Research Report February 2008 44295 v2 Economic Impacts of Sanitation in Cambodia A �ve-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) THE WORLD BANK Jakarta Stock Exchange Building Tower II/12th Fl. Jl. Jend. Sudirman Kav. 52-53 Jakarta 12910 Tel: (6221) 5299-3000 Fax: (6221) 5299-3111 Printed in 2008. The volume is a product of World Bank staff and consultants. The �ndings, interpretations, and conclusions expressed herein do not necessarily reflect the views of the Board of Executive Directors of the World Bank 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 of acceptance of such boundaries. WSP-EAP Research Report February 2008 Economic Impacts of Sanitation in Cambodia A �ve-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) EXECUTIVE SUMMARY Executive Summary Until 2004, only about 17% of Cambodian people had access to improved sanitation meaning that there were still more than 11 million Cambodians living with an unimproved or no latrine. Although the �gure given by the Cambodia Demographic and Health Survey (CDHS) in 2005 indicates the increase of access coverage to nearly 22% in 2005, it is estimated that about 204,000 people need to gain access each year to improved latrines if Cambodia is to achieve its Millennium Development Goal target of reducing by half in 2015 those without improved sanitation from a base year of 1990. Most of the people without improved sanitation (84%) are in rural areas which are home to more than 90% of the Cambodian poor. While there is a consensus that lack of access to clean water and improved sanitation has a variety of impacts, there is often a lack of evidence to affirm that poor sanitation imposes a signi�cant burden on society. This in turn hampers the implementation of the required investments in the sanitation sector. In response, the “Sanitation Impact� study, initiated by the World Bank Water and Sanitation Program, aims to generate sound evidence on the negative impacts of existing sanitation conditions and the potential bene�ts of improvements in sanitation and hygiene in Cambodia. In this study, quantitative assessment of economic impacts of poor sanitation and hygiene on health, water resource, other welfare indicators, and tourism is conducted. The inclusion of health is based on the close links between sanitation and hygiene and disease incidence. Once the sanitation and hygiene related diseases occur, various costs will be generated such as health care cost, productivity cost, and, in more severe case, premature death. In addition, water resources are polluted by poor sanitation, with related economic costs. Among others, water pollution leads to costly avertive behavior in response to less usable water resources, and it affects the �sh population. Other welfare impacts are evaluated in this study because the absence of sanitary facilities affects people in terms of the time spent accessing facilities which incurs some costs to the users, among other things. Last but not least, the inclusion of tourism in the study is also deemed important given the signi�cance of this sector in Cambodia’s economy, accounting for almost 15% of Gross Domestic Product (GDP) in 2006. Although hard to quantify, poor sanitation facilities and lack of hygienic conditions affect the country’s attractiveness as a tourist destination. In addition to the quantitative evaluation, the study also discusses the effects of poor sanitation on various qualitative dimensions including health-related quality of life, intangible user preferences, life decisions, and the quality of the surrounding environment. Even though these qualitative impacts are not measurable monetarily, they do have implication on human welfare, for which populations may be willing to pay to avert. The study also presents some of the bene�ts of improved sanitation when impacts mentioned above are completely or partially eliminated. However, these bene�ts are not necessarily fully representative of the potential gains as other possible impacts are not included such as impacts on foreign direct investment, other sanitation and hygiene related diseases which are not included in the health impact assessment, and others. In the analysis, sanitation refers to activities related to human waste, which is the term used for human excreta. However, there are instances in which sanitation as it relates to gray water and solid waste are included. In measuring the impacts, the study uses a peer-reviewed methodology developed speci�cally for this study, which draws on established methods and, where these do not exist, develops new approaches to capture the impacts of poor sanitation. For improving policy interpretation of the results, the study distinguishes where possible between �nancial and economic impacts, and presents for rural/urban areas and different provincial groupings (zones). Overall, the study �nds that poor sanitation leads to economic losses of US$448 million per year which translates into per capita loss of approximately US$32. These economic losses are equivalent to 7.2% of the Cambodia’s GDP in 2005. This amount is roughly equivalent to the contribution of the �shery sector to the GDP, or twice the forestry’s contribution. While these economic costs are not all tangible, the immediate money ‘in the hand’ losses (�nancial losses) amount to about US$160 million per year, which is roughly 2.5% of the GDP, being equivalent to nearly US$12 Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 1 EXECUTIVE SUMMARY per capita. These �gures, however, exclude a number of other impacts, some of which are qualitatively assessed in the study, while others are omitted in the estimation. Economic loss of poor sanitation and hygiene (Total: 448 million US$) 16% 9% 42% Health costs Water costs Access time Tourism 33% In the economic cost estimation, it is found that health impact is the largest contributor of quanti�ed costs. It amounts to US$187 million accounting for 42% of the total economic costs. The next main contributor of the quanti�ed economic costs is the water costs attributed to the cost for accessing cleaner drinking water, the cost for accessing other domestic water uses, and the loss in �sh production due to polluted water. According to the estimation, water costs share roughly one third of the total economic losses being nearly US$150 million. Moreover, the tourism sector which may be also affected by poor sanitation and hygiene practice in the country is also estimated to lose about US$74 million per year making 16% of the total costs. This loss is actually a result of under-exploitation of tourism capacity (below optimal capacity) and that the envisaged tourism growth is not achievable. The economic loss caused by the loss of time due to unimproved sanitation is roughly US$38 million being 9% of the total economic costs. This cost incurs among those who practice open defecation (journey time) and shared toilet users (waiting time). Economic losses by impacts 200 180 Tourism loss 160 Access time 140 Fish production Million US$ 120 Domestic water 100 Drinking water 80 Premature death 60 Productivity 40 Healthcare 20 0 Health Water resource Access time Tourism Impact categories Economic Impacts of Sanitation in Cambodia 2 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) EXECUTIVE SUMMARY While health cost is the main contributor to the total economic costs, the cost of premature death is estimated to be the biggest source of health costs which amounts to US$169 million annually contributing to more than 90% of the health impacts. This huge cost is due to the high number of deaths among children from diarrhea, and from diseases related to malnutrition which results from diarrhea. It is also noted that there are various methods for estimating value of human life which result in signi�cantly different values. However, the technique used in this study – the human capital approach – is among the most conservative of estimation methodologies. Besides premature death, there are two other components of health costs making up total health cost. Cost of health care amounts to over US$13 million per year, including the costs of consultation, treatment, medicine, and transportation. Productivity cost, which refers to the productive time loss of patients both adults and children as well as their carers, amount to US$5 million per year. Three components make up the water costs which are access to clean drinking water, access to water for other domestic purposes, and �sh production loss. Of the US$150 million of water access costs, accessing clean drinking water accounts for approximately 63%, while domestic water use (excluding drinking water) accounts for over 7%. Moreover, the cost due to lower �sh catch affected by reduced levels of dissolved oxygen, part of which is attributed to poor sanitation, is about US$44 million accounting for nearly 30% of total water access costs. Tourism impacts are computed on the assumption that visitors to Cambodia are sensitive to sanitation conditions. Based on some crude assumptions, it is estimated that poor sanitation and hygiene leads to the tourism loss of about US$74 million being 7% of the tourism income in 2006. This cost is anticipated by the fact that the potential growth of tourism may not be fully achievable if sanitation and hygiene conditions in the country are not properly managed. This estimation, nevertheless, does not account for the sanitation and hygiene-related sickness, and welfare loss due to the sickness. The economic cost associated with time loss is estimated to be about US$38 million. While poor sanitation may explain the drop-out of school girls, the quanti�ed cost of other welfare impacts does not include the loss that may be incurred by drop out from school which will influence career seeking and income levels in the future. Moreover, the fact that women may be absent from or not willing to go to work due to poor sanitation facilities is not accounted for. Also, there are other dimensions of welfare impacts (intangible user preferences) which are not quanti�ed such as the issue of privacy, prestige, convenience, etc. It is highlighted in the study that hygiene and cleanliness are the main perceived bene�ts of most latrine users in Cambodia. These bene�ts are followed by comfort, health improvement, safety, convenience, privacy, and prestige. In terms of the impact of poor sanitation on environmental surrounding and related human welfare impacts, this impact is assessed by focusing on the problem of household solid waste. Based on literature sources and interviews with stakeholders, it is clear that solid waste in urban and rural areas of Cambodia is not yet properly handled. Although in some urban areas where the collection service is available, the scattering of waste in public places such as at markets and along the streets is still common. This practice damages the sight of the areas, and also produces bad odors to the people, affecting the welfare of the local people as well as attractiveness for tourists. Moreover, the designated dump site, particularly in Phnom Penh, is already operating at over capacity, thus impacting air pollution and water resources. Having estimated the impacts, the study also evaluated the bene�ts associated with improved sanitation and hygiene practices. Better hygiene practices and improvements in toilet systems were linked to a reduction in health costs, while improved physical access and treatment/disposal can reduce the other costs components. The results showed that improved hygiene practices such as hand washing can reduce health costs by approximately US$84 million. Improved physical access to sanitary toilets can reduce economic costs associated with time use by about US$38 million, and contribute US$0.3 million to input markets (latrine builders and materials). Improved toilet systems can reduce health costs by US$60 million and contribute to input market of US$1.2 million. Improvement in the treatment or disposal of waste has a large impact on water resources and tourism which can reduce costs totaling US$223 million where US$149 million is from mitigated water impacts and US$74 million is from tourism Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 3 EXECUTIVE SUMMARY gain. While the bene�ts from pursuing all the improvements will not necessarily lead to gains which are equal to the sum of the values above, the results nonetheless suggest that the gains can be signi�cant. Hence the different sanitation options need to be examined in terms of what economic and �nancial bene�ts they have on health, water resources, the environment, other welfare, tourism, and local markets. For example, it is estimated that the bene�ts from sanitation market in terms of input and output market for the reuse of human excreta waste amounts to US$1.8 million per year. This bene�t is from the construction of biodigester plants for digesting animal and human waste for biogas and fertilizer, and from the cash saving associated with the use of biogas. While the �gures mentioned above are the base case estimation, sensitivity analysis was conducted to explore variation in estimated costs using different data sources and assumptions. The sensitivity analysis indicates that the base case results are sensitive to data inputs, in particular the value of life for premature deaths, which could be considerably higher. Input ranges for other variables tested in one-way sensitivity analysis do not individually affect the overall results signi�cantly. The �ndings of this study support widely though poorly substantiated beliefs that poor sanitation has signi�cant economic costs. Consequently, it has also shown that the gains of improving sanitation could be substantial. On the basis of these �ndings, the study recommends the following: • Given broad impacts of poor sanitation, decision makers from various sectors covering health, water resources, environment, rural and urban planning and development, �sheries, and tourism are advised to act now as the negative impacts of poor sanitation will increase over time incurring higher economic costs. • Governments should give priority to the populations with no latrine, recognizing that effective demand may be low in these groups due to low incomes and poor awareness of the bene�ts of investing in sanitation. • Sanitation investments should not be made just in latrine extension programs, but in improved sludge, water and solid waste management, and in hygiene programs to raise population awareness on personal and community hygiene issues. • Since health impacts incur the highest cost related to poor sanitation, health aspects of sanitation programs deserve central focus. Governments should focus on the easy health wins from improved sanitation, through targeting children and focusing on safe but simple latrine designs, improved excreta isolation measures, and improved hygiene practices. • Governments should urgently implement sanitation standards that reduce the release of waste matter into water resources given that water is partly polluted by poor sanitation. In this case, the focus should not be just on excreta, but also solid waste, household, agricultural and industrial wastewater. • Given broader impacts of poor sanitation, sanitation cannot be only the responsibility of an individual sector/ ministry, nor of a single level of government. The fact that sanitation touches on many sectors and line ministries should be used as a strength rather than hampering progress, and clear roles and responsibilities need to be de�ned. • While economic impacts of poor sanitation vary across regions in the country due to demographic situation, environment, and sanitation coverage, local studies should be commissioned to better inform local policy makers on the impacts. In this sense, the methodology used in this study can be of good use. Moreover, local level cost-bene�t studies will inform decision makers how to invest efficiently in sanitation • Future survey and research work is a key in monitoring progress of improving sanitation in the country. Surveys and government reporting systems should be assessed for extension to include behavior and outcomes related to sanitation. The link between poor sanitation and tourism and foreign direct investment losses is poorly understood, and merits further assessment. Economic Impacts of Sanitation in Cambodia 4 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) FOREWORD Foreword Cambodia is developing in all sectors, gradually lifting its people out of poverty as well as improving their living standards. Along with such development, the Royal Government of Cambodia, led by Samdech Akka Moha Sena Padei Techor Hun Sen, Prime Minister of the Kingdom of Cambodia, acknowledged in the National Forum on Rural Sanitation and Hygiene on 13th – 14th November 2007 the issue of poor sanitation coverage and hygiene practices of the Cambodian people living in rural areas. The Prime Minister Samdech Techor said, “In Cambodia, poor sanitation and hygiene is one of the factors contributing to the poverty of Cambodian people and blocking the efforts of the Royal Government in national economic development.� The lack of good sanitation and hygiene practices severely affects the lives of rural people, especially poor households and vulnerable people who are at higher risk from water- borne and hygiene-related diseases. Sanitation and hygiene has received very limited attention from relevant institutions within Cambodia. Very limited information exists on the impacts of poor sanitation and hygiene, or the institutional policy options and strategies to improve sanitation and hygiene. Sanitation and hygiene improvement is one of the priorities of the Ministry of Rural Development, who needs better understanding of the impacts of poor sanitation and hygiene both in the present and in the future. Ultimately this will enable the implementation of pro-poor strategies in line with the policies of the Royal Government of Cambodia. For the above reasons, the Water and Sanitation Program of the World Bank, East Asia and the Paci�c region, supported the research program “Economic Impacts of Sanitation� in Cambodia. The study aim is to provide scienti�c evidence and information related to economic bene�ts of improved sanitation and hygiene options. The principal focus of this study is to examine the economic and social losses associated with poor sanitation and hygiene, and conversely, the potential economic and social gains of improving sanitation and hygiene. On behalf of the Ministry of Rural Development, as the government institution in charge of rural water supply, sanitation and hygiene, I would like to express my sincere thanks to the Water and Sanitation Program of the World Bank, East Asia and the Paci�c region, for including Cambodia as one of the collaborating countries in this useful research program. The results of the research will be valuable for inclusion in the National Strategy on Rural Sanitation and Hygiene Promotion in Cambodia, which the Ministry is planning in the year 2008. I would encourage concerned institutions to use the data and information from this study to improve the planning of rural sanitation and hygiene programs in Cambodia. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 5 ABBREVIATIONS Abbreviations ADB Asian Development Bank ALRI Acute Lower Respiratory Infection BOD Biochemical Oxygen Demand CATA Cambodia Association of Travel Agents CBA Cost-Bene�t Analysis CDHS Cambodia Demographic and Health Survey CFR Case Fatality Rate CIPS Cambodia Inter-censal Population Survey CMDG Cambodia Millennium Development Goals CNMC Cambodian National Mekong Committee COD Chemical Oxygen Demand CSES Cambodia Socio-Economic Survey DHS Demographic and Health Survey DO Dissolved Oxygen DPWT Department of Public Works and Transport EAP East Asia and the Paci�c EASAN East Asia Sanitation Conference Ecosan Ecological Sanitation EIC Economic Institute of Cambodia ESI Economics of Sanitation Initiative FAO Food and Agriculture Organization FDI Foreign Direct Investment GDP Gross Domestic Product GNP Gross National Product GRP Gross Regional Product HH Household HIS Health Information System HRQL Health-Related Quality of Life IPD Inpatient Day JMP Joint Monitoring Program (WHO,UNICEF) Kg Kilograms MAFF Ministry of Agriculture Forestry and Fisheries MDG Millennium Development Goal MoE Ministry of Environment MoEYS Ministry of Education Youth and Sports Economic Impacts of Sanitation in Cambodia 6 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ABBREVIATIONS MoH Ministry of Health MoLVT Ministry of Labor and Vocational Training MoWA Ministry of Woman Affair MoWRM Ministry of Water Resource and Meteorology MPP Municipality of Phnom Penh MRC Mekong River Commission MRD Ministry of Rural Development NBP National Biodigester Program NGO Non-Governmental Organization NSDP National Strategic Development Plan OECD Organization for Economic Co-operation and Development OPV Outpatient Visit PPP Purchasing Power Parity PPWSA Phnom Penh Water Supply Authority SEAR-B WHO Southeast Asia Region epidemiological strata B SEI Stockholm Environment Institute UNEP United Nations Environment Program UNITAR United Nations Institute for Training and Research WHO World Health Organization WPR-B WHO Western-Paci�c Region epidemiological strata B WSP Water and Sanitation Program WTP Willingness To Pay Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 7 ACKNOWLEDGEMENTS Acknowledgments The Sanitation Impact Study was conducted under the Economics of Sanitation Initiative (ESI) in four countries: Cambodia, Indonesia, the Philippines and Vietnam. A study is ongoing in Lao PDR. The study was led by the East Asia and Paci�c office of the World Bank’s Water and Sanitation Program (WSP-EAP), with the contribution from WSP teams in each of the participating countries. The study took one year to complete, and has undergone two major peer review processes. Guy Hutton (WSP-EAP Senior Water and Sanitation Economist) led the development of the concept and methodology for ESI, and the management and coordination of the country team. The study bene�ted from the continuous support of other WSP-EAP staff. Isabel Blackett was the task team leader, Jemima Sy, Brian Smith, Almud Weitz and Richard Pollard provided input to the concept development and study execution. Bjorn Larsen (WSP consultant) contributed to the study methodology and provided the �gures for malnutrition-related health effects of poor sanitation. The country team in Cambodia consisted of Phyrum Kov (Economic Institute of Cambodia, country lead), Hach Sok (EIC director), Sophanara Roth (EIC) and Kongkea Chhoeun (EIC). Jan Willem Rosenboom (WSP Cambodia) contributed importantly to the study. Sovannra Aing (EIC) assisted with report �nalization. The ESI was �nanced by the regional component of the Sustainable Sanitation in East Asia (SUSEA) program, which is funded by the Swedish International Development Agency (SIDA). The Philippines study received co-funding from the United States Agency for International Development (USAID) Environmental Cooperation-Asia (ECO-Asia) program. WSP and the report authors are grateful to the funding agencies for supporting this study. Elena Strukova, Caroline van den Berg, Anjali Archarya, and Tracey Hart reviewed the methodology study before its implementation. Peer reviewers of the Cambodia country draft report were: Hilda Winarta (UNICEF), Jan Lam (SNV), Chea Samnang (Ministry of Rural Development), Ruud Corsel (Niras-Scanagri, Vietnam), Oun Syvibola (Plan International), Jan Willem Rosenboom (WSP Cambodia), Isabel Blackett (WSP-EAP), and Brain Smith (WSP-EAP). The Cambodia study bene�ted from peer review conducted simultaneously on other country reports as well as the synthesis report. Peer reviewers of the synthesis draft report were (World Bank staff unless otherwise stated): Eddy Perez, Anjali Acharya, Pete Kolsky, Elena Strukova (consultant), Bjorn Larsen (consultant), and Peter Feldman (Plan International). Peer reviewers of the Philippines country draft report were: Jema Sy and Andy Robinson (consultant). Peer reviewers of the Vietnam country draft report were: Samuel Leibermann, Doan Hong Quang, Pham Khanh Toan (Ministry of Construction), Nguyen Viet Anh (University of Civil Engineering), Nguyen Kim Thai (University of Civil Engineering), Nguyen Van Thuan (Australian Agency for International Development), and John Collett (Plan International). WSP and the country team appreciate the inputs of local stakeholders – Department of Rural Health Care of Ministry of Rural Development, Department of Planning of Ministry of Health, Phnom Penh water supply authority, and other institutions. A complete list of key informants is given in Annex E. Reference for citation: Economic impacts of sanitation in Cambodia. Kov P, Sok H, Roth S, Chhoeun K, Hutton G. World Bank, Water and Sanitation Program. 2008. This technical report is available in shortened form from WSP offices and from http://www.wsp.org/pubs/index. asp. Economic Impacts of Sanitation in Cambodia 8 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ACKNOWLEDGEMENTS References for other reports • Economic impacts of sanitation in Southeast Asia. Hutton G, Rodriguez UE, Napitupulu L, Thang P, Kov P. World Bank, Water and Sanitation Program. 2008. • Economic impacts of sanitation in Indonesia. Napitupulu L and Hutton G. World Bank, Water and Sanitation Program. 2008. • Economic impacts of sanitation in the Philippines. Rodriguez UE, Jamora N, Hutton G. World Bank, Water and Sanitation Program. 2008 • Economic impacts of sanitation in Vietnam. Thang P, Tuan H, Hang N, Hutton G. World Bank, Water and Sanitation Program. 2008. