This paper addresses two approaches to guiding the inclusion of the poor and vulnerable in clean water and sanitation programmes in the framework of Sustainable Development Goals. It proposes an operational and adapted participatory approach for identifying the poorest and most vulnerable, which has become a central interest for inclusive development policies worldwide post Millennium Development Goals. Then, it proposes an inclusive water poverty index for the inclusion of the poor and vulnerable. This index, with six components, materializes the spatial and temporal equity in the distribution of water and sanitation services through a given district or municipality, particularly in developing countries.

Introduction

The issue of sustainable and equitable management of water resources is of the greatest concern worldwide, because water resources are speedily decreasing due to demographic pressure, which has engendered an increase in their utilization for several purposes. This issue is even more important when considering the current problem of climate change, for which resilience strategies are necessary. In this sense, Wilk & Jonsson (2013) have suggested that a better understanding of the poverty or prosperity of a country or a district in relation to water is imperative to targeting strategic actions for the efficient management of water resources. This approach is particularly necessary for arid and semi-arid areas, where the access to and reliability of water sources have a large influence on promoting sustainable livelihoods and where the environmental impact associated with inadequate resource management is significant (Giné & Pérez-Foguet, 2009). However, a major challenge is knowing how to relevantly determine water prosperity or poverty. In other words, what reliable indicators have to be used to capture water poverty, and how can we comprehensively compile them into a synthetic indicator that can capture the dimensions of water resource management? These questions are important because, in the framework of the Sustainable Development Goals post-2015, one of the main concerns of policy makers, development agencies and professionals of water resources is taking into account the specific needs of the poor and vulnerable in the provision of water and sanitation services, particularly in Burkina Faso (Directorate General of Water Resources (DGRE) & World Bank, 2014). This paper aims to (i) propose an operational approach to characterizing and identifying the poor and vulnerable groups because they have to be taken into account in future development programmes and (ii) propose an innovative theoretical tool to help in the planning of the provision of water and sanitation services. To this end, it is necessary to first review the concept of water poverty or prosperity and its operational measurement. Second, we focus on an operational approach for the characterization and identification of the poor and vulnerable on the basis of past experience. Lastly, a theoretically compiled indicator of water poverty is developed to conduct a water poverty mapping, which is helpful for the spatial and temporal planning of the provision of water and sanitation services.

Water poverty

The water poverty concept and index

Water poverty is a concept based on Pigou's work (Pigou, 1920) on the fundamental needs of humans. When a person is deprived of one of these fundamental needs, that person qualifies as poor. Moreover, the most recent works of Sen (1981a, 1981b, 1995, 1999) stated that poverty is a result of lacking at least one of the fundamental prerequisites for a full and accomplished life. On this basis, some authors, such as Scoones (1998) and Carney (1998), have identified the lack of water as being one of the deprivations that characterize poverty because water has an impact on the basis of human livelihoods, particularly financial, physical, natural, social and human assets. To Sullivan et al. (2003), the lack of water is equivalent to a lack of these prerequisites and can have several additional consequences. For example, the poor availability of water, in quantity and quality, can directly affect health because body and food hygiene may be less regarded. The concept of water poverty has evolved from the world scale (Meigh et al., 1999) to the local or community scale (Lawrence et al., 2003; Sullivan et al., 2003; Giné & Pérez-Foguet, 2010; Wilk & Jonsson, 2013).

A water poverty index was created by Sullivan (2002) to measure water poverty. The index was developed as a holistic tool for measuring water stress at the household and community levels, to help national, central and community decision makers, as well as development agencies, to determine the needs for interventions in the sector of water. To Sullivan & Meigh (2003), a water poverty index is an interdisciplinary tool that integrates the key issues relating to water resources, combining physical, social, economic and environmental information associated with people's ability to gain access to water and use water for productive purposes. Sullivan et al. (2003) highlighted that the water poverty index is strongly linked to the deprivation of a prerequisite. The index then measures the water status by focusing on poverty and assets that generate a livelihood for the poor. To formalize the water poverty index, several other indicators from various sources have been utilized, such as the joint programme for access to water index of the World Health Organization (WHO)/United Nations Children's Fund (UNICEF) (2000) on the state of health and education, the human development index (HDI) (United Nations Development Programme (UNDP), 2002), and environmental aspects based on the environmental sustainability index (World Economic Forum, 2001). These indicators are supposed to be indirect measures of certain aspects of water needs (Sullivan et al., 2003). On this basis, water poverty index components have been identified to measure water needs and related to assets that provide livelihoods to populations. It is a matter of water access; water quality for domestic, food and productive uses; water management capacity and environmental aspects. However, a critical analysis of these components was conducted by Sullivan et al. (2003) regarding the measurement aspect. Indeed, to measure water access, the standard measure is to estimate the distance required for an individual to reach a water point or the time spent on the site for water collection. However, experience on the ground shows that water points are often located approximately 50 metres from household living areas, particularly in urban areas, with the duration on site often being very long because those water points are insufficient compared to the population size (Sullivan et al., 2003). In addition, according to these authors, it is important to know that in certain arid countries, the dry season entails a decrease of water flows, which means that rich or influential persons in a community tend to have priority in water access. This also increases the duration of time spent by the poor and vulnerable in situ. However, the authors denounce the problem of land tenure, particularly in the outskirts or surrounding non-developed areas, justifying the fact that authorities are often reluctant to increase the water supply through the installation of water infrastructure (Sullivan et al., 2003). Regarding the water quality component, the model developed by Sullivan et al. (2003) suggests that water quality requirements must go beyond simply improved water sources, i.e. water that contains a low dose of chemicals such as fluorine, arsenic, etc. Then, the authors highlight, in relation to quality, the concept of water variability, which is a factor often ignored or omitted. To them, variability is strongly dependent on the inter-seasonal and inter-spatial variation of rainwater, with low water flows and increasing household recourse to bad quality water sources during dry periods (e.g. polluted water and river water) to meet consumption needs. In addition, non-adequate, under-financed or poorly maintained water infrastructures are perceived to be a factor of uncertain water availability. The authors also argue for not limiting water to that being used for domestic purposes, as already proposed by WHO/UNICEF (2000). Therefore, they suggest also taking into account water for food production and the other types of production, such as industrial production, which contribute to enhancing individuals' livelihoods and reducing extreme poverty.

