The current lack of knowledge about small towns, the great diversity of such settlements and the pressure to ‘scale up’ interventions make it difficult for policy makers and practitioners to develop models which are suitable for these towns. Currently, principles and practices informing models for water services in urban and rural areas are applied in small towns without question. This paper highlights how these principles that are engrained in the sector, may be pernicious for expanding water services in small towns as the realities of these towns may not be conducive for models incorporating these principles. Often these models are adapted to the realities of small towns to benefit water services provision. However, these adaptations are rarely documented. The little documentation of these adaptations is a consequence, we argue, of two main factors. First, such adaptations question the suitability of the models present in the sector and alter the principles on which these models are based. Secondly, the great range of adaptations linked to the diversity of small towns. Taking such adaptations seriously then limits the replicability of models, making them less suitable for scaling up interventions. As a result, certain dogmas underlying water services models continue to be reproduced.
The Water services sector has long been characterized by a strong urban-rural dichotomy. The foundation for the urban-rural distinction lies in the belief that ‘the urban’ and ‘the rural’ require different service modalities in terms of technology used for supplying services, the organization and management of service provision and the financing regime for providing services. Underlying this distinction is that the urban and rural water services markets are perceived as being substantially different. The urban market is generally characterized as high population density and a relatively high level of income of consumers who demand a relatively high level of service. The rural market concerns an area with low population density and lower levels of economic development. Whereas urban water services would be provided through a structured network with in-house connections, rural services would be provided through either a simple network or point sources. The different types of service infrastructures also demand different capacities of the provider. In the case of urban services, an autonomous service provider operates the system. In the case of rural services community management was promoted as the main management model (Moriarty et al., 2013). Over the past decades the urban water services sector has received considerable attention, particularly as a result of the heated privatization debates (Schwartz, 2008; Tutusaus et al., 2018). At the same time, a distinct group of academics and practitioners focused on the rural water sector and thoroughly documented the history and development of current approaches to rural water supply (Moriarty et al., 2013).
1.1. The bias towards large urban centres
Although urban water services have been subject to study over the past decades, the academics and practitioners studying the urban water services sector have largely limited themselves to studying large urban centres, whilst neglecting small urban centres. This bias towards large cities has a long tradition. Under colonial rule, ‘small cities were only encouraged to expand if they were convenient administrative centers through which colonial governments could extend political control over the hinterlands or if they could serve as transfer points in transportation systems designed to exploit raw materials and agricultural resources’ (Rondinelli, 1983: 43). Following this logic, development policies focused on the expansion of regions with exploitable resources and favored large urban centers, whilst largely ignoring small towns (Rondinelli, 1983; Owusu, 2005). Following Independence, the uneven development of basic water services continued as a result of the way in which governments tend to allocate resources. Gugler (as cited in Bakker, 2003: 333) highlights that governments reallocate ‘resources with three goals: improving the immediate environment (of decisionmakers); assuring the continued collaboration of the middle class; and placating strategically elements of labour’. The result is that these ‘public resources are disproportionately spent on the privileged consumption of the few and conspicuous investment for the few in cities’.
As a result, residents of smaller cities1 suffer from lower levels of access to basic public services such as piped water, waste disposal, electricity and schools in comparison to residents of larger urban centers (Cohen, 2006; Ferre et al., 2012)2. In fact, ‘small urban centers in many rural regions lack the infrastructure and facilities as well as transportation and communications linkages needed to promote regional growth, social transformation and spatial integration’ (Rondinelli, 1983: 379). There are specific characteristics of small towns that seem recurrent in explaining the lack of progress in these settlements. First, the cost of providing services to a widely scattered population is relatively high. Generally, small towns are characterized by relatively low population densities which hamper the realization of economies of density. As a result, centralized systems may not represent a viable service modality as the costs of such infrastructure cannot be recovered from the revenue that the provider can obtain from the users (Rondinelli, 1983; Adank & Tuffuor, 2013). Second, there is a presumed mismatch between the technical and financial capacity available at the local level and the requirements demanded by increasingly complex water supply systems (Njiru & Samson, 2002; Mara & Alabaster, 2008). Last, the dynamic and relatively fast growth of small towns often requires the implementation of more flexible structures rather than fixed designs of treatment plants and distribution networks. Lack of information on current developments in small towns, as well as poor planning strategies do not facilitate the effective development of infrastructure (WaterAid, 2010; Moriarty et al., 2013; World Bank, 2017a, 2017b).
