WATER GLOBAL PRACTICE DISCUSSION PAPER J U LY 2 01 8 Water Supply in a War Zone Naif Mohammed A Preliminary Analysis of Two Urban Abu-Lohom, Yasuo ­Konishi, Yogita Water Tanker Supply Systems in the Mumssen, Bilkis Zabara, and Scott Michael Moore Republic of Yemen About the Water Global Practice Launched in 2014, the World Bank Group’s Water Global Practice brings together financing, knowledge, and implementation in one platform. By combining the Bank’s global knowledge with country investments, this model generates more firepower for transformational solutions to help countries grow sustainably. Please visit us at www.worldbank.org/water or follow us on Twitter at @WorldBankWater. Water Supply in a War Zone A Preliminary Analysis of Two Urban Water Tanker Supply Systems in the Republic of Yemen Naif Mohammed Abu-Lohom, Yasuo Konishi, Yogita Mumssen, Bilkis Zabara, and Scott Michael Moore © 2018 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington, DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Please cite the work as follows: Abu-Lohom, Naif Mohammed, Yasuo Konishi, Yogita Mumssen, Bilkis Zabara, and Scott Michael Moore. 2018. “Water Supply in a War Zone: A Preliminary Analysis of Two Urban Water Tanker Supply Systems in the Republic of Yemen.” World Bank, Washington, DC. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Cover photo: © Anadolu Agency / Getty Images. Cover design: Jean Franz, Franz & Company, Inc. Contents Acknowledgments v Abbreviations vi Summary 1 Introduction 2 Basic Features of Urban Water Supply Systems in Sana’a and Aden 5 Analysis of Private Tanker Truck Water Supply Systems in Sana’a and Aden 8 Potential Implications for Household Livelihoods, Health, and the Environment 16 Conclusions, Overall Implications, and Recommendations 20 Notes 24 Bibliography 25 Box 1. Sampled Water Quality in the Well-to-Household Water Supply Chain 11 Figures 1. Population without Clean Water, 2015–17 2 2. Sample Prices for Water in Select Channels, Sana’a and Aden (YER/m ) 9 3 3. Sample Value Chain Maps for Tanker Truck Fleet Operations in Sana’a and Aden, by District 13 4. Sample Value Chain Map for Independent Tanker Truck Driver in Sana’a, by District 15 5. Sana’a Tanker Drivers Contracted by International Agency 15 6. Household Utilization of Delivered and Collected Water 17 7. Household Utilization of Municipal Water Services and Private Tanker Trucks 17 Table 1. Recommendations to Improve the Private Water Tanker Value Chains in Sana’a and Aden 21 Water Supply in a War Zone iii Acknowledgments This discussion paper summarizes the results and The team would like to thank Asad Alam (Country implications of a study, “Initial Value Chain Analysis Director, Egypt, Yemen, and Djibouti,  World Bank), for Private Water Tankers in Yemen,” prepared for the Sandra Bloemenkamp (Country Manager, Yemen, World Bank by Global Development Solutions, LLC, World  Bank), Poonam Gupta (Country Program Alexandria, VA, USA, and the Water Environment Coordinator, World Bank), and Steven Schonberger Center, Sana’a University, Sana’a, Yemen, July 2017. (Practice Manager, World Bank) for their guidance and encouragement. The team would like to thank the Yogita Mumssen (Senior Infrastructure Economist, Global Solutions Group (GSG) on Water, Poverty, and the World Bank) was the Task Team Leader for this paper, Economy for its support, in particular Richard Damania and her coauthors were Naif Mohammed Abu-Lohom (Senior Economic Advior, World Bank) and Scott (Senior Water Resources Management Specialist, Michael Moore (Water Resource Management Specialist, World Bank), Yasuo Konishi (Global Development World Bank). The team would also like to thank the GSG Solutions), and Bilkis Zabara (Sana’a University). on Water Security and Integrated Resource Management for its support, in particular Amal Talbi (Lead Water Resource Management Specialist, World Bank). Water Supply in a War Zone v Abbreviations AU autonomous utility GARWSP General Authority of Rural Water Supply Projects INGO international nongovernmental organization LC local corporation MWE Ministry of Water and Environment NGO nongovernmental organization NWRA National Water Resources Authority NWSA National Water and Sanitation Authority NWSSIP National Water Sector Strategy and Investment Program TDS total dissolved solids UNICEF United Nations Children’s Fund WASH water supply, sanitation, and hygiene WBG World Bank Group WEC Water and Environment Centre vi Water Supply in a War Zone Summary T his discussion paper summarizes the results the only detailed, survey-based assessments of water and implications of a study commissioned by supply and sanitation in an active war zone. While it the World Bank to conduct a rapid assessment is based on a partial rather than a systematic survey, it of the state of private water tanker supply systems in highlights a number of key aspects of Yemen’s urban two Yemeni cities, Sana’a and Aden. The study ema- water supply and sanitation situation. First, as a result nated from the World Bank’s Water, Sanitation, and of pre-existing challenges exacerbated by recent con- Health (WASH) Poverty Diagnostic for the Republic flict, there is a considerable gap in the urban water of Yemen, which identified serious gaps in access to supply sector. Second, this gap is being filled largely basic water and sanitation services (World Bank 2017). by private tanker trucks, on which urban Yemenis are A combination of poverty, water scarcity, armed con- increasingly dependent. Third, while the tanker truck flict, and warfare has produced serious challenges for system plays a critical role in filling this gap in the both water supply and sanitation in Yemen. Despite formal water supply  system, it raises serious ques- this urgency, little recent data is available on the state tions with respect to affordability, health, environ- of the country’s water supply and sanitation systems, ment, and water resources management. Overall, the in large part because key parts of the country remain study suggests the need for urgent interventions to inaccessible due to active conflict. The study described sanitation in Yemen’s two improve water supply and ­ in this discussion paper accordingly presents one of largest cities. Water Supply in a War Zone 1 Introduction B eset by poverty, water scarcity, and conflict, the privately-operated tanker trucks, kiosks, and private Republic of Yemen is facing one of the world’s wells. Today, many of the public services are no lon- most severe water supply and sanitation crises. ger operating, as the Central Bank of Yemen has been The already high share of Yemen’s population lacking unable to consistently pay salaries for civil servants access to clean water grew from 48 percent in 2015 to since the summer of 2016.4 At the same time, water 69 percent in 2017 (figure 1). Today, approximately 19.4 scarcity has risen rapidly throughout the country due million people in Yemen, including 10.2 million chil- to population growth, climate change and unsustain- dren, lack clean drinking water and sanitation, con- 1 able water resource management practices, including tributing to the continued prevalence of disease.2 As of the over-use of groundwater mainly in the agriculture the drafting of this report, more than 320,000 people in sector including the widespread cultivation of the Yemen have contracted cholera since an outbreak in late narcotic qat (Moore and Fisher 2012). April 2017. The problem is especially acute in Yemen’s 3 The decline in formal urban water service provision cities. In 1990, over 96 percent of urban residents had ­ has created a gap that has been filled to a significant access to an improved water source, but by 2010 the per- degree by informal market players. Specifically, pri- centage had fallen to 72 percent (World Bank 2017). vate tanker trucks have become more prominent in Prior to the onset of conflict since 2010, Yemen’s the water delivery supply chain as the availability of urban population gained access to water from a vari- formal municipal water services has declined. While ety of sources, including both formal municipal water the presence of private tanker trucks has improved supply systems as well as informal sources including the availability of water services particularly in areas FIGURE 1. Population without Clean Water, 2015–17 Yemen water crisis: population without clean water 30 80 70 25 Population (million people) 60 20 Percentage of total 50 15 40 30 10 20 5 10 0 0 2015 2016 2017 (est) Total population Without clean water and sanitation Percentage of total population Source: World Data Indicators, https://data.worldbank.org/indicator, and World Bank 2017. 