Improving Water Access Indicators in Developing Countries: a Proposal Using Water Point Mapping Methodology A

Improving water access indicators in developing countries: a proposal using water point mapping methodology A. Jime´ nez and A. Pe´ rez-Foguet

ABSTRACT

The Millennium Development Goals set the target to “halve by 2015 the proportion of people A. Jimenez Ingenierıa Sin Fronteras-ApD. without sustainable access to safe drinking water and basic sanitation”. The current international C/Cristobal Bordiu, definition of indicators of access is insufficient for measuring this target. Furthermore, the lack of 19-21.48 D, 28006 Madrid, Spain an internationally agreed definition and measurement methodology is causing confusion and E-mail: [email protected] uncertainty regarding the figures that are disseminated worldwide. Moreover, the current A. Jimenez A. Perez-Foguet context, in which almost 70% of funds for the sector are channelled through national Universidad Politecnica de Catalunya, governments, emphasises the importance of a monitoring system for national water sectors in C/Jordi Girona 1-3, C2-206, ETSECCPB, MA3 Department, developing countries. From this, an improvement in investment efficiency is expected. The water Campus Nord UPC, 08034 Barcelona, point mapping methodology, promoted and widely developed by the international organization Spain E-mail: [email protected] WaterAid, is presented as an alternative way of defining water access indicators. The present paper describes its potential for defining new indicators and making improvements. Our research is based on the results of a pilot study carried out by Ingenierıá Sin Fronteras (Engineering without Borders, Spain) in the Same district, Tanzania. The paper concludes with a discussion of the challenges posed by the implementation of this methodology and proposes further steps for improvement. Key words | access, indicators, mapping, Millennium Development Goals, monitoring, quality