The urban bias is also reflected in the academic literature on water services. Much academic literature dealing with water provision in urban areas mainly depicts certain types of cities. The cities that absorb most of the attention of academics are the large economic and political centers of a country. In case other, relatively smaller, cities or towns appear in literature they are often studied in relative terms to those larger economic centers (Bunnel & Mariganti, 2010). Moreover, the literature on water services in small towns that does exist ‘shows a striking degree of unanimity, indeed it seems that […] there is a tendency for multiple authors and institutions to draw and recycle a limited number of primary sources’ (Ryan & Adank, 2010: 4). Illustrative of the lack of recognition of small towns are the reductionist, yet broad, definitions used to describe such settlements. According to such definitions a small town is ‘a settlement that is sufficiently large and dense to benefit from the economies of scale offered by piped systems, but too small and dispersed to be efficiently managed by a conventional urban water utility’ (WSP, 2002; Pilgrim et al., 2004). What these definitions do is merely highlight that small towns' water supply systems exceed the requirements of a rural approach (WaterAid, 2010), whilst also do not enjoy the scale and densities required to make a typical urban approach economically viable (Pilgrim et al., 2004). In doing so, such definitions perpetuate the urban-rural dichotomy widely present in the water services sector.
The incompatibility of water services models with the realities of small towns
‘A first class water works man in a small town needs the wisdom of a Solomon, the patience of a Job, the political tack of a Jim Farley, the tenacity of a General McArthur, the diplomatic ruthlessness of a Von Papen, the strength of a Pittsburg steel worker and a doctor's degree in engineering.’
Dillery (1945: 1185)
As highlighted above, small towns are frequently described as being neither urban nor rural, whilst having characteristics of both urban and rural areas. In reality, these small urban centers will typically appear to have rural socio-economic features while requiring urban-type technology for basic services (Hopkins, 2003). The different characteristics also indicate that within small towns considerable diversity may exist as to the more rural or urbanized features of specific areas within a town. More densely populated, urban-like centres may be bordered by less densely populated, rural-like areas. The absence of a clear and common definition of small towns reflects the challenge in capturing such diversity in one definition. Settlements considered small towns may have a few thousand inhabitants in some countries or up to 100,000 in others (Otiso, 2005). As such, there is no common threshold in population numbers that clearly defines a small town. The size of a city is only based on where the city limits officially cover under its administrative boundaries, but this is ‘quite arbitrary’ and may not include large numbers of people living adjacent to the ‘proper’ city (Cohen, 2006: 66). ‘Levels of urbanization even in major urban areas seem to be on a continuum between purely urban and the purely rural and this point complicates the derivation of hard and fast definitions’ (Choguill, 1989: 268). The diversity in size and population distribution is complemented by a diversity in economic activities that characterize the towns. Whereas some towns may be characterized by agriculture-based economies, others may have more mixed economic activities or act as regional trading centers (WaterAid, 2010).
As a consequence of the little research undertaken concerning water services in small towns, practitioners and policy makers have little information or knowledge on approaches to service provision in small towns. Instead, what we know about development of water services in small towns is understood through either ‘big theories’ on urban development or through well-documented approaches to rural water services. Settlements are mostly qualified and subsequently categorized as being either urban or rural (UN-HABITAT, 2006). However, working in these two apparently distinct categories blurs the reality of those people living in those areas that do not clearly fit in such distinct categorization (Cardone & Fonseca, 2006; Satterthwaite, 2006). These urban or rural approaches lack reflection on the actual characteristics and complexities of small towns, making the application of these approaches to small urban centers problematic, if not impossible (Bell & Jayne, 2009: 684). The result is that standardized models for water service provision are implemented in these areas where perhaps other approaches are required (Choguill, 1989).
Models of water service provision
Models are essentially constituted of three elements, namely, the physical infrastructure used to provide the service, the organization that manages and operates that technology, and the financial arrangements that maintain that specific service. The importance of water service models lies in the fact that such generic approaches allows practitioners and policy makers to upscale development of water services by applying a similar model in different places (Cleaver & Franks, 2008). For such models to be easily upscaled, they need to be relatively simple and sufficiently broad. Such reductionist models allow for easier transfer. Models, which have been presented or are perceived as being very successful, may then be implemented across a wide variety of locations (Rap, 2006). Despite the increased relevance of small towns, very few new models for water services have been developed in recent years, particularly in relation to small towns. This inertia for developing new models and how these have been shaped for small towns can be traced to a series of dominant discourses and principles that have become engrained in the water services sector, largely adopted by practitioners and policy-makers alike. These discourses and principles have become so dominant that they largely dictate the design of current water service models, and with it stifle any innovation. Moreover, these principles are considered applicable to water services across the rural-urban continuum and are well established almost to the point of being unquestionable, also for small towns. The incompatibility of models based on these principles is elaborated on below by highlighting somewhat interrelated factors.