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 9 CONTENTS Contents Executive Summary 1 Foreword 5 Abbreviations 6 Acknowledgments 8 Contents 10 1. Introduction 15 1.1 Human development and sanitation 16 1.2 Sanitation in Cambodia 16 1.3 Measuring the economic impact of sanitation 19 1.4 Study goal and target audience 19 2. Study Methodology Overview 21 2.1 Levels and units of analysis 22 2.2 Scope of sanitation 23 2.3 Impact identi�cation and classi�cation 24 2.4 Estimation methods for �nancial and economic costs of poor sanitation 27 2.5 Impact mitigation 29 2.6 Uncertainty 30 3. Economic Impact Results 31 3.1 Summary of economic impacts of poor sanitation 32 3.2 Summary of economic gains from improved sanitation 35 3.3 Health impacts 37 3.4 Water resource impacts 41 3.5 Environmental impacts 47 3.6 Other welfare impacts 48 3.7 Tourism impacts 51 3.8 Sanitation markets 52 3.9 Sensitivity analysis 53 4. Discussion, Conclusions and Recommendations 57 4.1 Discussion 58 4.1.1 Overview and interpretation of main results 58 4.1.2 Policy implications of the study results 59 4.1.3 Study weaknesses 60 4.1.4 Gender 60 4.2 Conclusions 61 4.3 Recommendations 61 Annexes 65 Annex A. Study Methods 66 A1 Health 66 A1.1 Selection of diseases 66 A1.2 Disease burden from diseases directly related to poor sanitation 69 A1.3 Disease burden from diseases indirectly related to poor sanitation 70 Economic Impacts of Sanitation in Cambodia 10 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) CONTENTS A1.4 Health care cost estimation 75 A1.5 Health-related productivity cost estimation 78 A1.6 Premature death cost estimation 80 A2 Water resources 83 A2.1 Water quality measurement 84 A2.2 Contribution of poor sanitation to water pollution 84 A2.3 Cost implications of water pollution for drinking water supply 86 A2.4 Water quality and domestic uses of water 88 A2.5 Water quality and �sh production value 89 A3 Environment 92 A3.1 Aesthetics 93 A3.2 Land quality 93 A4 Other welfare 93 A4.1 Intangible user preferences 94 A4.2 Access time 94 A4.3 Impact on life decisions and behavior 95 A5 Tourism 97 A5.1 Introduction to tourism 97 A5.2 Tourism and sanitation 98 A5.3 Estimation of tourist losses due to poor sanitation 99 A5.4 Tourist sickness 100 A6 Impact mitigation associated with improved sanitation and hygiene 100 A6.1 Health impacts mitigated 101 A6.2 Other economic losses due to poor sanitation 101 A6.3 Market for sanitation inputs 102 A6.4 Market for sanitation outputs 103 A7 Uncertainty analysis 103 Annex B. Algorithms 106 B1. Aggregating equations 106 B2. Health costs related to poor sanitation and hygiene 106 B3. Water related costs associated with poor sanitation and hygiene 106 B4. Other welfare impacts 107 B5. Tourism losses 107 B6. Variable de�nition summary 107 Annex C. Recommendations for research 110 Annex D. Sanitation coverage by region and rural/urban grouping 111 Annex E. List of Contributing Stakeholders 112 Annex F. References 113 Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 11 CONTENTS List of Tables Table 1. Improved sanitation coverage statistics for Southeast Asian countries versus other developing 16 world regions Table 2. Comparison of sanitation types and coverage values measured in different national surveys in 18 Cambodia Table 3. Population size and provincial make-up of regions in Cambodia in 2005 23 Table 4. De�nition of ‘improved’ and ‘unimproved’ sanitation and water supply 24 Table 5. Aspects of sanitation included in the present ‘Sanitation Impact’ study 24 Table 6. Justi�cation for choice of impacts included in the study 26 Table 7. Categorization of impacts measured in the present study 27 Table 8. Financial and economic costs of poor sanitation measured in the study 28 Table 9. Potential bene�ts of different sanitation improvement options 29 Table 10. Financial and economic losses due to poor sanitation, 2005 32 Table 11. Table showing rural / urban breakdown for the main impacts, 2005 33 Table 12. Other non-quanti�ed impacts of poor sanitation 34 Table 13. Predicted �nancial and economic gains from improved sanitation 36 Table 14. Summary health impacts by disease due to poor sanitation and hygiene 37 Table 15. Total health care costs by disease, 2005 38 Table 16. Total productivity costs1, 2005 38 Table 17. Total costs of premature death, using human capital approach, 2005 39 Table 18. Total health-related costs 40 Table 19. Water resources in Cambodia 41 Table 20. Daily release of polluting substances to inland and ground water bodies 42 Table 21. Selected river water quality measurements in Cambodia from PPWSA sources, 2006 43 Table 22. Selected water quality measurements in Cambodia, from the Mekong River Commission, 2005 44 Table 23. Drinking water access costs 45 Table 24. Water access costs for domestic uses (drinking water excluded) 46 Table 25. Fish catch value – current actual and estimated loss 47 Table 26. Annual time spent accessing latrines 49 Table 27. Volume and importance of tourist sector in Cambodia 51 Table 28. Economic impact of lower tourist numbers 52 Table 29. Annual sanitation input market values 53 Table 30. Annual sanitation output market values 53 Table 31. Results of one-way sensitivity analysis – economic variables 54 Table 32. Results of one-way sensitivity analysis – sanitation links 55 Table 33. Results of one-way sensitivity analysis – impact mitigation 55 Table A1. Diseases linked to poor sanitation and hygiene, and primary transmission routes and vehicles 67 Table A2. Importance of sanitation and hygiene-related diseases, total cases and total deaths (2005) 68 Table A3. Distribution of morbidity and mortality by age group, year 2005 69 Table A4. Diarrheal disease incidence in Cambodia 70 Table A5. Current and estimated counterfactual underweight prevalence rates in children under 5 71 Table A6. Relative risk of mortality from mild, moderate and severe underweight in children under �ve 72 Table A7. Relative risk of illness from moderate and severe underweight in children under �ve 72 Table A8. Estimated cause-speci�c annual deaths in children under �ve in 2005 73 Economic Impacts of Sanitation in Cambodia 12 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) CONTENTS Table A9. Demographic and mortality data in 2005 74 Table A10. Estimated annual cases of illness in children under �ve (thousand cases) 74 Table A11. Percent of total under-�ve child mortality attributable to poor sanitation 75 Table A12. Percent of cases of illness in children under �ve attributable to poor sanitation 75 Table A13. Treatment seeking behavior, by provider 76 Table A14. Estimated numbers of disease cases attributed to poor sanitation and hygiene seeking care 76 from different providers, year 2005 Table A15. Health service use and unit costs associated with outpatient care 77 Table A16. Health service use and unit costs associated with inpatient care 78 Table A17. Variables for estimating amount of time lost from disease 79 Table A18. Comparison of alternative sources of time value (year 2005) 80 Table A19. Estimated number of annual deaths from poor sanitation & hygiene 81 Table A20. Unit values for cost of a premature death 83 Table A21. Selected drinking water quality standards 84 Table A22. Proportion of untreated sewage discharged to water bodies 85 Table A23. Waste load production per capita per day, subdivided by gray water and sewage, for urban 86 households with pipe connection Table A24. Numbers of households by main source of drinking water 87 Table A25. Unit cost of water 88 Table A26. Lack of latrine – indicators of defecation conditions 94 Table A27. Toilet access, by geographical location sub-type 95 Table A28. Water and sanitation coverage in schools (2005-2006) 96 Table A29. Male / female participation rates in school and workplace (2005-2006) 97 Table A30. Inputs for calculating the �nancial losses in tourist receipts 100 Table A31. Summary of meta-analysis results on WSH intervention efficacy for diarrheal disease reduction 101 Table A32. Unit prices of sanitation improvement options (per household) 102 Table A33. Input values for estimation of returns to re-use of human waste 103 Table A34. Alternative assumptions and values used in one-way sensitivity analysis 104 Table A35. Alternative assumptions for links between poor sanitation and impacts 105 Table A36. Alternative assumptions for impact mitigation 105 Table B1. Subscripts used in algorithms 107 Table B2. Variables used in algorithms 108 Table B3. Parameters used in algorithms 109 List of Figures Figure 1. Primary and economic impacts associated with improved sanitation options (human waste) 26 Figure 2. Economic cost of premature death at different unit values for premature death1 41 Figure 3. Contribution of different costs to total cost, by disease 42 Figure 4. Perceived importance of improved latrine to households 50 Figure 5. Female school drop-out rate vs. school sanitation 51 Figure 6. Modeled relationship between dissolved oxygen levels and �sh production (with lower and 93 upper range) Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 13 Basic Information on Cambodia Population (2005) Total population 13,806,974 Rural population 11,565,127 Urban population 2,241,847 Under 5 population (% of total) 12.3% Female population (% of total) 51.5% Urban population (% of total) 16.2% Currency Currency name Riel Year of cost data presented1 2005 Currency exchange with US$ 4,050 Exchange rate date 1-Jun-07 GDP per capita (US$) 447 GDP per capita (PPP) 2,929 Sanitation (2005) % improved rural 15.7% % improved urban 56.1% % urban sewage connection 28.9% 1 Except tourism loss where the cost is estimated based on 2006 �gure WSP-EAP Research Report February 2008 1 Introduction INTRODUCTION 1.1 Human development and sanitation Pollution and poor health are still major problems in Cambodia, hindering human development in the country. Among others, pollution of water resources and the surrounding environment due to poor sanitation and the improper management of solid waste is still a major challenge. The release of sewage and gray water into water bodies and the seepage of latrines into ground water have contaminated water sources making them unsafe for untreated consumption and affected productivity of water resources. Rates of disease due to poor sanitation and hygiene are amongst the highest in the region. In addition, the mismanagement of solid waste has damaged the aesthetics particularly in urban areas and caused environmental concerns among inhabitants. 1.2 Sanitation in Cambodia Unarguably, sanitation is lagging behind other global development goals. In 2004, 59% of the world’s population had access to improved sanitation, which represents a 10% increase from 49% global coverage in 1990. However, due to population growth, the global population without improved sanitation has decreased only marginally from 2.7 to 2.6 billion over a 14 year time period [1]. Table 1 shows, according to sanitation coverage data from the WHO/UNICEF Joint Monitoring Program (JMP), that Cambodia has the lowest sanitation coverage in Southeast Asia. Table 1. Improved sanitation coverage statistics for Southeast Asian countries versus other developing world regions Rural (%) Urban (%) Total (%) Country 1990 2004 1990 2004 1990 2004 SOUTHEAST ASIA Cambodia - 8 - 53 - 17 Indonesia 37 40 65 73 46 55 Lao PDR - 20 - 67 - 30 Malaysia - 93 95 95 - 94 Myanmar 16 72 48 88 24 77 Philippines 48 59 66 80 57 72 Singapore - - 100 100 100 100 Thailand 74 99 95 98 80 99 Timor-Leste - 30 - 66 - 33 Vietnam 30 50 58 92 36 61 TOTAL 40 56 70 81 49 67 OTHER REGIONS East Asia 7 28 64 69 24 45 South Asia 8 27 54 63 20 38 West Asia 55 59 97 96 81 84 Oceania 46 43 80 81 54 53 Latin America & Caribbean 36 49 81 86 68 77 North Africa 47 62 84 91 65 77 Sub-Saharan Africa 24 28 52 53 32 37 CIS 63 67 92 92 82 83 Source: http://www.wssinfo.org/ Economic Impacts of Sanitation in Cambodia 16 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) INTRODUCTION In 2004, only 8% of the rural population in Cambodia had access to improved sanitation while other countries reached at least 20%. In addition, urban areas of Cambodia have improved coverage of 53% compared to more than 70% in other Southeast Asian countries. In 2004, it is estimated that more than 11 million people are living without access to improved sanitation in Cambodia. Comparing the sanitation coverage with water supply coverage, it is noted that water supply coverage at 41% exceeds sanitation coverage by more than two times. Given that rural coverage of improved sanitation is very much lower than urban coverage, the national policy on rural water and sanitation sets the vision: “Every person in rural communities has sustained access to safe water supply and sanitation services and lives in a hygienic environment by 2025�. Obviously, this vision emphasizes the need for more investment in sanitation in rural areas. What is more, the Cambodia Millennium Development Goal (CMDG) target for sanitation coverage is set to be 30% for rural areas and 74% for urban areas, to be achieved by 2015 [2]. Based on current coverage and existing resource allocations to sanitation, however, it is unlikely that these targets will be reached by 2015. It is stated that at the current rates of latrine construction, it will take about 24 years to reach the 2015 target, and another 130 years to reach universal rural sanitation coverage [3]. Therefore, the sanitation improvement efforts need to be sped up if the universal rural sanitation target set by the national policy is to be achieved by 2025. It should be noted that the statistics reported by the JMP diverge from national statistics. JMP statistics are compiled according to international standards of what is ‘improved’, and are based on a comparison of different coverage sources which use different de�nitions and are often at variance with each other. It is worth noting that the JMP estimate for 2004, which gave 8% in rural areas in Cambodia, was based on the 1998 Census and the 2000 Demographic and Health Survey. In Cambodia, various data sources are available for sanitation coverage each of which may have different interpretation of improved and unimproved latrine. Three sources of data are listed in Table 2, which are the Cambodia Socioeconomic Survey (CSES) 2004, Cambodia Inter-censal Population Survey (CIPS) 2004, and Cambodia Demographic and Health Survey (CDHS) 2005. It is observed that different surveys give different results of sanitation coverage. However, the CDHS tends to be more reliable compared to other source of data. It can be observed that CDHS (2005) has utilized the JMP de�nitions of improved and unimproved sanitation1. Moreover, the sanitation coverage �gure given by CDHS (2005) is considered as the official �gure to be used in Cambodia. According to CDHS, the sanitation coverage also varies by regions in the country. While improved sanitation is largely available in Phnom Penh zone (84% improved), the unimproved sanitation is highest in the Plateau zone (16% improved). 1 The latrine is improved if it is used only by household members and not shared with others, and if the system can separate human waste from human contact. The types of facilities that are likely to achieve this may consist of flush or pour flush to piped sewer/septic tank/pit latrine, ventilated improved pit (VIP) latrine, pit latrine with slab, and composting toilet. The unimproved sanitation facility, however, refers to the shared latrine and other types of latrine which do not effectively separate human waste from human contact. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 17 INTRODUCTION Table 2. Comparison of sanitation types and coverage values measured in different national surveys in Cambodia Improved sanitation (%) Unimproved sanitation (%) Survey Septic Pit Public Pit House Total Open Other Total tank latrine toilet latrine CSES (2004) Rural 1.5 12.7 2.3 16.5 0.3 - 45.7 37.4 83.4 Urban 30.2 28.5 2.4 61.1 0.5 - 15.6 22.9 39.0 Total 7.3 15.8 2.4 25.5 0.3 - 39.7 34.5 74.5 CIPS (2004) Rural 4.5 8.0 3.2 15.7 - - - 84.3 84.3 Urban 33.5 15.1 5.8 54.4 - - - 45.6 45.6 Total 8.6 9.0 3.6 21.2 - - - 78.8 78.8 CDHS (2005) Rural 1.1 12.6 2.0 15.7 - 0.9 78.1 5.2 84.2 Urban 28.9 25.8 1.4 56.1 - 0.7 32.3 10.8 43.8 Total 5.2 14.5 1.9 21.6 - 0.9 71.4 5.9 78.2 Despite low access to improved sanitation, the National Strategic Development Plan 2006-2010 (NSDP), a well- prepared document on the strategies for development of Cambodia, slightly mentions about sanitation in its priority areas. It is noted that while the targets for achieving better access to improved sanitation are clearly de�ned in NSDP, there is no clear action plan to achieve those targets. Moreover, the budget allocation in sanitation sector is not explicitly given. Based on NSDP, while the budget for sanitation improvement is partly included in water and sanitation sub-sector, the budget for rural sanitation seems to be included in rural development sub-sector. According to the classi�cation, the water and sanitation sub-sector covers more than just water supply and sanitation2, whereas rural development sub-sector does not clearly include rural sanitation3. In this sense, it is hardly seen how much the budget is really allocated to sanitation. According to NSDP, the total allocated budget between 2006 and 2010 for water and sanitation and rural development is respectively US$150 million and US$350 million. The former amounts to 4.3% of the total budget of the NSDP, while the latter is roughly 10%. Despite these amounts, it is generally perceived that water supply and sanitation investment lags behind the investment in other sectors, which can be explained by the low political pro�le of sanitation in terms of government prioritization and funding, limited government budget, the lack of recognition of the many costs to society of poor sanitation, and a higher demand for investments in domestic water supply. As well as lack of top-down investment in the sector, the opportunities for attracting private sector engagement in the �nancing and provision of sanitation services are not sufficiently enabled, especially the potential for contribution by small-scale entrepreneurs. To date most sanitation investment has been by household self-provision. In this sense, it is important that the sanitation impact study is conducted to provide sound evidence to policy makers about the cost of neglecting the sanitation issue in the country, and the bene�ts of investing in sanitation. 2 Water and sanitation: sector policy and planning, water supply and sanitation, river development, waste management, education and training, and other (Source: The Cambodia Aid Effectiveness Report 2007) 3 Rural development and land management: rural sector policy and administration, forestry, land management and spatial planning, landmine clearance, and other (Source: The Cambodia Aid Effectiveness Report 2007) Economic Impacts of Sanitation in Cambodia 18 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) INTRODUCTION 1.3 Measuring the economic impact of sanitation To generate concrete evidence, the present study will examine some of the major negative economic consequences of poor sanitation. Until now, many of these consequences are understood at a general level, but there is often a lack of data to support assertions that poor sanitation imposes a considerable burden on society, or else data are context speci�c. Without such information, policy makers are unable to act. Furthermore, while some impacts of sanitation are now better understood, such as health impacts, many of the stakeholders that need to become convinced of the importance of sanitation are not directly concerned with health. Hence a range of potential impacts need to be examined and presented so that stakeholders see the multiple negative impacts of sanitation, and thus become convinced that concerted action is needed from several sectors. The study, therefore, attempts to look into the impact on health, water resource, environment, tangible and intangible user preference, and tourism. Despite the attempt, this study has faced several challenges in attempting to both meet scienti�c criteria, as well as present evidence that is useful for national as well as local policy makers. Studies examining the economic impact of pollution rely on a whole range of different types of evidence in order to quantify the impact related to one speci�c cause of pollution. First, there is a severe lack of routine information systems or research studies that indicate the overall level of the impact. Therefore, in the absence of generalisable �eld studies, relationships must be modeled and assumptions must be made. Second, valuation of impacts in economic terms must rely on sometimes crude techniques, using both market and shadow prices, which can change over time thus adding uncertainty to results. Third, when there are multiple causes of pollution, then a portion of the overall economic impact estimated must be apportioned to the component of pollution (i.e. poor sanitation in the case of the present study). In this regard, methodologies needed to be developed for this present study that allowed estimation of economic impacts that are both realistic and scienti�cally sound. 1.4 Study goal and target audience The speci�c goal of the present ‘Sanitation Impact’ study is to provide decision makers at country and regional level with better evidence on the negative economic impacts of poor sanitation, and to provide tentative estimates of those negative impacts that can be mitigated by investing in improved sanitation. The target audience is primarily national level policy makers with influence the overall allocation of resources to sanitation, including central ministries (budgeting, economics, �nance), line ministries (infrastructure, sanitation, water, rural development, urban planning) and external funding and technical partners (multilateral, bilateral and non-government agencies). The study is also targeted at sub-national decision making levels where results and conclusions of this study are also relevant. The study results disaggregate impacts by provincial groupings for each country, as well as providing a rural-urban breakdown. However, to inform local decisions, further studies are needed that disaggregate at provincial, city, and district levels, and below. Hence, the study presented here is a situation analysis, whose primary aim is to mobilize the different stakeholders and partners inside and outside the sanitation sub-sector to use better quality and comprehensive evidence in allocating resources to the sanitation sub-sector. In order to provide timely evidence, the study uses an evaluation methodology that draws largely on existing data sources available from governments, donors, non-government agencies and the scienti�c literature. The data gaps and weaknesses identi�ed in this study enable recommendations for future strengthening of routine information systems and priority areas for scienti�c research to allow better estimation of sanitation impact in the future. Separate reports and policy briefs are available for each country (www. wsp.org/pubs/index.asp). This current report provides a synthesis of the major �ndings from the four participating countries4. 4 A study on-going in Lao PDR will be published later. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 19 INTRODUCTION The results of this �rst study will contribute to the design and execution of a second study under the Economics of Sanitation Initiative, whose primary purpose is to evaluate the comparative costs and bene�ts of alternative sanitation improvement options in a range of country contexts. This second study is based on the rationale that decision makers need to know which sanitation improvements provide the best value for money, what the overall costs and bene�ts are, and who is willing or able to �nance the improvements. These studies together will provide an improved evidence-base for the efficient planning and implementation of sustainable sanitation options in the East Asia and Paci�c region. A focus in this present study on sanitation, and not water per se, is justi�ed for two main reasons: 1. Water has historically received greater emphasis than sanitation, in terms of research, policy development, programmatic support, as well as resource allocation. The WHO / UNICEF Joint Monitoring Programme estimates that in the 1990s, water received US$12.6 billion annually, while sanitation received US$3.1 billion, a factor difference of 4 times [4]. In Asia, the factor difference between spending on water and spending on sanitation in the same period is 5.5 times. As a result of this skewed spending, sanitation is lagging behind other global development goals: 59% of the world’s population had access to improved sanitation in 2004, compared to 83% for access to improved water supply [1]. 2. Poor sanitation practice is the starting point for many of the observed negative impacts of poor water and sanitation. For example, water quality is affected by poor sanitation; hence by improving sanitation, the quality of water for human consumption and productive purposes is improved. Also, the major share of water, sanitation and hygiene-related diseases are fecal-oral in nature, which means that they are transmitted because the sanitation practice fails to isolate the pathogens from contact with humans. Hence, this present study is a �rst attempt to comprehensively evaluate the impacts of poor sanitation in Cambodia. Many of these impacts are quanti�able in economic terms. Other impacts which are less tangible or less easy to evaluate are also potentially important for economic development, quality of life, and political decision making. This study is the �rst application of a comprehensive sanitation impact evaluation methodology developed by the World Bank Water and Sanitation Program [5]. Based on the experiences of this present study, the methodology will be revised for application in other countries and regions of the world. Economic Impacts of Sanitation in Cambodia 20 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) WSP-EAP Research Report February 2008 2 Study Methodology Overview Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) STUDY METHODOLOGY OVERVIEW The Sanitation Impact study follows a standardized peer reviewed methodology in all �ve countries [5]. Differences in the quality and level of detail of data in the �ve countries required adaptations to the methodology to arrive at the same output data on economic impacts. This section describes: 1. The levels and units of analysis (2.1). 2. Which aspects of ‘sanitation’ are included in this study (2.2). 3. How impacts are classi�ed and which are included and excluded in the study (2.3). 4. An overview of how the different economic impacts of unimproved sanitation are measured (2.4). 5. The methods used for predicting the economic bene�ts associated with improved sanitation (‘Impact mitigation’) (2.5). Annex A describes the detailed methods for estimating the economic impacts of unimproved sanitation, and how methodological weaknesses and uncertainty in input variables are evaluated in sensitivity analysis. 