2017 population estimated from average population growth rates 2015–16. Data compiled by Global Development Solutions, LLC. 2 Water Supply in a War Zone where public services have collapsed, it raises poten- of these changes to key development outcomes in the tial concerns over issues such as water quality, mar- country. It is organized into three main sections. The ket pricing, service delivery, sustainable water first describes the urban water supply system in Sana’a resource management and the impact of private and Aden, and briefly notes the factors that have tanker use on the country, especially on the poor. caused an increased reliance on private water tanker Households that no longer have access to reliable truck supply systems in the two cities. The second sec- water from water pipe networks or which cannot tion details how this system operates, including a afford to purchase tanker truck water increasingly description of the basic value chain from water must rely on free water provided by mosques or sources, which are primarily private wells, to house- other informal means, such as tapping standpipes. holds. The third section summarizes how this water The proliferation of private water service providers supply system affects households in terms of water and the absence of any government monitoring sys- use and potential implications for livelihoods and tem makes it especially challenging to ensure water health. The discussion paper concludes with some rec- quality and prevent the spread of waterborne illness ommendations on the basis of this analysis for improv- such as the cholera epidemic currently occurring in ing water supply services in light of the prevalence of Yemen (World Health Organization 2018). water tanker delivery. In light of the current situation, the World Bank Group Before proceeding, some additional description of the (WBG) contracted Global Development Solutions, LLC methodology employed in this study and its limita- (GDS), in partnership with the Water and Environment tions is warranted. Given limited availability of time Centre (WEC), Sana’a University, to conduct a study and resources, this report reflects a rapid assessment to understand the current role of water tanker trucks of Yemen’s water delivery services: data collection, within the broader context of urban water services in analysis, and reporting for the assessment were all per- Yemen, and highlight changes that may have taken formed in the period from April 20 to June 30, 2017. place since the conflict began. The study builds on a Time limitations due to the month of Ramadan and pre-conflict analysis of tanker services completed by uncertainty due to armed conflict prevented collection WEC (Zabara et al. 2010) and the recent Water Supply, of a representative or statistically significant sample. Sanitation and Hygiene (WASH) Poverty Diagnostic for Instead, the field team captured a non-random “snap- the Republic of Yemen (World Bank 2017), and comple- shot” sample of water delivery facilities and house- ments the WASH diagnostic efforts by focusing on pri- holds in each target region, with participants and the vate water tanker supply services. The study c ­ onsisted resulting analytic data acting as proxies for the of field surveys in two cities, Sana’a and Aden, that socio-economic and market conditions among a wider collected a wide range of data from households, water range of water delivery sector stakeholders and house- tanker truck drivers and owners, well owners, and holds in the target regions. An important caveat is that other water supply sector stakeholders. The study also the results of this study cannot necessarily be general- collected water quality data at initial source collection ized to other cities, whether in Yemen or elsewhere. points as well as in tankers and upon delivery at house- Nonetheless, given the general paucity of data related holds.5 It represents one of the only such studies to be to water supply in conflict zones, this study offers a conducted in an active conflict zone. unique view of the effect of conflict on urban water supply in certain areas of Yemen. This discussion paper presents some key findings from the study, with a focus on understanding the impact of This study also covers only the two largest cities in conflict on water supply in Yemen and the relationship Yemen, Sana’a and Aden. These cities were selected Water Supply in a War Zone 3 for several reasons. First, Sana’a and Aden have suf- Within each city, sampling for the study was conducted fered a decline in municipal water services since the at district level. This was done for several reasons. The armed conflict began in 2010. The decline accelerated settlement of internally displaced persons in major cit- considerably after March 2015, after which the munic- ies such as Sana’a and Aden has created additional ipal water supply system in both cities has suffered stress on water resources, and water demand has severe damage. Second, both cities have public water reportedly been observed to have increased, yet no cen- utilities—water and sanitation local corporations sus of water demand is available on the district level.6 (LCs)—which have almost collapsed due to the ongo- Observations of the rate of access to free water in cer- ing armed conflict. As a result, the citizens of Sana’a tain districts moreover suggest that citizens are unable and Aden have become highly dependent on private to access water from other, more convenient sources, water tanker truck delivery. Prior to the onset of further necessitating a district-based study to assess the conflict, Aden’s LC used to cover over 90 percent access points by district. Moreover, in Yemen water of the city’s residents, and Sana’a’s LC less than prices differ considerably by district, necessitating a 50 ­percent. Third, existing World Bank and other sur- study of price by district. Finally, an understanding of vey data exists for these cities, and in the case of private water tanker service areas can help inform Sana’a these include a 2010 baseline survey which future interventions.7 For the household and tanker sur- facilitates an understanding how the water supply veys, all ten districts in Sana’a city and all eight districts situation has changed. in Aden were sampled. 4 Water Supply in a War Zone Basic Features of Urban Water Supply Systems in Sana’a and Aden T o understand the role that private water tank- sanitation delivery. It sets tariff policy, including ers play in urban water supply in Yemen, it is approving tariff levels charged by urban utilities, and necessary to first understand the challenges must approve senior appointments to local utility that the sector has long faced, and the additional com- boards. MWE subsidizes capital costs and, in some plications that have arisen as a result of Yemen’s recent cases, local utility operating and maintenance conflict. The county’s water supply and sanitation sec- expenses. MWE is also responsible, through GARWSP, tor has long faced significant financial, administrative, for developing rural water supply and sanitation technical, and other constraints, but these have grown ­ services. However, direct urban water supply and san- significantly worse as a result of the ongoing conflict itation service delivery is by law the responsibility of that has affected Yemen since approximately 2010. local utilities in the form of 23 local corporations (LCs) These have critically affected virtually all important and 10 “annexed autonomous utilities” (AUs) that actors in Yemen’s urban water supply sector. together provide water supply and sanitation services in most Yemeni towns. Each LC is established by Broadly speaking, there are three elements of Yemen’s decree, and features its own board with representa- water supply and sanitation sector. The first pertains tives from central and local government as well as from to formal municipal water services, which are deliv- the community. Water prices are set by the LCs with ered via LCs under the purview of the Ministry of management oversight from their respective Boards of Water and Environment (MWE). The General Authority Directors. MWE provides guidance to LCs on price- and of Rural Water Supply Projects (GARWSP) is responsi- tariff-setting. ble for water supply delivery in rural areas. The second relates to overall water resource management, includ- Approximately 5 percent of Yemen’s towns are not ing water use and allocation, which is primarily the served by local utilities, and are instead served by the responsibility of the National Water Resources National Water and Sanitation Authority (NWSA). Authority (NWRA). The third and final dimension con- MWE also plays a role in water resource management cerns water quality and sanitation. While national via cooperation with an intermediate-level entity, the WASH activities are coordinated by MWE and the NWRA. NWRA has wide-ranging legal powers to imple- United Nations Children’s Fund (UNICEF) through ment water laws and regulations, allocate water rights, GARWSP, Local Corporations, NWRA and, in the case approve permits for drilling wells, and to undertake of multi-lateral development projects, Urban Project various other water resource management functions. Management Units, ultimately water and sanitation However, in practice NWRA functions in a decentral- services are locally managed via decentralized local ized manner, operating through seven branch offices utilities. The institutional landscape for urban water and five river basin committees established in Sana’a, supply is briefly outlined below, followed by similarly Taiz, Saada, Turban, and Abyan. Apart from MWE, the abbreviated descriptions of the regulatory framework Ministry of Agriculture and Irrigation plays a signifi- and how urban water supply actors have been affected cant role in agricultural water management.8 by the ongoing conflict. Regulatory Framework of the Water Sector in the Institutional Landscape for Urban Water Supply: MWE is Republic of Yemen: The water sector in Yemen is guided the principal line ministry for water supply and by two primary regulations, namely the Water Law Water Supply in a War Zone 5 and the National Water Sector Strategy and Investment health and recommending specific water quality Program (NWSSIP). The Water Law was ratified in standards.11 The General Authority and Department 2002, with the primary objective being: 9 for Awareness and Environmental Health under the Ministry of Public Works and Highways is responsi- to regulate, develop and ration the exploitation of ble for supervising water quality in urban areas. water resources, as well as the protection thereof Finally, Yemen Authority for Specification, Standards from depletion and pollution, the improvement and Quality Control is responsible for setting water of the efficiency of conveying and distributing their quality standards. In practice, actual water quality uses and the proper maintenance and operation of monitoring is conducted by several laboratories the installation thereof and participation of the ben- under MWE and NWRA. The MWE laboratory in eficiaries thereof in their management in the various Sana’a currently employs 30 lab technicians and is states of their development, investment and conser- the largest