INTRODUCTION

Goal 7 of the Millennium Development Goals focuses on 2006). Simultaneously, donors are making efforts to environmental sustainability and directly addresses the improve the effectiveness of aid, as demonstrated by the issue of water supply. One of its targets, Target 10, is to Rome Declaration on Aid Harmonization of February 2003 “halve by 2015 the proportion of people without sustainable and the Paris Declaration on Aid Effectiveness of March access to safe drinking water and basic sanitation” (WSSD 2005. The European Union has adopted its own commit- 2002; UN 2003). The year 1990 has been established as the ment via the European Consensus on Development (EU baseline year. There has been an increase in interest from 2006). Improving the effectiveness of aid is based on the international donors to the water and sanitation sector. principles of ownership, alignment and harmonization. In Moreover, the United Nations has declared the decade practical terms, at least 85% of aid flows will be reported on 2005–2015 the “International Decade for Action: Water for in national budgets and will use public financial manage- Life” (UN 2004) and states that the main goal should be a ment systems (Paris Declaration). This will mean that a greater focus on water-related issues at all levels and on the large proportion of aid will be channelled through sector or general budget support and that ministries’ budgets will implementation of water-related programmes in order to increase considerably. It is expected that funds for the water achieve internationally agreed water-related goals (UN sector, channelled through national governments in fulfilment of the Millennium Development Goals. The most aid-receiving countries, will increase. According to our used and suitable indicator for Target 10 is the number of estimates, this increase could represent around 70% of total people with “access to improved” water sources (WHO/ funding for the water and sanitation sector in recipient UNICEF 2000, 2005). “Access” is usually evaluated by house- countries, which is around 20,000 million dollars a year hold surveys and includes personal interpretation about what (Jime´ nez 2006). Thus, the ability to monitor national it means and is therefore not as objective as policy provisions governments is crucial in efforts to fight water poverty and claim. “Improved” water sources is better defined (Table 1). increase access to services. Research indicates that up until However, the coverage figures from technological indicators now budget support has not significantly improved national do not provide enough information about the quality of the accountability (de Rienzo 2006): the last revision of GBS for water provided or about its use (WHO/UNICEF 2000). Tanzania (1995–2005) states that “poverty impacts remain Moreover, no information is collected regarding the sustain- uncertain for the last half decade, the most relevant period, ability of the service. because there has been no household survey since 2001” There are several ways in which “Access” can be (Lawson & Rakner 2005). interpreted. In rural Tanzania, for example, it is stated Sector budget support for water and health is usually that “the basic level of service for domestic water supply in based on annual reviews carried out jointly by donors, rural areas shall be a protected, year-round supply of 25 governments and other actors (e.g. private enterprise and litres of potable water per capita per day, through water civil society) in which performance is assessed. The main points located within 400 meters from the furthest home- problem is the lack of reliable and objective indicators for stead and serving 250 persons per outlet” (Government of carrying out this assessment. To continue with the example Tanzania 2002). However, in Mozambique this water point above, the Joint Water Sector Review in Tanzania 2006 was would serve 500 people in a radius of no more than 500 m produced without an appropriate set of indicators in place, (Government of Mozambique 1995). This highlights how the which made it impossible to accurately measure results. indicators must not only be accurately defined but also Something else that should be highlighted is that a too long standardized and internationally applied. The use of time-lag between the disbursement of funds and the equivalent indicators across different nations would lessen outcomes’ measurement should be avoided, because this confusion by facilitating comparisons of performance, situation facilitates corruption and compromises political uniform sector information collection systems, and the accountability regarding poverty reduction decisions. The avoidance of misinterpretations of definitions. importance of tracking the performance of water sectors on This paper addresses the issue of monitoring water an annual basis makes it crucial to include sector-specific poverty in developing countries, a process that must data collection routines that provide annual outputs; these involve establishing EASSY indicators: Easy to get at the are in fact implemented in other basic social sectors such as local level, Accurately defined, Standardized and inter- health. Therefore, in the short term information has to be easily available at the local level and at a reasonable cost, Table 1 | Improved and not improved water sources (WHO/UNICEF 2005) even if some aspects must necessarily be oversimplified. The inclusion of data collection routines at the lowest appro- Water supply Improved Not improved priate level would simultaneously improve the tracking of transparency and accountability at all levels, while it would Piped connection into dwelling, Unprotected well plot, or yard also raise awareness of the importance of systematic data Public tap or standpipe Unprotected spring collection at the national level. Borehole Vendor provided water The most important monitoring task in the water sector is Protected dug well Bottled water carried out at the international level by the WHO and the Protected spring Provision by tanker truck UNICEF Joint Monitoring Programme for Water Supply Rainwater River, stream, pond, or lake and Sanitation (JMP), whose main goal is to track the nationally applied, Scalable at all administrative levels, The information is displayed using digital maps.” (WaterAid Yearly updatable (Jime´ nez et al. 2007). Water service is ODI 2005). WPM’s main function is to simply and provided at different water points distributed across the objectively demonstrate how water points are distributed territory and by many different actors. This then requires within a territory; thus it serves as a valuable analysis and that indicators be integrative from the lowest level so as planning tool for decentralized governments that improves to include all the activity that takes place in a certain area efficiency and accountability. Moreover, it helps to define and allow for local and regional trends. Water point reliable indicators of access constructed from the lowest mapping is proposed as an option for establishing water geographical level with the data available. By using an access indicators, and new improvements, which include example, the following section explores the results of WPM issues of quality and sustainability, are defined within it. and the challenges it faces in compiling effective indicators Challenges to its effective implementation are discussed in of sustainable access to safe drinking water. Evidence is the conclusions. taken from field work carried out in the rural Same district, Tanzania, during the second semester of 2006. In this case study, the Standard Water Point Mapping campaign, as it has been defined, was completed with quality assessments. METHODOLOGY: WATER POINT MAPPING Portable water kits were used to test all the functional water For over a decade, a variety of water points mapping systems in the rural Same district and networks were activities have been carried out, the scope and objectives of examined at either one or two points, depending on their which have been diverse (WaterAid ODI 2005). In the size. All the individual functional water points were majority of cases, these activities have addressed the lack of analysed. The parameters that were measured include pH, accurate, reliable local data that international programmes turbidity, chlorine, electrical conductivity and concen- and local governments require to plan investments. This is tration of thermotolerant (faecal) coliforms. A total of 723 enormously important since many countries are currently water points were mapped and 138 water quality tests were going through a process of decentralization that will undertaken. The field work lasted 29 days, covering an area transfer the responsibility of resource allocation to local of 5,186 km2 where 185,169 people live in rural planners. Furthermore, the problem of inefficiency in communities. international programmes often stems from the lack of coordination with other initiatives in a particular zone (Birdsall 2004). Moreover, the MDGs tend to target those RESULTS without access. Success will not come simply by achieving a certain level of balanced investment in a region, but by Target 10 of the Millennium Development Goals advocates targeting those areas where service is below minimum an increase in “sustainable access to safe drinking water and requirements. As demonstrated by Stoupy & Sudgen (2003), basic sanitation” and covers the three aspects listed below. given enough investment, untargeted allocations due to † Access: Access can be divided into “physical access”, unreliable information at the local level can make the defined in national policies that establish maximum difference between achieving the MDGs or not. values for the distance to a water point and the number Water point mapping (WPM) can be defined as an of people served by a water point (see examples above) “exercise whereby the geographical positions of all and “socio-political access”, which includes aspects that improved water points in an area are gathered in addition influence access, such as the affordability of the service. to management, technical and demographical information. Furthermore, it calls for no discrimination on the This information is collected using GPS and a questionnaire grounds of sex, age, ethnicity, etc. located at each water point. The data is entered into a † Quality (safe): Potable water is defined by quality geographical information system and then correlated with standards, which vary between nations. Nevertheless, available demographic, administrative, and physical data. the more recent concept of safe water is not being measured directly using indicators, but indirectly, administrative structures is usually quite well documented. assuming that improved sources (Table 1) provide safe Thus, the first measuring option mentioned may be the most water. appropriate. Of course, this reduces the accuracy of the † Guarantee of service (sustainable): Sustainability is a methodology since inequity is only considered up to the broad and complex concept. Related to a water system, it administrative level, at which the population information is stresses the permanence in time of that service. There are aggregated. Moreover, the availability of defined adminis- many factors that affect sustainability and the majority trative boundaries could hinder the level of detail of our are interdependent and can be environmental and/or analysis: in the case of the Same district in Tanzania, social. They may be complicated by a political or information on population distribution is available at the economic context and require institutional arrangements village and hamlet level, but administrative boundaries are for the effective management of the service (Harvey & only defined at the ward level and thus determine the spatial Reed 2004). resolution of the analysis. With this information, the percentage of access in an area can be accurately estimated. The first indicator of access defined is Improved Commu- Defining access nity Water Point Density (ICWPD), which is equal to the number of ICWP per 1,000 inhabitants. If we continue with An Improved Community Water Point (ICWP), as defined the example of Tanzania, a certain area would have access if in the WaterAid methodology (Stoupy & Sudgen 2003), is a its density were four or more water points per inhabitant. place with some improved facilities where water is drawn The percentage of people not served in an area would be for various uses such as drinking, washing and cooking. The proportional to the lack of water points available compared types of water points considered as improved are consistent to that threshold. However, it is simple to further improve with those accepted internationally and are presented in ICWP because information on functionality for each water Table 1. As previously explained, access is normally defined point is also collected during the survey. The difference by establishing a ratio of the maximum distance and number between in-place water points and functional ones would of people served by each water point. In the case of normally amount to more than 30% and is thus an Tanzania, this ratio would be one water point for 250 important factor for consideration. Consequently, Func- people within a radius of 400 m. At this stage there are three tional Community Water Point Density (FCWPD) is used possibilities for defining this measurement: by WaterAid as the real access indicator. 1. The number of people served per water point, consider- Figure 1 shows the FCWPD for the Same district at the ing that one water point serves 250 people, regardless of end of 2006. Information is displayed by ward, with whether their households are further than 400 m from between 10,000 and 20,000 people in each. The legend the water point. represents ward access status based on a colour code: red 2. The number of people served, including families living represents the most underserved wards (less than 1 FCWP/ less than 400 m from the water point, regardless of 1000 people), while dark green represents wards with more whether the number of people is more than 250. than four FCWP/1000 people (above the official threshold 3. A case-specific approach combining both of the above for access). This demonstrates the potential for this conditions and applying the most restrictive one in each methodology to identify underserved areas and improve scenario. planning. In order to accurately assess the number of people An important point to highlight is that the percentage of served using distance as a criterion, the population the population with access to water in the Same district, distribution at the household level is required, which assessed using this methodology, is 42.74%. This is lower might be problematic in the near future for the majority of than the percentage found based on household surveys in countries involved. However, due to the concept of the the same area (51.64%) (Tanzania Ministry of Water, based periodic sociological census, population distribution in on Household Budget Census 2002). contaminated water. A total of 20 villages out of 67 had vulnerability of water services to dry or high demand quality problems in their systems. The definitions of seasons, though additional information must be con- Bacteriological Acceptable Water Point Density, defined sidered when one is dealing with pastoralist and nomad as the amount of FCWP providing water with an acceptable populations, or when there is competition for water use in concentration of faecal coliform at the time of the test the area. (Tanzanian standards), have reduced water coverage from 42.74% (when only functionality is considered) to 31.37%. Defining sustainable access to safe drinking water

