike engineer Ramesh, who emphasized policy alignment, Dinesh highlighted how contextual conditions and the SLBs’ commitments to pro-poor service delivery shaped discretionary practice by explaining,

People are habituated to this system. You will have 10 houses in a housing compound and just 1 or 2 connections. All 10 houses need to line up and collect water during supply, so they will ask for a 2-hour duration. We have planned the duration from their point of view, not the single-family with one connection [as is the policy assumption]. In between, during the Corona period [COVID-19 lockdowns], we changed the duration [and frequency]. We would supply water for 1 hour once every 4 days. Back then, Tiruppur’s population reduced with the migrant workers returning home, and there weren’t as many people collecting water in compounds. But today, where there were 250 persons [during lockdowns], there are 450. So, the 1-hour duration is too ‘short’ for shared connections. (Interviewee #TIRENG07, personal communication, 14 June 2022)

In another supply area, waterman Pichai elaborated how they had arrived at the duration through a negotiated process that involved a failed experiment,

We experimented once. We supplied water for one hour every 2 days instead of 2 hours every 4 days. But we received SO many complaints… You see these [housing] compounds have just one connection and many tenants. [In one hour], each tenant could fill just 2-4 pots of water […] These [water-deprived] households complained, and it became a political issue. So, we reverted to our old system. Now, each household can fill 8 to 10 pots of water each time and get by with that stored water until the next supply. (Interviewee #TIRWO06, personal communication, 27 December 2019)

The above quotes reveal how SLBs’ discretionary actions draw on their deep observational knowledge of water connection, storage, and use practices in low-income tenant communities across Tiruppur, as well as their negotiations with water users and politicians (albeit through conflictual, failed experiments) to arrive at a mutually acceptable supply duration. These observations additionally show that SLBs are sensitive to the unique scarcity burdens that low-income residents experience, including the emotional labour of negotiating access with their neighbours and the physical exhaustion involved in storing water safely in many containers.

Variable supply frequency invariably meant that each connection received an unequal volume of water every month. Flat-rate tariffs further implied that every connection (unfairly) paid the same price regardless of how much water they received (or consumed). Further, households within a single valve area also received unequal quantities of water depending on variations in topography, network condition, and water pressure in their valve area’s d-system. Tiruppur’s SLBs often complained that labour shortage prevented them from addressing these inequalities. However, SLBs also improvised to compensate for unequal pressure (and, therefore, unequal water volumes). First, the engineers and tap inspectors re-zoned valve areas to reduce the number of connections within each area, and subsequently minimize pressure variations. Second, at their level, the watermen provided a ‘grace period,’ i.e., they increased supply duration by about 15 to 20 minutes in large valve areas where residents complained about low pressure. Decisions about these extended durations were ad hoc and made by individual valve operators for their supply areas; they also did not deviate noticeably from the citywide norm of two hours (Interviewee#TIRENG20, personal communication, 4 December 2019). Despite their modest gains, these improvisations demonstrate the SLBs’ commitment to fair and equitable allocations despite labour and infrastructural constraints.

Since the supplied piped water was insufficient to meet the daily water needs of low-income tenants with shared connections, Tiruppur’s water department supplemented this with groundwater supply through neighbourhood-scale decentralized borewell water networks. Unlike piped water supply, borewell water supply is untreated. However, it is supplied free of cost to the user and is available every day at specified, convenient hours each morning and evening. For reasons of space, details of borewell water supply and governance are elaborated elsewhere, but some of these borewell networks were managed by SLBs in the water department (Subramanyam, 2023). Borewell water is a lifesaver for low-income tenants who share water connections and have limited access to piped water. Whereas borewells allow the city to meet policy norms for water quantity, they introduce and legitimize unevenness in water and service quality. Not only do low-income women disproportionately carry the burden of fetching borewell water, but their families also consume a greater share of lower-quality water compared to wealthier residents with private piped water connections.

Thus, Tiruppur’s SLBs strive to distribute the burdens of water scarcity somewhat equally per connection while adhering to policy norms. They also try to be inclusive of informal housing residents. However, as noted, a lack of specificity in supply timings creates unreliable supply and contributes to poor service quality across the city, inconveniencing users and increasing their water access costs. The next section details how SLBs encounter and address the uncertainties impacting water supply reliability and predictability, key service quality dimensions.