water lab in Yemen, followed by the vation thereof. NWRA-Sana’a lab (with 7 technicians), and GARWSP The NWSSIP was developed through a multi-stake- (3 technicians). holder initiative led by MWE to prepare a consolidated strategy, action plan and investment program for the Yemen’s ongoing conflict has exacerbated long-stand- water sector. The NWSSIP has several major objectives ing challenges facing the country’s urban water supply for sector management:10 sector and its accompanying regulatory framework. The LCs that supply formal municipal water systems • To ensure coordination among all partners work- are increasingly in debt and unable to afford to operate ing in urban and rural water supply and sanitation or maintain water networks, or to conduct regular sub-sectors, within and outside MWE; water quality tests. Tariffs levied for municipal water • To ascertain that policies in the urban and rural supply to households vary by quantity used, and range water supply sub-sectors are unified and that invest- from 130 to 374 Yemeni rials (YER) per cubic meter (m3) ments are equitably allocated among governorates in Sana’a, equivalent to approximately US$0.52–1.50, according to unified rules and that no projects are and YER31–120/m3, or US$0.12–0.48 in Aden. Although duplicated, especially in rural areas, to ensure that some LCs steadily increased tariffs prior to the conflict, investments complement each other; in most LCs water tariff rates for municipal water sup- ply have remained the same for years—in the case of • To integrate water policies and national policies of Aden they have remain unchanged since 2000—and sustainable growth and poverty reduction; remain far below cost recovery rates.12 Tariffs in Sana’a • To ensure that sector financing effectively supports and Aden do not cover operating costs, but at the same sector goals; and time widespread unemployment since the onset of the current conflict in Yemen has made it difficult for most • To monitor and evaluate the performance of water households to pay even minimal tariffs. In any case, supply utilities. the conflict has resulted in major disruptions to formal The primary responsibility for monitoring and regu- water supply systems in Sana’a and Aden. In Sana’a, lating water quality falls under NWRA, LCs and the municipal piped water network operates only GARWSP. In addition, other ministries and agencies intermittently, and usually only long enough to fill play an important role in monitoring and regulating household water tanks in one district at a time. Aden’s water quality. Other stakeholders include the piped water supply network operates more reliably, Ministry of Health, which is responsible for public but is subject to major challenges including silt and 6 Water Supply in a War Zone sodium buildup and salinization of source water and war damage that has apparently rendered the lab- supplies. oratory inoperable. Consequently, Aden LC is at the mercy of international nongovernmental organiza- The conflict has also substantially disrupted water qual- tions (INGOs) such as International Rescue Committee, ity monitoring and supervision efforts. In a reflection of the International Committee of the Red Cross, Oxfam the increasing importance of informal water supply net- and Mercy Corp to undertake chlorination treatment works, multiple NGOs as well as the Ministry of Health of wells.14 Furthermore, Aden LC faces salinity issues distribute chlorine tablets free of charge to private tanker in its Dar Almanasirah groundwater field, where some truck drivers to combat cholera. However, utilization is of the wells have a high concentration of salt, so the not systematic, and the predominance of informal actors water is mixed with water from the other two fields to results in little consistency or reliability in assessing and make it potable.15 ensuring water  quality. There is no apparent protocol for traceability in the water supply chain or chain of cus- The effective collapse of formal municipal water supply tody with respect to water quality, and little oversight and water quality systems in Sana’a and Aden has cre- and no formal registration or monitoring of tanker truck ated a substantial gap in water service provision. delivery services. Because of the security situation, it is Specifically, according to an interview, Aden authorities difficult to assess water quality with any degree of currently supply around 24 million m3 of water annu- ­accuracy.13 However, anecdotal reports indicate that the ally, whereas the demand is estimated to be more than conflict has severely and negatively impacted water 39 million m3. Before the war, pumps operated 22 hours quality in both Sana’a and Aden. According to Sana’a LC per day, but currently, pumps are operating less than officials, well water consistently exhibits high levels of 8 hours per day, and not on a regular basis. The water iron and nitrates. Moreover, aerial bombing has appar- shortage problem continues to be further exacerbated ently affected several water treatment facilities. by increasing numbers of refugees from Africa as well as Consequently, the Sana’a LC at times is unable to treat or internally displaced people. In response, illegal well filter water before it is distributed, which has resulted in drilling has apparently increased along with reliance on water quality and equipment degradation. private water tanker trucks. In Aden, small-scale private desalination plants also operate to fill the water supply The situation is reportedly similar in Aden. The city’s gap. The following section describes one of these infor- LC water testing laboratory is no longer functioning, mal adaptations, the private water tanker supply sys- due to both insufficient funds to cover operating costs tems operating in Sana’a and Aden in greater detail. Water Supply in a War Zone 7 Analysis of Private Tanker Truck Water Supply Systems in Sana’a and Aden A s noted in the Introduction, the core purpose accounted for 82 percent of the water sold by the of the study on which this discussion paper 40 tanker truck drivers surveyed in Sana’a and 85 per- is based was to gather key data on the private cent of the water sold by the 40 tanker truck drivers tanker truck water supply system that has grown in surveyed in Aden, with customers such as businesses, importance for Yemen’s cities since the onset of conflict. government offices, construction sites, factories, This section summarizes some of the data gathered as farms, cafeterias and restaurants, and supermarkets part of the study, and is divided into three sub-sections. accounting for the remaining tanker truck water sold. The first describes the well-to-tanker supply chain, In the case of supermarkets, water is then resold to including data on prices and costs at each major stage households. In Sana’a, some tanker drivers (up to of the chain. The second section presents data on well 25  percent of surveyed tanker water trucks in three owners, and the third and final section highlights major districts) reported being contracted by UNICEF to findings from surveys of private water tanker truck driv- deliver water free to poor and needy households, ers, including practices commonly employed to ensure rather than for direct sale to households.17 In Aden, or protect water quality. Box 1, meanwhile, describes about half of the well water purchased by surveyed the results of water quality sampling conducted as tanker trucks is desalinated at water purification shops part of the study. The analysis overall shows that while prior to delivery to customers (based on customer the water tanker system fills a key gap in formal water request). Just over half (53 percent) of tanker drivers supply services, there are few if any safeguards with surveyed in Sana’a reported that they charge house- holds higher prices than they do other customer types respect to ensuring water quality to end-users. (e.g., commercial or government clients). In Aden, no Well-to-Tanker Supply Chain: The supply chain for private price differentials by customer type were reported. water distribution in Sana’a and Aden runs from private well owners to household consumers, with private water Sellers, both well owners and tanker truck companies tanker owners and drivers acting as delivery agents. The and operators, set water prices based primarily on the majority of well water distributed through private water prevailing market price and the distance traveled to percent in Sana’a supply networks, approximately 56  ­ deliver water. Reported prices for water vary widely by and 57 percent in Aden, is eventually sold to household city and distribution channel. In Sana’a, interviewed consumers.16 In Sana’a, the 4 private well owners inter- well operators sold water to tankers for YER113–500/m3 viewed sell 60–80 percent of well water to tanker trucks, (US$0.45–2.00), while surveyed tanker truck drivers while in Aden, tanker trucks purchase 40–100 percent sold water to households for a very wide  range of of the water extracted from the 3 private wells studied. prices, ranging from YER226–1,000/m3 (US$0.9–3.99). In both Sana’a and Aden, interviewed well owners In Aden, surveyed tankers buy saline water at a rela- commonly donate the balance of their water (up to ­ tively low rate, averaging approximately YER112/m3 40 percent) to mosques and/or charities. (US$0.45) from private wells, which is then re-sold to The same general value chain exists with respect to households at a considerably higher average price of private water tanker trucks. On average, households YER1,832/m3 (US$7.3). Most such water sold in Aden is 8 Water Supply in a War Zone not directly potable, but instead requires desalination, Besides obtaining water from private tankers or which tanker truck drivers undertake as an additional piped water delivery, households may buy water service. If requested by the customer, tanker truck from purification shops. This method of obtaining drivers take water to a third-party purification shop for water is commonplace for households that cannot desalination treatment which costs approximately afford a full tanker delivery or lack a home storage YER2,000/m3 (US$7.98) and after an additional markup, tank. Such sales generally are in small quantities as desalinated tanker water may be sold to households by compared with those sold by tanker trucks, and tanker truck drivers for YER4,000–YER5,000/m 3 prices reflect a much higher unit cost, the equiva- (US$15.96–19.95), based on the study sample. 