Defining sustainable If we consider a single indicator that includes information on both quality and seasonality, we can define the The fact of the guarantee of service provided has up to now Bacteriological Acceptable and Year-round Functional been overlooked in the indicators. Factors affecting this Water Point Density. This indicator reduces water access aspect are numerous and interdependent. WPM provides figures in the Same district to 25.29%. Table 2 summarizes valuable information collected from questionnaires that the indicators provided by central governments, WaterAid include information on seasonality, frequency and reactivity WPM and the ISF proposal. The bold style represents the to breakdowns, the financial status of the system and access indicator used by each methodology and the third institutional arrangements in place for management column shows the results obtained in terms of access for the (Jime´ nez et al. 2007). Nevertheless, despite the information case of the Same district. In this case, the difference in available it remains difficult to measure sustainability in an coverage obtained is significant. Basic quality and season- objective and standardized fashion. It is important that ality reduces access from 42.74% to 25.29% and thus institutional arrangements and financial system status reduces adequate coverage by 40.8%. Although it can be undergo a detailed analysis for each individual case. argued that the difference between surveys and mapping Reactivity to breakdowns could be used as a proxy to (from 51.64% to 42.74%) is due to statistical error during assess the concept, but more research is needed on how to sampling, the introduction of quality and seasonality in the measure and standardize this aspect before indicators can access indicator gives a reduction of 40%. be proposed. In this first approach, ISF has analysed the seasonality of water points, reported by water users, as a precondition for sustainability. One water point is not considered DISCUSSION functional all year round if water users report a seasonality of more than one month. With this concept we can define After presenting how the information acquired through the Year-round Functional Water Point Density, which in WPM can be converted into reliable indicators, this section the Same district was 30.78%, compared to 42.74% when discusses the applicability of this methodology at a higher only functionality was considered. This approximates the scale for monitoring access at the national level.