Communicative Strategies Informing Water Supply Planning

Communicative strategies like relationship building, dialogues, and negotiations underlie the SLBs’ discretionary actions to devise local norms for water supply, particularly supply duration and frequency. These strategies also help the SLBs manage deviations from the local norm when supply is disrupted or delayed due to factors beyond their control, like power cuts, pipe bursts, and water leakages, which happens about 20%-25% of the time. A field encounter illustrates one such service disruption and how the SLBs used communicative strategies to address ensuing delays.

In July 2022, my interview with tap inspector Selvan was interrupted by an unexpected phone call from waterman Muthusamy. Muthusamy had called Selvan to inform him that he was stopping supply to one of the valve areas scheduled to receive supply that day. Selvan anxiously tried to discover what happened, including who would need to be informed about this sudden, unplanned stoppage. He instructed Muthusamy to meet him at a landmark in the supply area to decide next steps. Once he hung up, Selvan turned to me and explained,

See, this is how our plans get disrupted. Last night, the sewer construction contractor started digging in a new location [on the main road] without informing us. The main [water] pipe burst from all the vibrations. Muthusamy started supply but had to stop as soon as he noticed the leak […] Now, we will have to call the ward councillor so that he can inform everyone in the area. (Interviewee #TIRENG23, personal communication, 21 July 2022)

The above encounter shows how SLBs’ supply plans are vulnerable to disruption by the ad hoc actions of other urban actors. It equally illustrates how the tap inspector and watermen exercise discretion on-the-fly to stop supply, change the scheduled supply order, communicate the sudden disruption to users and ward councillors, and mitigate issues. Established communication channels with councillors through WhatsApp groups and with users through relationships built on monthly cash tips help SLBs reason with all those impacted by the disruption and rationalize supply delays. As one waterman clarified, “We can delay the supply so long as we explain it to people […] Only if supply is delayed by 2-3 days do people get anxious, start making calls, and complain. In that case, we go to each area and warn them so that they can be judicious in water consumption (and complaints)” (Interviewee #TIRENG17, personal communication, 22 November 2019). If supply was delayed beyond 2-3 days in low-income areas where many households shared water connections (which occurred rarely), tap inspectors would arrange for water trucks to be sent to these areas, and/or watermen would increase the frequency and duration of borewell water supply to alleviate water scarcity.

On their part, water users and their elected representatives also held the SLBs accountable for their discretionary actions by monitoring if they complied with locally established norms for supply duration and frequency. Whereas it is beyond this paper’s scope to elaborate on how users and politicians engaged the SLBs to demand water supply improvements and if these were effective, I observed that nearly a third of the surveyed water users (n = 110) noted that they called their area waterman whenever water was not supplied on the scheduled day, and escalated their complaints to the ward councillor or the Corporation office if the delay (or SLBs’ reasons for it) seemed unreasonable. In interviews, councillors stressed that they always “watched” (using the English term) supply frequency in their respective wards and stayed in touch with the Corporation administrators, who included assistant engineers, via ward-specific WhatsApp groups to ascertain reasons for supply delays beyond the local norm and communicate the same to their constituents. One councillor in particular noted taking to the streets to protest prolonged supply delays (those that extended beyond 2-3 days) and reflected on the unsustainability of this approach by sharing,

Suppose we gather a crowd and demand water once every 4 days from the administration [referring to the bureaucracy]. The next 2-3 supplies will be once every 4 days. After that, they will revert to their ‘once in 8 days’ schedule. The administration tires us out… (Interviewee #TIRPOL04, personal communication, 28 July 2017)