18 lent of up to YER20,000/m3 (US$79.8). When sold and  delivered by the purification shop in bulk, The need for desalination, along with increasing dis- prices may in contrast be as low as YER4,000/m3 tances traveled to reach productive wells as those closer (US$15.96). to the city become over-drawn, contributes to signifi- cantly higher water prices in Aden, where households Well Operations: In Sana’a, the average age of the may pay 8.1 times more for tanker truck drinking water studied private wells is 59 years, with some reported than do Sana’a households (see figure 2). Further, the as having been in operation for over 100 years, with deterioration of the formal municipal water supply net- depths ranging from 60 to 600 meters, albeit with work in both cities means that Sana’a households pur- reported declining water depth and volume. Due to chasing water from private tanker trucks pay, on deteriorating public services, demand for private well average, 3.8 times more for water delivered via private water is high, and tanker trucks in Sana’a endure long water tanker truck than they do for piped water, whereas wait times (90 percent of surveyed Sana’a tanker driv- in Aden, households pay 18 times more for non-desali- ers wait in line 30 minutes to more than 1 hour to nated water and 45 times more for desalinated water access wells). Compared to the Sana’a wells studied, delivered via private tanker truck versus pipe. Aden wells in the study are newer (average has been FIGURE 2. Sample Prices for Water in Select Channels, Sana’a and Aden (YER/m3) Sample water prices (YER/m3) Sample Water Prices (YER/m3) Sana’a Aden Sales channel Min Max Average Min Max Average Tanker to HH Well to tanker 113 500 240 100 400 112 Sana’a Well to tanker Desalination NA NA NA 2,000 2,000 2,000 process MWS to HH NDW 1,000 3,000 1,832 Tanker to HH 226 1,000 554 Tanker to HH (DW) DW 4,000 5,000 4,500 Tanker to HH (NDW) MWS (local Aden 130 374 146 31 120 100 corporation) to HH Well to tanker Puri cation shop 5,000 20,000 NA 4,000 20,000 NA MWS to HH to HH 0 1,000 2,000 3,000 4,000 5,000 Sources: Global Development Solutions, LLC and Water and Environment Centre. Note: DW = desalinated water; HH = household; NA = not applicable; NDW = non-desalinated water. Water Supply in a War Zone 9 in operation for 9 years, according to interviewed well trucks for more than five years. Even in Sana’a, where owners) and are shallower, with typical depths rang- private water tanker delivery services are more ing from 13 to 40 meters. The majority (75 percent) of established, there are many new entrants (55 percent interviewed well owners in Sana’a reported being reg- of surveyed tanker drivers have delivered water for istered with GARWSP or the MWE, compared with less than five years). The maturity of the Sana’a none in Aden, where illegally dug wells are more market is further demonstrated by the customer ­ common. Less than 3 percent of the Sana’a well own- ­structure. In the Sana’a sample, 63 percent of deliver- ers and no interviewed Aden well owners ration or ies are on an on-call basis and 35 percent are deliv- otherwise limit access to their wells, despite reported ered according to standing contracts, while in the water table decline due to continued and increased Aden sample, 98 percent of deliveries are based on water extraction. All well owners interviewed in calls from customers. Sana’a and Aden reported adding chlorine to the well. Notably, three of the four well owners interviewed in The use of credit payment is another indication of a Sana’a, compared to only one in Aden, reported test- more mature market for tanker truck water delivery ing their water quality, though all expressed interest service in Sana’a than in Aden. According to surveyed in doing so. drivers in Sana’a, 25 percent of household customers paid cash upon delivery for their water, and 23 percent Sampled wells in Aden serve a higher number of pri- of household clients purchased water on credit. In con- vate water tanker trucks than those in Sana’a (15–30 trast, in Aden, drivers reported that nearly all house- tankers/day in Aden versus 5–30 tankers/day in hold customers paid drivers cash on delivery for water Sana’a), reflecting the smaller average tank capacity (only 1 driver surveyed in Aden reported having cus- of trucks operating in Aden. Of the major well operat- tomers who paid on credit). ing expenses (fuel; oil and filter; and repair and main- tenance), fuel is the largest operating cost for well The operational characteristics of the water tanker owners in both cities (70–80 percent of total cost in fleets in the two cities also demonstrate considerable Aden and 50–90 percent of total cost in Sana’a). Data differences. The four fleet owners interviewed in Sana’a provided by interviewed well owners indicated profit served fewer districts but more households than the margins ranging from 8 to 40 percent in Sana’a and 19 four owners interviewed in Aden (30–180 households 33 to 56 percent in Aden. in Sana’a versus 20–40 households in Aden). High demand for water delivery in both cities is evidenced Providers of Private Water Tanker Truck Delivery by the long working hours of surveyed tanker truck Services: Survey findings suggest that independent drivers, with drivers in both Sana’a and Aden working, drivers dominate tanker truck water delivery ser- on average, 11–12-hour days, 6–7 days per week. Trucks vices in Sana’a (95 percent of surveyed drivers in in the Aden sample are relatively new (68 percent built Sana’a), while in Aden, water delivery services more between 2000 and 2016), while trucks in the Sana’a often are performed by tanker truck fleet owners who sample tend to be older (68 percent built between 1980 operate several trucks using hired drivers (100 per- and 1999), in line with the more established service in cent of surveyed drivers in Aden). Given the contin- Sana’a. There currently is no official entity that con- ued availability of piped water in Aden, tanker truck ducts vehicle inspections in Yemen, raising potential water delivery services are a relatively new phenom- concerns regarding vehicle safety. enon (43 percent of surveyed drivers have been driv- ing for less than one year), whereas in Sana’a, 45 A greater concern, however, may stem from the poten- percent of surveyed drivers have been driving tanker tial effects of private water tanker delivery on public 10 Water Supply in a War Zone BOX 1. Sampled Water Quality in the Well-to-Household Water Supply Chain As part of this study, GDS/WEC drew water samples to assess water quality and determine at what point(s) along the supply chain water quality becomes degraded as it progresses from private wells to households in both Sana’a and Aden (World Bank 2017). Testing consisted of physical, chemical and microbiological (also referred to as bacteriological) analysis. Despite testing authorities in both cities listing WHO “standards” for properties tested, and operating under the authority of Yemen Authority for Specification, Standards and Quality Control (YASSOC), which is responsible for setting water quality standards, each laboratory tested according to individually-adapted parameters and specifications (World Bank 2017). For example, the testing laboratory in Aden did not test for free chlorine concentration; chlorine is commonly used globally to kill bacteria in drinking water. In Sana’a, tested water exceeded limits for total dissolved solids (TDS) and coliform (including fecal E. coli) but featured below-standard concentrations of iron, ammonia and chlorine. Given the presence of contaminants across the entire supply chain in Sana’a, and the outbreak of cholera in Yemen, the low concentration of free chlorine in tested Sana’a supply chains may be a cause for concern. However, the Sana’a test results did not enable pinpointing of the specific points of origin for contamination or degradation, since tested parameters failed to meet various standards at each point in the supply chain. Thus, the problems may have originated at the source well and continued untreated throughout the supply chain, or may alternatively have been treated but later reintroduced at multiple points in the supply chain. In Aden, tested water failed to meet many physical, chemical and microbiological standards for water quality. Although Aden test results showed higher nonconformity to laboratory testing standards than did those in Sana’a (World Bank 2017), which tested positive for both total and fecal coliforms at every point in the supply chains tested, Aden results indicated total coliforms tested negative at the well but tested positive at subsequent points in both chains. Fecal (E. coli) coliforms were not found at any point in Aden water supply chains at the sites tested. Such findings are consistent with the siting of cholera outbreaks, which are more prevalent in Sana’a than in Aden (World Bank 2017). Nevertheless, 28 percent of the 80 surveyed Aden households and 13 percent of the 80 households surveyed in Sana’a reported that HOUSEHOLD members have become ill from drinking water at home. health (see box 1). Among the sampled population of of waterborne disease, and that an outbreak of tanker fleet owners and independent drivers, no tanker disease, such as the ongoing cholera epidemic, could water was tested with a public agency such as an LC or potentially be spread through the private tanker truck central health laboratory. The absence of