Table 2 | Comparison among different methodologies and access indicators provided, Same District results, 2006

Methodology Indicators provided Access

Governmental Household Surveys Aggregated Access Indicator 51.64% WaterAid Water Point Mapping Improved Community Water Points Density 75.02% Functional Improved Community Water Points Density 42.74% ISF Water Point Mapping Bacteriological Acceptable Functional ICWP Density 31.37% Year round Functional ICWP Density 30.78% Bacteriological Acceptable and Year round functional ICWP Density 25.29% Can water point mapping indicators be described as administrative level with accepted legal spatial bound- EASSY? aries. In any case, the methodology has a bottom-up approach that allows simple integration from the lowest As we have argued in the introduction, indicators used for level upwards. monitoring the water sector should be EASSY (Easy to get † Standard and internationally applicable: There are no at local level, Accurately defined, Standardized and inter- internationally agreed access indicators. As explained nationally applicable, Scalable at all administrative levels, above, countries have different definitions in their Yearly updatable). In the following section the WPM policies for access (related to distances) and different indicators previously presented are analysed based on five quality standards. The WPM methodology helps define characteristics. and measure access indicators in an objective manner; † Easy to get at local level and Yearly updatable: After the these indicators can then be applied everywhere and baseline is established it is possible to update data at the allow water access situations to be effectively compared local level. New water points and updated information between countries. on existing ones (such as functionality, seasonality and management) should be reported by implementers and is