Notably, the surveyed users and councillors rarely demanded more frequent or longer supply or questioned unequal supply quantity or service quality. All they expected was that SLBs adhere to the locally established norms for supply frequency (i.e., roughly once a week) and duration (about 2 hours per supply) and communicate any deviations promptly. To ensure SLBs promptly communicated any changes or disruptions, residents voluntarily tipped watermen or gave them “festival bonuses” to strike a relationship. In this context of discretionary planning, the monthly ‘tip,’ which was equivalent to the cost of a cup of tea, created a space for communication and dialogue (Anjaria, 2011), allowing the tip giver to demand accountability from the watermen, build in more certainty around supply timings, and/or occasionally influence supply duration or timings in case of contingencies like funerals or weddings when they needed more water. However, not all residents possessed the socio-cultural capital or know-how to participate in the local tipping culture (Gupta, 2012). Out-of-state, non-Tamil migrants neither paid tips nor availed of the communication channels they created, increasing their vulnerability to unexpected disruptions or supply delays. They usually relied on non-piped, market-supplied water sources to meet their water needs.

Consequently, all the interviewed SLBs stressed communications and ‘relationship management’ as an important part of their daily job. In general, all watermen saw themselves as first responders in solving water problems and facing residents’ ire whenever supply was delayed. Often, in the middle of an interview, as if on cue, watermen would answer phone calls from residents and patiently explain reasons for delays, pacify them, and provide an indication of the next water supply to the caller’s area. After the call, they would turn to me and say, “Whenever people call, day or night, we have to answer the phone and provide ‘proper’ convincing responses when they ask about delays in water supply!” Then, we would continue with the interview.

It is worth underscoring that even if the SLBs’ communicative strategies helped improve one dimension of service quality (predictability) and manage uncertainties, they did not increase supplied water volumes, supply frequency, and duration. At best, by providing timely information, these strategies assuaged councillors and users, enabling them to develop individual- or household-level coping strategies.

Discretionary Actions and Inequalities in Everyday Water Supply

Intermittent water supply in water-scarce cities of the Global South imposes two burdens on users: insufficient water quantity and poor service quality that impact how residents experience and cope with water scarcity. This paper sought to understand how frontline planners exercise discretion in everyday supply planning to alleviate these unevenly experienced scarcity burdens. I traced how they develop distributive strategies to allocate water volumes equitably according to situated notions of deservingness, even if there is no guidance around these ‘norms’ in official water policies. They use communicative strategies involving observations, experiments, negotiations, and dialogues to deduce these locally acceptable ‘norms’ and reduce scarcity burdens that inevitably arise when they exercise discretion to address uncertainties and supply unpredictability.

Findings from Tiruppur show that discretionary planning at the street-level (or the valve area level in this case) can produce progressive, pro-poor outcomes in contexts marked by water scarcity, institutional and infrastructural deficiencies, and high levels of informality. Without clear policy prescriptions for IWS, Tiruppur’s SLBs improvised local norms for supply frequency and duration to achieve policy targets for daily per capita water volumes. Instead of simply checking off policy targets, they also tried to improve service delivery in a context-sensitive manner, aiming to serve the intended beneficiaries of piped water supply (cf. Gupta, 2012). The locally prevalent ‘norm’ for supply duration (approximately 2 hours per supply) was developed to alleviate scarcity burdens and conflicts for low-income tenants in informal housing where the practice of sharing water connections, although informal, is rampant. This finding contrasts other cases from the Global South where SLBs either succumb to political pressures (Anand, 2017) or redirect greater water flows (volumes) to higher-income areas (Alda-Vidal et al., 2018). In showing that discretion can help improve performance outcomes, the findings, thus, help temper overly cynical takes on discretion and the abuse of power in the planning literature.