water quality distribution system. testing, coupled with the lack of truck inspections, a fleet of old and aging tanker trucks (particularly in The sample of private water tanker truck drivers also Sana’a), and the inconsistent application of chlorine throws into sharp relief Yemen’s growing localized suggests that the tanker truck water delivery services water scarcity. Twenty percent of surveyed Sana’a driv- continue to operate in an unregulated environment ers observed source wells running dry in recent history, that is ripe for water contamination and rapid spread while only 2.5 percent of the surveyed Aden drivers Water Supply in a War Zone 11 indicated the same. One of the most striking differ- (US$1.64–2.08), while interviewed Aden fleet own- ences between private tanker truck operations in Sana’a ers reported paying YER100–400/m3 (US$0.4–1.6) and Aden is the wait time experienced by surveyed for water, having it desalinated for YER2,000/m3 drivers at a well to pump water. For example, approxi- (US$8.0), and selling it for YER4,000–5,000/m3 mately 80 percent of drivers surveyed in Aden waited (US$16–20), resulting in profit margins of 17–44 per- less than 10 minutes in line to pump water, while in cent in Sana’a and 34–35 percent in Aden. Poor access Sana’a 90 percent of surveyed drivers waited from to diesel fuel and traffic congestion led concerns 30 minutes to more than 1 hour to access a well. Once of interviewed tanker fleet owners in both cities. at the well, no surveyed drivers in Aden reported that Notably, private water tanker truck fleet owners did well owners were rationing water usage, while in not cite high or rising water prices as a concern in Sana’a, only one surveyed driver reported that well either Sana’a or Aden, perhaps because truck owners owners were exercising water rationing. simply pass price increases through to customers. Well depletion is also reflected in the higher water costs Similar results were found for independent tanker in the Sana’a sample as compared to the Aden sample. truck drivers in Sana’a. Of the major operating Well water prices are rising rapidly in both cities; nearly expenses of independent tanker truck drivers sur- 58 percent of surveyed independent drivers in Sana’a veyed (fuel, water, and maintenance and repair), indicated that the price of water paid to well owners water is the largest operating cost (60–90 percent of had increased between 10 and 50 percent in the past total cost in Sana’a) (see figure 4). Independent driv- 12  months, and 48 percent of drivers reported price ers reported profit margins ranging from 1 percent to increases of at least 25 percent in the past 12 months. In 49 percent in Sana’a, with some drivers potentially Aden, meanwhile, 48 percent of surveyed hired drivers indicating net loss, due to poor or non-existent book- indicated that the price paid for water increased more keeping. Independent tanker services marked up than 50 percent in the past 12 months. Indicative of the water prices for water bought from private well own- daily hardship incurred by households in Sana’a, a high ers and sold to households by 43–169 percent. number (83 percent) of the surveyed Sana’a tanker driv- However, these markups do not necessarily translate ers reported losing customers in the past 12 months due into accompanying profit margins, particularly for to customers’ loss of income and/or water price drivers operating older trucks or who travel to more increases. Together, these two reasons accounted for distant source wells. 65 percent of reported customer loss.20 The survey results also provide some indication Of the major tanker fleet operating expenses (fuel, that private water tanker truck distribution does water, maintenance and repair, driver wages, and, play some role in providing water to poor and disad- in the case of Aden, desalination), water purchases vantaged populations. Besides selling water to account for the largest portion of operating cost in households, some drivers in the Sana’a sample Sana’a (42–64  percent of total cost), while desali- reported being contracted by UNICEF to deliver nation is the largest operating cost for those sur- water. Specifically, the study found that 25 percent veyed in Aden (75–87 percent of total cost) (see of drivers, all of them independent, surveyed in Old figure  3). In Sana’a, interviewed water tanker fleet ­ Sana’a,  Al Wahdah, and Maein Districts were con- owners reported paying YER113–208/m (US$0.45–3 tracted by UNICEF in their respective districts to 0.83) of water and selling water for YER410–520/m 3 deliver water, particularly to the poor and needy 12 Water Supply in a War Zone FIGURE 3. Sample Value Chain Maps for Tanker Truck Fleet Operations in Sana’a and Aden, by District Sana’a a. Al-Sa a b. Maein Tanker Truck Company Value Chain Tanker Truck Company Value Chain Tanker Truck Margins for Water Sales Tanker Truck Margins for Water Sales Water purchase price by driver (YER/m3) 208 Water purchase price by driver (YER/m3) 113 Water selling price to households (YER/m )3 410 Water selling price to households (YER/m3) 450 Est. Pro t Margin 20% Est. Pro t Margin 17% Diesel Water Maint Labor Other Diesel Water Maint Labor Other 14.6% 63.7% 4.3% 13.8% 3.6% 2.5% 41.6% 11.0% 24.5% 6.0% Diesel Purchase Diesel Purchase Wage YER/liter Price YER/liter Price YER/yr 240 YER/m3 240 YER/m3 960,000 208 113 c. Shaoub d. Al Thawrah Tanker Truck Company Value Chain Tanker Truck Company Value Chain Tanker Truck Margins for Water Sales Tanker Truck Margins for Water Sales Water purchase price by driver (YER/m3) 192 Water purchase price by driver (YER/m3) 176 Water selling price to households (YER/m ) 3 520 Water selling price to households (YER/m3) 420 Est. Pro t Margin 44% Est. Pro t Margin 36% Diesel Water Maint Labor Other Diesel Water Maint Labor Other 1.5% 59.3% 11.0% 14.8% 13.4% 1.2% 62.5% 13.0% 21.6% 1.7% Diesel Diesel Purchase Purchase YER/liter Wage YER/liter Price Price 240 YER/yr 240 YER/m3 YER/m3 1,080,000 192 175.9259 figure continues next page Water Supply in a War Zone 13 FIGURE 3. continued Aden a. Ayman Abdu Alhakim (Zamzam) b. Almnsora Tanker Truck Company Value Chain Tanker Truck Company Value Chain Tanker Truck Margins for Water Sales Tanker Truck Margins for Water Sales Water purchase price by driver (YER/m3) 200 Water purchase price by driver (YER/m3) 300 Water selling price to households (YER/m3) 4,000 Water selling price to households (YER/m3) 4,000 Est. Pro t Margin 33.6% Est. Pro t Margin 34.9% Diesel Water Maint Labor Other Desal Diesel Water Maint Labor Other Desal 3.9% 7.5% 4.5% 6.2% 2.6% 75.3% 2.9% 9.9% 1.3% 3.3% 1.4% 80.2% Purchase Diesel Purchase Wage Desal: Diesel Desal: YER/liter Price Price YER/yr YER/m3 YER/liter YER/m3 240 YER/m3 YER/m3 960,000 2,000 240 2,000 300 200 c. Al Mansoura d. Al Slam Tanker Truck Company Value Chain Tanker Truck Company Value Chain Tanker Truck Margins for Water Sales Tanker Truck Margins for Water Sales Water purchase price by driver (YER/m3) 100 Water purchase price by driver (YER/m3) 100 Water selling price to households (YER/m ) 3 4,000 Water selling price to households (YER/m3) 4,000 Est. Pro t Margin 35.3% Est. Pro t Margin 33.6% Diesel Water Maint Labor Other Desal Diesel Water Maint Labor Other Desal 3.1% 2.3% 1.3% 5.9% 0.7% 86.7% 3.7% 2.2% 1.4% 6.3% 1.5% 84.9% Diesel Purchase Wage Desal: Diesel Purchase Desal: YER/liter Price YER/yr YER/m3 YER/liter Price YER/m3 240 YER/m3 933,333 2,000 240 YER/m3 2,000 100 100 Source: Global Development Solutions, LLC (GDS). Note: Aden operations shown pertain to desalinated water. (see figure 5). In Aden, no surveyed tanker drivers readily apparent how the tankers identified suitable were contracted by  UNICEF, but 20 percent of the recipients of free water from UNICEF; however, drivers surveyed in Al Buraiqeh District reported interviews with INGOs suggest that beneficiary being contracted by INGOs  in their district to groups are identified during monthly  INGO deliver water to the poor and needy. It was not meetings. 14 Water Supply in a War Zone FIGURE 4. Sample Value Chain Map for Independent Tanker Truck Driver in Sana’a, by District a. Al-Sa ia b. Azal Independent driver pro itability analysis (lowest margin) Tanker truck value chain (highest margin) Tanker truck margins for water sales Tanker truck margins for water sales Water purchase price by driver (YER/m ) 3 500 Water purchase price by driver (YER/m3) 233 Water selling price to households (YER/m3) 1,000 Water selling price to households (YER/m3) 580 Est. pro it margin 1% Est. pro it margin 49% Diesel Water Maint Other Diesel Water Maint Other 4.6% 70.7% 21.2% 3.4% 28.0% 68.6% 2.8% 0.6% Diesel Purchase Diesel Purchase YER/liter price YER/liter price 240 YER/m3 240 YER/m3 500 233 Source: Global Development Solutions, LLC. FIGURE 5. Sana’a Tanker Drivers Contracted by International Agency Other UN No UNICEF INGO agency (percent) (percent) (percent) 100 (percent) Overall 92.5 7.5 0.0 0.0 80 % of tanker trucks 1 Old Sana’a 75.0 25.0 0.0 0.0 60 2 Shaoub 100.0 0.0 0.0 0.0 40 3 Azal 100.0 0.0 0.0 0.0 4 Al Sa ia 100.0 0.0 0.0 0.0 20 5 Al Sabein 100.0 0.0 0.0 0.0 0 6 Al Wahdah 75.0 25.0 0.0 0.0 Sa ll Sh a’a ub Al al Al Sab a W ein Ta h Al Ma ir Ba Th ein H h ith i Al da Al ra ld era hr Az Al Sa ao ar n ni aw 7 Al Tahrir 100.0 0.0 0.0 0.0 ah v O O 8 Maein 75.0 25.0 0.0 0.0 9 Al Thawrah 100.0 0.0 0.0 0.0 District 10 Bani Al Harith 100.0 0.0 0.0 0.0 INGO Other UN agency UNICEF No Sources: Global Development Solutions, LLC and Water and Environment Centre. Note: INGO = international nongovernmental organization; UN = United Nations; UNICEF = United Nations Children’s Fund. Water Supply in a War Zone 15 Potential Implications for Household Livelihoods, Health, and the Environment T he study results show significant changes in greater importance was placed on water availability household water use as a result of the conflict, (26 percent in Sana’a vs. 40 percent in Aden). and their consequent dependence on tanker