Regarding water management components, Sullivan et al. (2003) noted the necessity for the community and central administration to have good capacity in water management. Thus, the community must have real expertise in the efficient management of water and strongly lobby for improvements in this matter. To this end, the education and income levels and the presence of water user associations or organizations may be considered indicators of managerial expertise at the community level (Sullivan et al., 2003). At the central level, the authors identified the necessity of districts, administrations and municipalities having adequate institutional capacities to suitably implement government policies and then meet local needs in a constructive manner.

Finally, the last component taken into account in the determination of the water poverty index is maintaining environmental integrity. To Sullivan et al. (2003), sustainability cannot be achieved if the improvements in water systems do not contribute to maintaining the environment, particularly because ecosystems also need water to be able to provide necessary services.

Several approaches that allow us to estimate the water poverty index have been reviewed by Sullivan (2002), the first developer of this index, such as the time analysis approach, the matrix approach, and the conventional approach of the composite index. The simple approach of time analysis considers time as a currency for the purpose of estimating water poverty. The index is then obtained by determining the necessary time per person to access a particular quantity of water fixed to 1,000 cubic metres (Sullivan, 2002). However, this simple method has some weaknesses because the obtained index is not participative, reflects only the domestic use of water and does not take into account the needs for ecosystem and industrial issues. The matrix approach combines the main characteristics of water stress and human welfare in two dimensions. This approach involves the identification of key indicators representing appropriate series of characteristics combined at a suitable scale. The conventional approach of the composite index uses components such as water availability, access to clean water, hygiene and sanitation and the time required to collect domestic water. These indicators can be estimated by using data at national and local scale (Sullivan, 2002).

Limitation of the water poverty index

With regard to the utility and scope of the application of the water poverty index (Sullivan, 2002), several concerns in the index-development process have been highlighted by both researchers and practitioners. These concerns are considered the real weaknesses of the index and include conceptual weaknesses, such as how to combine the data on the index components and its statistical properties (Giné & Pérez-Foguet, 2010). The first weakness mentioned by Giné & Pérez-Foguet (2010) is the ad hoc character in the selection of indicators that compose the initial water poverty index of Sullivan (2002). To Booysen (2002), the availability and exactitude of data on these indicators should guide the selection process. However, the author admits that the approach is acceptable only in the context of data scarcity, particularly in developing countries. To Sullivan & Meigh (2003), the better way to use this index is with national and official data in the sector.

The second weakness of the water poverty index is the pondering and aggregating techniques used to obtain the value of the index because they influence the coherence and interpretability of final values (Nardo et al., 2005). Indeed, according to Feitelson & Chenoweth (2002), the weights attributed to the index's components are submitted to personal judgement or subjective judgement. In addition, Molle & Mollinga (2003) criticized the index based on the fact that it is a compound of disparate parts that is correlated among them and is based on information with weights that are arbitrarily defined. In addition, according to Nardo et al. (2005), when a component is not performing enough because of its weight, there is a possibility that this performance weakness is compensated by sufficiently high values from other components with regard to the aggregation technique used in Sullivan's (2002) model.

The third weakness of the index is that it seems to be inadequate for assessing water poverty because the issues related to water are often the most complex issues facing humans (Komnenic et al., 2009). In addition, the fact that some indicators of the water poverty index are correlated with the gross domestic product or HDI reduces the utility of the index as a relevant tool on an international scale (Jiménez et al., 2008). Finally, Shah & van Koppen (2006) found that the ‘access component’ is relevant and easy to understand in the development of the index because hydraulic resource endowment does not have any relationship with water poverty. In a poor nation, there are no means to exploit resources.

In the current context, with the specific interest of public authorities and their partners in providing clean water, hygiene and sanitation services to the population, particularly to the poor and vulnerable in developing countries like Burkina Faso, can such a composite water poverty index be adequate in its current form? This question is very important because the concept of poverty that underlies the water poverty index can be contended (Sullivan, 2002; Lawrence et al., 2003; Sullivan et al., 2003; Giné & Pérez-Foguet, 2010, etc.) because the poor here are individuals who are deprived of water, which is a prerequisite for a full and accomplished life. Viewed from the angle of a transversal use of water to meet domestic needs (drinks, cooking, etc.), production needs (in breeding, agriculture, industries, etc.), and ecosystem needs, we agree with these authors that the water poverty index is relevant, as already mentioned. However, if we consider the context of the water and sanitation sub-sector, which aims to provide clean water and a hygienic life framework to populations, can we really classify a person or a household as poor who does not have access to these services, particularly water? This concern is justified by the fact that the concept of poverty is variable according to people's living areas and cultures through the world, particularly in Burkina Faso. Indeed, in the Participatory Assessment of Development project activities reports (PADev reports, 2008–2012) and according to Pouw & Kini (2014), access to clean water and sanitation services does not often figure among the local criteria of poverty or vulnerability. Therefore, it is necessary to find an approach that allows us to characterize and identify the poor and vulnerable from the bottom up, which is compatible with the concept of poverty in the sub-sector of water and sanitation. We should keep in mind that the targeted populations have to be included in the process because they are the key and final actors of any intervention activities in the area.