The analysis of everyday water supply planning in Tiruppur also demonstrates that experimentation, dialogue, and negotiations are key to improvising policy through discretionary actions in concurrence with the planning literature (Forester et al., 2023). In other words, discretionary actions are not just shaped by the individual planner’s biases, competencies, or capacities (cf. Biggar & Siemiatycki, 2023; Lipsky, 1980), but they are equally informed by their relationships with service users and/or local politicians forged through diverse communicative practices. However, I argue that these seemingly enabling communicative practices also expose the limits to policy improvements through discretionary actions; they reflect and reinforce social differences in access to citizenship entitlements. Their communications with ward councillors shape Tiruppur SLBs’ discretionary actions to manage supply delays and (un)predictability or maintain service quality. Consequently, only Tamil ethnic residents who organise through local political society can influence and benefit from these communications (and SLBs’ actions) (Benjamin & Bhuvaneswari, 2001; Chatterjee, 2004; Heller et al., 2023). Many Tamil-speaking residents also build relationships with the SLBs, particularly watermen, by voluntarily paying monthly cash tips. The tips help them access the local state (Anjaria, 2011; Gupta, 2012) and claim better service quality, i.e., improved supply predictability and updates in case of supply disruptions or delays. Out-of-state, circular migrant workers lack the socio-cultural capital or know-how to access the local state (and improved service quality) by participating in these tipping cultures. Thus, communicative practices that help one marginalised group mitigate the uncertainties associated with discretionary practices can nonetheless be exclusionary towards other groups, even if they otherwise advance policy improvements. The ‘illicit’ nature of these monthly tips might suggest that discretionary actions encourage corruption (cf. Forsyth, 1999). However, these illicit practices open spaces for dialogue between water users and SLBs, alleviating users’ water scarcity experiences. Considering that users and SLBs alike do not consider these tips ‘illicit’ but see them as instrumental to accessing better services suggests that we need to challenge homogenising discourses of corruption and, instead, attend to situated notions of ethics in infrastructure development and urban governance when analysing plan implementation (Rankin et al., 2024).

The study also suggests that the local state, particularly SLBs in the water department and ward councillors, are relatively more accessible in Tiruppur compared to larger cities, even if it is through illicit practices like ‘tips.’ This access allows residents to discern, if not always influence, discretionary practices underlying their access to citizenship entitlements like piped water supply. It also enables the local state to exercise discretion in context-sensitive ways by observing and learning from user practices. Unlike India’s megacities, where urban governance is characterised by complex institutional settings with a bewildering mix of federal, state-level, and city-level agencies or parastatals with overlapping jurisdictions and bureaucratic fragmentation (Heller et al., 2023), smaller cities like Tiruppur have fewer agencies and actors governing water supply. These actors also represent a much smaller number of residents; Tiruppur’s ward councillors represent about 15,000 residents each compared to Bengaluru’s, who cater to an average of about 50,000 constituents (Heller et al., 2023). In future, policymakers could leverage this proximity between the state and residents in small cities to design more equitable, pro-poor service delivery by supporting more inclusive community engagement strategies in discretionary decision-making and/or policy implementation. Ongoing support from the Indian Smart Cities Program could be deployed smartly to augment SLBs’ capacity to use emerging technologies like dashboards, social media, and text alerts to reach the most vulnerable, water-stressed populations in the event of service disruptions and deviations from schedules (e.g., Kumar et al., 2018).

Finally, SLBs’ discretionary actions are circumscribed by the water department’s scope and the fragmented nature of the Indian state and urban governance (Gupta, 2012; Shatkin & Vidyarthi, 2013), where multiple actors shape the built environment and service delivery. Tiruppur SLBs seek to incrementally improve water supply quantity and quality to connections within the piped network rather than pursue structural change or question who is excluded by the network and how best to extend piped water connections to them. SLBs do not challenge the prevalence of informality in the housing market, where landlords subdivide their property to house multiple tenants without obtaining formal approvals (or the required number of water connections). They consider housing and property issues as lying beyond the water department’s scope. SLBs address informality by tweaking supply duration so tenants in informal housing can obtain a minimum amount of piped water or by informally supplementing piped water supply with untreated but easy-to-install borewell water. However, as the total amount of piped water to go around is limited, there are material limits to the SLBs’ ability to address ‘informality’ through everyday supply planning. Thus, discretionary actions by frontline utility workers in and of themselves are no silver bullets for solving water scarcity (and socio-spatial exclusions) produced by infrastructure planning.

Similarly, findings show how the actions of other actors (e.g., the sewer contractor) introduce unanticipated disruptions, provoking sudden discretionary actions in response, whose impacts need to be managed and carefully justified to users and their political representatives. The actions of actors in different public agencies might, thus, work at cross-purposes with each other instead of in coordination (Gupta, 2012), undermining the water department’s ability to leverage discretionary powers to improve service delivery outcomes. Future research can investigate how SLBs in water utilities build networks and collaborate with other frontline actors across the bureaucracy in leveraging discretionary powers to improve service delivery.