Increased dependence on informal private water dis- truck deliveries for water supply in Sana’a and Aden. tribution networks seems to have also been associated Survey results show that concerns regarding water with a decline in household water use. In both Sana’a availability are greater in Aden, while water quality is and Aden, the average size of surveyed households a greater concern in Sana’a. In both cities, however, pri- included nine members, who together consumed an vate water tanker trucks play an increasingly important average of 8.2 m3/month/household, implying average role in water supply. However, given the price differ- water use of 30 liters/person/day for all domestic uses. entials documented in the previous section, it seems Many households, including 60 percent of the Sana’a likely that the lack of dependable coverage for munic- sample and 44 percent of the Aden sample reduced ipal water supply and sanitation services has a regres- water consumption in the past five years, due primar- sive impact on the poor, while the lack of regulation ily to higher prices (cited by 45 percent of surveyed for tanker truck operators poses its own risks. Further, Sana’a households) and inadequate availability of as private tankers do not currently seem to have great water (cited by 35 percent of surveyed Aden incentives for water resources stewardship, there is households). a concern that their actions, together with the well owners from which they source, could exacerbate the More concerning from a health and sanitation per- already very tenuous water scarcity situation in Yemen’s spective is the fact that surveyed households in both cities. This section is divided into three sub-sections Sana’a and Aden also reported decreasing use of deliv- that discuss potential implications of the private water ered water whether by public or private pipe network tanker truck supply system for, respectively, household or by tanker truck. In Sana’a, the share of households water use, and attendant potential implications for live- relying solely on delivered water fell from 91 percent lihoods, health, and the sustainable management of conflict to 64 percent at the time of research, pre-­ while in Aden the share of households relying solely Yemen’s very scarce water resources. percent on delivered water fell from 68 percent to 54 ­ (figure 6). This finding suggests that since the onset of Household Water Use: Survey data conducted for this the conflict more households face greater difficulties study suggest that increased dependence on infor- in regularly accessing water supplies for drinking and mal private water distribution networks is associ- sanitation. ated with distinct changes in household water use, which in turn could have distributional and At the same time, households survey results con- health-related implications. In general, the study firmed that private water tanker trucks are increas- indicates that the water usage rate was very similar ingly filling the gap in water service provision caused in both Sana’a and Aden, although households sur- by decreasing use of formal municipal water networks veyed in Sana’a placed greater importance on water figure 7). The transformation has been particularly (­ safety (68 percent in Sana’a compared with marked in Sana’a, where municipal water service uti- percent in Aden) compared to Aden, where 48  ­ lization fell from 44 percent of surveyed households 16 Water Supply in a War Zone FIGURE 6. Household Utilization of Delivered and Collected Water Increased reliance on water collection to fill delivered water supply and price gaps 100 91 80 68 64 60 54 Percent 40 26 24 23 20 18 15 10 6 3 0 Sana’a Aden Sana’a Aden Sana’a Aden Delivery Collection Both delivery and collection Pre-conflict 2017 Sources: Household surveys, Global Development Solutions, LLC and Water and Environment Centre. Note: Delivery includes pipe networks (public or private) and tanker trucks. Collection includes free (e.g., mosque, standpipe) and purchased sources (e.g., purification shop). FIGURE 7. Household Utilization of Municipal Water Services and Private Tanker Trucks Private tanker trucks fill the water service gap 80 75 71 69 70 61 60 50 46 44 Percent 40 38 30 28 20 10 0 Sana’a Aden Sana’a Aden Municipal water services Private tanker truck Pre-conflict 2017 Source: Household surveys, Global Development Solutions, LLC and Water and Environment Centre. Note: Multiple responses accepted; households commonly rely on multiple means for water supply, including sources not shown (e.g., private pipe network, collection). Water Supply in a War Zone 17 pre-conflict to 28 percent currently, while 69 percent However, while private water tankers may be filling a of surveyed households have come to rely on private critical market gap, the study results also suggest that tanker trucks for water. The convenience of readily they do not necessarily serve the poorest populations available tanker truck delivery service nonetheless in Sana’a or Aden.23 High water prices and income scar- comes at a high price. Surveyed Sana’a households pay city have translated to increasing household reliance YER400–1,000/m3 (US$1.6–3.99) for water purchased on free water from standpipes, mosques and charity from tanker trucks versus YER146/m (US$0.58) for 3 organizations. In the Sana’a sample, there is a high municipal piped water. 21 In Aden, surveyed house- concentration of households with no income, particu- holds pay YER1,000–3,000/m (US$3.99–11.97) for non-­ 3 larly in Old Sana’a (38 percent), Azal (25 percent) and desalinated water deliveries and YER4,000–YER5,000/ Maein (38 percent) Districts, and which are conse- m (US$15.96–19.95) for desalinated tanker truck water, 3 quently unable to pay for water. In Aden, 4 percent of while piped municipal water costs for households surveyed households reported having no income, and range from YER31 to 120/m (US$0.12–0.48). However, 3 22 8 percent reported a low income between YER1–5,000 despite this cost differential, reliability issues prevent (US$0.004–19.95). Such households must rely on col- some households from being able to depend on the lected free water or sources such as family and friends formal municipal water supply network. Some house- who offer remittances, gifts, or loans, in order to holds using piped water have been unable to pay water acquire water. bills, resulting in service shutoff. To restore water delivery, some households have made private agree- The burdens of collecting water appear to be consid- ments with LCs to pay a flat-rate water charge toward erable based on the survey results. In Sana’a, 36 per- the accumulated water bill, further demonstrating the cent of the surveyed households must now collect challenges facing the municipal network. water for some or all of their household water needs compared with only 9 percent prior to the conflict, Implications for Access: The study results show that while in Aden, nearly half (46 percent) of the sur- while private water tanker deliveries are considerably veyed households must allocate time and resources more expensive than municipal piped water, there is to water collection for some or all of their household some evidence of market competition and efficiency water, up from 33 percent pre-conflict. Overall, sur- with respect to the responsiveness of private service veyed households in both Sana’a and Aden that col- providers in delivering water to households. Private lect water allocate at least 2 hours daily as a water tanker truck delivery service appears more effi- household to do so. Among surveyed households, cient in Sana’a, where 41 percent of surveyed house- the leading source for water collection in Sana’a is holds indicated that water is delivered within 1 hour of standpipes (tapped by 26 percent of households), making a call for delivery, than in Aden, where only while in Aden, water collectors rely on both mosques 21  percent of surveyed households received delivery (33 percent of households surveyed) and standpipes within an hour of calling for water, and 5 percent (21 percent of households surveyed). waited one to two days for tanker truck water delivery. Given these reported response times private water Apart from the direct additional burdens of water col- tanker truck delivery services appear to be consider- lecting water, these data suggest additional impacts on ably more reliable than the extant piped municipal development outcomes, especially in Sana’a. These water supply network, which currently operates only data indicate that on average surveyed households in intermittently. both Sana’a and Aden spend at least 2 hours per day 18 Water Supply in a War Zone collecting water, and at least one-third of the members undergoes desalination treatment prior to delivery of a surveyed household are required to devote to households. upwards of 4 hours each day to collect water. Although not conclusively demonstrated by the survey data, this Even so, survey results indicate that water quality time commitment may prevent these household mem- remains poor in Aden. Households were asked as part bers from engaging in remunerative employment. of the survey whether any household members had Given the deteriorating water service infrastructure become ill from drinking water at home. The incidence and depleting water resources, even if the current con- of waterborne illness was relatively high, with flict is resolved, the ability of households in Sana’a to percent of surveyed households overall reporting 28  ­ undertake employment may be limited by the need to illness. Ash Shaikh Outhman and Dar Sad Districts in designate a large number of household members to the Aden reported that at least 50 percent or more of sur- task of collecting water. veyed households had a household member become ill from drinking water, while in Al Buraiqeh and Attawahi Implications for Health: It is apparent from the survey Districts, 30 percent of surveyed households had data that the overall