The Promotion of commercial finance
Although different assessments show considerable variation, projections concerning the amount of funds required to finance the realization of the Sustainable Development Goals (SDGs) are significant and increasing (Winpenny, 2015). Global estimates range from US$ 6.7 trillion by 2030 to US$ 22.6 trillion by 2050. A more recent study that quantifies only the recurrent expenditure (such as operations and maintenance, staffing, or electricity) suggests that the required expenditure will increase from US$ 4.2 billion per year to US$ 31.1 billion per year between 2015 and 2030, outweighing the costs of capital investments by about 1.5 times (Hutton & Varughese, 2016). At the moment the sector is only able to cover about one-third of these needs (Development Finance, 2016).
Despite an annual increase of the government budget allocation available for water services of 4.9%, over 80% of the surveyed countries report insufficient financial means to meet national targets for drinking water and sanitation in both urban and rural areas (GLAAS, 2017). Typically, the shortcomings encountered in national budgets have been filled in by Overseas Development Aid through grants and/or loans. However, aid commitments by bilateral donors and multilateral development banks have decreased in the last years (21% from 2012 to 2015, in Sub-Saharan Africa the reduction amounts to 44%) (GLAAS, 2017). Additionally, the external support that is still received is changing its target moving away from supporting governmental budgets for water supply and sanitation to supporting projects that seek to develop ‘self-reliance and sustainability’. As such, many objectives collected in donor agendas are geared to ‘supporting the sustainability of WASH services’, ‘Strengthening policy/institutions for sustainable WASH service delivery’, ‘WASH financing and leveraging private sector investments’, ‘WASH financing and mobilizing private sector investment’ or ‘innovative technological solutions’ (GLAAS, 2017). This highlights the efforts made in the sector to relieve national governments from responsibilities to directly provide basic services, shifting the attention to other parties. Moreover, the figures as currently collected obscure imbalances on how available funds are used for development of water services and highlight another shift in attention away from water and sanitation. When assessing allocations from external aid for ‘basic WASH’ services rural areas receive considerably less assistance than large urban areas (see Humphreys et al., 2018). Other development priorities such as health, refugees or humanitarian assistance are being currently prioritized ahead of water and sanitation issues (GLAAS, 2017; OECD, 2017).
Over the past decades, donors have heavily promoted commercial finance as a means to bridge the financing gap. They argue that commercial finance will instil the discipline and efficiency that is required to achieve results in the sector and that public funds or operations have not been able to deliver: ‘Increasing the level of private financing for the sector would allow service providers to borrow and invest in expanding service and improving quality, without having to wait for scarce public resources to be made available or rely on limited concessional financing’ (World Bank, 2017a, 2017b: 24). By pursuing commercial finance, the sector also reinforces the idea that the funds required for development of water services is mainly to come from revenue generation by the water provider (cost-recovery) and less so from governmental support. It suggests that governments are not able and should not be burdened with the financing development of the water supply and sanitation sector. The dominance of this paradigm, however, may particularly impact utilities operating in small towns. These utilities, which not only have difficulties attracting public funding, are also likely to face challenges in accessing commercial finance (Humphreys & Schwartz, 2018). These utilities have often limited technical and financial capacity and are challenged in achieving creditworthy ratings as demanded by commercial lenders. As such, they do not instil sufficient confidence in private investors concerning the returns that their investment may generate. Moreover, the relative market size that the utilities serve and the limited economies of density mean that it is unlikely, or at least challenging, that the utility can operate on the basis of full cost-recovery. This makes it difficult to realize the short- to medium-term results that commercial bank lending is looking for (GLAAS, 2017).
Water providers as commercial businesses
Linked to the importance given to commercial finance, is the emphasis placed on of having water providers operate as business-like entities. Models that are currently endorsed in the water services sector have been heavily inspired by the privatization decade (1993–2003). Although this decade failed to lead to a proliferation of private sector organizations active in the water services sector, it did lead to the firm anchoring of private sector practices and principles within public water utilities. Public utilities, either at their own initiative or under (donor) pressures, emphasized efficiency gains, performance management, results orientation and other approaches usually associated with the private sector. This emphasis on private sector principles is not something that is limited to water utilities servicing large urban centres.
A model for water user associations developed by Water Aid (n.d.), for its implementation in semi-rural settings, promotes a management system that is based on a business principle of cost recovery and profit-making. Rusca et al. (2015) highlight how this model for water user associations providing water services has largely been molded to mimic private sector organizations. Similarly, the need to improve efficiency and professionalism has also been ‘largely’ reported on typically rural management models such as community-based arrangements (Moriarty et al., 2013). As a result, water providers, regardless of the city, town or village in which they operate, are encouraged to operate as autonomous companies on the basis of full-cost recovery, be entrepreneurial and strive for ever more efficient utility operations through elaborate systems of performance management.