Conclusions

This study on street-level discretionary action in everyday water supply planning in Tiruppur’s IWS system reveals its pragmatic and iterative nature. I demonstrated the potential of frontline discretionary action to achieve water supply performance improvements in pro-poor ways, while exposing the limits to relying on micro-scale adjustments through everyday water supply planning to correct structural issues. SLBs seek to continually improve water distribution and service quality for network users. However, they do not question exclusionary policy assumptions underlying network designs or water scarcity. Nor do they seek to extend network coverage to excluded populations in informal housing. Although restricted to the piped network, SLBs’ frontline knowledge, discretionary practices, and their outcomes – measured as improved water quantity and service quality for the water poor – can inform higher-level water policies pursuing the ‘modern infrastructural ideal’ or continuous water supply goals. For instance, policy norms and performance-tracking benchmarks should be revised to define and operationalize concepts of sufficiency and improved service quality in IWS systems, drawing from SLB experiences. To avoid known pitfalls of discretionary action like exclusionary communicative practices, policies can provide funding, resources, and technical support to incentivize robust community engagement strategies, which extend to migrants and newcomers. Policies can also provide guidance on creating efficient and equitable schedules that equitably distribute water quantities, supply duration, timings, frequency, and reliability while providing SLBs with sufficient resources to execute such schedules. This might mean identifying neighbourhoods or valve areas with poor service quality (see Meyer et al., 2023 for examples), investigating if this deficit is linked to socio-political conditions or physical and/or bureaucratic infrastructure gaps, and prioritizing investments to close these gaps. Similarly, policy successes should not just celebrate continuous water supply cases as is common in India but also highlight progressive practices in IWS systems where everyday innovations have helped to create liveable cities for water-stressed users.

Finally, with increasing urbanization, the gap between available water, infrastructural capacity, and water demands from populations is widening, deepening the experiences of water scarcity. However, national and state-level policies in India (and elsewhere) focus on closing these gaps by investing in long-term network expansions, at the cost of overlooking frontline discretionary plans and practices that aim to bridge these gaps and reduce scarcity burdens on shorter, everyday time scales. Unfortunately, planners or policymakers seeking to address structural inequalities do not consider implementation issues or incremental change as drivers of innovation (Sanyal, 2005). This paper has argued for examining and improving everyday schedules, distribution plans, and discretionary practices to achieve water security for vulnerable populations in contexts where large networks are yet to materialize. In doing so, it reimagines what water infrastructures and supply planning might be in the present in water systems and city-regions representing the global majority.

Acknowledgements

In addition to informants in Tiruppur, the author is grateful for generous comments on ideas and drafts from Neema Kudva, Lori Leonard, David Meyer, Katharine Rankin, Jim Spencer, and Mildred Warner. She thanks the editors and anonymous reviewers for thoughtful and constructive feedback that strengthened the paper. The usual disclaimers apply.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author acknowledges generous funding support from the Mario Einaudi Center for International Studies at Cornell University in the form of an International Research Travel Award and the Einaudi-SSRC Doctoral Dissertation Proposal Development Award. She is equally thankful for financial support from a Catalyst Grant awarded by the University of Toronto’s Data Sciences Institute.

Notes on contributors

Nidhi Subramanyam

Nidhi Subramanyam is an Assistant Professor in the Department of Geography and Planning at the University of Toronto and a steering committee member of the School of Cities India. Her research investigates how planning, policies, and governance intersect with and enhance water security and environmental justice for socially marginalized communities in fast-growing regions in India and Canada.

Notes

1 In 2019, the flat-rate monthly tariff for each domestic connection was INR 120 (roughly about INR 12 to 15 per m³). Some peripheral parts of the city continued to pay an older water tariff of INR 30 each month. At this time, 1 INR = ∼0.015 USD.

2 Data from the 2013 Draft Detailed Project Report for Water Supply Improvement Scheme to Tiruppur City Municipal Corporation.