quality of water supply in Sana’a members become ill from drinking water. and Aden is low, and potentially poses health risks. Surveyed households overall considered safety as As noted above with respect to declining water deliveries, the most important issue in water supply, (68 per- the survey data further indicate that the cost and diffi- cent in Sana’a and 48 percent in Aden), followed by culty of obtaining water have caused many urban Yemenis availability (26 percent in Sana’a and 40 percent in to reduce water use considerably, potentially posing a Aden). However, there appear to be considerable dif- health and sanitation risk. In Aden, 25 percent of surveyed ferences in the quality of water between the two cit- households indicated that household water usage has ies, especially that delivered by private water tanker declined more than 50 percent since prior to the conflict. trucks. Surveyed households in Sana’a reported a This is particularly true in Ash Shaikh Outhman District, rapid decline in perceived water quality, as assessed where 80 percent of surveyed households indicated that by taste, color, smell and other characteristics, since their water usage rate has declined by more than 50 per- the current conflict began, while some Aden sur- cent. Conversely, between 10 and 30  percent of house- veyed households reported that such attributes holds surveyed in Al Mualla, Attawahi, and Khur Maksar improved. This discrepancy may imply that, com- Districts increased their water usage by over 10 percent. pared to municipal piped water supply, the quality This discrepancy suggests that local differences in water of tanker water from private wells is inferior in availability may be producing differential effects on Sana’a and superior in Aden. A possible explana- health and sanitation outcomes, although such a link is tion  for this is the fact that much water in Aden beyond the scope of this limited study. Water Supply in a War Zone 19 Conclusions, Overall Implications, and Recommendations W hile the presence of private tanker trucks and  Aden. However, it also clearly indicates that will improve the availability of water ser- informal, private networks are becoming increasingly vices particularly in areas where public important for water supply in both Sana’a and Aden, services have collapsed, it raises potential concerns suggesting that aid agencies and development insti- over issues such as water quality, market pricing, ser- tutions may need to adjust interventions in Yemen’s vice delivery, and sustainable water resource manage- urban water supply and sanitation sector. Table 1 ment. Water vendors typically operate outside or at the summarizes some key challenges identified by the margins of established legal frameworks. Water tanker study, along with some possible interventions and operators will always be able to purchase water from an impressionistic ranking of priority for Sana’a farmers as long as groundwater is available in the shal- versus Aden. Recommendations focus primarily on low urban/peri-urban agricultural wells because selling levers to mobilize and facilitate private sector devel- water to tanker operators is more profitable than farm- opment of private tanker truck delivery services, ing. This uncontrolled, unregulated informal tanker which is expected to continue to be a leading source markets may well lead to a steep fall in groundwater in of household water supply into the foreseeable urban and peri-urban areas, potentially affecting live- future in Yemen. Placing an emphasis on this spe- lihoods and causing conflicts between farmers, water cific segment of Yemen’s water supply and sanitation sellers and water end users. Moreover, while landown- sector, rather than on the sector as a whole, may be ers who sell water profit from informal water markets, more productive and pragmatic given the ongoing conflict. Going forward, it will be critical to expand non-selling farmers suffer from falling groundwater the knowledge base on Yemen’s water supply and levels, and landless laborers suffer from a loss of liveli- sanitation situation, especially with respect to pri- hood due to the elimination of agriculture activities. In vate tanker truck supply systems. Ideally, further addition, the increasing role of informal water markets research would expand on this research and conduct in supplying water to cities may well encourage the more systematic sampling as well as gather more expansion of illegal well-drilling, which may in turn data on pricing and water quality. These data could lead to serious negative impacts on the groundwater inform interventions, including possibly supporting aquifer, including a falling water table, land subsidence mobile-based platforms to publicize water prices or sea water intrusion. All these effects threaten to neg- and facilitate household-level water treatment, that atively impact the overall water security and sustain- should be urgently adopted to help Yemen address ability of Yemen’s limited scare water resources. one of the world’s most pressing water supply and As this paper and the study on which it is based have sanitation crises. Further emphasis also needs to be highlighted, Yemen’s water supply and sanitation placed on awareness of and approaches to managing challenges are complex, and vary between Sana’a the use of Yemen’s extremely scarce water resources. 20 Water Supply in a War Zone TABLE 1. Recommendations to Improve the Private Water Tanker Value Chains in Sana’a and Aden Priority for Key challenge Possible interventions Sana’a Aden Well owners 1 Inability to meet daily demand for Take stock of available water reserves in target areas to help High Not indicated water due to low ground water define a water management strategy in sample levels and poor well capacity ­ 2 High incidence of illegal well Register existing and future well owners through an independent/ Not indicated High ­drilling NGO-facilitated well owner’s association, to coordinate water in sample management and provide technical support such as access to chlorine, water quality testing and well drilling support 3 No pumping limits imposed on users Include pump limits requirements in the overall water manage- High High of wells ment strategy, to reduce stress on wells 4 Low rate of well water testing Independent/NGO-facilitated district-level well owner’s associa- High High tion could coordinate regular well water testing program, water management strategy, and chlorination programs with INGO/ WASH cluster 5 Insufficient number of qualified NRWA could partner with NGOs/INGOs to train additional tech- Medium Not indicated pump maintenance engineers nicians/engineers qualified to repair pumps in sample available in the market ­ 6 Monopoly among pump spare parts Conduct a rapid value chain analysis of the pump spare parts Medium Not indicated suppliers industry to identify ways to improve industry competitiveness in sample Tanker truck fleet owners 1 Low rate of water testing Coordinate regular water testing and chlorination programs with High High INGO/WASH cluster; effectiveness could be enhanced through something like an independent/NGO-facilitated district-level tank- er truck association (for owners and operators)a 2 Unregulated environment (e.g., lack Registration of water tankers is already planned by NWRA under High High of registration, inspection) poses the support of UNICEF, but may need additional support for public health and safety risk implementation. Registration could be linked with a successful vehicle safety inspection 3 Blackouts prevent pumps at well Partner with GIZ and other INGOs to develop a matching grant Not indicated Medium head from operating program to install solar panels to operate pumps during black- in sample outs Independent tanker truck drivers 1 Absence of organization among An independent/NGO-facilitated district-level tanker truck asso- High Not indicated large number of independent tanker ciation (for owners and operators),a with direct ties to a similar in sample truck operators association at the governorate level, could encourage the indus- try to adopt voluntary (self-regulating) guidelines to improve water quality and safety 2 Coliforms present at each step in Partner with the Ministry of Health and the Central Health High Not indicated the well-to-household water supply Laboratories to conduct regular-random water quality tests, both in sample chain increases risk of waterborne at the water source (prior to pumping), and at the point of delivery diseases Provide simple water testing systems for both self-regulation High Not indicated and oversight by independent/NGO-facilitated inspection body in sample and water testing body, linked with UNICEF project to register or renew registration of trucks table continues next page Water Supply in a War Zone 21 TABLE 1. Continued Priority for Key challenge Possible interventions Sana’a Aden 3 Fleet of old and aging tanker Establish a matching grant and financing program to replace old Medium Not indicated trucks creates a ripe environment water tanks from vehicles built prior to 1990 with the original in sample for water contamination and rapid tanks. The truck owner could be required to make a pre-de- spread of water borne disease termined down payment towards a replacement tank, which is matched by the program, and the remaining amount is paid back to the program through monthly installment plan based on a concessional interest rate 4 Rapid water table decline due to An independent/NGO-facilitated district-level association of Medium Not indicated continued and increased water well owners, with direct ties to a similar association at the gov- in sample extraction ernorate level, is a possible way of organizing the industry to help introduce voluntary (self-regulating) guidelines to improve water resource management practices, and monitor water ­quality 5 High price to households for water Directory of service providers and their contact information