2.1 Levels and units of analysis The primary aim of this study is to describe and quantify sanitation impacts at national level, in order to inform policy makers about the overall negative impacts of poor sanitation and the potential bene�ts of implementing different types of sanitation improvement in Cambodia. The ultimate usefulness of these overall economic impacts of poor sanitation is to serve as the basis for estimation of what impacts can be mitigated from improving sanitation. It is key to note in the interpretation of the results of this study that the gains from improving sanitation will be less than the losses from unimproved sanitation, given that (a) sanitation interventions do not have 100% effectiveness to reduce adverse health outcomes associated with poor sanitation, (b) poor sanitation is one of many causes of water and environmental pollution. The aim of the study is to present impacts in disaggregated form, to aid interpretation and eventually policy recommendations. Geographical disaggregation of results is presented for some types of economic impact, i.e. at the regional level. Rural/urban disaggregations are made for all countries for impacts where feasible. Furthermore, health impacts are disaggregated by age groups for selected diseases and descriptive gender analyses are conducted for selected impacts. The study uses a modeling approach and draws almost exclusively on secondary sources of data. The study presents impacts in terms of both physical units, and converts these to monetary equivalents using conventional economic valuation techniques. Results on economic impact are presented for a single year – the latest available data were for 2005 for most variables, while for some variables 2006 was the latest year. Overall impacts are presented in terms of per capita impacts in United States Dollars. For those impacts where quanti�cation in economic terms is not feasible using secondary data sources, the impacts are examined and reported descriptively. Table 3 shows the population size and provincial make-up of each region of Cambodia. Economic Impacts of Sanitation in Cambodia 22 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) STUDY METHODOLOGY OVERVIEW Table 3. Population size and provincial make-up of regions in Cambodia in 2005 Population size (‘000) Provinces contained Zone Rural Urban Total Phnom Penh 567 747 1,314 Phnom Penh Plains 5,336 247 5,584 Kg Cham, Kandal, Prey Veng, Svay Rieng, Takeo Banteay Meanchey, Battambang, Kampong Thom, Siem Tonlé Sap 3,607 649 4,255 Reap, Kampong Chnange, Pursat Coastal 738 303 1,041 Kampot, Kep, Sihanoukville, Koh Kong Plateau/ Kampong Speu, Kratie, Modulkiri, Preah Vihear, 1,318 296 1,614 Mountain Ratanakiri, Stung Treng, Oddormeanchey, Pailin Total 11,565 2,242 13,807 Source: Population projection 1998-2020, National Institute of Statistics It is noted that two of the �ve zones – the Plains zone and Tonlé Sap zone – account for roughly two-thirds of the Cambodian population of 13.8 million in 2005. The national population growth rate from 2004 to 2005 is estimated at nearly 2%. It is also observable that Phnom Penh zone has had the highest population growth rate (3.3%) followed by plateau and mountain zone (2.5%), and coastal zone (2.3%). 2.2 Scope of sanitation In conducting an impact study of poor sanitation, it should be clear what aspects of sanitation are being assessed given that ‘sanitation’ has relevance for many aspects of life. Furthermore, what actually constitutes improved sanitation will vary across countries and cultural contexts. In the international arena, the sanitation target adopted as part of the Millennium Development Goals focuses on the disposal of human waste, thus leading to a narrower understanding of the term ‘sanitation’. Table 4 presents de�nitions used by the WHO/UNICEF Joint Monitoring Programme for improved and unimproved water supply and sanitation. However, this study has recognized that other aspects of sanitation are also relevant to the economic impacts being measured in the present study, and in line with the broader de�nition of sanitation used in this study: “the hygienic disposal or recycling of waste, as well as protection of health through hygienic measures�. The broader de�nition of sanitation refers to management of human and animal excreta, solid waste, other agricultural waste, toxic waste, wastewater, food safety, and associated hygiene practices. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 23 STUDY METHODOLOGY OVERVIEW Table 4. De�nition of ‘improved’ and ‘unimproved’ sanitation and water supply Intervention Improved Unimproved 1 Sanitation • Flush or pour-flush to: • Flush or pour-flush to elsewhere • Piped sewer system • Pit latrine without slab or open pit • Septic tank • Bucket • Pit latrine • Hanging toilet or hanging latrines • Ventilated Improved Pit-latrine • No facilities or bush or �eld • Pit latrine with slab • Composting toilet Water supply • Piped water into dwelling, plot, or yard • Unprotected dug well • Public tap / standpipe • Unprotected spring • Tubewell/borehole • Cart with small tank/drum • Protected dug well • Tanker truck • Protected spring • Bottled water • Rainwater collection • Surface water (river, dam, lake, pond, stream, canal, irrigation channels) Source: This table reflects the updated de�nition of improved and unimproved sanitation and water supply presented in the 2006 JMP report [1]. 1 De�ned as being unimproved due to being unsafe or costly While it is understood that sanitation is often more broadly de�ned than the components listed above, it was not possible to apply a broader de�nition in the present study due to time and resource constraints. Hence, issues such as drainage, flood control measures, hospital waste, industrial waste, and broader environmental health such as food hygiene, air pollution and vector control, were not included. Table 5 summarizes the aspects of sanitation included and excluded from this study. Table 5. Aspects of sanitation included in the present ‘Sanitation Impact’ study Included Excluded • Practices related to human excreta: • Drainage and general flood control • Quality, safety and proximity of latrine system measures • Disposal or treatment of waste and impact on the • Industrial effluents, toxic waste and (inhabited) outdoor environment medical waste • Hygiene practices • Air pollution unrelated to human excreta • Practices related to disposal or treatment of gray water • Vector control • Practices related to disposal or treatment of household solid waste • Broader food safety • Practices related to use or disposal of animal excreta • Broader environmental sanitation 2.3 Impact identi�cation and classi�cation Poor sanitation has many actual or potential adverse impacts on populations as well as national economies. On the reverse side of the same coin, different measures for improving sanitation can go some way to mitigating those negative impacts, hence stimulating economic growth and reducing poverty. Figure 1 presents a range of possible impacts of sanitation, as they relate to �ve key aspects of human excreta management: latrine access, latrine system, hygiene practices, waste disposal, and waste re-use. The major links are shown with arrows: links between the sanitation option and the primary impact (between left-hand and central boxes); and links between the primary impact and the resulting economic impact (between central and right-hand boxes). In the context of Cambodia, Economic Impacts of Sanitation in Cambodia 24 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) STUDY METHODOLOGY OVERVIEW not all impacts are relevant in the present study. The impacts such as entry fee, house price rises, cottage industry, and Foreign Direct Investment (FDI) will not be considered in the study because although important, they are very minor compared to other major impacts of sanitation. Figure 1. Primary and economic impacts associated with improved sanitation options (human waste) 1 Comfort, convenience, security, privacy 2 Visual effects, smells 3 HRQL: health-related quality of life Based on the exhaustive set of impacts shown in Figure 1, a shortened list of negative impacts of poor sanitation to be included in the present study was selected, shown in Table 6 above. These impacts are classi�ed under �ve main categories: health impacts, water resource impacts, environmental impacts, other welfare impacts, and tourism impacts. Table 6 also provides further justi�cation for inclusion of these impacts in the study, showing the presumptions based on preliminary evidence of importance [6] and discussion with country partners. Based on available evidence, the major anticipated impacts of poor sanitation were on health and water resources, and therefore greater focus was given on data collection for these impacts in Cambodia. Annex A provides further background on these impact categories. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 25 STUDY METHODOLOGY OVERVIEW Table 6. Justi�cation for choice of impacts included in the study Impact Link with sanitation Justi�cation for inclusion in the present study Health - Poor sanitation and - Scienti�c evidence is available on the causal pathways hygiene cause diseases, between unimproved sanitation/hygiene and the causative which lead to a range disease pathogens/hosts of direct and indirect - Health information systems, household surveys and economic effects economic studies testify to the diseases suffered by the population and the associated costs of disease Water - Released human and - Unregulated sewage release into water bodies is a proven animal excreta pollutes signi�cant contributor to inland (and marine) water resource water resources which pollution affects its usability or - Water is treated or purchased by households, and productivity and leads to undergoes costly treatment by piped water providers for costly averting behavior domestic and commercial purposes and/or production impact - Households hauling water themselves travel further to access a cleaner, safer water supply - Fish are unable to reproduce and survive in heavily polluted water. At lower levels of pollution, �sh numbers are affected by oxygen depletion and micro-bacteria - Humans are affected when they eat �sh that have been exposed to raw sewage External - Neighborhoods with - Land and building prices are highly sensitive to environment poorly managed sanit- environmental factors ation are less pleasant to - Poor people tend to live on marginal land live in, and population welfare is thus affected - As income rises, households are willing to pay more for better sanitation services Other - Poor sanitation results - Household members have to spend time accessing toilet in welfare from cultural barriers, low the open (nature) or queuing to use shared or public facilities awareness, lack of design - Privacy and convenience are underestimated ‘intangible’ options, low income, and aspects in sanitation choices lack of home ownership - There exists an income gradient in latrine ownership - Poor sanitation in institutions affect life - Sanitation is more important to people who lack voice in choices, or lead to household or community decisions – women and children absenteeism at school or the workplace Tourism - Poor sanitation affects the - Tourism is an important source of national income and attractiveness of tourist employment, offering high returns on investment destinations and tourist - The most popular tourist destinations have clean arrivals; and can lead to environments, good toilet facilities, and a lower risk of holiday sickness getting sick Table 7 details the speci�c impacts to be examined under health, water resources, external environment, other welfare impacts and tourism. The columns indicate the �ve key components of sanitation assessed (refer to Table 6) for the different impacts. Human excreta management is relevant for all impact areas. Poor hygiene mainly affects health. Gray water mainly affects water resources. Solid waste mainly affects the external environment and tourism. Economic Impacts of Sanitation in Cambodia 26 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) STUDY METHODOLOGY OVERVIEW Also, potential impacts of improved sanitation – the stimulation of local markets for sanitation inputs (labour, materials) and the reuse of waste for productive purposes – are also included in the Table 7. Table 7. Categorization of impacts measured in the present study1 Human Hygiene Gray Animal Solid Impact Sub-impacts waste practices water waste waste 1. Health Health status √ √ Disease treatment costs √ √ Productive time loss √ √ Premature death √ √ 2. Water resources Water quality √ √ √ Drinking water √ √ √ Domestic uses of water √ √ √ Fish production √ √ √ 3. External Aesthetics √ √ √ environment Land use and quality √ √ √ 4. Other welfare Intangible aspects √ Time used for toilet access √ Life choices √ 5. Tourism Tourist numbers √ √ √ √ Tourist sickness √ √ √ 6. Sanitation markets Sanitation ‘inputs’ √ Sanitation ‘outputs’ √ √ 1 A tick shows which impacts were measured in this study. The absence of tick does not indicate that no empirical relationship is anticipated; only that it was not evaluated in this study. 2.4 Estimation methods for �nancial and economic costs of poor sanitation Policy makers are interested to understand the nature of the economic impacts being measured. For example, do the impacts have immediate implications for expenditure and incomes by households or governments, or are the effects non-pecuniary or longer-term in nature? The answer will naturally affect how the results are interpreted, and what level of support there will be for impact mitigation measures. Hence, while recognizing the difficulties in distinguishing different types of economic impact, this present study attempts to distinguish broadly between two different types of impact – �nancial and economic: • Under �nancial costs, those costs which are most likely to affect quanti�ed indicators of economic activity in the short term were included. Financial costs include changes in household and government spending as well as impacts likely to have real income losses for households (e.g. health-related time loss with impact on household income) or enterprises (e.g. �shery loss). It should be noted that, while these ‘�nancial’ costs affect economic activity indicators in the short term, these impacts are not expected to directly affect Gross Domestic Product, due to substitution effect, transfer payments, and so on. • Under economic costs, other costs were added to the above �nancial costs to approximate the overall population welfare impact of poor sanitation. These include the longer-term �nancial impacts (e.g. less educated children, loss of working people due to premature death, loss of usable land, long-term tourist losses), as well as non- �nancial implications (value of loss of life, time use of adults and children, intangible impacts). Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 27 STUDY METHODOLOGY OVERVIEW Table 8 describes which cost components were included for �nancial and economic de�nitions of cost for each sub- impact. It should be noted that costs were those attributed to poor sanitation using an attribution factor (variable by impact). Some costs were non-quanti�ed, as indicated in Table 8. The detailed methods of impact estimation are described in Annex A. Table 8. Financial and economic costs of poor sanitation measured in the study Sub- Impact Financial costs attributable Economic costs attributable to impacts category to poor sanitation poor sanitation evaluated 1. Health Health care Marginal health seeking costs, Full costs of health seeking, including costs including patient transport, full health care and patient transport medication cost in public sector, and costs private sector tariffs Productivity Income loss due to lost adult working Welfare loss due to adult and child costs days due to sickness sickness time Premature Short-term household income loss Discounted lifetime income losses for mortality due to adult death (1 year) adult & child death 2. Water Drinking Financial costs of water treatment Financial + Time spent hauling water resources water costs and distribution from safe water sources Domestic Additional expenditure sourcing Financial + Time spent hauling water water uses water from non-polluted sources from less polluted sources Fish losses Lost sales value due to reduction in Lost sales value due to reduction in �sh catch �sh catch 3. External Land quality - Economic value of land made environment unusable by poor sanitation 4. Other Time loss - Welfare loss due to adult & child welfare latrine travel/waiting time Work/school - Temporary absence of women from absence work and girls from school 5. Tourism Tourism - Revenue loss from low occupancy costs rates and failure to exploit long term potential tourist capacity 2.5 Impact mitigation Having estimated the �nancial and economic costs of poor sanitation, from a policy viewpoint it is important to know by how much these costs can be reduced by implementing improved sanitation options. Indeed, while this study initially presents total costs attributed to poor sanitation, it is unlikely that this total value can be averted by improving sanitation. While there are many types and con�gurations of sanitation improvement available, this present study aims to estimate potential bene�ts obtainable for a selected number of features of sanitation improvements. This study provides an initial tentative estimate of the likely gains possible from improving sanitation using different options. It is the aim of the second study of ESI to estimate the costs and bene�ts of speci�c sanitation options, which are the most relevant policy options in each country context. Table 9 shows the �ve main features of sanitation improvement (in columns) assessed in this study, and the relevance of these for each sub-impact category (in rows). The features are described in the table footnotes. The impact mitigation estimation methods are described in Annex A6. Economic Impacts of Sanitation in Cambodia 28 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) STUDY METHODOLOGY OVERVIEW Table 9. Potential bene�ts of different sanitation improvement options A B C D E F Impacts Latrine Improved Hygiene Waste Waste Tourism physical toilet practices3 treatment reuse5 access1 system2 or disposal4 Health √ √ √ Water resources √ Environment Aesthetics √ √ Land quality √ Other welfare Intangible effects √ √ √ Access time √ Life choices √ √ √ Tourism √ Sanitation markets Sanitation inputs √ √ √ √ Sanitation outputs √ 1 Close and improved latrine for those using open defecation; improved population:toilet ratios through increased coverage of latrines (less queuing time) 2 Improved position or type of toilet seat or pan; safe, private and secure structure: walls / door / roof; improved & safe collection system (tank vault, pit); improved ventilation; improved waste evacuation 3 Availability of water for anal cleansing; safe disposal of materials used for anal cleansing; hand washing with soap; toilet cleaning 4 Improved septic tank functioning and emptying; sealed top of pit latrine to withstand flooding; household connection (sewerage) with treatment; sewers with non-leaking pipes and a drainage system that can handle heavy rains; wetlands or wastewater ponds 5 Urine separation, composting of feces, hygienization; use of human excreta products in commercial aquaculture, composting (fertilizer); biogas production (anaerobic digestion) 2.6 Uncertainty This study has faced several challenges in attempting to both meet scienti�c criteria and present evidence that is useful for national as well as local policy makers. In order to provide timely evidence on sanitation impact, the present study is based on entirely secondary information collected from a variety of sources, and combined with assumptions where necessary input data were missing. Therefore, in order to �ll the gaps in evidence, several innovative and not previously tested methodologies were developed for this present study. Quantitative information were combined using the methodology outlined above and in Annex A to estimate the impacts of poor sanitation and the potential bene�ts of improving sanitation. A number of impacts were excluded from quantitative estimation, which are described in Chapter 3. Three major types of uncertainty surround the quantitative �gures presented in this study: 1 Uncertainty in the input values for the estimation of overall economic impacts, such as in the epidemiological variables (for health) and economic variables such as market prices and economic values. In fact, there is a severe lack of data available from routine information systems or research studies to feed into the quantitative model. Hence, in the absence of these data, relationships were modeled and assumptions made. 2 Uncertainty in the attribution of the overall impact to poor sanitation. For example, when there are multiple sources of pollution, a portion of the overall economic impact estimated must be apportioned to the Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 29 STUDY METHODOLOGY OVERVIEW component of pollution being examined (e.g. domestic waste contribution to overall water pollution). A second example is the importance of poor sanitation in keeping away tourists from a country. 3 Uncertainty in the actual size of impact mitigation achievable. The variables with greatest importance for the quantitative results were evaluated further in one-way sensitivity analysis by varying a single input value over a reasonable range, to assess the impact on overall �ndings. Alternative values used in the sensitivity analysis are provided in Annex A7. Economic Impacts of Sanitation in Cambodia 30 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) WSP-EAP Research Report February 2008 3 Economic Impact Results Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ECONOMIC IMPACT RESULTS 3.1 Summary of economic impacts of poor sanitation Table 10 summarizes the estimated �nancial and economic impacts of poor sanitation and hygiene for each impact category in 20055. The total annual �nancial loss due to poor sanitation and hygiene is about US$160 million which is equivalent to US$12 per capita. It is noted that a substantial part of the �nancial loss due to poor sanitation and hygiene is accounted for by the water access costs (US$147 million) sharing more than 90% of total �nancial costs. The rest of �nancial losses are attributed to health impacts which amount to US$13 million. The annual economic loss from poor sanitation and hygiene is estimated at around US$448 million amounting to about 7.2% of the Cambodia’s GDP in 2005. The per capita economic cost is around US$32 compared to the GDP per capita of about US$450. It should be noted that in contrast to �nancial losses where health-related losses are much less than water access costs, health cost (US$187 million) is the main contributor (42%) to the economic costs of poor sanitation and hygiene. The water costs are the next major loss in economic impacts of poor sanitation amounting to about US$149 million and making one third of the total costs. The economic loss of tourism due to poor sanitation and hygiene is roughly US$74 million which shares about 16% of the total economic costs. The economic value loss in time use attributed to poor sanitation access is US$38 million, sharing about 9% of the total costs. Table 10. Financial and economic losses due to poor sanitation, 2005 Financial losses Economic losses Impact Value Per Value Per (million capita % (million capita % US$) (US$)1 US$) (US$)1 Health costs 13.3 1.0 8.3 187.1 13.6 41.8 Health care costs 10.7 0.8 6.7 13.4 1.0 3.0 Productivity costs 2.5 0.2 1.5 5.1 0.4 1.1 Premature death costs 0.2 0.0 0.1 168.6 12.2 37.6 Water costs 146.8 10.6 91.7 149.0 10.8 33.3 Drinking water 92.7 6.7 57.9 93.8 6.8 20.9 Domestic water uses 9.8 0.7 6.1 10.9 0.8 2.4 Fish production 44.4 3.2 27.7 44.4 3.2 9.9 Other welfare - - - 38.2 2.8 8.5 Tourism - - - 73.7 5.3 16.4 TOTAL 160.1 11.6 100 448.0 32.4 100 1 Per capita refers to the total value divided by the total population in 2005 Table 11 shows the �nancial and economic losses of poor sanitation by rural/urban breakdown. However, since not all impacts are available for the breakdown some costs are non-assigned to rural/urban (i.e. health, �sh, and tourism). 5 Except tourism loss where 2006 �gure is used. Economic Impacts of Sanitation in Cambodia 32 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ECONOMIC IMPACT RESULTS Table 11. Table showing rural / urban breakdown for the main impacts, 2005 Financial losses Economic losses Impact Per Per Value (million Value (million capita1 % capita1 % US$) US$) (US$) (US$) Health costs 13.3 1.0 8.3 187.1 13.6 41.8 Water costs 146.8 10.6 91.7 149.0 10.8 33.3 Rural 74.6 5.4 46.6 76.6 5.5 17.1 Urban 27.8 2.0 17.4 28.1 2.0 6.3 Non-assigned 44.4 3.2 27.7 44.4 3.2 9.9 Other welfare - - - 38.2 2.8 8.5 Rural - - - 35.3 2.6 7.9 Urban - - - 3.0 0.2 0.7 Tourism - - - 73.7 5.3 16.4 TOTAL 160.1 11.6 100.0 448.0 32.4 100.0 Rural 74.6 5.4 46.6 111.8 8.1 25.0 Urban 27.8 2.0 17.4 31.0 2.2 6.9 Non-assigned 57.7 4.2 36.0 305.1 22.1 68.1 1 Per capita loss for rural-urban breakdown and non-assigned is equal to loss divided by total population. As well as quanti�ed, monetized impacts, there are a number of other key impacts which have not been valued in this present study, and which should be taken into account in interpreting the quantitative impacts discussed above. These non-monetized impacts include suffering from disease, intangible aspects of environmental impacts (aesthetics) and other welfare, time loss from seeking private place for urination (especially women), loss from marine �sheries, the non-use value of clean water resources such as ‘existence’ and ‘bequest’ values, and the losses to wildlife from polluted water resources and an unclean environment. Other impacts with less clear linkages with poor sanitation include the use of water for irrigation purposes and hence agricultural productivity, the impact of poor sanitation on foreign direct investment, and impact of unimproved sanitation (and running water) in institutions which affect life decisions of the population, especially the decision of women to take employment and of girls to enroll in or complete school. Table 12 below shows the non-quanti�ed impacts of poor sanitation. Together, the quanti�ed and unquanti�ed �nancial and economic losses will affect the overall economic situation in a country, including economic growth. The main effects are likely to be through sickness time and income loss associated with premature death, and household expenditure on health care and clean water, including water treatment. The production and sale of sanitation options can also give a stimulus to the local economy through local employment; and re-use of human and animal excreta can lead to cost savings and higher productivity at household level. Property prices also may rise due to better living standards brought by improved sanitation. However, given the weak empirical evidence on the direct economic effects of improved sanitation, this study did not move beyond a partial equilibrium analysis to examine redistributive effects. Therefore, the empirical link between poor sanitation and macro-economic indictors such as gross domestic product (GDP) is still not known. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 33 ECONOMIC IMPACT RESULTS Table 12. Other non-quanti�ed impacts of poor sanitation Impact Sub-impact Excluded items 1. Health Quality of life Sanitation-related diseases cause pain and suffering beyond the purely economic effects. Disability-adjusted life-years (DALY), which attempt to capture quality of life loss, indicate Other sanitation- The following disease and health conditions have been excluded, but in related diseases some contexts are potentially important: 1. Helminthes 2. Reproductive tract infections for women bathing in dirty water 3. Dehydration resulting from low water consumption from lack of access to private latrines (especially women) 4. Speci�c health problems suffered by those working closely with waste products (sanitation workers, dump scavengers) 5. Health impacts due to flooding that are not in reported national health statistics 6. Education impacts of childhood malnutrition 7. Food poisoning due to contaminated �sh (e.g. E Coli) 8. Animal and insect vectors of disease (e.g. rodents, mosquitos) 9. Animal health related to human sanitation 10. Avian influenza 2. Water Household water Household time spent treating drinking water, including boiling, resources use maintaining rain water collection systems, replacing �lters, etc. Fish production The study included only the value of recorded freshwater �sh production. However, the following are not included: 1. Non-recorded marketed freshwater �sh 2. Marine �sh 3. Subsistence �shing losses 4. Nutrient losses Water Economic losses associated with flooding from lack of drainage management Irrigation Polluted surface water may lead to extraction of scarce groundwater, or use of polluted water has implications for plant growth, animal health, and eventually human health 3. External Aesthetics Welfare loss from population exposure to open sewers and open environment defecation Land value Economic value of land made unusable by poor sanitation 4. Other Intangible impacts Welfare loss from lack of comfort, privacy, security, convenience, status, welfare prestige Time loss Time for urination, especially women Life decisions and Poor sanitation in schools and the workplace affect daily attendance, absence from daily especially of girls and women activities 1. Loss of time from temporary absence of women from workplace 2. Welfare loss from school absence 3. Work decisions and early drop-out of girls from school 5. Tourism Tourist sickness Expenditure by tourists becoming sick and welfare loss of sick tourists 6. Other Foreign direct Companies seeking to make investments may be affected in their investment decision, among other factors, by the sanitation situation in a country; tangible secondary evidence is however very limited. Economic Impacts of Sanitation in Cambodia 34 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ECONOMIC IMPACT RESULTS Economic growth results from the combination of many of the bene�ts listed above. The main effects are likely to be through lower disease rates, increased time availability, easier access to and reduced treatment of suitable water sources for domestic, agricultural and industrial use, and more tourism and foreign direct investment. The production and sale of sanitation options can also give a stimulus to the local economy through local employment; and re-use of human (and animal excreta) can lead to cost savings and higher productivity at household level. Property prices also may rise due to better living standards brought by improved sanitation. However, given the weak empirical evidence on the direct economic effects of improved sanitation, and the lack of studies linking local effects to the macro-economy, the empirical link between sanitation and macro-economic indictors such as gross domestic product (GDP) is not known. 3.2 Summary of economic gains from improved sanitation Besides looking at the economic loss due to poor sanitation and hygiene, it is also important to look into the possible economic gain from improved sanitation and hygiene. Five options are given in the Table 13 below where each option corresponds to the bene�t only to some impact categories. In general, all of the costs measured in Section 3.1 can be averted from one or more or the improvement options; except that is, the health bene�ts, given that basic improved sanitation or hygiene only reduces by a proportion of 30-50% of the overall sanitation and hygiene related diseases. In addition, the results from the sanitation market studies (inputs and outputs) are presented in more detail here. From the table, it can be seen that: • Hygiene practices bring US$6 million �nancial gain and US$84 million economic gain, through reducing by 45% the measured health impacts. • Better physical access of latrines and more private as opposed to shared latrines and open defecation practice bring nearly US$38 million economic gain, through saving time for those whose time access is not already minimized. Moreover, the input market for construction of latrine leads to �nancial and economic bene�ts of US$0.3 million. • Improved toilet system leads to more than US$4 million �nancial gain and US$60 million economic gain, mainly through the reduction by 32% of the measured health impacts. Sanitation markets are included under category C, leading to both signi�cant �nancial and economic bene�ts of about US$1 million. • Treatment or disposal of human excreta leads to US$147 million �nancial gain and close to US$223 million economic gain, reflecting the total reversal of the estimated losses due to water and tourism impacts of poor sanitation. • The reuse of human excreta is estimated to lead to nearly US$2 million �nancial and economic gains where US$1.3 million is the input market bene�t (biodigester construction) and the rest is the gain from the use of biogas for cooking and lighting. This estimate is based on relatively conservative assumptions about the numbers of households using biodigesters, and utilizing the commodity prices shown in Annex A6.3 and A6.4 Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 35 ECONOMIC IMPACT RESULTS Table 13. Predicted �nancial and economic gains from improved sanitation A B C D E Impact area Hygiene Latrine physical Improved Treatment or Reuse (million US$) practices access toilet system disposal Fin. Econ. Fin. Econ. Fin. Econ. Fin. Econ. Fin. Econ. Health 6.0 84.2 4.3 59.9 Health care1 4.8 6.0 3.4 4.3 Productivity1 1.1 2.3 0.8 1.6 Premature 0.1 75.9 0.1 54.0 death1 Water 146.8 149.0 Drinking water 92.7 93.8 Domestic uses 9.8 10.9 Fish production 44.4 44.4 Other welfare - 38.2 Time use2 - 38.2 Tourism - 73.7 Tourist - 73.7 numbers3 Sanitation - - 0.3 0.3 1.2 1.2 - - 1.8 1.8 markets Input markets4 - - 0.3 0.3 1.2 1.2 - - 1.3 1.3 Output markets 0.6 0.6 TOTAL 6.0 84.2 0.3 38.6 5.5 61.1 146.8 222.7 1.8 1.8 1 Improved sanitation generally means improved physical access, improved toilet system and treatment or disposal, all of which have implications for health status. For the purposes of reporting, the sanitation bene�ts are included only under improved toilet system, to avoid confusion over double-counting. 2 Other welfare impacts discussed and presented in Chapter 4.5 are all presented here under ‘latrine physical access’ but some of these are also likely due to improved toilet system. 3 Tourist numbers will also be related to hygiene practices and toilet systems used, but the bene�ts are reported here under treatment and disposal, as this has the major environmental implications. 4 All interventions involve a market value Given that sanitation improvements (see Table 4 in section 2.1) have several features of the above categories, it is possible to add together the savings associated with the categories above. For example, by installing a private sanitary pit latrine would lead to bene�ts A and B; and with sewage treatment would add D. Ecosan could lead to A, B, D and E. Adding hygiene interventions enhances (increases) the health effect from sanitation improvements alone. For example, research shows that hygiene interventions have an average 45% reduction in diseases, compared to 32% reduction for sanitation (latrine) interventions alone (see Annex A6.1). In practice, there exists uncertainty in the extent of savings, because in some cases the improvement is not fully effective in mitigating the costs, especially in the case of water resources: • Water for drinking and domestic uses. The study apportioned sanitation-speci�c costs of water pollution using an attributable fraction, based on the release of BOD from different sources. However, this is an imperfect indicator, especially of household behavior in relation to mitigation measures concerning domestic water supply. Various extreme arguments could be put forward concerning the degree of costs mitigated. Economic Impacts of Sanitation in Cambodia 36 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ECONOMIC IMPACT RESULTS o On the one hand, it could be argued that households will still treat their water even in the absence of human waste and bacterial risk, due to habit, taste, and other water pollutants, both natural (e.g. silt) and manmade (e.g. pesticides, industry). Hence very few costs may in fact be mitigated through complete isolation or treatment of sewage. o On the other hand, the bacterial content of water is one of the major health risks from the majority of water sources, including groundwater. The absence of sewage release may make it less necessary for households to treat their water, depending on other types of pollutants in the water that can be effectively removed at the household level. It should also be noted that water sources used by treatment plants that contain sewage can increase considerably the unit costs of treatment, which are passed onto the consumer. • Fish production. This study has found (a) considerable uncertainties in the link between water pollution from poor sanitation, dissolved oxygen levels, and �sh production, and (b) weak routine information systems linking monitoring stations for water quality indicators with physical locations of �sh farming. Hence, �sh production impact was difficult to estimate scienti�cally; and by implication, it is difficult to know the expected gains to �sh production from reducing the fecal load in �shing waters. On the other hand, the more direct link – �sh contamination through exposure to sewage and its �sh and human health implications – was not measured in this study, but could be a considerable impact associated with the isolation of human excreta. 3.3 Health impacts Poor sanitation and hygiene in Cambodia remains a substantial impact on health causing morbidity and deaths as shown in Table 14. It is estimated that the number of cases associated with poor sanitation and hygiene in 2005 totaled roughly more than 9.5 million cases. Of those cases, 97% are diarrheal diseases. Moreover, it is observed that the total deaths related to poor sanitation and hygiene is conservatively close to 10,000 in 2005, most of them are resulted from diarrheal diseases (67%), while the rest are from ALRI (18%), malaria (10%), and measles (4%) respectively. In addition to data on cases and deaths, these diseases can have other signi�cant impacts on the quality of life of sufferers, in the sense that the pain and suffering of the diseases make people uneasy and uncomfortable. In more severe cases, the diseases can restrict the daily activities and cause economic costs. Other long-term effects of associated health problem can be also foreseen as having economic costs. For instance, malnutrition which is also sourced from diarrheal infections attributed to poor sanitation and hygiene may also affect children educational attainment and work performance in the future. Table 14. Summary health impacts by disease due to poor sanitation and hygiene Cases Deaths Disease 1 Reported Estimated Total1 Reported 2 Estimated Total1 Diarrheal diseases 621,353 9,364,210 99 6,600 Skin disease 101,393 144,596 - - Malnutrition 574 852 - - ALRI 50,164 159,706 926 1,786 Measles - - 1 420 Malaria 1,295 19,108 282 1,033 1 With adjustment for attribution to poor sanitation and hygiene 2 Without adjustment for attribution to poor sanitation and hygiene Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 37 ECONOMIC IMPACT RESULTS Table 15 presents the total costs of health care related to treatment seeking for diseases related to sanitation and hygiene. From the table, the health care cost related to poor sanitation and hygiene is nearly US$11 million for �nancial loss and more than US$13 million for economic loss. It is observed that health care cost involving private clinics is the highest loss, both �nancially and economically. It shares about 66% of the �nancial loss and about 53% of the economic loss. What is more, the diarrheal diseases which are strongly related to sanitation and hygiene appear to incur the largest economic cost among other diseases. It incurs �nancial losses of up to approximately US$10 million and economic losses of US$12 million. The latter shares roughly 92% of the total health care cost (economic) attributed to poor sanitation and hygiene. Table 15. Total health care costs by disease, 2005 Financial costs (thousand US$) Economic costs (thousand US$) Disease Public Private Trans- Public Private Trans- Self- Total Total clinics clinics port clinics clinics port treatment Diarrhea 789 6,810 2,288 9,887 2,041 6,810 2,288 1,161 12,300 Skin disease 130 36 198 364 239 36 198 5 477 Malnutrition 5 - 3 8 27 - 3 - 30 ALRI 55 184 121 360 136 184 121 22 462 Measles - - - - - - - - - Malaria 41 27 22 90 69 27 22 - 118 Total 1,021 7,056 2,632 10,709 2,511 7,056 2,632 1,188 13,388 Table 16 presents the total costs of productivity related to days off daily activities for diseases related to sanitation and hygiene. It is noted that costs are incurred for all age groups. For children, the diseases may keep them away from school which has some economic costs. For adult population, the diseases will have �nancial and economic costs as they may lose their working time when getting sick. However, not all adults suffer direct �nancial implications by losing income due to sickness, as some may still be able to go to work despite the disease. In this case, only 70% of sick adults are assumed to directly lose income due to day taken off sick (see Annex A1.5 for further details). It can be observed from the table that the productivity loss due to poor sanitation and hygiene related diseases is roughly US$2.5 million �nancial loss, and US$5 million economic loss. It is also noteworthy that the economic cost induced by the loss of productivity mainly occurs among adults of over 15 (US$2.8 million) which accounts for nearly 55% of the total loss. This loss is followed by the productivity losses among under-�ve children and their caretakers which contribute about 37% to the total economic costs. In terms of disease, diarrhea appears to be the main disease incurring the highest economic cost amounting to 98% of total productivity cost. Table 16. Total productivity costs1, 2005 Financial costs Economic costs, by age group (thousand US$) Disease (age group 15+) (thousand US$)2 0-4 5-14 15+ Total Diarrhea 2,459 1,760 437 2,775 4,972 Skin disease - - - - - Malnutrition - 1 - - 1 ALRI - 84 - - 84 Measles - - - - - Malaria - 12 - - 12 Total 2,459 1,858 437 2,775 5,070 1 Value of time is average wage approximated by average compensation of employees 2 Perspective of the household only – actual earnings lost of working adults Economic Impacts of Sanitation in Cambodia 38 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ECONOMIC IMPACT RESULTS Table 17 presents the total costs of premature death from diseases related to sanitation and hygiene. The value of premature death is based on the human capital approach. The �nancial costs of premature death are assumed to be one year loss of income, for adults only. In this sense, it is estimated that premature death associated with poor sanitation and hygiene costs about US$168,000 in �nancial losses, and US$169 million in economic losses resulting from deaths of both children and adults. The �nancial loss of premature death is only caused by diarrhea as this involves only adults over 15 years of age, with malnutrition and malnutrition-related diseases for adults being omitted from the study. For the economic loss, however, the main disease leading to high premature death cost is diarrhea sharing about 67% of the total economic cost. The diseases which follow are those associated with malnourished children under �ve years such as ALRI (18%), malaria (11%), and measles (4%). Additionally, it can be noted that the premature death among under-�ve children imposes the highest economic cost among other age groups. According to the table below, it shares about 92% of the total economic cost. Regarding diarrheal deaths, the under-�ve age group appears to also impose the highest economic loss compared to other age groups amounting up to US$100 million while it is only US$13 million for other age groups combined. One of the reasons for the high cost of under-�ve death from diarrhea is the fact that the prevalence of diarrheal disease in the under-�ve is signi�cantly higher than the over-�ve age group, while the value of life estimated from human capital approach is in a comparable range. Table 17. Total costs of premature death, using human capital approach, 2005 Financial costs Economic costs, by age group (thousand US$) Disease (age group 15+) (thousand US$) 0-4 5-14 15+ Total Diarrhea 168 100,030 9,103 3,716 112,849 Skin disease - - - - - Malnutrition - - - - - ALRI - 30,767 - - 30,767 Measles - 7,234 - - 7,234 Malaria - 17,793 - - 17,793 Total 168 155,824 9,103 3,716 168,643 Figure 2 shows the variation in cost of premature death by using different values for premature death. The losses estimated using the human capital approach give the most conservative value. The willingness-to-pay approach using the value-of-statistical-life (VSL) estimated at purchasing power parity (PPP) with income elasticity 1.0 gives the highest estimate which is up to US$1.4 billion compared to only US$0.17 billion for human capital approach. The Willingness to Pay (WTP) approach estimated at official exchange rate (OER) with income elasticity of 0.6, 0.8, and 1.0 give the value of premature death costs of US$1.25, US$0.5, and US$0.2 billion respectively. Therefore, the losses are highly sensitive to the method used. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 39 ECONOMIC IMPACT RESULTS Figure 2. Economic cost of premature death at different unit values for premature death1 1,600 VSL (OER) with I.E. 1.0 1,400 VSL (OER) with I.E. 0.8 Total cost of premature death VSL (OER) with I.E. 0.6 1,200 VSL (PPP) with I.E. 1.0 (million US$) 1,000 Human capital approach 800 600 400 200 0 1 Refer to Table 19 for unit values used. VSL = Value of saved life; I.E. = income elasticity; OER = official exchange rate; PPP = purchasing power parity Table 18 shows a summary of the �nancial and economic costs of different health impacts due to poor sanitation and hygiene in Cambodia. The estimation shows that total health-related �nancial cost is US$13 million, while the economic cost is US$187 million. Health care cost accounts for more than 80% of the �nancial loss, while it accounts for only 7% in the economic loss. The majority of economic cost is accounted for by premature death which shares up to 90% (US$169 million) of the health-related economic costs. In terms of disease, diarrhea is the main �nancial loss accounting to 94% of the total �nancial losses. In addition, it is also a major contributor to economic costs amounting to nearly 70% of total costs. ALRI and malaria are the next contributors to the total economic losses sharing 17% and 10% of the total costs respectively. Table 18. Total health-related costs Total �nancial costs (million US$) Total economic costs (million US$) Disease HC PROD DEATH Total HC PROD DEATH Total Diarrhea 9.9 2.5 0.2 12.5 12.3 5.0 112.8 130.1 Skin disease 0.4 - - 0.4 0.5 - - 0.5 Malnutrition 0.008 - - 0.008 0.030 0.001 - 0.031 ALRI 0.4 - - 0.4 0.5 0.1 30.8 31.3 Measles - - - - - - 7.2 7.2 Malaria 0.1 - - 0.1 0.1 0.0 17.8 17.9 Total 10.7 2.5 0.2 13.3 13.4 5.1 168.6 187.1 Note: HC = Healthcare costs; PROD = Productivity costs; DEATH = Premature death costs Figure 3 shows the contribution of different costs to overall cost, by disease. It is seen that premature death is the main economic cost for diarrhea and diseases associated with malnutrition (indirect effects). More notably, the premature death takes the extremely high share for malnutrition-related diseases compared to other health-related costs. However, health care costs are the most important costs for malnutrition and skin diseases. It is also remarked that the losses which take patients or caretakers away from productive activities have mere contribution to the economic costs of any disease. Economic Impacts of Sanitation in Cambodia 40 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ECONOMIC IMPACT RESULTS Figure 3. Contribution of different costs to total cost, by disease 100% 80% 60% 40% 20% 0% Diarrhea Skin disease Direct Malnutrition Malnutrition related diseases DEATH PROD HC 3.4 Water resource impacts The economic impacts of polluted water resources depend on three main factors: the extent of water resources in the country, the release of polluting substances in water resources, and the actual or potential uses of water in the country. Table 19 presents a summary of the water resources in Cambodia. The Mekong River and Tonlé Sap lake are the main sources of water for both drinking water supply and freshwater �shing in the country. Table 19. Water resources in Cambodia Rivers & major canals Regions Length (km) Surface area (km2) Flow (m3 per second) 66,700 (wet) Mekong River Plains 540 N/A 1,250 (dry) 1,125 (wet)1 Sesan River Plateau 252 N/A 213 (dry)1 2,292 (wet) Sekong River Plateau N/A N/A 428 (dry) 924 (wet) Sre Pok River Plateau N/A N/A 241 (dry) 13,000 (wet) Tonlé Sap Lake Tonlé Sap N/A 70 billion m3 total volume 2,500 (dry) Source: Ministry of Environment, Institute of Geography of Vietnamese Institute of Technology, Mekong Secretariat; 1 Voeun Sai Station Despite the fact that Cambodia has many water resources – surface water, ground water and rain water – most land- based water resources suffer from pollution due to human activities. The water bodies near the cities or populated areas are usually more polluted than the remote water body due to excessive discharge of pollutants generated by densely populated human settlements. To estimate the total pollution of water sources both surface water and ground water, Table 20 below gives the volume of the major polluting substances associated with human excreta. Based on the assumption of sewage leakage discussed in Annex A2.2, it can be estimated that there is more than 230 tons of feces and more than 2,300 m3 of urine released to the water bodies every day. The release of human Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 41 ECONOMIC IMPACT RESULTS waste generates about 500 tons of BOD per day to the water bodies in the country. Moreover, more than 8,000 m3 of gray water is discharged to water bodies every day through sewage system. Table 20. Daily release of polluting substances to inland and ground water bodies Total release (volume)1 Polluting substances Region Feces (Tons) Urine (m3) Graywater (m3) BOD (Tons) Phnom Penh 29 287 607 58 Plains 71 713 1,777 196 Tonlé Sap 77 765 3,100 150 Coastal 26 257 1,305 37 Plateau 31 314 1,365 57 Total 234 2,335 8,154 497 1 Assumption on sewage leakage is given in Table A22 All of the substances given above are considered the main pollutants to water resources from poor sanitation. However, given the large volume of water in the country, these pollutants may be partly diluted naturally. Despite this, the water quality of the surface water body is still unsafe for untreated consumption due to the presence of bacteria. Table 21 and 22 show the water quality measurement in Cambodia from various surface water sources from two different institutions, namely Phnom Penh Water Supply Authority (PPWSA) who supply drinking water to Phnom Penh citizens, and the Mekong River Commission (MRC) who tests water quality at different locations along the Mekong river and its tributaries. According to PPWSA data presented in Table 21, the DO levels at 3 different locations appear to vary from 3.4 to 5.5 mg/l. This low DO indicates the sign of pollution in the surface water bodies. Moreover, the presence of high thermotolerant coliform in the water proves that the water sources in Cambodia are contaminated by bacteria originating from human and animal feces. While the standard requires thermotolerant coliform be zero to make water safe to drink (See Annex 2.1 Table A21), the water bodies that are used to supply Phnom Penh citizens have the thermotolerant coliform ranging from 400 to 3,500 cfu/100ml in dry season and from 1,300 to 9,000 cfu/100ml in wet season. The amount of thermotolerant coliform also varies by location of water sources, and the closer the water sources to the cities the more is thermotolerant coliform concentrated. Economic Impacts of Sanitation in Cambodia 42 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ECONOMIC IMPACT RESULTS Table 21. Selected river water quality measurements in Cambodia from PPWSA sources, 2006 Water quality indicators Water body Total Location characteristics Thermotolerant DO TSS coliform pH coliform (cfu/ (mg/l) (mg/l) (cfu/ 100ml) 100ml) Chroy Mekong River (Upstream Changva Phnom Penh) Water level: 2-10m Wet season 7.4 5.5 174.5 17,121 1,317 Flow: 2800-36100m3/s Water level: 2-5m Dry season 7.7 4.4 30.8 59,51 416 Flow: 1700-5900m3/s Bassac river (Downstream Chamkamon Phnom Penh) Wet season N/A 7.6 N/A 186.0 24,750 2,994 Dry season N/A 7.4 N/A 68.2 58,871 1,241 Phum Prek Tonlé Sap (Dual flow)1 Water level: 2-9m Wet season 7.4 3.4 119.8 23,509 8,693 Flow: 85-8900m3/s Dry season N/A 7.3 4.4 54.9 71,162 3,476 Source: PPWSA 2006 for water quality, and MoWRM for river characteristics 1 The flow direction of Tonlé Sap is from the Mekong toward the lake in wet season, and from the lake toward the Mekong in dry season. N/A refers to the fact that data are not available. Based on data from the Mekong River Commission (MRC) presented in Table 22, the DO levels at various water body locations vary from 5.1 to 7.5 mg/l in dry season and from 6.3 to 7.4 mg/l in wet season. Although this DO is still lower than the standard requirement (See Annex 2.1 Table A21), the values measured by MRC appear to be higher than that of PPWSA for the same water source location. This can be caused by the different time of test or location of sampling. The similar pattern that can be observed in the two data sources is that the Total Suspended Solid (TSS) in all water locations is higher in wet season than in dry season. This can be explained by the fact that in the wet season, due to flood and rain, more soil is flushed into the Mekong river thus leading to increased TSS. It is noted that TSS, although contributed by the release of human excreta, may be caused by the erosion as a result of deforestation in Cambodia and the upstream. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 43 ECONOMIC IMPACT RESULTS Table 22. Selected water quality measurements in Cambodia, from the Mekong River Commission, 2005 Water quality indicators Location and Water body date Characteristics DO CODMN TSS pH (mg/l) (mg/l) (mg/l) Phnom Krom Tonlé Sap Lake Lake Wet season 6.9 6.5 4.6 661.3 Dry season 6.9 5.1 5.3 214.0 Prek Kdam Tonlé Sap (Dual flow) River Water level: 2-9m Wet season 7.0 6.4 2.8 78.8 Flow: 85-8900m3/s N/A Dry season 6.7 5.9 6.0 48.3 N/A Kratie Mekong (upstream) River WL: 17-21m Wet season 7.2 6.3 2.6 92.6 Flow: 28500-42800m3/s WL: 6-11m Dry season 7.3 7.3 4.8 26.2 Flow: 3300-10290m3/s Chroy Changva Mekong (midstream) River Water level: 2-10m Wet season 7.2 7.4 1.6 99.7 Flow: 2800-36100m3/s Water level: 2-5m Dry season 7.2 7.5 6.3 20.0 Flow: 1700-5900m3/s Takhmao Bassac River Wet season 7.0 6.6 3.2 94.5 Dry season 6.8 6.8 5.6 23.3 Khaom Samnor Mekong (downstream) River Wet season 7.1 7.3 3.7 100.0 Dry season 7.3 7.1 5.4 13.8 Source: Mekong River Commission, 2005 One of the major impacts of polluted water in wells, springs, rivers and lakes is that populations and water supply agencies will have to treat water, or treat water more intensively, for safe human use. Alternatively, populations and water supply agencies can access cleaner water from more distant sources, thus increasing access costs. Those who do not take precautionary measures are exposed to higher risk of infectious disease, or poisoning due to chemical content. Table 23 shows the costs attributed to poor sanitation of access to drinking water, based on an assumed minimum daily intake of 4 liters per capita. It is also important to note that this attribution may overestimate the costs actually associated with poor sanitation, as in the absence of poor sanitation, the costs may not be reduced given there still exist other sources of water pollution. The table below indicates that, due to poor sanitation, the drinking water access costs nearly US$93 million �nancially and about US$94 million economically in 2005. Of all those costs, household water treatment costs have the largest share in both �nancial and economic costs, which are more than 85% of the total. The costs for purchased non-piped water appear to be the next contributors to �nancial and economic costs amounting to close to US$12 million (13% of total costs). The welfare loss of hauling water due to the polluted local source is roughly US$1 million. Economic Impacts of Sanitation in Cambodia 44 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ECONOMIC IMPACT RESULTS Table 23. Drinking water access costs Financial (million US$) Economic (million US$) Water source Value % Value Purchased piped water 0.2 0.3 0.2 0.3 Rural 0.2 0.2 0.2 0.2 Urban 0.06 0.1 0.06 0.1 Purchased non-piped water 11.6 12.5 11.6 12.4 Rural 4.9 5.3 4.9 5.2 Urban 6.8 7.3 6.8 7.2 Household water treatment 80.8 87.2 80.8 86.2 Rural 62.0 66.9 62.0 66.1 Urban 18.9 20.3 18.9 20.1 Hauled water - - 1.1 1.1 Rural - - 1.0 1.0 Urban - - 0.1 0.1 Total 92.7 100.0 93.8 100.0 Rural 67.0 72.3 68.0 72.5 Urban 25.7 27.7 25.8 27.5 As well as water for drinking, water is required by households for basic living – cooking, laundry, household cleaning and washing up of utensils, personal hygiene activities (showering, bathing), and sanitation (if water is used to flush waste away). The study used an estimated minimum requirement of 28 liters per person per day. However, since the volume of water used also depends on how people get the water, it is sensible to assume that those purchasing non-piped water need only 14 liters per person per day. This assumption will lead to a more conservative estimate of the domestic water access costs (see Annex A2.4). Households may in fact use more than this for these basic items, as well as for other purposes such as watering plants, domestic animals, leisure activities, and rituals. Table 24 shows the costs attributed to poor sanitation of accessing domestic water from improved water sources. Although there are not the same strict requirements for water quality, households still may walk further for improved water and pay companies to deliver or piped water for non-drinking domestic uses. In spite of this, it is uncommon for Cambodian people to treat their water for non-drinking purpose. From the table, it can be seen that the water access costs for domestic uses is roughly US$10 million in �nancial term and US$11 million in economic term. The purchase of non-piped water accounts for the majority in both �nancial costs and economic costs which shares about 88% and 79% successively of the respective costs. Purchased piped water constitutes 12% of �nancial losses and 11% of economic losses. Hauling water from distant sources also induces the welfare losses which worth of US$1.1 million contributing 10% to the total economic costs. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 45 ECONOMIC IMPACT RESULTS Table 24. Water access costs for domestic uses (drinking water excluded) Financial (million US$) Economic (million US$) Water source Value % Value % Purchased piped water 1.2 12.3 1.2 11.0 Rural 1.0 9.8 1.0 8.8 Urban 0.2 2.4 0.2 2.2 Purchased non-piped water 8.6 87.7 8.6 78.7 Rural 6.6 67.8 6.6 60.8 Urban 1.9 19.9 1.9 17.9 Hauled water - - 1.1 10.3 Rural - - 1.0 9.2 Urban - - 0.1 1.1 Total 9.8 100.0 10.9 100.0 Rural 7.6 77.7 8.6 78.8 Urban 2.2 22.3 2.3 21.2 The third aspect of water quality assessed quantitatively is the potential impact of poor water quality on �sh catch in inland (freshwater) water bodies. As stated above, Cambodia has abundant water resources which provide favorable conditions for �shing and aquaculture, and which makes the �shery sector a key economic sector in the country, with high contribution to GDP as well as to subsistence living. In Cambodia, all inland water sources are amenable for �sh production. The Tonlé Sap lake is the water body with the highest yield in �sh catch. According to the Economics Institute of Cambodia (EIC), the �shery sector contributes about US$525 million to the GDP in 2006. With the exported volume of 12,000 tons of fresh water �sh in 2006, the export value is estimated to be worth about US$20 million. Table 25 below shows the actual value of �sh catch in different regions compared with the predicted �sh catch under optimal water conditions. A portion of �sh catch is not reflected here due to the omission of subsistence �shing and �sh caught and bartered locally. To evaluate the �sh catch loss due to poor sanitation and hygiene, it is crudely assumed that the contribution of poor sanitation to water pollution is 65%. It is important to note that the DO value given in Table 25 is only indicative as it is based on the average DO for the latest available year. Indeed, MRC data on dissolved oxygen reflects a single day measurement per month; since the DO level may vary from one day to another, the monthly DO data are not necessarily fully reflective of average DO levels. Also, impact on �sh is also determined by minimum DO levels reached, and for how long these levels remain, and therefore average DO levels are only a very crude indicator of the state of the water bodies for �sh production. Furthermore, �sh are affected by more than just DO, but other parameters such as temperature, water flow, water depth, and the availability of food also affect �sh production. From the table, the �sh loss from pollution due to poor sanitation is worth of roughly US$44 million. Economic Impacts of Sanitation in Cambodia 46 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ECONOMIC IMPACT RESULTS Table 25. Fish catch value – current actual and estimated loss Estimated Actual value of Fish catch Dissolved Fish catch value of �sh lost compared Region Oxygen in 2005 �sh catch from poor to optimal (mg/l) (Tons) (million sanitation (%) US$)1 (million US$) Plains Region 50,500 84.3 2.7 Phnom Penh 7.46 9,000 15.0 100 - Kandal 7.46 21,500 35.9 100 - Kampong Cham 6.80 10,000 16.7 88 1.5 Prey Veng 6.67 5,000 8.3 87 0.8 Takeo 7.21 5,000 8.3 93 0.4 Tonlé Sap Great Lake Region 78,500 131.1 40.9 Kampong Thom 6.15 9,000 15.0 72 3.8 Siem Reap 5.77 15,000 25.0 65 8.8 Banteay Meanchey 5.77 3,000 5.0 65 1.8 Battambang 5.77 15,000 25.0 65 8.8 Pursat 5.77 15,000 25.0 65 8.8 Kampong Chnang 6.15 21,500 35.9 72 9.1 Upper Mekong Region 5,000 8.3 0.7 Stung Treng 6.80 2,500 4.2 88 0.4 Kratie 6.80 2,500 4.2 88 0.4 Total inland �shery production 134,000 223.7 44.4 Source: MAFF for �sh catch; MRC for Dissolved Oxygen 1 Estimation of �sh value is based on Consumer Price Index 2005 3.5 Environmental impacts Based on interviews with stakeholders, it is con�rmed that solid waste management in Cambodia has gradually been improved, although there is a lot more to be done to keep the country environmentally clean. Many areas of Cambodia’s cities, including Phnom Penh, are still without adequate waste collection service. Many tons of wastes are dumped into rivers and ponds, burned, or left uncollected to be scattered by animals, thus blocking the drains and creating unsanitary conditions. Waste collection is relatively weak in outlying areas of the cities, and unplanned settlements that are home to thousands of the city’s poorest families. Moreover, it is very common to see piles of waste at many market places dumped by the sellers and the households nearby. Besides, the official designated dump site of solid waste, particularly in Phnom Penh city where nearly 1,000 tons of waste is dumped everyday, has reached its capacity. This mismanagement of solid waste is known to cause an unpleasant living environment for many inhabitants. In addition, there are health hazards for the population from poorly disposed of solid waste, which until now has not been quanti�ed. The present study, therefore, attempts to qualitatively assess the impacts of solid waste in terms of aesthetics and land quality. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 47 ECONOMIC IMPACT RESULTS The greatest perceived impact of solid waste on aesthetics is the fact that waste produces odor, and spoils visual appearance, especially in towns and cities. In most towns and cities of Cambodia, household solid waste is usually disposed of in front of houses, on sidewalks, or in some cases on open land. Those wastes sometimes decompose prior to being picked up by waste collectors, thus producing bad smells to the surrounding environment. This polluted air quality creates unpleasant atmosphere to not only the households nearby, but also the pedestrians, the travelers, and particularly the tourists passing by the areas. In addition to the odor, the scattered wastes have damaged the visual aesthetics of many cities of the country, which make the cities less attractive to tourists (see Section 3.7). Besides household solid waste, the management of waste at most marketplaces has been very poor. In most cases, the wastes are untidily scattered around the sellers and at best loaded on the edge of their stall or in the entrance to the market. More often than not, those wastes produce bad odors and transform the market place into an unregulated dump sites for the households situated nearby. While a market should be a pleasant place which needs to be attractive to customers, the market with improper waste disposal keeps away its customers. This may have some economic losses to the sellers in the market. In addition to scattered household and market wastes, the impact of the so-called designated dumpsite on the nearby residents is even more severe. The ten-hectare dumpsite in Phnom Penh is situated not far from the residential areas. While bad odor from the dumpsite is disturbing the livelihood of the residents, the smoggy air pollution due to burning of waste can be harmful to the health of the residents as well as dumpsite scavengers. In addition, the dumpsite may contaminate ground water quality, and damage local land quality through the penetration and spillover of waste and chemically-contaminated water. It should be also noted that in Cambodia the designated dump sites are only available in urban areas. In rural areas, however, the waste is normally burnt or buried into the ground for disposal. It is argued that the land used for dump site is normally hard to convert to agriculture land. This is because the waste is usually mixed between the composted waste and the non-composted waste such as plastics and other materials. However, the conversion of the dump site for other selected non-agriculture purpose may be possible. Therefore, although the environmental impact is not easily quanti�able, its effects on human well-being, livelihood, and health are still considerable. Moreover, the loss of land value due to the unproductive post-dump site land may be signi�cant. 3.6 Other welfare impacts According to a recent study by Water and Sanitation Program (WSP) in three selected provinces and Phnom Penh, it is found that hygiene and cleanliness is the main perceived bene�ts of latrine [7]. It is noted that although the study covered only a small number of provinces, it is stated that the rural and urban areas in those selected study sites are broadly representative of rural and urban areas of Cambodia considering socio-economic and environmental conditions. Based on the study, more than 80% of urban and 70% of rural citizens recognize that improved latrine will provide better hygiene and a generally clean environment for living (Figure 4). In addition, comfort, health improvement, safety, and convenience are the next most perceived bene�ts of improving latrine. It is also claimed that privacy, improved family status and prestige are other advantages of having a latrine at home. Referring to Figure 4 below, rural and urban people tend to have similar pattern of perception regarding the bene�ts of latrine. Based on the �gure, however, there are less proportion of people in urban areas perceiving ‘improved health’ and ‘improved status/prestige’ than those in rural areas. For other perceived bene�ts, however, there is higher proportion of urban people than rural people. Economic Impacts of Sanitation in Cambodia 48 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ECONOMIC IMPACT RESULTS Figure 4. Perceived importance of improved latrine to households Hygiene/clean More comfortable Improved health Improved safety Convenience/save time %Urban More privacy % Rural Improved status/prestige 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% Source: Demand assessment for sanitary latrines in rural and urban areas of Cambodia, 2007 [7] As indicated in the WSP study, a latrine is perceived to save time for around 40% of Cambodian households [7]. Based on the time saving assumptions used in this present study, the total time spent in accessing for both open defecation and shared facility is calculated. According to Table 26, it is estimated that the total annual (2005) economic value lost of time spent accessing open defecation is about US$37.5 million and of shared facility is roughly US$0.7 million. These costs include both the adults and children’s welfare losses assuming adult’s time value is 30% of income, and children’s time value is 50% of adult’s. It should be noted that the evaluation of toilet access time provided below does not include the time for urination which can be a time-consuming daily activity for women. If this has been included, the access time cost would be much higher than the current estimate. Table 26. Annual time spent accessing latrines Population size Total access time Value (million US$) (million) (million hours) Location Shared Open Shared Open Shared Open Total facility1 defecation facility defecation facility defecation Rural 0.5 9.0 8.7 549.5 0.5 34.7 35.3 Urban 0.2 0.7 2.9 44.1 0.2 2.8 3.0 Total 0.6 9.8 11.6 593.5 0.7 37.5 38.2 1 Refers to population with shared facilities who are assumed to have inadequate toilet. Using by many people, the shared facilities cause waiting time to users. Besides access time loss, poor sanitation in schools also has potential implications for participation of girls in education – rates of enrolment, rates of school drop-out, and absenteeism. While there are many different reasons for school drop-out among school girls, the lack of toilet facility at school is potentially one of the reasons. However, according to the information collected, this factor has not been examined previously in Cambodia. Hence this current study attempts to explore this relationship by comparing across provinces the drop-out rate with absence of latrines in schools. According the Figure 5 below, it is interesting to note that the drop-out rates among school girls are high in the province with high rate of without-toilet schools. It can be noticed also from the �gure that the Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 49 ECONOMIC IMPACT RESULTS impact tends to be more sensitive for secondary school students as the drop-out rate is higher than that of the primary school students. This can be explained by the fact that when the girls are getting older, more privacy for toilet going is needed. It can be noted that in addition to Cambodia’s cultural values towards toilet-going, the lack of latrines at schools causes difficulties for female adolescents after menstruation has begun; hence lack of privacy for toilet-going during their the menstruation period may cause absences and eventually be a determining factor, among others, for female drop-out of school in Cambodia. Therefore, although this is not easily quanti�ed, the fact that girls cannot participate actively in the society will likely undermine their ability to independently make income in the future leading to various social problems and economic slow-down. Figure 5. Female school drop-out rate vs. school sanitation 50 Primary school 45 40 Secondary school 35 Drop-out rate (%) 30 25 20 15 10 5 0 0 10 20 30 40 50 60 70 80 % of school without latrine Source: Ministry of Education Youth and Sports, 2006 Besides school sanitation problem, there are many other factors such as budget allocation by government, school distance, poor age-appropriate entry, etc. which may explain the drop-out of school among pupils. The inadequate budget allocation to the education sector has led to little improvement in education system discouraging students from going to school. The number of schools is yet to be enough to accommodate the increasing number of population making classroom overcrowded with the pupil-class ratio of 43 nationwide in 2005. In addition, many schools being distant from the communities coupled with poor road infrastructure contribute to high drop-out rate of students, especially among female. Not only does poor sanitation negatively impact on education sector, but also on workplace where adequate sanitation and hygiene facility is important. It is recognized that poor sanitation affects the health of workers or employees, which in turn reduces the productivity of workers6. Therefore, while poor health negatively affects the quality of life of workers, the reduced productivity may have some economic costs for the employers and the country as a whole. In addition to poor toilets at schools and workplaces, public toilets in market places are often unpleasant for their unhygienic condition and odor. This unfavorable situation mainly affects the welfare of the sellers who spend most of their daytime in the market, and frequently use those public toilets. 6 An interview with an official at the Ministry of Labor and Vocational Training Economic Impacts of Sanitation in Cambodia 50 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ECONOMIC IMPACT RESULTS 3.7 Tourism impacts The growth in the number of tourist arrivals has contributed to the recent high economic growth in Cambodia [8]. The share of tourism in GDP has expanded from 11% in 2004 to nearly 15% in 2006. In absolute term, income from tourism has quickly grown from only US$580 million in 2004 to more than US$1 billion in 2006. In addition, the tourism sector has provided employment to 225,000 people in Cambodia contributing to about 3% of total employment in 2005. It is also seen that in addition to the increasing number of tourist arrivals, the average daily expenditure per tourist has been also increasing from US$87 in 2004 to US$95 in 2006. The average length of stay has also increased from 6.3 days per tourist in 2004 to 6.5 days in 2006. In addition, private investment in tourism sector has experienced sharp increase during the past few years. The total private investment reached US$352 million in 2006 which is more than three times the investment in 2005 and more than six times that in 2004. Table 27 below shows the importance of the tourism sector in Cambodia. Table 27. Volume and importance of tourist sector in Cambodia Variable 2004 2005 2006 Visitors arrivals to Cambodia by country of 1.05 1.42 1.70 residence (million) ASEAN 183,362 219,579 328,459 Asia & Oceania 413,000 574,443 702,967 Europe 242,811 310,006 314,194 The Americas 122,169 152,328 159,429 Africa & Middle East 26,017 76,644 80,301 Others 6,000 Preah Vihear1 67,843 88,615 108,691 Domestic tourists 4,251,270 5,278,113 7,901,039 Av. Expenditure per tourist (US$) 87 93 95 Average length of stay (days) 6.3 6.3 6.5 Tourist income – total (million US$) 578 832 1049 As % of GDP 10.9% 13.3% 14.6% Private sector investment in tourism- �xed 55.87 102.57 352 assets (million US$) Establishments supporting tourism Hotel 299 317 351 Guesthouse 615 684 742 Restaurants 713 719 747 Massages 56 56 53 Sporting Clubs 17 17 53 Souvenir shops 40 40 40 Travel agency and Tour operator 302 336 382 Employment in tourism 180,000 225,000 - As % total employment - 3% Source: Ministry of Tourism 1 Preah Vihear is a tourism attraction (temple) on the Cambodian-Thai border. Most tourists come to this area through Thailand as road infrastructure in Cambodian side is less accessible. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 51 ECONOMIC IMPACT RESULTS Given the tourism growth and its potential growth in the near future, it is important to address some of the challenges facing the tourism growth in Cambodia. One of them is sanitation. Poor sanitation in the country generally, and in tourist sites speci�cally, can have important implications on number of tourists visiting the country and their length of stay. Also, once tourists are on-site, they may get sick from a sanitation or hygiene-related disease, and thus experience a reduction in enjoyment of their holiday. Getting sick is a bad experience in itself, but it also wastes time in their holiday, and may incur some expenses related to treatment. Besides this, tourists having their enjoyment with the environment may be also spoiled due to bad sights and smells, if the area is polluted by human waste or solid waste. So, the tourists will be discouraged to come again, or the bad experience is spread among their friends and families which, in the long-run, results in less tourists coming to the country. According to Cambodia Association of Travel Agents (CATA), it is important that attractive places with high tourism potential be developed �rst. If the tourist places are not clean, less tourists will visit or tourists will be discouraged to come back again, so people living in that area will earn less income for their living. It is also highlighted that a common health problem occurring among most tourists in Cambodia is diarrhea which is related to unhygienic practice and poor food preparation in the country. Table 28 below shows some estimates for potential economic impacts nationwide of lower tourist number attributed to poor sanitation and hygiene. However, there is no information on the attribution of tourism to sanitation. In this case, it is assumed that the attribution of lower than optimal tourist hotel occupancy rates to poor sanitation is 10%. The annual tourist number increase is 20% and the target occupancy rate is 80% (see Annex A5.3 for further details). The economic cost is estimated using the gap between the current and potential tourist numbers through a faster growth rate in tourist numbers due in part to improved sanitation. Moreover, the annual discount rate of 3% is assumed in the estimation of present value of economic impacts. In this case, the economic cost is about US$74 million. It should be noted that the economic cost of the number of tourists getting sick attributed to poor sanitation is not evaluated in this study due to lack of data. Yet, it can be assumed that the fact that tourists falling sick will undermine Cambodia’s tourism prospects which, in the long-run, will affect the country’s economy as a whole. Table 28. Economic impact of lower tourist numbers Current Hotel occupancy rate Potential tourism (%) value Attribution to Annual economic Region value (Million sanitation loss1 (million US$) (million Current Target US$) US$) Cambodia 1,049 54.8 80 1,786 10% 73.7 Source: Ministry of Tourism 1 Calculated as the gap between current and potential, multiplied by the attribution to sanitation 3.8 Sanitation markets Table 29 below shows the potential annual market size for sanitation inputs, based on market prices. From the table, the total annual input market value is about US$2.8 million including both labor and materials. It is noteworthy that the input market value of the EcoSan (biodigester plant construction) is very high compared to others sharing about 45% of the total input market value. It is also important to note that there is more economic gain than just the �nancial gain given in Table 29 below. This can be explained by the fact that once there is a need to improve latrine, more employment is generated for latrine providers and builders. This employment generation will provide income opportunity for local people which bene�ts the community and the country as a whole. Economic Impacts of Sanitation in Cambodia 52 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ECONOMIC IMPACT RESULTS Table 29. Annual sanitation input market values Market value (thousand US$) EcoSan Piped Variable Simple pit Septic (Biodig- sewer VIP Total latrine tank ester connect- Plant)1 ion Households receiving each 16,171 8,085 8,085 4,225 8,085 improvement Total value 331.5 250.6 363.8 1,254.8 622.6 2,823.4 Superstructure 226.4 113.2 153.6 - 153.6 646.8 Slab 56.6 28.3 80.9 - 80.9 246.6 Underground 48.5 109.2 129.4 - 388.1 675.1 1 This involves only with the construction of the biodigester plant. The biodigester plant functions with the use of animal dung and human excreta. Table 30 below shows the potential annual bene�ts of sanitation outputs (biogas), based on the cash saving by using biogas. This gain is mainly through the own use of biogas and not from the sale, as biogas is currently not commercially practical. In general, the total annual �nancial gain from biogas (cooking and lighting) is about US$577,500. This value is also considered as the economic gain of biogas from EcoSan. It is noted that economic bene�t should be actually more than the �nancial one if the time spent on collecting �rewood, especially in rural areas, and the replacement of chemical fertilizer by biodigester’s products are included in the estimation. Table 30. Annual sanitation output market values Biogas (thousand US$) Annual total value for biogas (thousand US$) Sales of gas and sludge Own use Financial Economic - 577.5 577.5 577.5 Considering the potential input and output market, it is estimated that the annual economic gains from combined input and output market may be US$3.4 million. This gain is spread between input US$2.8 million and output US$0.6 million. While input market, in this sense, refers to the possible construction of sanitation facilities and biodigester plants, output market is for the reuse of biogas produced by animal and human waste. 