relatively simple to do. The integration of an efficient Can WPM be adopted? routine information collection system is crucial if the use of indicators is to be successful; these systems are often The information provided by WPM is more accurate and in place but are ineffective. The problem is the difficulty easier to present than the indicators currently in use. for the reporting authority in collecting reliable infor- Despite this, WPM has not yet been adopted widely as a mation from users. This problem does not stem from the sector-monitoring system in any of the countries where pilot indicators used for this WPM, but it is common for every studies have taken place. Arguments usually made on monitoring system to be put in place. However, the issue technical grounds against WPM are the following. of quality measurements requires further discussion. It is † The baseline is expensive at a cost of 12–15 dollars/ unreasonable to expect users to update quality measure- water point for standard water point mapping (Stoupy & ments but it cannot be overlooked. With a global picture Sudgen 2003) and around 20 dollars when quality and an established baseline, the responsibility of updat- analysis is included. This might appear high, but it is ing quality measurements could be assigned to a certain not so if important investments are foreseen in the body (e.g. basin organs). A number of key measures sector. About 2 million dollars for data collection would should be included in the yearly routine (from tank be needed for Tanzania, for example, while 950 million distribution to large-scale projects, and groundwater dollars will be invested in the sector from 2007 to 2011. measurements where there is a high risk of underground Moreover, the whole process, including quality measure- contamination, etc.) and be accounted for in the ments, may cost considerably less if the methodology is indicators. scaled up, a process which is explained elsewhere † Accurately defined and Scalable at all levels: Once the (Jime´ nez & Pe´ rez-Foguet 2007). level of service is defined, the methodology provides a † Data treatment is expensive and complicated. It is well simple way to calculate the access to improved water known that the most costly part of setting up databases is points. The results are objective and comparable among information collection. Once developed, a database is administrative levels and countries. The limitation for easy to use and to update. GIS software licences can also representing data is the availability of geographical be avoided, since several open source programmes are information. For example, in Malawi, digital maps exist available and widely used. at the enumerator area level (500 to 1,500 people), † The capacity for managing information has to be in whereas in Tanzania they are only available at the place. This may be the most problematic aspect of ward level (10,000 to 20,000 people), which is the establishing the system. Technical and human resources must be placed at the lowest possible level, depending on water points can be separated, giving a more precise the particular conditions of each country. Ideally, the picture of the situation, especially when more than 30% of decentralized body responsible for water service plan- the constructed water points become non-functioning ning and delivery should be able to manage this (Government of Tanzania 2002). Consequently, Functional information. Alternative solutions are possible, however. Improved Community Water Point Density in a particular Standard information packages (such as maps displaying territory, displayed via digital maps, provides a much better the density of water points per area and others) could be representation of the access situation than ever before. prepared and sent to both these bodies and users and be Despite these important advances, some important aspects used as tools for planning and accountability. Planners at of access continue to be overlooked: the quality of water the local level could benefit from these services from served (“safe”) and to what extent the service provided by a upper level bodies (e.g. the Ministry). Once the strategy is certain water point is reliable (“sustainable”). The research defined, targeted capacity building should be put in place presented in this paper assesses both aspects and includes to enable technicians to use these tools and to allow them in a new indicator, defined using the same WPM users to understand them. methodology: Bacteriological Acceptable and Yearly-round