Water providers, thus, not only have to ensure reliable provisioning of a vital resource, but also have to address commercial objectives (Tutusaus et al., 2018). Sustainability is often reduced to a perspective that largely limits itself to equate sustainability at the lowest possible level, usually perceiving the need to deliver financially viable services at the level of individual systems avoiding any interference or support from external systems. The system that is able to close the circle, mainly in financial terms, is perceived as the ultimate achievement of success and will most likely be portrayed as a ‘best practice’ (see Rautanen & White, 2018).
These two trends deeply engrained in the sector are of particular significance for small towns. The relatively smaller market sizes, lower population densities and economies profiles potentially clash with the ‘as desired’ implementation of these principles. Although we acknowledge the importance that lays on the efficient management of water services and financial sustainability, as well as the burden of overwhelmed governments we do question what these pursue of these principles does to ensuring universal water provision.
Special issue on water supply in small towns
Despite the moderate attention from academia, the importance of small towns is gaining increased recognition as a result of two developments. The first development concerns the possible role of small towns in migration flows and urbanization. Development of small towns, in the form of improved basic services and other amenities, has been promoted in order to abate the impact of urbanization on large urban centres and to alleviate service provision pressures in major urban centers, whilst stimulating rural economies and eventually prompting social transformations (Otiso, 2005; Bell & Jayne, 2009; Roberts, 2016). The second development concerns the targets set by the SDGs in 2015. SDG6 requires countries to ensure universal and equitable water services by 2030. This inclusive target requires that the water services needs of small towns are considered in the expansion of sustainable and equitable water services. In this Special Issue, we aim to contribute to the study of water services in small towns by critically examining different approaches and experiences of water supply in small towns. The special issue brings together empirical testimonies of how the implementation of reductionist models and the perseverance of certain principles underlying these models in the water sector have yielded sub-optimal results.
Explaining what works
In the articles of this special issue we see how the credo that private operators are synonymous of efficiency, accelerated access, increased transparency and more sustainability trickles down to different contexts in small towns of Mozambique or Nepal. In Mozambique, the government agency in charge of expanding services to small towns insists on engaging in contracts with local private operators. Despite the fact the local private sector is mostly inexistent in the country, the governmental agency in charge makes it part of their mandate to expand and capacitate these yet to be developed private sector in order to develop small-scale private operators able to operate in the context of small towns. The delegation of water operations to small-scale private operators does, however, not automatically lead to successful private operators. Rather it asks for tailored efforts from governmental agencies and for a significant reinterpretation of the contractual relationship between public (governmental agencies) and private (operators) entities (Tutusaus et al., 2018). In Nepal, only one operator in semi-rural areas in the country has managed to expand services and operate efficiently. However, the case suggests that professionalization and commercialization in themselves do not fully explain the success of service expansion in semi-rural areas of Nepal (Rautanen & White, 2018). Both cases show that explaining what works needs to go beyond the reductionist explanations engraved in the models applied to most small towns. Without necessarily doing away with the focus on efficiency, professionalization and commercial operations of small-scale providers the cases show that by themselves these factors are not enough to explain why a particular model is working. Other additional factors are required to make the model work. Rather than assume small-scale providers have certain capacities, it may be more important to document what is required and possible in these settings and which demands can be made of such operators.
The heterogeneous landscape
The need for working with models leads to a necessary simplification which requires a fair amount of assumptions made about local realities. The diversity of small towns, however, may not lend itself to such broad assumptions. Humphreys & Schwartz (2018) elaborate on the role of public finance in developing water infrastructure in small towns in Burkina Faso. While national government budget allocations exist for the expansion of water services in small towns, most of these allocated funds do not reach small towns. The little local capacity to generate funds, as well as the limited capacity to claim funds from the state budget leaves towns searching for alternative funds such as project financing and donor funds. As a result, small towns resort to accessing funds from alternative sources such as donors or private sector donations. As the ability of small towns to access such funds differs, however, different small towns have varying degrees of success in accessing such funds. What results is an uneven landscape of water infrastructure development across small towns. This finding suggests that models currently promoted for small towns may be more suitable for some types of small towns, whilst other small towns may need alternative models and approaches.