Approaches for characterization of the poor and vulnerable

Review of poverty characterization in Burkina Faso

Poverty, a polysemic concept, is becoming the subject of more attention. To better understand poverty in order to alleviate it, several approaches have been developed and used by researchers and practitioners (Sawadogo et al., 1995; Sawadogo, 1997; Lachaud, 2002a, 2002b). The money-metric approach is usually used by Breton-Woods institutions, such as the World Bank and International Monetary Fund. This approach seems to be appropriate for characterizing poverty and vulnerability through the world because of its statistical robustness. In Burkina Faso, this approach has been used several times by the national institute for statistics and demography (INSD, 1994, 1998, 2003, 2008). However, this approach remains limited in developing countries such as Burkina Faso because it does not take into account certain specificities of those countries related to social and cultural factors. The non-monetary approach to poverty is used by some institutions such as the United Nations Development Programme and takes into account several social aspects of human life. Education, health, access to clean water, etc. are included in poverty estimations in this approach. From the HDI, this approach has evolved towards the current multidimensional poverty index (MPI) developed by Alkire & Foster (2011). Other approaches, such as Household Economy Analysis (HEA) (Save the Children UK, 1995), are also used to capture and measure the concept of poverty. However, those approaches remain limited because they do not allow an operational identification of the poor and vulnerable people on the ground. In other words, people who are often considered poor through such approaches sometimes do not recognize themselves as such because the perceptions of the phenomenon vary according to their living areas and cultural values (Wetta et al., 2011). Consequently, these approaches seem to be more inefficient for identification of the poor and vulnerable on the ground. On the other hand, the approach developed by Dietz et al. (2013a), the PADev, overcomes these limitations to make an operational categorization or characterization of wealth groups at the local or community scale. In fact, a comparative analysis of the different approaches was undertaken by Pouw & Kini (2014), who revealed the following. First, based on an inclusive and holistic participatory identification of the poor in rural zones, this approach proposes a characterization of poverty on the basis of the local population’s self-perception of the phenomenon. Therefore, for example, contrary to the MPI, which seems to underestimate poverty (Pouw & Kini, 2014), the PADev approach presents the advantages of being more inclusive than the MPI and better revealing the socio-relational and psychological or subjective aspects of poverty. In addition, the money-metric poverty index underestimates poverty more than does the PADev approach according to the authors. Another advantage of PADev is that poverty criteria are identified by communities, making it possible to directly locate all people who meet these criteria for an individual interview if necessary. Finally, the PADev approach is different from HEA, which provides a detailed picture of socio-economic groups’ livelihoods per area during a reference year, called the normal one. It also makes it possible to know which household strategies are developed to get food and increase their income. Therefore, the HEA approach is directed towards the welfare of economic agents. The PADev approach draws up its source from the new theory of inclusive development because its focuses on well-being according to Pouw & McGregor (2014). Finally, because the PADev approach does not exclude any social category, it considerably reflects the inclusion of the poor and vulnerable in the process of elaboration and implementation of post-2015 development programmes. Thus, in relation to the access of the poor and vulnerable to water and sanitation services, some indicators related to social and cultural aspects should be taken into account in the design of a more adapted water poverty index because it will be more inclusive and pro-poor.

Experience in using PADev for the characterization of the poor and vulnerable in the water and sanitation sub-sector

The use of the PADev approach in characterizing the poor and vulnerable consists of organizing workshops at the local or community level with participants from all the socio-professional categories (see Dietz et al., 2013a). According to Kazimierczuk (2010), these workshops are a kind of focus group but go into more depth under the theme defining the criteria of the local perception of wealth. Based on these criteria, the participants are asked to estimate or evaluate the number or proportion of each wealth group (see PADev reports, 2008–2012 at www.padev.nl). In this sense, this approach seems to be limited in practice compared to the ordinary focus groups used when implementing participatory rural-appraisal methodologies. However, on the ground, it is an exciting exercise that gives a voice to all the socio-professional categories of people in a consensual debate relating to their living conditions, particularly in terms of their criteria for characterizing community wealth (Dietz et al., 2013b).

This exercise has been carried out in the framework of a pilot test of the characterization of the poor and vulnerable groups and how to include their needs in post-2015 water and sanitation national programmes financed by the World Bank local agency beginning in 2014 in Burkina Faso. This study tested the criteria locally identified during the PADev implementation covering the period of 2008 to 2012 in the Niabouri, Silly and To municipalities in Burkina Faso. The purpose was to use these criteria to determine whether some indicators measuring poverty and vulnerability can be drawn up from the PADev approach and at what scale these are extendable. During the PADev implementation period, several criteria describing wealth groups have been listed in Burkina Faso (see Pouw & Kini, 2014). These criteria were explored during a brainstorming workshop organized by the World Bank local agency in Leo during April 2014, and some were selected by participants because they seemed to better describe the concept of poverty and vulnerability. These participants have great expertise in water and sanitation, such as in non-governmental organizations (NGOs) intervening in this sub-sector, and local public services providing people with clean water, and sanitation directors (see the list in the report of the brainstorming workshop).