Medium Not indicated purchased from tanker trucks should be created and made widely available allowing house- in sample hold consumers to contact multiple water delivery services to get the most competitive price. Creation of such a directory could be tied to an independent/NGO-facilitated district-level tanker truck association (for owners and operators)a or to truck registration Hired tanker truck drivers 1 Absence of organization among An independent/NGO-facilitated district-level tanker truck High High individuals who own multiple tanker association (for owners and operators),a with direct ties to a trucks with hired drivers to deliver similar association at the governorate level could be a way of water organizing the industry to help introduce voluntary (self-regu- lating) guidelines to improve water quality and safety and build capacity to do so (training, access to finance) 2 Fleet of old and aging tanker Establish a matching grant and financing program to replace old Medium Medium trucks creates a ripe environment water tanks from vehicles built prior to 1990 with the original for water contamination and rapid tanks. The truck owner could be required to make a pre-de- spread of water borne disease termined down payment towards a replacement tank, which is matched by the program, and the remaining amount is paid back to the program through monthly installment plan based on a concessional interest rate Households 1 High concentration of households Conduct household census in selected districts to estimate the High High with no or very low incomeb current and future (free) water demand among the poor and needy 2 Exceptionally low and low availabil- ity of water for consumptionc 3 Relatively high incidence of report- Funding support and awareness raising campaign through NGOs Not indicated High ed illness from drinking water at and INGOs could help to distribute chlorine, and conduct regu- in samplee homed lar water quality test through NRWA, Ministry of Health, Central Health Laboratories, and other institutions table continues next page 22 Water Supply in a War Zone TABLE 1. Continued Priority for Key challenge Possible interventions Sana’a Aden 4 Rapid decline in perceived water Partnerships between (a) tanker truck owners (independent High Medium quality (taste, color, smell and fleet) and drivers in affected districts and (b) Ministry of Health overall water quality) from tanker and the Central Health Laboratories could be used to collect trucksf random water samples on a weekly basis and conduct tests to monitor water quality and to identify required interventions to improve water quality, and to prevent possible spread of diseases 5 Dramatic decline in MWS through Rapid assessment of municipal services to identify key binding High High LC and private pipe network,g and constraints limiting their ability to deliver water services, and high cost of water sold through estimate the cost of recommissioning services tanker trucks 6 Need to recommission the water Conduct rapid assessment to determine the cost of rehabili- Medium Medium pipe network, particularly in dis- tating, and the volume of water required to recommission the tricts where a high proportion of water pipe network households have traditionally been dependent on the water pipe net- work to access water, and remain connected to networks, but have had to increase reliance on tanker trucks and water collection since start of conflict and warh 7 Rise in household dependence on Regular, free water quality testing for tanker trucks that ­ deliver Not indicated Medium mosques and charity organizations 10 percent or more of total water delivery to mosques and in sample of water charity organizations free of charge could reduce likelihood of spread of disease through these channels Source: Global Development Solutions, LLC. Note: GIZ = Deutsche Gesellschaft für Internationale Zusammenarbeit; NGO = nongovernmental organization; INGO = international nongovern- mental organization; MWA = Municipal Water Services; WASH = water supply, sanitation, and hygiene; NWRA = National Water Resources Authority; UNICEF = United Nations Children’s Fund; LC = local corporations. a. Independent supervision (e.g., by WBG or another INGO) to draft standard bylaws and facilitate operation of such an association would be recommend- ed as a best practice, to mitigate risk of cartel formation. b. Household surveys indicated high proportion of households with no income in Sana’a districts of: Old Sana’a (37.5 percent), Azal (25 percent) and Maein (37.5 percent) and in Aden districts of: Dar Sad (20 percent) and Al Mansoura (9.1 percent). Household surveys indicated high proportion of house- holds with very low income in Aden districts of: Al Mualla and Attawahi (20 percent), Al Buraiqeh (11.1 percent), and Craiter (10 percent). c. Per household surveys, in at least 18.8 percent of the households in Sana’a overall and also very low availability of water for household consumption in Aden districts of: Khur Maksar, Craiter, Attawahi, Al Mualla, and Ash Shaikh Outhman districts. d. Household surveys indicated high incidence of illness from drinking water at home in sample population, particularly in Aden districts of Ash Shaikh Outhman, Dar Sad, Al Buraiqeh, and Attawahi. e. Priority not indicated in Sana’a by the household survey, but is of concern in Sana’a given the situs of cholera outbreak. f. Especially throughout Sana’a overall and in Aden districts of Al Buraiqeh and Ash Shaikh Outhman. g. Especially throughout Sana’a overall and in Aden districts of Al Mansoura, Dar Sad, and Attawahi. h. Household surveys indicated that more than 87.5 percent of households surveyed in Sana’a districts of Old Sana’a, Shaoub, and Al Wahdah traditionally depended on the public water pipe network to access water have had to increase reliance on tanker trucks and water collection since start of conflict and war. ­ Water Supply in a War Zone 23 Notes 1. Yemen Humanitarian Response Situation Report (Save the Children 13. Reportedly, registration of water tankers is planned by NWRA under 2016), as cited in World Bank 2017. the support of UNICEF; however, this had not yet been imple- mented at the time of field mission or data analysis. 2. Estimated 12.9 million lacked clean drinking water as of March 2015. UN OCHA 2014, 2017. 14. Reconstruction of the Aden LC water testing facility began as of early July 2017. 3. Report draft submitted 17 July 2017. From 27 April to 11 July 2017, 320,199 suspected cholera cases and 1,742 deaths have been 15. Officials at Aden LC indicated that there have not been any inci- reported in 91.3 percent (21/23) of Yemen governorates and dents of illness from the use of municipal water. percent (292/333) of districts (WHO 2017). 87.7 ­ 16. The number of private wells currently operating in Sana’a and Aden 4. Situation Report: Yemen’s Public Salary Disbursement Efforts, is unknown. Embassy of the Republic of Yemen—Washington, DC, March 2017. 17. UNICEF was the only I/NGO reported by tanker drivers surveyed 5. Target areas for study were determined by WBG; rationale for as hiring tanker water delivery services. The household selection selection is detailed in the project Inception Report and in section 2 ­ criteria were unknown. ­ Methodology. 18. The treatment is for desalination only and does not filtrate other 6. As of September 2016, Sana’a reported 149,994 IDPs, and 26,658 residues and harmful particles. IDPs in Aden. Resilience Performance Indicators Jan-Sept 2016, GIZ 19. Profit is defined as revenue minus cost. Profitability herein refer to Water Sector Program. profit margin, defined as profit as a percent of revenue. 7. Water and Environment Centre, Sana’a University (2017). 20. Drivers surveyed in Aden did not report customer loss, since many 8. Agriculture utilizes about 90 percent of available water resources. drivers have not been operating long enough to have an established customer base. 9. Article 3 of the Water Law, Republic of Yemen, July 2002. 21. Reported average MWS LC rate per household survey. 10. National Water Sector Strategy and Investment Program (NWSSIP), Ministry of Water and Environment, the Republic of Yemen, 22. Aden household water prices are assessed in YER/liter; the equiva- December 2004. lent YER/m3 is shown here for comparison. 11. The General Authority for Lands, Survey and Urban Planning is in 23. It should be recognized that this study relies on very limited data. charge of regulating private drinking water purification stations. A more comprehensive analysis of the relationship between water, poverty, and health was conducted as part of the Yemen WASH 12. Political resistance to raising water tariffs was generally high, but Poverty Diagnostic (World Bank 2017). with the support and encouragement of aid agencies several LCs did manage to increase tariffs. 24 Water Supply in a War Zone Bibliography Moore, Scott, and Joshua Fisher. 2012. “Challenges and Opportunities in World Bank. 2017. “Dire Straits: The Crisis Surrounding Poverty, Conflict, GRACE-Based Groundwater Storage Assessment and Management: An and Water in the Republic of Yemen.” WASH Poverty Diagnostic. Example from Yemen.” Water Resources Management 26: 1425–53. Washington, DC: World Bank. WHO (World Health Organization). 2017. “Yemen: Cholera Outbreak UN (United Nations) Office for the Coordination of Humanitarian Affairs Daily Epidemiology Update: 12 July 2017.” World Health Organization, (OCHA). 2014. 2015 Yemen Humanitarian Needs Overview. http://www​ Geneva. .humanitarianresponse.info/system/files/documents/files/2015_HNO​ _­Yemen_Final_0.pdf. Zabara, Bilkis, Abdulla Babaqi, Naif Abu Lohom, Abdulrahman Al-Eryani, and Fadhl AL-Nozaily. 2010. Analysis of Private Water Providers in Urban ———. 2017. Yemen Conflict: How Bad Is the Humanitarian Crisis? http:// and Peri-Urban Areas in Sana’a. Yemen: Water and Environment Center, www.bbc.com/news/world-middle-east-34011187. Sana’a University. Water Supply in a War Zone 25 SKU W18027