3.9 Sensitivity analysis The present study is based on secondary information, which was combined in a model to estimate the impacts of poor sanitation and the potential bene�ts of improving sanitation. Two major types of uncertainty surround the �gures presented above: • Uncertainty in the values and assumptions used for the included variables (data uncertainty) • Uncertainty due to the fact that some hypothesized impacts were not included (‘model’ uncertainty) Table 31 and 32 below show the sensitivity analysis of the economic impact results based on the uncertainty of economic variables and the attributable fractions. For Table 31, the ‘high’ estimate is assumed to be the highest value corresponding to each of the economic variables evaluated for uncertainty, while ‘low’ estimate is considered as the lowest range of the selected economic variables. In Table 32, the ‘high’ and ‘low’ estimates correspond to the highest and lowest values respectively of economic impacts given the assumed highest possible of attributable fraction. The variable values used in the sensitivity analysis are given in Table A34 and A35 in Annex A7. Based on Table 31, the estimate of �nancial loss of health attributed to poor sanitation and hygiene ranges from US$11 million to US$15 million with base case US$13 million, and the economic loss ranges from US$149 million to Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 53 ECONOMIC IMPACT RESULTS nearly US$1.3 billion with base case US$187 million. This range actually varies based on the assumption of diarrheal incidence, time value of adults and children as well as the estimation of premature death cost. For water pollution losses, the sensitivity analysis is based on different assumptions of the relationship between �sh catch and dissolved oxygen levels. In this case, while the lowest estimate of economic impact is US$108 million for the case where �sh is less affected by low DO, the highest estimate is US$195 million where �sh is more affected by low DO. The base case of economic cost of water access is about US$149 million. It is also seen from the table that the economic cost of time loss for improper toilet going ranges from US$20 million to US$57 million (base case US$38 million). This variation is largely influenced by the assumption on time spent for journeying to open defecation site, and the value of time used for adults and children. In addition, the tourism (economic) impact may range from US$18 million to US$89 million with base case of US$74 million. This range is attributed to the assumption on the prospects of tourism growth, and hotel occupancy rate in the country. Among all the above variables, health impacts have the largest variation of economic costs. This is mainly influenced by the different methods used for the estimation of premature death which provides signi�cant variation in death costs. Table 31. Results of one-way sensitivity analysis – economic variables Financial (million US$) Economic (million US$) Variables selected Base Base Low High Low High case1 case1 Health Diarrheal incidence 10.7 13.3 14.7 148.9 187.1 189.1 Hourly value of non-income earning 13.3 13.3 13.3 186.7 187.1 188.3 time – economic only Hourly value of productive time for 13.3 13.3 13.3 184.8 187.1 189.4 children Premature death 13.3 13.3 13.3 172.6 187.1 1,276.4 Water Fish production impact 105.8 146.8 192.3 108.0 149.0 194.5 Other welfare Time access - - - 19.5 38.2 57.0 Value of time - - - 35.3 38.2 47.1 Tourism Future growth in tourist number - - - 17.9 73.7 88.6 1 Base case shows impact-speci�c results – i.e. under the health impacts, the base case shows the total health impacts; under the water impacts, base case shows total water impacts only In Table 32, ‘high’ estimates show the economic impact resulting from assuming the highest attribution of each variable to poor sanitation and hygiene. Similarly, ‘low’ estimates are those resulting from assuming the lowest attribution (see Table A35). It is seen that different assumptions on attribution give different impact values. Health impact costs vary from US$11 million to US$14 million �nancial cost, and US$149 million to US$202 million economic cost. The water pollution related costs may range from US$138 million to US$153 million in economic values. The impacts of inadequate sanitation and hygiene on tourism also vary from US$37 million to US$111 million depending on how sanitation and hygiene situation affects tourists visiting Cambodia. Economic Impacts of Sanitation in Cambodia 54 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ECONOMIC IMPACT RESULTS Table 32. Results of one-way sensitivity analysis – sanitation links Financial (million US$) Economic (million US$) Variables selected Base Low High Low Base case High case Health Disease incidence attributed to poor 10.7 13.3 14.4 148.9 187.1 201.9 sanitation and hygiene (diarrhea) Water Water pollution attributed to poor sanitation 136.2 146.8 150.3 137.9 149.0 152.7 Tourism Tourist numbers impact attributed to poor - - - 36.8 73.7 110.5 sanitation Table 33 below shows the range of economic gains of impact mitigation. The ‘high’ estimate refers to the highest economic gain based on the highest effectiveness of mitigation measure, and vice versa for ‘low’ estimate (see Table A36). According to the table below, the economic gains from improving sanitation and hygiene practice strongly depend on how mitigation measure can reduce the impact. Based on different assumption of achievable impacts, it is estimated that the gain in health improvement ranges from nearly US$2 million to US$8 million �nancially, and from US$24 million to US$112 million economically. For water cost gain, the range is between US$119 million to US$147 million in �nancial gain, and between US$121 million and US$149 million in economic gain. The economic gain of tourism is estimated to range from US$52 million to US$74 million. The economic gain in sanitation market also values from US$3 million to US$3.5 million according to the number of households using bio-digester. Table 33. Results of one-way sensitivity analysis – impact mitigation Financial (million US$) Economic (million US$) Variables selected Base Base Low High Low High case1 case1 Health Impact mitigation sanitation 1.7 4.3 6.3 24.3 59.9 87.9 Impact mitigation hygiene (relative risk) 3.3 6.0 8.0 46.8 84.2 112.3 Water Water pollution impact on drinking water 119.0 146.8 N/A 120.9 149.0 N/A Water pollution impact on �sh production 133.5 146.8 N/A 135.7 149.0 N/A Tourism Tourist numbers increased with improved - - N/A 51.6 73.7 N/A sanitation Sanitation markets Sanitation output coverage (households using 3.0 3.4 3.5 3.0 3.4 3.5 biodigester) 1 Base case shows impact-speci�c results – i.e. under the health impacts, the base case shows the total health impacts; under the water impacts, base case shows total water impacts only N/A – Not tested Based on sensitivity analysis above, it is clear that the economic impacts of any impact-speci�c variable fall somewhere in the range provided in the tables above – between the low and high range. This range undoubtedly affects the overall economic impacts meaning that the estimated economic impacts are largely affected by the underlying assumption of some variables which are, due to lack of scienti�c evidence, uncertain in nature. In the impact mitigation analysis, the range of economic cost mitigated is also given in a range which is heavily based on the effectiveness of mitigation measures. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 55 WSP-EAP Research Report February 2008 4 Discussion, Conclusions and Recommendations DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS 4.1 Discussion 4.1.1 Overview and interpretation of main results The impact evaluation given in the study has clearly highlighted the economic costs of poor sanitation and hygiene in Cambodia. The impacts which are included in the study are health impacts, water resource impacts, other welfare impacts, and tourism impacts. It is estimated that the total �nancial loss and economic loss due to poor sanitation and hygiene cost the country about US$160 million and US$448 million respectively in 2005. This economic loss amounts to nearly 7.2% of the GDP in 2005. Per capita loss due to poor sanitation and hygiene is over US$32 in economic terms and nearly US$12 in �nancial terms. While the per capita economic loss is equivalent to �ve weeks of average wage loss, the �nancial loss represents about nearly two weeks of wage loss of a working adult. Moreover, at the household level, the annual economic and �nancial loss amounts roughly to US$160 and US$60 per household respectively. It should be noted that although the study uses the best available secondary data to estimate impacts of poor sanitation and hygiene, there is considerable uncertainty associated with these results. In this sense, to avoid overestimation of the losses associated with poor sanitation and hygiene, the study uses conservative input values. For some impacts, the basic data was missing or contained uncertainties for the estimation of the total impact. Therefore, the values such as some attributable fractions are more a matter of opinion and expert judgment than one of scienti�c proof. In addition, it should also be noted that the applied methodology was developed for the present study, with only limited peer review and experience to validate the results. The majority of economic loss is due to the health problem which costs the affected people the spending on health care, the loss in productive time, and the loss due to premature death. It can be remarkable that the economic cost due to premature death alone is about US$170 million taking about 38% of the total economic cost meaning that a lot need to be done to prevent the death from sanitation and hygiene related diseases. In the �nancial loss, however, premature death cost appears to be the lowest which amounts to only US$0.2 million. This is because the �nancial loss of premature death refers only to the short-term household income loss of adults for one year. It is also important to note that the health problem associated with poor sanitation and hygiene is normally much more diverse than those included in the study. The non-inclusion of other diseases such as helminthes, hepatitis A, trachoma, etc will underestimate the impacts of sanitation on health-related costs presented in this study. Besides health-related costs, water costs are the next major contributor of economic costs followed by tourism loss, and access time loss. The water pollution related cost is estimated to be nearly US$150 million annually. It can be noted that poor sanitation has accumulated the pollution in the water source which is not safe for drinking and other domestic use, and not favorable for �sh production. Polluted water costs the consumers and users the energy for treatment or the purchase of other sources as well as the time accessing cleaner water sources which may be farther than the local sources. In this regard, the cost accessing drinking water and domestic water is totaling US$105 million being 70% of the water costs. Besides, the cost of lower �sh production is also signi�cant which amounts to nearly 10% of the total economic cost and 30% of the water costs. It should be noted that if the nutrient loss from the loss of �sh production is included in the estimation, the economic cost would be higher. In this sense, while �sh is the main nutrient for most Cambodian people, this cost highlights the importance of the impacts of sanitation on water resource. In estimating �nancial losses, it is revealed that the water resource impacts of sanitation stand to be the highest in the total �nancial costs which amounts to US$147 million sharing more than 90% of total �nancial losses. This huge loss is mainly due to the impacts on drinking water followed by �sh production. Despite the results, the overall assumption of contribution of sanitation to water resource pollution largely influences the results, which is 65% in this study. The losses from lower tourism potential are the other economic losses, which accounts for over 16% of the total economic cost (US$74 million). It is important to also note that the impacts of tourism loss would be more signi�cant, Economic Impacts of Sanitation in Cambodia 58 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS if the welfare loss of tourists is included. A point that should be noted in this study is that the estimation of economic impacts of tourism due to sanitation are mainly made based on an assumption of optimal occupancy rate and the potential scenario of tourism growth. Moreover, the estimation is drawn based heavily on an assumption of 10% attribution of poor sanitation and hygiene on tourist number reduction. The economic cost of poor access (access time lost) to sanitation is about US$38 million sharing nearly 9% of the total economic costs. It is noted that this economic cost of other welfare impacts refers only to the time used for toilet going for those who do not have their own toilet. There are other dimensions of welfare losses which have not been quanti�ed in this study such as privacy, security, prestige, and so on. If these losses are quanti�ed, then the economic cost of welfare impacts would be more signi�cant. The economic gains from improved sanitation and hygiene can be achieved through different improvement options: 1) hygiene practice, 2) latrine physical access, 3) toilet system, 4) treatment or disposal, and 5) reuse of waste (animal dung). However, it is important noting that a program aiming at improving sanitation and hygiene may cover a number of options above. In other word, the options above may be implemented together. However, this study identi�es which component of sanitation and hygiene improvements are expected to bring the greatest bene�ts. In option 1 where the bene�t is mainly on health improvement, the economic bene�t from the improvement is about US$84 million and the �nancial bene�t is US$6 million. For option 2, however, the bene�t in terms of time use for toilet going and input markets is totaling US$39 million. While more than US$38 million (99%) in this cost comes from the time use loss, the input market is only US$0.3 million. Option 3 involving the improvement of toilet system can induce economic gains for health and input market which amounts to US$61 million for economic gains and US$6 million for �nancial gains. Option 4 which mitigates impacts on water pollution and tourism can provide economic gains of up to US$223 million and �nancial gains of US$147 million. Option 5 relates to the input and output market of waste can induce gains of US$1.8 million both economically and �nancially. This gain is from the input market for biodigester’s builders and household saving by using biogas. It should be noted that, except health bene�ts, those gains may be over estimated due to the assumption that the improvement will lead to 100% reduction of loss, which in reality may not be achievable. An important point to note in the present study is that throughout the presentation of results, distinction has been made between �nancial losses and economic losses. While attempts were made to follow conventions used in economic evaluation techniques, it was in practice difficult to distinguish between �nancial and economic. Whether a loss is felt as a real �nancial cost (involving immediate monetary transaction) or as affecting resource use and long- term behavior, and hence income, is largely context-speci�c, which could not be fully reflected in a methodology that was developed. Hence, policy makers should interpret with particular caution the �nancial cost estimates, which would not necessarily lead to a changed �nancial situation in the short term if sanitation is improved. 4.1.2 Policy implications of the study results As shown in the present study, sanitation (and hygiene) are strongly linked or associated with other development issues related to Cambodia Millennium Development Goals (CMDG). Improved sanitation has positive effects on child and adult health, gender equality, hunger, environmental sustainability, and water resources (clean drinking water). It is proved that poor sanitation and hygiene cause wide range of and huge economic costs to the country. Therefore, it is important that sanitation receives greater attention from Cambodian policy makers as well as other stakeholders. Since lack of appropriate latrine is mainly the rural phenomenon, these losses are concentrated in rural areas which host more than 90% of the poor. While those people already have very small income, the economic costs attributed to poor sanitation and hygiene will impose more burdens hampering them moving out of poverty. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 59 DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS Given that poor sanitation has multi-dimensional impacts, various policy makers of different institutions need to work together. In this sense, programs to improve sanitation should involve various stakeholders working on different sectors such as health, rural development, water resource management, education, tourism, and environment, among others. Although these sectors work on different issues, they are highly complimentary. For instance, to improve hygiene practice among people, it is important that health sector, education sector, and rural development sector jointly work on promoting hand washing. In this way, the cooperation across sectors will lead to optimal achievement of improved sanitation and hygiene in Cambodia. Therefore, a clear strategy and actual implementation of that strategy to improve sanitation, particularly in rural areas are very much needed. More importantly, it is also crucial that improving sanitation and hygiene be part of the strategies to sustainable poverty alleviation in Cambodia. In this regard, strongly integrating sanitation in the National Strategic Development Plan (NSDP) is one the key priority area. 4.1.3 Study weaknesses Although the study attempts to follow scienti�c methods, this study is conducted using mainly official secondary data and statistics as well as a number of key assumptions, which is likely to lead to imprecision in economic impact estimation. Moreover, some sub-impacts were not included in the quantitative estimates due to the lack of information and data. Therefore, the results presented in this study may be both overestimated (due to incorrect data inputs and assumptions) as well as underestimated (due to omission of impacts). In terms of health impact, there are more diseases than those included in this study due to the fact that they are not available in the government reporting system. In addition, the attribution fraction of skin disease to poor sanitation and hygiene is crudely assumed which can lead to bias. For water resource impact, �sh production impact is largely based on a modeled relationship between �sh production and DO level. On the other hand, the attributable fraction of water pollution to poor sanitation used in the estimation is also based on expert opinion. In the estimation of impact of time use in other welfare impacts, the present study uses the productive time value in the evaluation of economic loss of time loss. In other word, the study tends to assume that the time that people are journeying to toilet can be used in a productive way although it is not necessarily true in real life. This time can instead be used for other leisure activities, rather than the productive ones, which may have less economic value. While estimating impact on tourism, the study attempts to assume the optimal occupancy rate of tourism which is not likely achievable despite the presence of adequate sanitation condition. To offset these limitations, the study has included several sensitivity analyses to examine the consequence of different assumptions. 4.1.4 Gender Besides economic cost, poor sanitation is also likely to have impacts on gender in Cambodia as it may constrain women from getting a better employment. While many more men are moving into a better-paid position, women are still limited to garment industry and other informal sectors7 requiring less skill. This phenomenon can be explained by the fact that most men are better educated than women so that the chance for women to compete with men for a well-paid job is very low. In this sense, it is important to see what factors influence girls to be absent from or drop out of school, and to what extent poor sanitation influences their choices. This impact may be more apparent when girls get older (secondary school girls). With the cultural value, Cambodian girls and women �nd themselves more difficult than men to go to the open �eld for practicing defecation, especially at the day time at schools or workplace. In this case, in addition to 7 http://web.worldbank.org/WBSITE/EXTERNAL/COUNTRIES/EASTASIAPACIFICEXT/EXTEAPREGTOPSOCDEV/ 0,,contentMDK:20264079~menuPK:502969~pagePK:34004173~piPK:34003707~theSitePK:502940,00.html Economic Impacts of Sanitation in Cambodia 60 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS the family’s poverty and distance from school, the lack of latrine at school can worsen the girls’ motivation to go to school. The less girls going to school, the less chance they get for better job opportunity and the greater the gender inequality becomes visible. Given this sequence of reasons, it is always worth to address the poor sanitation problem, particularly in schools, work places, and public places, if Cambodia is to achieve greater gender equality in the society. 4.2 Conclusions The study highlights the signi�cance of the economic costs of the poor sanitation with the range of US$448 million, which currently lags behind other development in Cambodia. While the traditional thought of the impact of poor sanitation is mainly on health, this study shows that there are other dimensions beyond health impact, i.e. water, welfare, and tourism. Those impacts are also substantial in economic costs although not comparable to the health impact. It is found that, except health, the next major impacts of poor sanitation are water resource and tourism impact. Another impact which is not signi�cant in economic terms but is important to the welfare of the people is access time loss and other welfare impacts. Moreover, the study also analyzes the flip side of the coin in showing the potential bene�ts of improving sanitation and hygiene through various options. It is seen that each option will provide different bene�ts in relation to the impact it mitigates. In the light of the �ndings, the study �ndings should emphasize the point that sanitation is not a sole health issue, but a cross-sectoral issue which need participation of all stakeholders. As well as the health sector, there is a role to play for those involved in other sanitation-related sectors such as rural development, water resource management, education, tourism, and gender, among others. In this sense, a policy to address sanitation effectively can be done through a well coordinated effort among the institutions mentioned. 4.3 Recommendations The central aim of this present study was to generate an evidence base to enable recommendations to be made for improved sanitation policies. This study has identi�ed a broad range of impacts of poor sanitation, and quanti�ed those impacts most amenable to secondary analysis. The following policy recommendations are based on eight major �ndings of the study: Major �nding 1. Poor sanitation causes signi�cant losses to the national economy This study has found that poor sanitation is responsible for at least US$448 million economic losses per year in Cambodia – or an average of 7.2% of annual GDP. Of these costs, at least US$160 million are �nancial in nature (2.5% of GDP), involving additional expenditure or actual income loss for the population. In addition to these quanti�ed impacts, a range of other negative economic and social effects of poor sanitation result. By improving sanitation, signi�cant proportion of socio-economic impacts is mitigated. Recommendation 1. Decision makers from various sectors are advised to act now Sanitation ‘players’ are advised to act now, otherwise the negative impacts of poor sanitation will increase over time. The Government of Cambodia and other stakeholders should jointly reassess the current and planned spending levels in the sanitation and related sectors, covering health, water resources, environment, rural and urban planning and development, �sheries, and tourism. Increased political importance and budget allocations should be given to sanitation. Sanitation decision makers should use an evidence-based approach to design efficient sanitation policies and implementation strategies, to increase value-for-money from public and private investments into sanitation. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 61 DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS Major �nding 2. Poor sanitation has greater impact on the poor and vulnerable A considerable socio-economic burden of poor sanitation falls on the population currently without improved sanitation – health impacts, time access, water pollution, aesthetics, and land use – hence causing inequities in society. The population group unserved with improved sanitation tends to be the poorer and more vulnerable members of society. A disproportionate share of the burden falls on women, children and the elderly, especially health burden, intangible welfare impacts and life decisions. Recommendation 2. Governments must de�ne and target the needs of priority groups Governments should give priority to the populations with no latrine, recognizing that effective demand may be low in these groups due to low incomes and poor awareness of the bene�ts of investing in sanitation. As well as stimulating demand through public health and latrine advocacy messages, governments should target programs, �nancing mechanisms and any subsidies to the most disadvantaged population groups. Major �nding 3. Negative impacts result from several poor sanitary practices Economic impacts occur not just through the use of unimproved latrines (CMDG target), but also through poor hygiene practices, poor isolation of wastewater from the environment and water sources, and poor broader environmental sanitation. Cambodia tends to fall short of meeting broader environmental standards. Recommendation 3. Players should broaden the scope of sanitation beyond latrines Investments should not be made just in sanitation hardware programs, but in improved sludge, water and solid waste management, and in hygiene programs to raise population awareness on personal and community hygiene issues. Major �nding 4. Health related economic impacts take a signi�cant toll on society This study has con�rmed that the major and most tangible impact of poor sanitation is an increased risk of infectious disease and premature death. In Cambodia, of those diseases included in this study, at least 10 million disease episodes and 10,000 deaths are attributed to poor sanitation annually. One third of these deaths are from the indirect diseases resulting from poor sanitation through childhood malnutrition. This study has shown that economic losses of over US$187 million result from health care costs, health-related productivity costs and premature mortality costs, or US$14 annually for each and every person in Cambodia. Recommendation 4. Health aspects of sanitation programs deserve central focus The Government of Cambodia should focus on the easy health wins from improved sanitation, through targeting children and focusing on safe but simple latrine designs, improved excreta isolation measures, and improved hygiene practices. Given the key role of hygiene practices in health improvement, high-impact hygiene components should be integrated in the planning and implementation of sanitation programs. In this case, in addition to Ministry of Rural Development, the Ministry of Health should play a central role in the health aspects of sanitation programs. Major �nding 5. High water pollution levels are partially caused by poor sanitation The majority of human excreta eventually �nd their way to water bodies; so do gray water, animal excreta, solid waste and industrial wastewater. Together these cause signi�cant water pollution in Cambodia with associated high economic losses. Quanti�ed economic losses associated with polluted water from domestic sources reach US$149 million in Cambodia (US$11 per person per year), while other impacts include loss of aesthetics, leisure activities, and tourism. Recommendation 5. Sanitation solutions should focus on reducing water pollution Governments should urgently implement sanitation standards that reduce the release of waste matter into water resources. Low technology, low cost and effective options should be explored as a matter of priority. Focus should not be just on excreta, but also solid waste, household, agricultural and industrial wastewater. The contamination of groundwater with microbiological pathogens should be averted through better planning, increased resource allocation, and awareness raising. Water quality monitoring should be conducted to assess the extent and nature of water pollution and to inform populations of which water sources are safe to use. Economic Impacts of Sanitation in Cambodia 62 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS Major �nding 6. Sanitation is linked with sustainable development in many ways Sanitation has a major role in sustainable development, due to its links to other development goals (e.g. CMDGs). Sanitation plays a key but unrecognized role in population welfare, economic growth and poverty reduction. Impacts not fully explored in this study – in particular the investment climate – are potentially major arguments for improving sanitation in countries, and suggest the adoption of a broader understanding of the term ‘sanitation’. Recommendation 6. Several coordinated measures are needed to improve sanitation Sanitation cannot be only the responsibility of an individual sector/ministry, nor of a single level of government. The fact that sanitation touches on many sectors and line ministries should be used as a strength rather than hampering progress, and clear roles and responsibilities need to be de�ned. The development of a policy and regulatory framework for environmental and health protection is crucial and imperative in the context of rapid industrialization and high economic growth in Cambodia. While further progress is needed at the highest levels to ensure political support and resource allocations for sanitation, greater emphasis is needed on the implementation levels where sanitation demand must be stimulated and affordable and attractive solutions for sanitation must be available. Major �nding 7. Variability is expected in the actual impacts of poor sanitation The national per capita costs in rural and urban areas in Cambodia have signi�cant difference. In addition, there will exist signi�cant variation in the impacts of poor sanitation between different geographic locations depending on sanitation coverage, demographics, environment, and practices related to health and water consumption. Recommendation 7. Local as well as national studies should inform sanitation policy The study indicates that impacts are likely to vary between different population groups, varying based on geographical and physical features, sanitation coverage, demographics, and practices related to health, hygiene and water consumption, among others. Major �nding 8. Existing data sources are weak for quantifying sanitation impact This study has used a number of available data sources, but has been limited by lack of speci�c information on outcomes related to sanitation. With the exception of basic latrine coverage indicators, surveys tend not to include questions related to sanitation, such as expenditure, preferences, access time, health-related time loss, sanitation and hygiene practices, and gender. Questions related to broader sanitation ‘coverage’ (e.g. waste disposal, environmental quality) are largely left out. Routine government reporting systems such as health indicators and health service use, and water quality monitoring, only imperfectly capture the substantial impacts of poor sanitation. Water quality is known to be important for �sh reproduction, growth and safety for human consumption, but little is known about the exact relationships, and the role poor sanitation plays. Recommendation 8. Future survey and research work is key in monitoring progress Surveys and government reporting systems should be assessed for extension to include behavior and outcomes related to sanitation. Selected research studies could �ll important gaps in knowledge about the economic and welfare effects of poor sanitation. Further research is required on the population bene�ts of improved sanitation, and what levels of bene�t different types of sanitation option can deliver. A gender perspective is key in understanding the effectiveness of different sanitation options. The link between poor sanitation and tourism and foreign direct investment losses is poorly understood, and merits further assessment. Country-speci�c studies on the value of time and the value of life will allow a better understanding of the importance of the identi�ed health impacts Detailed recommendations for research are outlined in Annex C. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 63 WSP-EAP Research Report February 2008 Annexes ANNEXES Annex A. Study Methods A1 Health Health impacts are usually considered to be one of the most signi�cant impacts associated with poor sanitation and hygiene. There are many diseases associated with poor sanitation and hygiene practices, among them diarrhea, dysentery, cholera, salmonellosis, shigellosis, typhoid fever, hepatitis A, trachoma, and some parasitic diseases (ascariasis, trichuriasis, hookworm, schistosomiasis). Not only do diseases have direct implications for population welfare through health-related quality of life (HRQL), but diseases also have �nancial and economic impacts, which include spending on health care, loss of income or production, and in the case of premature death, the value of loss of life. A1.1 Selection of diseases There are many diseases associated with exposure to human waste due to poor sanitation and poor hygiene practices. These are presented in detail in Table A1. Diseases related to poor sanitation and hygiene can be viral, bacterial, parasitic, protozoal, helminth, and fungal in nature, and have many pathways: fecal-oral, urine-oral, and fecal-eye; the main one being fecal-oral [9, 10]. According to the F-diagram, pathogens can be passed from the feces through fluids, �elds, flies and �ngers [11]. In addition, food can act as an intermediary for all of these four direct transmission pathways. The principle ‘poor practices’ which support heightened transmission of disease from human waste include an unsanitary toilet area, poor personal hygiene practices following toilet-going, open defecation in the �elds or water sources, lack of protection or treatment of drinking water, poor food preparation practices, and lack of latrine and water-source protection in flood-prone areas. Furthermore, exposure to household solid waste, agricultural and industrial wastes can also lead to disease and premature death, from contact with toxic materials or otherwise dangerous substances. Poor sanitation is directly and indirectly affecting population health. Directly, poor sanitation causes diarrheal infections and other health effects which in turn lead to mortality especially in young children. Indirectly, poor sanitation contributes to child malnutrition through the effect of diarrheal infections on nutritional status. Malnutrition, or poor nutritional status, increases the risk of child mortality from disease as well as increases the incidence of disease (Fishman et al., 2004). This indirect effect of sanitation mainly affects children under the age of �ve years old, while the direct effect of sanitation affects the whole population. Given the large number of diseases and health effects due to poor sanitation, this present study selected the key health impacts based on their epidemiological and economic importance. The availability of health data from national statistics, local research studies and international sources also played an important role in determining which diseases to include. Table A2 below presents data available from the national health information system (HIS) in Cambodia on number of cases and deaths from key sanitation and hygiene-related diseases. Although these data are not representative of the total disease burden at national level due to underreporting, these data do provide an indication of which diseases are of most signi�cance nationally to aid selection of diseases to include in this present study. Economic Impacts of Sanitation in Cambodia 66 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES Table A1. Diseases linked to poor sanitation and hygiene, and primary transmission routes and vehicles Primary trans- Disease Pathogen Vehicle mission route Diarrheal diseases (Gastrointestinal tract infections) Rotavirus diarrhea Virus Fecal-oral Water, person-to-person Typhoid/ Bacterium Fecal-oral and Food, water. + person-person Paratyphoid urine-oral Vibrio cholera Bacterium Fecal-oral Water, food Escherichia Coli Bacterium Fecal-oral Food, water. + person-person 1 Amebiasis (amebic dysentery) Protozoa Fecal-oral Person-person, food, water, animal feces Giardiasis Protozoa1 Fecal-oral Person-person, water (animals) Salmonellosis Bacterium Fecal-oral Food Shigellosis Bacterium Fecal-oral Person-person. +food, water Campylobacter Enteritis Bacterium Fecal-oral Food, animal feces Helicobacter pylori Bacterium Fecal-oral Person-person. + food, water Protozoa Other viruses2 Virus Fecal-oral Person-person, food, water Malnutrition Caused by diarrhoeal disease and helminthes Helminths (worms) Intestinal nematodes3 Roundworm Fecal-oral Person-person. + soil, raw �sh Digenetic trematodes (e.g. Flukes (parasite) Fecal/urine-oral; Water and soil (snails) Schistosomiasis Japonicum) fecal-skin Cestodes Tapeworm Fecal-oral Person-person. + raw �sh Eye diseases Trachoma Bacterium Fecal-eye Person-person, via flies, fomites, coughing Adenoviruses (conjunctivitis) Protozoa1 Fecal-eye Person-person Skin diseases Ringworm (Tinea) Fungus Touch Person-person (Ectoparasite) Scabies Fungus Touch Person-person, sharing bed and (Ectoparasite) clothing Other diseases Hepatitis A Virus Fecal-oral Person-person, food (especially shell�sh), water Hepatitis E Virus Fecal-oral Water Poliomyelitis Virus Fecal-oral, oral- Person-person oral Leptospirosis Bacterium Animal urine- Water and soil – swamps, rice �elds, oral mud Sources: WHO http://www.who.int/water_sanitation_health/en/ and [12, 13] Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 67 ANNEXES 1 2 3 There are several other protozoa- Other viruses include: Intestinal nematodes include: based causes of GIT, including • Adenovirus – respiratory and • Ascariasis (roundworm - soil) • Balantidium coli – dysentery, gastrointestinal infections • Trichuriasis trichiura intestinal ulcers • Astrovirus – gastrointestinal (whipworm) • Cryptosporidium parvum - infections • Ancylostoma duodenale gastrointestinal infections • Calicivirus – gastrointestinal / Necator americanus • Cyclospora cayetanensis - infections (hookworm) gastrointestinal infections • Norwalk viruses – • Intestinal Capillariasis • Dientamoeba fragilis – mild gastrointestinal infections (raw freshwater �sh in diarrhea • Reovirus – respiratory and Philippines) • Isospora belli / hominus – intestinal gastrointestinal infections parasites, gastrointestinal infections Table A2. Importance of sanitation and hygiene-related diseases, total cases and total deaths (2005) Reported morbidity Annual reported Disease Rate (cases per Cases in 2005 deaths population) Diarrheal diseases (total) 706,083 0.05114 99 Diarrhea 395,364 0.02864 42 Dysentery 310,719 0.02250 8 Typhoid fever1 10,408 (IP) 0.00075 48 Cholera1 125 (IP) 0.00001 1 Skin diseases 202,786 0.01469 - 2 3 Malnutrition - underweight in 1,147 - 597,485 0.00068 – 0.3525 - children <5 Diseases associated with 1,029,205 0.07454 1,209 malnutrition in children <54 ALRI 964,688 0.06987 926 Measles 1,350 0.00010 1 Malaria 63,167 0.00458 282 Source: Health Information System 2005, MoH 1 The total morbidity cases are not available. So, the �gure indicates only the number of inpatients (IP). This �gure is not added up in the diarrheal diseases. 2 The �gure is based on Health Center and National Hospital data for underweight cases taken from HIS (2005). 3 Sourced from CDHS (2005) 4 Total cases, not yet attributed to poor sanitation and hygiene It can be seen from the table that the diarrheal diseases and skin disease amount to more than 0.7 million and 0.2 million cases, respectively. The malnutrition rate among under-�ve children has a close relationship with poor sanitation and hygiene. In 2005, the recorded number of malnutrition is 1,147 cases according to the HIS; and estimated over half a million according to the CDHS (2005). It is useful to note that in the case of malnutrition, only under-weight is the main focus in the study as it reflects the suffering of children from both chronic and acute malnutrition. The number of reported disease cases that are related to malnutrition such as ALRI (Acute Lower Respiratory Infection), measles, and malaria totaled roughly one million cases among under-�ve children. The fact that malnutrition and its associated diseases are considered in this study is because poor sanitation and hygiene indirectly contributes to malnutrition through the effects of diarrheal infections on nutritional status. More importantly, malnutrition increases the incidence of diseases and the risk of child mortality from diseases such as ALRI, measles and malaria. Economic Impacts of Sanitation in Cambodia 68 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES The reported deaths in Table A2 above reflect only the deaths in public health establishments as the data from private clinics is largely unaccounted for, and deaths outside medical establishments is excluded. Thus, this �gure is underreported in the sense that the actual number of death may be much higher than the �gure given. Given the importance of these diseases and the availability of data in HIS, only the diseases listed in Table A2 above are considered in present study. To understand better the burden of disease by age group in public health service, Table A3 below shows the distribution of morbidity and mortality by age group. This table, however, does not reflect the actual disease burdens in Cambodia. From the table, it can be observed that nearly half of diarrheal diseases occur among population of over �fteen years of age followed by under-�ve children where one third of diarrheal diseases occur. For the case of ALRI, it is observed that the morbidity is highest for the population above 15 years of age, followed by the under-�ve children. However, the death from ALRI is more predominant among under-�ve children than 15+ population. Table A3. Distribution of morbidity and mortality by age group, year 2005 Condition Morbidity (% cases) Mortality (% deaths) 0-4 5-14 15+ 0-4 5-14 15+ Diarrheal disease 31.2 19.9 49.0 36.4 25.3 38.4 Skin disease 22.0 24.3 53.7 0.0 0.0 0.0 Malnutrition1 100.0 0.0 0.0 0.0 0.0 0.0 ALRI 38.6 19.4 42.1 55.4 6.0 38.6 Measles 2.0 3.9 94.1 100.0 0.0 0.0 Malaria 6.9 17.5 75.5 14.5 20.9 64.5 Source: Health Information System, Ministry of Health, 2005 1 Note that malnutrition �gures are only available for the 0-4 category, but that malnutrition also occurs in older age groups From the table, it is important to note that although ALRI, measles, and malaria involve population of all age groups, the present study only focuses on the under-�ve population. This is because those diseases are linked with poor sanitation and hygiene through malnutrition attributed to diarrheal infections (See A1.3). A1.2 Disease burden from diseases directly related to poor sanitation In order to estimate the full impact of diseases on the daily activities of the afflicted person, it is necessary to estimate the total number of episodes, and not just the episodes seeking treatment, as indicated imperfectly from routine health information systems. As shown by household surveys such as the Cambodia Demographic and Health Surveys (CDHS) and the Cambodia Socio-Economic Survey (CSES), a proportion of the sick do not seek official care, and can approach an informal carer, or they may self-treat by visiting the local pharmacy, or they wait to see if the disease gets better with no action8. For diarrheal disease incidence in the under �ves – the age group on whom the major disease burden falls – CDHS data were used for under �ves. Given that CDHS does not normally report diarrheal disease prevalence for population over �ve years of age, available WHO regional data were used for this population (see Table A4). In this case, the WHO’s WPR-B region is applied for Cambodia. CDHS usually report prevalence during a two-week recall period based on self-diagnosis9. It is important to note that the national averages from CDHS data presented in the table 8 As shown by the regional and socio-economic disaggregations of these survey data, treatment seeking behaviour varies signi�cantly between different populations, and depends on such factors as geographical proximity to health services, out-of-pocket costs of treatment and transport, attitudes towards health providers, and cultural factors. 9 The calculation is to take the proportion of children with diarrhea in the past 2 weeks, and scale up to a year period by multiplying by 52 (weeks) and dividing by 2.5 (recall period 2 weeks, but cases recalled 2 weeks ago would have contracted diarrhea up to half a week before that). Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 69 ANNEXES hide a considerable variation by age group (from 2.04 for 48-59 month children to 6.61 for 6-11 month children) and geographical location (from 2.11 in Coastal zone to 4.56 in Plains zone). In estimating total cases, an attributable fraction of diarrheal disease to poor sanitation and hygiene of 88% was applied to the rates in Table A4 [9]. Table A4. Diarrheal disease incidence in Cambodia Sanitation Cases of diarrhea per person, by age group condition 0 to 4 1 5 to 142 15 to 592 60+2 Improved 0.33 0.16 0.16 4.06 Unimproved 0.52 0.26 0.26 1 CDHS 2005 data used for children under �ve years. Based on DHS data of 2-week recall, the calculation is to take the proportion of children with diarrhea in the past 2 weeks, and scale up to a year period by multiplying by 52 (weeks) and dividing by 2.5 (recall period 2 weeks, but cases recalled 2 weeks ago would have contracted diarrhea up to half a week before that). 2 WHO (WPR-B) regional estimates for population over �ve years of age. A1.3 Disease burden from diseases indirectly related to poor sanitation10 The approach used here to estimate the indirect health effects of sanitation (via malnutrition) in children is as follows: (a) the effect of diarrheal infections on children’s nutritional status is �rst determined from a review of the research literature; (b) counterfactual nutritional status is then estimated, i.e., the nutritional status that would have prevailed in the absence of diarrheal infections; and (c) health effects of currently observed nutritional status and health effects of counterfactual nutritional status are estimated. The difference in health effects of observed versus counterfactual nutritional status is then the indirect health effects of diarrheal infections, caused largely by poor sanitation. Commonly used indicators of poor nutritional status are underweight, stunting and wasting11. Underweight is measured as weight-for-age (WA) relative to an international reference population12. Stunting is measured as height- for-age (HA), and wasting is measured as weight-for-height (WH). Underweight is an indicator of chronic or acute malnutrition or a combination of both. Stunting is an indicator of chronic malnutrition, and wasting an indicator of acute malnutrition. Underweight status is most commonly used in assessing the risk of mortality and morbidity from poor nutritional status. A child is de�ned as mildly underweight if his or her weight is in the range of -1 to -2 standard deviations (SD) below the weight of the median child in the international reference population, moderately underweight if the weight is in the range of -2 to -3 SDs, and severely underweight if the child’s weight is below -3 SD from the weight of the median child in the reference population. The standard deviations are also called z-scores and noted as WAZ (weight-for-age z-score). 10 This section is largely based on Larsen B. Cost of environmental health risk in children under 5: Accounting for malnutrition in Ghana and Pakistan. Background report prepared for the World Bank study on malnutrition and environmental health. 2007. Washington DC: World Bank. 11 Micronutrient de�ciencies are not explicitly evaluated here, but are found in other studies to have a signi�cant cost (World Bank, 2006; Horton and Ross, 2003; Horton, 1999). Also, Alderman and Behrman (2006) �nd a signi�cant cost associated with low birth weight, which in part is caused by low maternal pre-pregnancy body mass index (Fishman et al, 2004). 12 The international reference population is de�ned by the National Center for Health Statistics (NCHS standard), United States or by the World Health Organization’s international reference population. Economic Impacts of Sanitation in Cambodia 70 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES Repeated infections, and especially diarrheal infections, have been found to signi�cantly impair weight gains in young children. Studies documenting and quantifying this effect have been conducted in communities with a wide range of infection loads in a diverse group of countries such as Bangladesh [14-16], Gambia [17, 18], Guatemala [19], Guinea-Bissau [20], Indonesia [21], Mexico [22], Peru [23], Philippines [24], Sudan [25], and Tanzania [26]. These studies typically �nd that diarrheal infections impair weight gains in the range of 20-50 percent. A mid-point – i.e., 35% of children’s weight de�cit - is here attributed to diarrheal infections to estimate the indirect disease burden from sanitation13. So in the absence of weight retarding infections, the weight-for-age z-score (WAZ) of an underweight child would be approximately 40 percent greater than the observed z-score (i.e., observed WAZ*(1- 0.4))14. For instance, if a child has a WAZ=-3, then in the absence of weight retarding infections, the child’s WAZ would be -1.8. Prevalence of underweight malnutrition rates in Cambodia are presented in Table A5. Current rates are estimated for the year 2005. It should be noted that CDHS report does officially report the prevalence of severe and moderate-to- severe underweight, but not mild underweight. Mild underweight is however important in relation to increased risk of child mortality [27]. This rate was then calculated from the original data in CDHS 2005. Counterfactual prevalence rates of underweight, i.e., prevalence rates in the absence of weight retarding infections were calculated using the original household data in the CDHS 2005. This was performed through the following procedure: Counterfactual WA z-scores were calculated for each underweight child in the survey using the formula discussed above (i.e., WAZ reported for each child in the survey multiplied by (1-0.4)). Counterfactual underweight prevalence rates were then tabulated using the counterfactual WA z-scores. The results are presented in Table A5 below. Table A5. Current and estimated counterfactual underweight prevalence rates in children under 5 Prevalence rate in Cambodia Current prevalence rates Severe underweight ( < - 3 SD) 6.6% Moderate underweight (-2 to -3 SD) 29.1% Mild underweight (-1 to -2 SD) 38.5% Non-underweight ( > -1 SD) 25.9% Counterfactual prevalence rates Severe underweight ( < - 3 SD) 0.07% Moderate underweight (-2 to -3 SD) 3.0% Mild underweight (-1 to -2 SD) 47.7% Non-underweight ( > -1 SD) 49.2% Source: Estimated based on CDHS 2005 Based on the table, in the absence of diarrheal infections, it is estimated that practically no children would be severely underweight and the prevalence of moderate underweight would be as low as 3%. The prevalence of mild underweight, however, would increase signi�cantly. Various health and debilitating effects from malnutrition are documented in the research literature. This includes long term chronic illnesses from low birth weight, effects of iodine, vitamin and iron de�ciencies, and impaired cognitive development (United Nations, 2004; World Bank, 2006). The focus here is on mortality and morbidity in under-�ve children years associated with underweight. 