The lessons learned from the use of the PADev approach in such conditions and context are presented as follows. The approach was well accepted by community members who participated in the workshops. The focus groups were very interactive, and each participant was interested in making a contribution. The theme on the local perception of wealth groups was very much appreciated by the participants of the workshops. The poverty definition was first given in the local language, and then the criteria for identifying each wealth group in the community were given after consensual comprehension of the participants. The main constraint was the shorter time authorized to the researchers for implementing the approach not at the municipality level but at the village level. Thus, for each village, researchers had only one day to organize those village-level PADev workshops. Despite these conditions and constraints, the results from this process were very impressive. Indeed, local participants in the workshops, after agreeing with the criteria selected during the brainstorming session and adding other criteria more adapted to their living areas, identified the poor and the vulnerable people and directed the researcher team to their houses. This approach allowed the researchers to conduct in-depth interviews with the poor and vulnerable as identified by their community members. The results of these interviews were very interesting. Approximately 97% of those interviewed recognized that they are poor and vulnerable according to the local criteria of poverty and vulnerability identified. The sample of the poor and vulnerable interviewed comprised approximately 832 people and covered five rural municipalities and the outskirts municipalities in the largest cities in Burkina Faso, such as Ouagadougou and Bobo-Dioulasso. From this experience, the study concluded that the PADev approach is an appropriate methodology for characterizing and identifying the poor and vulnerable because it works from the bottom up. This approach is also more inclusive and more suitable if development interventions have to be implemented to benefit the poor and vulnerable. The authors then suggest that any use of the PADev approach must follow the advice described in the PADev Guidebook (see Dietz et al., 2013a), but the scale of implementation depends on the purpose and the financial capacities. Thus, should we implement the approach only at the village or municipality level or at the regional or national level? The choice is important here because it will greatly influence the scale of the inclusive water poverty index.

Development of the inclusive water poverty index

In accordance with the ideas of Sullivan (2002), during the development of the water poverty index, for this tool to be largely accepted, it is important that it is conceived in collaboration with those who are able to use it. Suggesting a more inclusive approach (Sullivan, 2002), the current inclusive water poverty index is inspired by the PADev approach, which frames well with the expectations of the tool. The applicability of this tool also answers the concerns of scale, i.e. local or community level, district or regional level and national level, as suggested by Gibson et al. (2000), who found that enlarging or reducing the scale of the application of the tool may engender serious errors. This concern is real because the water environment is naturally heterogeneous and the physical availability of water can be very variable between two spaces separated by short distances (Sullivan et al., 2003). In relation to the question of scale, the PADev approach, on which our inclusive water poverty index is based, is really relevant because the approach can be perfectly adapted to village- or municipality-level analysis. In addition, in the current context of integral communalization in Burkina Faso and in several developing countries, the inclusive water poverty index is a suitable tool for monitoring and evaluating national water and sanitation policy implementation at the municipality level.

To formalize the inclusive water poverty index, we built on the works of Giné & Pérez-Foguet (2010), which is our analytical reference. Indeed, these authors, by developing a water poverty index at the local scale through the improved method, provide important data that one can gather locally. By highlighting the weaknesses of the water poverty index of Sullivan (2002) and its multiple applications in Sullivan et al. (2003) and so forth (see Giné & Pérez-Foguet, 2010), the authors of this study developed a water poverty index that has robust statistical properties and has statistically adequate aggregation techniques, particularly for its components. For example, the authors highlight that the weighting technique of the index components is performed not on a statistical basis but on the basis of experts’ opinions. In the same way, they find that the additive technique of the index component aggregation is not statistically adequate because there are possibilities of a compensation among those components, which is why they propose recourse to the multiplicative function as a technique for aggregating water poverty index components, only under the condition that official and credible data are available. Therefore, the authors used the basic indicators of the water poverty index of Sullivan (2002), which take into account factors that show the major concerns of developing countries in the matter of water and sanitation, including the physical availability of water resources (R), the extent of access to water and sanitation (A), the capacity of populations to sustainably manage the available water (C), the utilization of water for various purposes (U), and the environmental factors that have an impact on the water supplied for ecosystems (E). When a value is given to each indicator, based on available official data, the water poverty index can be calculated through the multiplicative function (Giné & Pérez-Foguet, 2010). To Salameh (2009), the water poverty index, as defined by Sullivan and used by several researchers already cited, does not take into account some very important aspects or factors that significantly influence the availability of water resources. The author identified the contribution of rain-fed agriculture and the various climatic conditions in arid and semi-arid areas in the index. However, the water poverty index developed by Salameh is different from that of Sullivan (2002) in several aspects. Salameh's water poverty index considers only the physical availability of water resources under various climatic conditions in arid and semi-arid areas. Salameh then built a ratio by dividing the amount of water of each climatic area by its population size. In contrast, Sullivan developed an interdisciplinary water poverty index with multidimensional aspects. Consequently, for this paper, the latter approach is found to be more suitable. However, during the data gathering, the diversity in climatic conditions in each area needs to be taken into account, as highlighted by Salameh (2009).

To conform with the concern of the inclusion of the poor and vulnerable (Pouw & Gupta, 2017) in water and sanitation services-providing programmes, we highlight that certain indicators or criteria of poverty and vulnerability are not taken into account in any of the existing water poverty indexes (Sullivan, 2002; Lawrence et al., 2003; Sullivan & Meigh, 2003; Sullivan et al., 2003; Cohen & Sullivan, 2010; Giné & Pérez-Foguet, 2010; Ty et al., 2010; Wilk & Jonsson, 2013). Indeed, Pouw & Kini (2014) have shown that the PADev approach highlights that psychological considerations, assets in social relationships, and cultural assets are some factors inextricably related to poverty and vulnerability at the local and community levels. For example, a person who is well known in a society and therefore has strong social relationships is not usually perceived to be poor in the community. As proof, when that person is facing a problem that requires the paying out of money, the mobilization of socially related people to solve this problem locally or at the community level is considered a good or service that cannot be estimated. In addition, when a vulnerable person, such as a non-active and poor older person without any assistance (see PADev reports, 2008–2012; Pouw & Kini, 2014; DGRE & World Bank, 2014), receives clean water from a member of their living community because of cultural values, which create solidarity with the poorest people, one can consider this local or endogenous system of solidarity to be an important indicator for capturing the inclusion of the poor and vulnerable in management system water resources. Finally, rich persons or households who possess their own clean water and sanitation infrastructures and then allow the poorest to use them are an indicator of a cultural value that can be taken into account in the process of the inclusion of the poor and vulnerable.