A considerable discrepancy exists between the models as they are described on paper and how they are implemented on the ground. Water services models are often adjusted to adapt to local conditions in small towns. Several papers in this issue analyse adjustments made to the policy models on paper in order to accommodate requirements and capacities in the small towns. In Adank et al. (2018) the modification of a framework for the evaluation of water service sustainability is presented. This framework had originally been designed for larger urban areas. During the translation of the framework to its implementation to small towns in Ethiopia, however, it was realized that not all parameters were applicable or reasonable to request from such settlements. In some instances, items that remained part of the eventual framework still do not provide an accurate picture of the actual service configurations in the towns. More often what seems to occur is that requirement adjustments are not collected in official documentation. What results is a dual system of actual operations and an associated level of performance as it is practiced, and a paper-based reality of how the model was initially designed and, by all official accounts, is still working. A striking example of this duality is the delegation of services to local private operators in Mozambique (Tutusaus et al., 2018). The case suggests contracting in small towns of Mozambique is much more relational in nature than the contracts and tendering procedure suggest. Similarly, Lele et al. (2018) describe how their case studies in India are small towns that officially are to source their water from surface water only. However, informally the water providers in these towns access alternative sources of water, such as groundwater, in order to facilitate the expansion of services. Moreover, Molle (2008) highlight how this sourcing of alternative sources increases reliability of service provision. This practice, however, is not formally acknowledged or officially collecting which prompts the authors to advocate a revision to the model to allow for ‘mixing’ sources of water to meet growing demand in small towns. That projects are adapted during implementation and adjusted to local conditions has been documented at large (Molle, 2008). We would like to concur with that point and furthermore add that these adjustments are not recorded for two reasons. As we have said before, the workings of models requires simplifications and acknowledging all adaptations does not help the process of scaling up (Cleaver & Franks, 2008). Furthermore, the recording of these adjustments could translate into an official acknowledgement that those principles that underlie the model contradict the local realities of small towns and may subsequently create the space to question these very founding principles.
Underlying the challenge of models for small towns and the requirement of local adaptations is the tension between the need for simple models and the complexity and diversity of small towns. On the one hand, the development of tailor-made models for each and every small town is impossible as the number of small towns is too large and the effort required to design such customized models too laborious. Policy makers and practitioners require relatively simple models which they are able to roll out over a large number of settlements if they are to achieve the desired rates of service expansion. At the same time, these settlements show such a vast diversity that any model would require some alterations to local specificities before it can be expected to function. The articles in this Special Issue highlight that in the water services sector, and in understanding water services in small towns, we are still far removed from the goal of reconciling this tension.
In this process, the lack of understanding of the needs, challenges, as well as opportunities in small towns also does not offer a very firm foundation on the basis of which to rethink models for expanding services in these type of settlements. The majority of the examples documented in this Special Issue are a result of implementing different models through trial and error. As some of the articles show, in some cases local adaptations, mostly implemented by water practitioners, have resulted in relatively successful cases of service provision. But more often, however, such adaptations failed to materialize. The limited capacity of water providers in small towns to absorb model malfunctions means that in such cases the impact on services provided will be both clearly visible and severe for consumers.
What is even more disheartening is that many adaptations that are made are not even documented or analysed. This lack of documentation has two important consequences. First, it leads to a situation where the principles and practices underlying the models currently being implemented in small towns are hardly challenged. Although these principles may be suitable in some cases, questions can be raised about the general applicability of these principles for small towns in general. Without documentation of adaptations that deviate from the established principles, it is unlikely that the general applicability of these principles will be contested. As long as those are not contested they will continue to inform and shape the models for water service provision in small towns of the future. Secondly, the adaptations provide valuable insights into making models more resilient and, thus, less likely to fail. The multiple sources of water presented in the Indian cases or the relational contracting approach argued for involving small-scale operators in Mozambique provide examples of how models are being made more flexible to allow for more generic models to be better customized to specific situations (such as water scarcity or of limited private sector capacity).
The inertia of developing models for water services more suitable for small towns has meant that these settlements find themselves at the tail end of water supply and sanitation development. Apart from the urban bias which favours spending public funds on basic services in large metropolitan areas, the neglect of being neither urban nor rural, and that very few models for development of water supply and sanitation have been developed specifically for small towns means that development of water service in such settlements lags behind. Although the SDG target to achieve universal service coverage by 2030 may provide a strong impetus to start taking the tail end of water infrastructure development seriously, much remains to be done before achieving universal service coverage in small towns is likely. In order to do that, we should start speaking of small towns as a category on their own and continue the work in elaborating further what these are and how they work.
Note that the smaller category according to this study is 100,000 inhabitants. Even though there is a mismatch in terms of population numbers in other studies used in this discussion, we could expect these trends to reflect equally disadvantageous situations in smaller urban centers.