Despite technical challenges, it is important to consider Functional Improved Water Point Density. This indicator that information is politically sensitive. An in-depth analysis includes new basic quality information collected during provides less optimistic figures than those given by central mapping campaigns (with a reduced total cost) and processes governments (Table 2). This can be perceived as a threat, seasonality data. making governments reluctant to adopt the system. Pressure Evidence from the Same district in Tanzania reveals from donors and civil society to increase accountability significant differences in coverage data when these aspects must be encouraged to effectively tackle this problem. are included (from 50% to 40% when doing a mapping in relation to usual household surveys, and from 40% to 25% when basic quality and sustainability are included). Given the fact that results from one district are not representative CONCLUSIONS of an entire nation, the aim of the indicator is to highlight International agreements aim to halve the number of people two aspects: firstly, that the common assumption that without access to safe drinking water and basic sanitation by improved water points give safe water may be too 2015. However, current indicators of access to water are optimistic, and secondly, that the vulnerability of rural insufficient to measure this in any reliable way. More water services both to climatic events (e.g. droughts) and to recently, with the ongoing decentralization processes, and inappropriate water use (e.g. source deviation for agricul- with more than 70% of funds for the sector expected to be ture, etc.) is usually high. Both aspects are sufficiently channelled through national governments in the next few important to be included. Water point mapping offers a years, the importance of monitoring national water sectors cost-effective and reliable way of integrating them into a using EASSY indicators (Easy to get at local level, Accurately single indicator. defined, Standard and internationally applied, Scalable at Despite some technical challenges required to adopt all administrative levels, Yearly updatable) has increased. this methodology and construct access indicators, political The water point mapping methodology is presented as an obstacles are the most significant. Internationally agreed alternative way of defining an EASSY water access indicator. basic indicators were long ago defined for other social Promoted and widely developed by WaterAid, the methodo- services such as health. The will of having a reliable logy enables geographically related indicators to be defined monitoring system in the water sectors should be high on and thus determine the level of inequity regarding the the agenda for international donors and for the govern- distribution of water points. Moreover, the questionnaire ments of developing countries. Furthermore, indicators will attached to every water point means that information be well updated as far as all stakeholders involved perceive from functioning water points and already non-functioning a certain degree of usefulness in them. Appropriate investments in capacity building and awareness up to the de Rienzo, P. 2006 Aid, budgets and accountability: a survey article. user level are required for effective implementation. More- Dev. Policy Rev. 24(6), 627 645. E.U. 2006 European Consensus on Development. Official Journal of over, the ability to use water point mapping to increase European Union, 24.02.2006. investment efficiency and accountability at the local level Government of Mozambique 1995 Ley de Aguas. will determine to what extent a reliable process for updating Government of United Republic of Tanzania 2002 National Water can be expected both from users and from decentralized Policy. Harvey, P. & Reed, O. 2004 Rural Water Supply in Africa: Building authorities. Blocks for Handpump Sustainability. WEDC, Loughborough In the definition of the methodology presented in this University, UK, pp. 6 10. report, geographical information and a bottom-up approach Jime´ nez, A. 2006 La inversio´ n internacional en el sector agua y saneamiento en los paı´ses en vıás de desarrollo. Trabajo de are included, thus allowing further improvements that Investigacio´ n Tutelado (MSc Thesis). ETSICCP. Universidad may benefit other access-related aspects that have not yet Polite´ cnica de Madrid. been considered. Further improvements to measure Jime´ nez, A. & Pe´ rez Foguet, A. 2007 Quality and sustainability quality and more sustainability related information are aspects in water access indicators: an example from Same District, Tanzania. 33rd WEDC International Conference. also needed. Submitted. Jime´ nez, A., Molinero, J. & Pe´ rez Foguet, A. 2007 Monitoring Water and Poverty: a vision from development practitioners. Presented at the 3rd Botıń Foundation Water Workshop. June ACKNOWLEDGEMENTS 2007. In press. Lawson, A. & Rakner, L. 2005 Understanding Patterns of The author/s would like to extend thanks to Ingenierıá sin Accountability in Tanzania: Final Synthesis Report. Fronteras, Universidad Polite´ cnica de Madrid and Univer- Stoupy, O. & Sudgen, S. 2003 Halving the Number of People sitat Polite` cnica de Catalunya, for their support to under- without Access to Safe Water by 2015 A Malawian Perspective. Part 2: New indicators for the millennium take the mapping project; to WaterAid Tanzania, for their development goal. A WaterAid report. continuous assistance during the process. Research grant UN 2003 Indicators for Monitoring the Millennium Development provided by the Generalitat de Catalunya, ACCD-U2006, is Goals. Publication number ST/ESA/STAT/SER.F/95. United also acknowledged. Special thanks are also given to ISF- Nations. New York. UN 2004 U.N. declaration 58/217. ´ ´ Same staff, and to Josep Costa, Angel Fernandez, Laura UN 2006 www.un.org/waterforlifedecade/ Last Visit August 2006. Montilla, Jordi Pascual, Carol Puig and Adria´ n Sua´ rez for Wateraid, ODI 2005 Learning for advocacy and good practice their dedicated work during and after mapping surveys. WaterAid water point mapping. Prepared by Katharina Welle, Overseas Development Insitute. Available at http://www. Same District Water Department collaboration is also wateraid.org/international/what_we_do/policy_and_research/ gratefully recognized. Last Visit 1th August 2007. WHO/UNICEF 2000 Global Water Supply and Sanitation Assessment 2000 Report. REFERENCES WHO/UNICEF 2005 Water for Life. Making it happen. WHO/ UNICEF Joint Monitoring Programme for Water Supply and Birdsall, N. 2004 Seven Deadly Sins: reflections on donors failings. Sanitation. World Health Organization Press. Geneva. Working paper 50, Centre for Global Development, W.W.S.D. 2002 Summit on Sustainable Development, Plan of Washington, DC. Implementation. Johannesburg, Sept. 2002.