Consequently, to complete the components of the inclusive water poverty index, we propose a sixth component entitled ‘Cohesion’ (Co) in the water area. The concept of cohesion involves the ‘force’ that allows a community to remain united regardless of the divergences of views and differences in socio-economic and professional categories. The first indicator of this component is the degree of solidarity within the community, measuring the extent to which the members of local communities are ready to help each other when they are facing a specific problem. The second indicator is the degree of solidarity towards poor and deprived persons. It captures the attention that the non-poor members of local communities pay vis-à-vis the poor and vulnerable. The third indicator is represented by the presence of local funds destined to compensate the loss of profit of water professionals from giving access to clean water services and sanitation to the poorest people. Finally, the fourth indicator is about the representativeness of the poor and vulnerable within the committees that manage the social funds destined to compensate the substantial poverty of the poor and vulnerable. Table 1 presents the components and their indicators of the inclusive water poverty index.

Table 1.

Components and operational indicators of the inclusive water poverty index (IWPI).

Components Indicators Definition Estimation framework or scale 
Resources (RWater quantity sufficiency* Degree in meeting water needs Always sufficient For human and livestock Only for human Not sufficient for human 
Reliability of supply* Time water supply system is non-operational <5% 5–10% 10–25% >25% 
Variability of water resources* Number of months during which water is available 12 9–11 7–8 <7 
Access (AAccess to clean water Proportion of people having sustainable access to clean water (%)  
Access to improved hygiene and sanitation Proportion of people having sustainable access to increased hygiene and sanitation (%)  
Distance from clean water source (m)* Distance from habitat grouping to clean water sources <250 250–500 500–1,000 >1,000 
Waiting time (min)* Time spent in situ before getting water <30 30–60 60–120 >120 
Cost of water (FCFA/20 l)* Cost per 20 l <10 10–20 20–50 >50 
Functionality of clean water sources Clean water sources functioning rate (%)  
Capacity (CManagement system: the ease in managing water sources at local level Proportion of water facilities managed at local level (%)  
Ownership over water source Proportion of water sources owned (%)  
Facilities managed by water users associations Proportion of water facilities managed by water users associations (%)  
Records kept Proportion of water entities1 with records (%)  
Financial control Proportion of water entities possessing a system of financial control (%)  
Funds audited Proportion of water entities regularly audited (%)  
Utilization (UDomestic water consumption rate (per capita)* Level of domestic water consumption per person (litres per day – lpd) > 20 l 10–20 l 5–10 l <5 l 
Conflict over water sources (human-human) Proportion of water sources in human-human conflict (%)  
Conflict over water sources (human-livestock) Proportion of water sources in human-livestock conflict (%)  
Use of local water treatment (boil water) Proportion of households boiling water before drinking (%)  
Livestock water use (m3 per day)* Quantity of water destined for animal consumption (m3 per day) <50 50–100 100–200 >200 
Environment (EAssessment of the quality of water* Protected sources Protected source Non-protected source but treated Non-protected source but locally treated Non-protected source and non-treated 
Protection of water sources Proportion of facilities protected (%)  
Pollution of water sources (PS)* Number of pollution sources around water point None 1 PS 2 PS >2 PS 
Environmental impact (EI)* Number of environmental impacts around water point None 1 EI 2 EI >2 EI 
Conflict over water sources (human-wildlife) Proportion of water sources in conflict (%)  
Cohesion (Co) Global solidarity* Degree of solidarity at the communities scale Good Acceptable Weak Rare 
Particular solidarity* Degree of solidarity towards the poor and vulnerable Good Acceptable Weak Rare 
Existence of social funds for solidarity* Opinion on the solidarity funds Formal and functional Informal and functional Informal and non-functional None 
Degree of representativeness of the poor and vulnerable within the management of the social funds Proportion of the poor and vulnerable members in the solidarity funds management committees (%)  
Note: *the values of these indicators are estimated according to the following scale: Good (1) Acceptable (0.66) Weak (0.33) Risky (0) 
Components Indicators Definition Estimation framework or scale 
Resources (RWater quantity sufficiency* Degree in meeting water needs Always sufficient For human and livestock Only for human Not sufficient for human 
Reliability of supply* Time water supply system is non-operational <5% 5–10% 10–25% >25% 
Variability of water resources* Number of months during which water is available 12 9–11 7–8 <7 
Access (AAccess to clean water Proportion of people having sustainable access to clean water (%)  
Access to improved hygiene and sanitation Proportion of people having sustainable access to increased hygiene and sanitation (%)  
Distance from clean water source (m)* Distance from habitat grouping to clean water sources <250 250–500 500–1,000 >1,000 
Waiting time (min)* Time spent in situ before getting water <30 30–60 60–120 >120 
Cost of water (FCFA/20 l)* Cost per 20 l <10 10–20 20–50 >50 
Functionality of clean water sources Clean water sources functioning rate (%)  
Capacity (CManagement system: the ease in managing water sources at local level Proportion of water facilities managed at local level (%)  
Ownership over water source Proportion of water sources owned (%)  
Facilities managed by water users associations Proportion of water facilities managed by water users associations (%)  
Records kept Proportion of water entities1 with records (%)  
Financial control Proportion of water entities possessing a system of financial control (%)  
Funds audited Proportion of water entities regularly audited (%)  
Utilization (UDomestic water consumption rate (per capita)* Level of domestic water consumption per person (litres per day – lpd) > 20 l 10–20 l 5–10 l <5 l 
Conflict over water sources (human-human) Proportion of water sources in human-human conflict (%)  
Conflict over water sources (human-livestock) Proportion of water sources in human-livestock conflict (%)  
Use of local water treatment (boil water) Proportion of households boiling water before drinking (%)  
Livestock water use (m3 per day)* Quantity of water destined for animal consumption (m3 per day) <50 50–100 100–200 >200 
Environment (EAssessment of the quality of water* Protected sources Protected source Non-protected source but treated Non-protected source but locally treated Non-protected source and non-treated 
Protection of water sources Proportion of facilities protected (%)  
Pollution of water sources (PS)* Number of pollution sources around water point None 1 PS 2 PS >2 PS 
Environmental impact (EI)* Number of environmental impacts around water point None 1 EI 2 EI >2 EI 
Conflict over water sources (human-wildlife) Proportion of water sources in conflict (%)  
Cohesion (Co) Global solidarity* Degree of solidarity at the communities scale Good Acceptable Weak Rare 
Particular solidarity* Degree of solidarity towards the poor and vulnerable Good Acceptable Weak Rare 
Existence of social funds for solidarity* Opinion on the solidarity funds Formal and functional Informal and functional Informal and non-functional None 
Degree of representativeness of the poor and vulnerable within the management of the social funds Proportion of the poor and vulnerable members in the solidarity funds management committees (%)  
Note: *the values of these indicators are estimated according to the following scale: Good (1) Acceptable (0.66) Weak (0.33) Risky (0) 

Source: Constructed by the author and inspired by the works of Giné & Pérez-Foguet (2010) and the clean water and sanitation Lexical of the ministry in charge of the sub-sector of Burkina Faso. FCFA = African Financial Community Franc.

1The water entities are hereby: water users associations, water delegated in Ouagadougou, water source ownership.

Once these indicators are identified, their values will be gathered during PADev workshops to be organized at the village or municipality level. In developing countries like Burkina Faso, there is a great lack of official and credible data to feed the model developed by Giné & Pérez-Foguet (2010). Such workshops can also be areas for gathering official data on the health and education in these communities. These indicators must be normalized as suggested by Giné & Pérez-Foguet (2010) to guarantee the uniqueness of their measurement unit. Therefore, the inclusive water poverty index will be calculated by applying the following theoretical equation: 
formula
The developed form of the index is as follows: 
formula
where is the value of the inclusive water poverty index according to the multiplicative function.

Xi refers to the i component of the index and takes the respective values of R, A, C, U, E and Co, where R = Resources, A = Access, C = Capacity of management, U = Utilization, E = Environment, Co = Cohesion (see Table 1), and w is the weight attributed to each component of the index. The weight of the component is in reference to the opinions of experts in the sub-sector of water and sanitation in accordance with the basic ideas of Sullivan (2002). This approach is necessary because, in developing countries like Burkina Faso, there is a great lack of data in the sector for sophisticated statistical analyses as performed by Giné & Pérez-Foguet (2010). The weight of each component is between 0 and 1, and the sum of those weights equals 1 .

In Burkina Faso, this theoretical model was made operational during the national workshop of validation of the pilot study for the poor and vulnerable. This workshop was organized by the local agency of the World Bank and Directorate General of water resources (DGRE). There were approximately 65 experts in the water and sanitation sub-sector, including participants in the public sector (district directors in water and sanitation, the national office of water and sanitation), the private sector, NGOs such as Water Aid, the German cooperation programme in water and sanitation sector, etc. From this workshop, the weights of the indicators of the inclusive water poverty index were estimated in a consensual way by the participants. Table 2 gives the different weights.

Table 2.

Weights of IWPI components.

Components Weights 
Resources (R0.30 
Access (A0.30 
Capacity (C0.15 
Utilization (U0.10 
Environment (E0.05 
Cohesion (Co0.01 
Components Weights 
Resources (R0.30 
Access (A0.30 
Capacity (C0.15 
Utilization (U0.10 
Environment (E0.05 
Cohesion (Co0.01 

Source: National workshop of validation, Léo, 24 October 2014.

Given these data, the operational inclusive water poverty index becomes the following: 
formula

The various values from the inclusive water poverty index calculations allow us to map the state of water and sanitation poverty in each municipality in Burkina Faso and several other developing countries. The mapping can be extended to the district and national levels because each of these is formed by several municipalities and villages. This mapping is intended to help in the spatial and temporal planning of intervention actions in a system that provides populations, including the poor and vulnerable, with clean water and sanitation services. This index also has a large scope because its indicators are a logical framework in the monitoring and evaluation of water and sanitation programmes at national, regional, municipality and village levels. For that, the index provides a permanent framework on data gathering for water and sanitation programmes post Millennium Development Goals (post-MDG).

Conclusion

This paper aimed to propose a new approach that is able to take into account the poor and vulnerable groups in water management in Burkina Faso and in other developing countries. It achieved two fundamental results. First, the PADev was a necessary approach for characterizing and identifying the poor and vulnerable to better include them in post-MDG water and sanitation development programmes. This choice was justified on the basis of experiences during the approach-testing period and during the pilot study ordered by the local agency of the World Bank Water and Sanitation programme to test the PADev approach and its applicability at various scales, such as urban and rural areas. The second result obtained is that a more holistic approach is available, the inclusive water poverty index. Based on the former water poverty index, the current index is innovative because it takes into account an aspect that was not considered earlier and uses the PADev approach, which considers some social, psychological and cultural aspects of a community. Indeed, the inclusive water poverty index considers social cohesion to be the sixth component, in contrast to the former index developed by Sullivan, which has been largely applied since 2002. This index is certainly of interest because it allows us to better perceive the equity in the distribution of clean water and sanitation services within a municipality, a district and the countryside. To that end, the inclusive water poverty index is a tool that helps in spatial and temporal planning related to the provision of water and sanitation services to populations. In addition, the index can be materialized through a map from the village level to the national level.

Acknowledgements

The author would like to greatly acknowledge the World Bank Agency in Burkina Faso for providing the opportunity to test the PADev approach as a method for identifying and characterizing the poor and vulnerable in 2014. It is during this period that the inclusive water poverty index was developed. Many thanks to Dr Claude Wetta and Dr Denis Akouwerabou for involving the author in this research. Janvier Kini is an environmental economist and researcher at the Health Science Research Institute (IRSS) in Burkina Faso.

1

The water entities are hereby: water users associations, water delegated in Ouagadougou, water source ownership

References

References
Alkire
S.
Foster
J.
(
2011
).
Counting and multidimensional poverty measurement
.
Journal of Public Economics
95
(
7–8
),
476
487
.
Booysen
F.
(
2002
).
An overview and evaluation of composite indices of development
.
Social Indicators Research
59
,
115
151
.
Carney
D.
(
1998
).
Sustainable Rural Livelihoods: What Contribution Can We Make?
DFID
,
London
.
Cohen
A.
Sullivan
C.
(
2010
).
Water and Poverty in Rural China: Developing an Instrument to Assess the Multiple Dimensions of Water and Poverty
.
Ecological Economics
69
(
5
),
999
1009
.
DGRE & World Bank
(
2014
).
Guidelines document for characterization of the poor and vulnerable groups and the covering of their specific needs in water, sanitation and hygiene. World Bank Water and Sanitation Program, Burkina Faso. Consultancy Report, November 2014
.
Dietz
A. J.
Bymolt
R.
Bélemvire
A.
van der Geest
K.
de Groot
D.
Millar
D.
Obeng
F.
Pouw
N.
Rijneveld
W.
Zaal
F.
(
2013a
).
PADev Guidebook
.
African Studies Centre, Leiden University
.
Dietz
A. J. T.
Bymolt
R.
Bélemvire
A.
van der Geest
K.
de Groot
D.
Millar
D.
Obeng
F.
Pouw
N. R. M.
Rijneveld
W.
Zaal
F.
(
2013b
).
The PADev story. Available at: www.padev.nl
(accessed 6 March 2017)
.
Feitelson
E.
Chenoweth
J.
(
2002
).
Water poverty: towards a meaningful indicator
.
Water Policy
4
,
263
281
.
Gibson
C. C.
Ostrom
E.
Ahn
T. K.
(
2000
).
The concept of scale and the human dimensions of global change: a survey
.
Ecological Economics
32
(
2
),
217
240
.
Giné
R.
Pérez-Foguet
A.
(
2009
).
Enhancing sector data management to target the water poor. Water, sanitation and hygiene: sustainable development and multisectoral approaches
. In:
Proceedings of the 34th WEDC International Conference
,
United Nations Conference Centre
,
Addis Ababa
,
Ethiopia
,
18–22 May 2009
, pp.
230
239
.
Giné
R.
Pérez-Foguet
A.
(
2010
).
Improved methods to calculate water poverty index at the local scale
.
Journal of Environmental Engineering
136
(
11
),
1287
1298
.
INSD (National Institute for Statistics and Demography)
(
1994
).
Burkinabe Survey on Households' Living Conditions (EBCVM) based on the Unified Questionnaire of Welfare indicators. Final report
.
INSD (National Institute for Statistics and Demography)
(
1998
).
Burkinabe Survey on Households' Living Conditions (EBCVM) based on the Unified Questionnaire of Welfare indicators. Final report
.
INSD (National Institute for Statistics and Demography)
(
2003
).
Burkinabe Survey on Households' Living Conditions (EBCVM) based on the Unified Questionnaire of Welfare indicators. Final report
.
INSD (National Institute for Statistics and Demography)
(
2008
).
Burkinabe Survey on Households' Living Conditions (EBCVM) based on the Unified Questionnaire of Welfare indicators. Final report
.
Jiménez
A.
Molinero
J.
Pérez-Foguet
A.
(
2008
).
Monitoring water poverty: a vision from development practitioners
. In:
Proc., Water Ethics, 3rd Botin Foundation Water Workshop
,
Taylor and Francis
,
London
.
Kazimierczuk
A.
(
2010
).
Participatory Poverty Assessment and Participatory Evaluation of the Impact of Development. Projects on Wealth Categories in Northern Ghana
.
MA Thesis
,
University of Amsterdam
.
Komnenic
V.
Ahlers
R.
van der Zaag
P.
(
2009
).
Assessing the usefulness of the water poverty index by applying it to a special case: can one be water poor with high levels of access?
Physics and Chemistry of the Earth, Parts A/B/C
34
(
4–5
),
219
224
.
Lachaud
J. P.
(
2002a
).
Pauvreté monétaire et privations de capacités au Burkina Faso (Money-Metric Poverty and Capacities Deprivation in Burkina Faso)
.
Research Series nr.6
,
Centre of Development Economy, University Montesquieu, Bordeaux IV
.
Lachaud
J. P.
(
2002b
).
Households Expenditures, Human Development and Poverty in Burkina Faso: Substitution or Complementarity? Working Paper nr.49
,
Centre of Development Economy, University Montesquieu, Bordeaux IV
.
Lawrence
P.
Meigh
J.
Sullivan
C.
(
2003
).
The Water Poverty Index: An international Comparison
.
Keele Economic Research Papers 2002/19
,
Keele University
,
Staffordshire
,
UK
.
Meigh
J. R.
McKenzie
A. A.
Sene
K. J.
(
1999
).
A grid-based approach to water scarcity estimates for eastern and southern Africa
.
Water Resources Management
13
,
85
115
.
Molle
F.
Mollinga
P.
(
2003
).
Water poverty indicators: conceptual problems and policy issues
.
Water Policy
5
(
5
),
529
544
.
Nardo
M.
Saisana
M.
Saltelli
A.
Tarantola
S.
Hoffman
A.
Giovannini
E.
(
2005
).
Handbook on Constructing Composite Indicators: Methodology and User Guide
.
OECD Statistics Working Paper
,
OECD, Statistics Directorate
,
Paris
.
Participatory Assessment of Development (PADev)
.
Activities reports in Burkina Faso and Ghana from 2008 to 2012. Available at: www.padev.nl
(accessed 6 March 2017)
.
Pigou
A. C.
(
1920
).
The Economics of Welfare
.
Clarendon Press
,
Oxford
.
Pouw
N.
Gupta
J.
(
2017
).
Inclusive development: a multidisciplinary approach
.
COSUST (Current Opinion in Environmental Sustainability)
628
,
1
5
.
Pouw
N. R. M.
Kini
J.
(
2014
).
Missing the Very Poor with Development Interventions: Results from the PADev Methodology in Burkina Faso. ASC InfoSheets nr.19, 2014
.
www.ascleiden.nl (accessed 6 March 2017)
.
Pouw
N. R. M.
McGregor
A.
(
2014
).
An Economics of Wellbeing: What Would Economics Look Like if it were Focused on Human Wellbeing? IDS Working Paper 436
.
Salameh
E.
(
2009
).
Redefining the water poverty index
.
Water International
25
(
3
),
469
473
.
DOI:10.1080/02508060008686855
Sawadogo
K.
(
1997
).
Poverty in Burkina Faso: A Critical Analysis of Local Interventions Policies and Strategies
.
ECDPM Working Paper nr. 51
.
Sawadogo
K.
Ouédraogo
J. B.
Thiombiano
T.
(
1995
).
Poverty Profile in Burkina Faso: A Qualitative and Quantitative Approach. Report for the World Bank
.
University of Ouagadougou
.
July 2005
.
Scoones
I.
(
1998
).
Sustainable Rural Livelihoods: A Framework for Analysis. IDS Working Paper No. 72
.
Institute of Development Studies, University of Sussex
,
UK
.
Sen
A. K.
(
1981a
).
Commodities and Capabilities
.
North-Holland
,
Amsterdam
.
Sen
A. K.
(
1981b
).
Poor Relatively Speaking. Oxford Economic Papers
, p.
35
.
Sen
A. K
. (
1995
).
Mortality as an Indicator of Economic Success and Failure. Discussion Paper 66
,
London School of Economics and Political Science
,
London
.
Sen
A. K.
(
1999
).
Development as Freedom
.
Clarendon Press
,
Oxford
.
Shah
T.
van Koppen
B.
(
2006
).
Is India Ripe for Integrated Water Resources Management? Fitting Water Policy to National Development Context
.
Economic and Political Weekly
41
(
31
),
3413
3421
.
Sullivan
C. A.
(
2002
).
Calculating A Water Poverty Index
.
Centre for Ecology and Hydrology
,
Wallingford
,
UK
.
Sullivan
C.
Meigh
J.
(
2003
).
Considering the water poverty index in the context of poverty alleviation
.
Water Policy
5
,
513
528
.
Sullivan
C. A.
Meigh
J. R.
Giacomello
A. M.
Fediw
T.
Lawrence
P.
Samad
M.
Mlote
S.
Hutton
C.
Allan
J. A.
Schulze
R. E.
Dlamini
D. J. M.
Cosgrove
W.
Delli Priscoli
J.
Gleick
P.
Smout
I.
Cobbing
J.
Calow
R.
Hunt
C.
Hussain
A.
Acreman
M. C.
King
J.
Malomo
S.
Tate
E. L.
O'Regan
D.
Milnerand
S.
Steyl
I.
(
2003
).
The water poverty index: development and application at the community scale
.
Natural Resources Forum
27
,
189
199
.
Ty
T. V.
Sunada
K.
Ichikawa
Y.
Oishi
S.
(
2010
).
Evaluation of the state of water resources using modified water poverty index: a case study in the Srepok river basin, Vietnam-Cambodia
.
International Journal of River Basin Management
8
(
3–4
),
305
317
.
United Nations Development Programme (UNDP)
(
2002
).
Human Development Report 2002
.
Oxford University Press
,
New York
.
Wetta
C.
Samuel
T.
Kibora
L.
Nikiema
A.
Kondé
M.
(
2011
).
Pauvreté chronique et Transitoire au Burkina Faso: une analyse de la dynamique à partir de données agricoles? (Chronic and Transitory Poverty in Burkina Faso: An Analysis of the Dynamics from Agricultural Data)
.
A Report of the Chronic Poverty Research Project
,
University of Ouaga 2
,
Burkina Faso
.
World Economic Forum, Yale Center for Environmental Law/Policy and Center for International Earth Science Information Network, Columbia University
(
2001
).
Environmental Sustainability Index
.
World Health Organization (WHO) & United Nations Children's Fund (UNICEF/WHO)
(
2000
).
Global Water Supply and Sanitation Assessment 2000 Reports
.
WHO, Geneva and UNICEF
,
New York
.