13 A child’s weight de�cit is the difference in weight between the child’s observed weight and the weight of the median child in the international reference population. 14 This is calculated using the WHO Anthro 2005 software. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 71 ANNEXES Fishman et al (2004) present estimates of increased risk of cause-speci�c mortality and all-cause mortality in under- �ve children with mild, moderate and severe underweight from a review of available studies. Severely underweight children (WA < -3 SD) are �ve times more likely to die from measles, eight times more likely to die from ALRI, nearly 10 times more likely to die from malaria, and twelve times more likely to die from diarrhea than non-underweight children (WA > -1 SD). Even mild underweight doubles the risk of death from major diseases in early childhood (Table A6). Table A6. Relative risk of mortality from mild, moderate and severe underweight in children under �ve Weight-for-age (WA) < - 3 SD -2 to -3 SD -1 to -2 SD > - 1 SD Pneumonia/ALRI 8.1 4.0 2.0 1.0 Diarrhea 12.5 5.4 2.3 1.0 Measles 5.2 3.0 1.7 1.0 Malaria 9.5 4.5 2.1 1.0 Other causes of mortality1 8.7 4.2 2.1 1.0 Source: Fishman et al (2004). 1 Not including mortality from perinatal conditions. Child underweight also increases the risk of illness. Fishman et al (2004) present estimates of increased risk in under-�ve children with moderate and severe underweight (WA < -2 SD). The largest increased risk of illness is for pneumonia/ALRI. No increased risk of measles is con�rmed (Table A7). Table A7. Relative risk of illness from moderate and severe underweight in children under �ve Weight-for-age (WA) < - 2 SD > - 2 SD Pneumonia/ALRI 1.86 1.0 Diarrhea 1.23 1.0 Measles 1.00 1.0 Malaria 1.31 1.0 Source: Fishman et al (2004). These relative risk ratios can be applied to the underweight prevalence rates in Table A5 to estimate attributable fractions (AF) of mortality and morbidity from diarrheal infections through their effect on nutritional status (underweight status)15. The following formula is used to calculate attributable fractions of ALRI, measles, malaria, and “other causes� of mortality, and attributable fractions of ALRI and malaria morbidity incidence from diarrheal infections: n n ∑ Pi RRi − ∑ Pi C RRi AF = i =1 n i =1 (1) ∑ P RR i =1 i i 15 The attributable fraction of mortality or morbidity from malnutrition is the percent of deaths or percent of cases of illness (e.g., percent of ALRI deaths or cases of ALRI) caused by malnutrition. Economic Impacts of Sanitation in Cambodia 72 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES where RRi is relative risk of mortality or morbidity for each of the WA categories (i) in tables A6-A7; Pi is the current underweight prevalence rate in each of the WA categories (i); and Pic is the counterfactual underweight prevalence rate in each of the WA categories (i). This formula is also called the “potential impact fraction� because it estimates the mortality or morbidity that would have been avoided for a different counterfactual population distribution (e.g., less children being underweight) exposed to those levels of risk of mortality or morbidity. For a further discussion of this formula, see Ezzati et al. (2004). For diarrheal mortality and morbidity the AF estimation procedure would be different because there are two risk factors, i.e. the direct effect of sanitation and the indirect effect through malnutrition. As already 88 % of diarrheal infections and mortality is estimated to originate from sanitation (or mediated from sanitation through water), the additional effect of malnutrition is minimal and is therefore ignored here16. Annual cases of mortality and morbidity from diarrheal infections caused by poor sanitation, through the effect of infections on nutritional status, are estimated as follows: j =m M = c ∑ AF j M 0 j (2) j =1 where AFj is the AF in eq. (1) for each cause of mortality or type of disease “j�, Mj0 is the current total annual cases of mortality or disease incidence in each of the categories in tables A2-A3, and “c� is the fraction of diarrheal infections caused by poor sanitation (88%). Most recent available estimates of annual cases of mortality (Mj0) in children under-�ve are presented in Table A8. These estimates reflect under-�ve child mortality rates in 2005, and the structure of cause-speci�c deaths is estimated from WHO country estimates of cause-speci�c mortality in 2002 [28]. Table A8. Estimated cause-speci�c annual deaths in children under �ve in 2005 Disease Annual deaths Diarrheal disease 6,600 ALRI 4,800 Measles 1,400 Malaria 2,600 PEM 800 LBW 3,500 Other perinatal conditions 6,000 Other causes 4,900 Total 30,600 Source: Adjusted to 2005 from WHO country estimates of mortality by cause in 2002 (WHO, 2004a), by applying child mortality rate in 2005. 16 See Larsen (2007) for methodology and estimation of environmental health effects from multiple environmental risk factors in Ghana and Pakistan. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 73 ANNEXES Table A9. Demographic and mortality data in 2005 Variable Cambodia Mortality rate, under-�ve children (per 1,000)1 83 Population, total2 13,806,974 Number of under-�ve children2 1,694,990 Estimated annual births3 369,682 Source: 1CDHS 2005; 2 Population projection 1998-2020, NIS; 3Estimated from the number of children u5. Complete records or statistics on annual cases of ALRI and malaria in under-�ve children are not available in Cambodia. This is due to many reasons, including incomplete reporting and record systems, cases never treated at health care providers, and incomplete or potentially incorrect case identi�cation and diagnostic. Annual cases therefore need to be estimated. WHO provides regional estimates of ALRI for the year 2002, the most recently available [29]. These data suggest that the incidence of ALRI in under-�ve children in Asia is on the order of 0.35 to 0.7 cases per child per year. In this regard, an annual incidence of 0.5 cases of ALRI is applied to Cambodia while it is only 0.35 for some ESI countries namely Indonesia, the Philippines and Vietnam. The reason for that is because Cambodia still faces more health challenges than many of the other countries in the region. Annual ALRI incidence in all under-�ve children is the incidence per child multiplied by the number of children (Table A10). The incidence of malaria is likely more uncertain than the incidence of ALRI. The regional WHO data for 2002 suggest that the incidence of malaria in SEAR-B is 0.07 cases of malaria per child per year. Indonesia holds a large share of the population in this region. The incidence of malaria in WPR-B is only 0.001 per child per year, as China constitutes more than 80 percent of the population in this region and has very low incidence of malaria. A recent paper by WHO estimates that the global incidence of malaria in 2004 was 6 times higher than recorded in national health information systems, and around 17 times higher in non-African countries [30]. The estimated country population incidence in Korenromp (2005) indicates that the incidence in under-�ve children could range from 0.16 cases per child per year in the Philippines, 0.27 cases in Vietnam, 0.39 cases in Indonesia, and 0.8 cases per child in Cambodia17. These estimates are, however, very uncertain. A much more conservative estimate would be to assume that the incidence in under-�ve children in Indonesia is 0.07 cases per child per year (as reported for SEAR-B for the year 2002) and that the incidence in the other countries are in the same proportion relative to the estimated incidence in Korenromp (2005). This approach gives an estimated incidence of 0.14 in Cambodia. Using this incidence rate, annual cases of malaria in under-�ve children in Cambodia are presented in Table A10. Table A10. Estimated annual cases of illness in children under �ve (thousand cases) Disease Annual cases ALRI 847 Malaria 240 Sources: Estimated from regional WHO incidence data (WHO, 2004b) and Korenromp (2005). Applying equation (2) to the cases of mortality and illness in Table A8 and Table A10 provides an estimate of mortality and morbidity from poor sanitation. Based on the estimates, in Cambodia, the under-�ve mortality directly attributable to poor sanitation (i.e., diarrheal mortality) constitutes 19% of total under-�ve child mortality. Mortality attributable to sanitation from malnutrition (i.e., the indirect effect of infections through malnutrition) constitutes 18% of total under-�ve child mortality. Total 17 Korenromp only present population incidence. The WHO regional data indicate that the incidence in under-�ve children in SEAR-B is 4.5 times higher than the population incidence. This ratio is applied to the estimated population incidence in Korenromp to estimate incidence in under-�ve children. Economic Impacts of Sanitation in Cambodia 74 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES attributable mortality to sanitation is 37% of total under-�ve child mortality (Table A11). For morbidity in children under �ve years, ALRI attributable to sanitation from malnutrition constitutes 19% of annual cases, and malaria attributable to sanitation constitutes 8% of annual cases (Table A12). Table A11. Percent of total under-�ve child mortality attributable to poor sanitation Category % of total under-�ve child mortality Directly attributable mortality to sanitation 19 Attributable mortality to sanitation from malnutrition 18 Total attributable mortality to sanitation 37 Table A12. Percent of cases of illness in children under �ve attributable to poor sanitation Diseases % of total child morbidity ALRI attributable from malnutrition 19 Malaria attributable from malnutrition 8 Despite the fact that poor sanitation causes several indirect diseases among under-�ve children, the subsequent calculation of economic costs of indirect diseases includes only diarrheal diseases, ALRI, measles, and malaria. The reason is that those diseases have more reliable data for estimation. A1.4 Health care cost estimation Health care costs result from diseases associated with poor sanitation and hygiene. In order to estimate health costs related to disease, it is necessary to compile information on disease rates for the selected diseases, treatment seeking rates, as well as health systems variables such as treatment practices and unit costs. Health care costs can fall on both the patient and the public health system, depending on where the sick person seeks care from and the tariff rates in public facilities. Private health care is assumed to be fully �nanced by the patient. Costs are both �nancial and economic in nature. Financial costs include the marginal cost to treat patients at public facilities (mainly drugs), patient transport costs, as well as the full costs of treatment in private clinics or self- treatment. In the absence of data on the actual production costs of health care provided by the private sector, the tariffs are taken to reflect the health care costs. Economic cost includes the �nancial costs plus the short-term �xed costs of public health facilities such as staff, capital items and overheads. In order to estimate the costs of health care, it is necessary to know the total number of cases seeking health care from different providers. Given that government statistics are often incomplete, public facility treatment seeking �gures were adjusted to reflect the total cases seeking care. Table A13 shows CSES data (2004), which shows where households seek care from, by diseases. It is assumed that these data are the most reliable given the lack of alternative data on treatment seeking by disease. The data from CSES appears to provide a different perspective on health treatment seeking behavior than the CDHS data (2005). While CDHS indicates the majority of treatment sought in private clinic (60%), CSES shows that most sanitation and hygiene related diseases patients seek treatment at public hospital or self treatment. Yet, since CDHS provides only the data of treatment seeking for all diseases or injuries combined, it can be assumed that CSES is more useful and reliable in this sense. In CSES, the information on treatment seeking behavior of various diseases can be extracted, where sanitation and hygiene related diseases are also included. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 75 ANNEXES Table A13. Treatment seeking behavior, by provider % seeking treatment from: No Disease Public Private clinic Private Self-treatment treatment Other (%) provider (formal) informal (pharmacy, other) (%) Diarrheal diseases 10.5 8.6 16.3 33.5 29.1 2.1 Skin diseases 76.0 5.7 1.7 7.0 9.3 0.3 ALRI 37.4 11.2 6.7 22.6 20.8 1.4 Measles 76.0 0.0 0.0 0.0 24.0 0.0 Malaria 54.1 10.7 4.1 19.7 9.3 2.1 Source: CSES 2004 Based on the number of reported cases in the public health system, and the place of treatment seeking (from Table A13 above), the total cases seeking treatment can be estimated for each health care provider, and for each region of the country. Table A14 below presents the �gures after they have been adjusted for attribution to poor sanitation and hygiene: 88% for diarrhea [9], 50% for malnutrition [10], and 50% for skin disease. For ALRI, measles, and malaria, the estimation method is mentioned in Annex A1.3 above. The reported cases have been adjusted by underreporting of the system due to poor record keeping and some health facility reports being delayed or not sent, and hence not included in national reporting. Table A14. Estimated numbers of disease cases attributed to poor sanitation and hygiene seeking care from different providers, year 2005 Public sector Private sector Reported Estimated Self- % under- Formal clinic Informal Disease cases actual cases treatment reported (%) (‘000) care (‘000) (‘000) (‘000) (‘000) Diarrheal 621.4 37 979.5 803.2 1,528.1 3,134.5 diseases1 Skin disease 101.4 8 109.9 8.2 2.4 10.1 Malnutrition 0.6 33 0.9 - - - ALRI2 50.2 16 59.7 17.9 10.7 36.0 Measles - - - - - - Malaria2 1.3 87 10.3 2.0 0.8 3.8 Source: HIS (2005) for reported cases; CSES (2004) for Health Seeking Behavior 1 Estimated based on CDHS 2005 incidence for under-�ve children and WHO for over-�ve population 2 Estimated based on WHO and other studies (see Annex A1.3) In order to calculate the costs associated with the cases seeking health care, it is necessary to know the treatment practices, the proportion of cases that are admitted for inpatient stay, and the costs associated with health care. Table A15 shows these variables for treatment seekers who receive their care on an outpatient basis from public providers, formal private providers (private hospitals and private clinics), informal carers (doctor/nurse’s home, traditional treatment), and self-treatment. Economic Impacts of Sanitation in Cambodia 76 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES Table A15. Health service use and unit costs associated with outpatient care Average Financial cost Full unit cost Other patient Provider and disease patient tariff (US$)1 (US$)1 costs (US$)1 (US$)1 Public provider Diarrheal diseases 0.2 0.6 1.6 1.8 Skin disease 0.2 1.2 2.2 1.8 Malnutrition 0.1 1.1 1.8 ALRI 0.2 0.8 1.8 1.8 Measles 0.2 0.4 1.4 1.8 Malaria 0.2 1.4 2.4 1.8 Formal private provider Diarrheal diseases 3.1 3.1 0.6 Skin disease 3.7 3.7 0.6 Malnutrition 2.6 2.6 0.6 ALRI 3.3 3.3 0.6 Measles 2.9 2.9 0.6 Malaria 3.9 3.9 0.6 Informal private provider Diarrheal diseases 1.8 1.8 0.3 Skin disease 2.4 2.4 0.3 Malnutrition 1.4 1.4 0.3 ALRI 2.0 2.0 0.3 Measles 1.6 1.6 0.3 Malaria 2.6 2.6 0.3 Self-treatment Diarrheal diseases - 0.4 0.4 - Skin disease - 0.5 0.5 - Malnutrition - - - - ALRI - 0.6 0.6 - Measles - - - - Malaria - 0.7 - - 1 All cost �gures reflect the cost per outpatient consultation Note: The exchange rate as of June 1, 2007: 1US$= 4,050 Cambodian Riels The tariff and �nancial cost of disease is obtained from a survey conducted by the Economic Institute of Cambodia (EIC) and interviews with staff from a health center. It is also assumed that drug provided in private care is exactly the same as the drug given in public service for all diseases. Moreover, the calculation of �nancial cost for private clinics also includes a consultation fee, which is 10,000 Cambodian Riels (US$2.47) for formal private care and 5,000 Cambodian Riels (US$1.23) for informal private care. The economic cost of public service facility is assumed to be 3991 Cambodian Riels (US$0.985) per outpatient visit, sourced from regional unit cost data from WHO18. It is assumed that the average number of outpatient visits per case is one. 18 http://www.who.int/choice/country/khm/cost/en/index.html Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 77 ANNEXES Table A16 below shows the same variables for inpatient care for public and formal private providers, including the proportion of cases admitted, average length of stay per patient, and associated costs per inpatient day. The tariff and �nancial cost of disease was obtained from an interview with staff of health center. It is also assumed that drug provided in private care is exactly the same as the drug given in public service. The only difference is the hospitalization fee per day in private care. The economic cost of public service facility is assumed to be 6,587 Cambodian Riels (US$1.626) per bed for inpatient19. Moreover, the ‘hotel’ component of formal private provider ranges from US$10 to US$20 – with an average of US$1520. Moreover, “other patient costs� refers to the transportation cost to and from healthcare service taken from CDHS (2005). Regarding the admission rate, due to the lack of information concerning private hospital, it is assumed that the admission rate of private hospital is the same as the public hospital. Table A16. Health service use and unit costs associated with inpatient care Per inpatient day (US$) Other % cases Days Average patient Provider and disease admitted admission Financial Full unit patient costs (%) per patient cost cost tariff (US$) Public provider Diarrheal diseases 4.5 4.0 0.2 1.2 2.8 1.8 Skin disease - 1.5 0.3 0.4 2.0 1.8 Malnutrition - 15.0 - 0.4 2.0 1.8 ALRI 4.5 5.0 0.2 0.5 2.1 1.8 Measles 1.0 5.0 0.2 0.4 2.0 1.8 Malaria 14.7 7.2 0.2 2.4 4.1 1.8 Formal private provider1 Diarrheal diseases 4.5 2.7 16.2 16.2 0.6 Skin disease - - - - - - Malnutrition - - - 15.4 15.4 0.6 ALRI 4.5 8.3 15.5 15.5 0.6 Measles 1.0 - - - - - Malaria 14.7 3.3 17.6 17.6 0.6 1 This includes private clinics and private hospitals The exchange rate as of June 1, 2007: US$1 = 4,050 Cambodian Riels A1.5 Health-related productivity cost estimation Disease takes people away from their occupations and daily activities, and regular sickness-related absences from school affects the ability of children to keep up with the curriculum and complete their education. Therefore, time lost from work, school or daily activities has a value. Given that time off work is determined by the severity of the disease, as well as whether the case was treated or not, assumptions were made on the proportion of cases that are severe, and the treatment seeking behavior associated with these cases. Table A17 below shows the proportion of severe and non-severe cases based on the categorization of disease and the severity level given in CDHS, and the day-off daily activities by severity of disease. The following assumptions are made to compute the table below: 19 http://www.who.int/choice/country/khm/cost/en/index.html 20 Obtained from an interview with private medical staff Economic Impacts of Sanitation in Cambodia 78 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES • For all diseases, except malnutrition, the severe and non-severe cases is assumed according to the CDHS 2005 where the severe cases are about 12% and non-severe cases (slight to moderate) are 88%. The severe case rate of malnutrition is taken separately from CDHS 2005 where there is 7% severe, and 93% non-severe. • It is assumed that all the severe cases will seek treatment at any healthcare service. So, there is no severe case which is not treated. More importantly, for treated cases, only severe cases will seek hospitalization. In this case, the day-off activities will be two days more added to the length of hospitalization to reflect the consultation and traveling to healthcare service. • The episode of disease for non-severe cases but treated is assumed to be 50% of the day-off activities of severe cases, and productive time loss from this non-severe disease is only 2 hours a day, except malnutrition where 4 hours a day (half day) is assumed. This productive time lost is applied to each day of disease episode. Moreover, another half day productive time loss is added to reflect the time spent for consultation and traveling for the non-severe cases. • For the untreated cases, the number of days off daily activities is assumed to be the same as the episode of the treated non-severe cases. Hence, the number of days off normal activities of untreated cases is equal to the days lost of treated non-severe cases less the traveling time to healthcare service of half a day. Table A17. Variables for estimating amount of time lost from disease % cases Days off daily activities (days) Disease Non- Treated Not treated Severe severe Severe Non-severe Non-severe Diarrheal diseases 12.3 87.7 5.8 1.2 0.7 Skin disease 12.3 87.7 3.5 0.9 0.4 Malnutrition 6.9 93.1 17 4.8 4.3 ALRI 12.3 87.7 8.3 1.5 1.0 Measles 12.3 87.7 7.0 1.4 0.9 Malaria 12.3 87.7 9.2 1.6 1.1 Given that time off work has an opportunity cost, and in some instances a real �nancial loss, time away from daily activities needs to be given a unit value to estimate overall �nancial and economic losses associated with disease. A commonly applied economic valuation technique for time loss is the human capital approach (HCA), which values time loss according to what the sick person could be earning in productive employment. Even when the person would not be earning income (especially in the case of children), time for leisure and other activities can be assumed to have a value greater than zero [31-34]. A second common approach, which measures the sick person’s willingness to pay to avoid disease, can more accurately reflect the welfare effects of disease, but due to lack of data on willingness to pay in Cambodia, this approach is not used in this study. Hence HCA is used as it is simple and it reflects the time loss component of disease. This study distinguishes between �nancial and economic cost. For some adults, time spent away from productive activities will have a direct income-loss, while for others the salary may be paid for a maximum number of sick days per year. Given the self-employed and/or agricultural nature of agrarian societies in Cambodia, loss of time from productive activities may not have immediate �nancial loss, but may lead to income-losses in the future unless a family member or business partner replaces their lost labor. In order to be conservative, �nancial cost is estimated as immediate income loss for those not paid their wage or earning an income from time lost due to sickness. In Cambodia, 70% of adults are crudely assumed to actually lose income by not working when getting sick. For those not directly losing income, there will also be a welfare loss, which may include longer-term income- earning potential as mentioned above. In estimation of economic cost, this study recognizes the value of time lost from daily activities, whether productive working time, school time, or leisure time. Given that value of time varies Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 79 ANNEXES according to what the person is doing with their time, economic ‘welfare’ losses are valued at less than the �nancial losses described above. Research studies have shown a whole range of results on the value of time. This present study takes the economic value of time as 30% of the unit value of time. Furthermore, this study distinguishes between the value of adults and of children’s time, given that children do not generally have the same values as adults. On the other hand, children’s time is not worthless, given that children are or should be at school learning and hence time away from school would mean lost education and eventually lower income levels [35]. Also, for young children of non-school age, sickness will involve more time input from a carer, and hence incur a cost. In study countries, caring for a child is mostly the mother’s task and thus ill children are more likely to take the time of women than men, thus hindering women from working. Given the limited empirical work on the value of children’s time, and very few precedents in terms of valuing children’s time, a time value of 50% of adults time is given in this present study [36]. Table A18 shows some alternative sources of economic value by region, comparing Gross Regional Product (GRP) per capita, compensation of employees, minimum wage and average wage. The annual value was converted to hourly value by assuming 8 working hours per day, and 236 working days (public holiday subtracted). Hourly minimum and average wages were converted to annual �gures by using the reverse calculation. Compensation of employees was considered the most appropriate global �gure to reflect the average value of time, given that it reflects the amount paid to all formal employees. Compensation of employees per capita was calculated at regional level by multiplying national compensation of employees by the ratio of GRP per capita at regional compared to national level, and dividing by the total employment. Table A18. Comparison of alternative sources of time value (year 2005) Average GDP per capita Minimum wage Average wage over compensation of (US$) (US$)1 all sectors (US$)2 employees (US$) Annual Hourly Annual Hourly Annual Hourly Annual Hourly Phnom Penh 1,421 0.75 1,555 0.82 600 0.32 933 0.49 Plains 323 0.17 353 0.19 600 0.32 212 0.11 Tonlé Sap 351 0.19 384 0.20 600 0.32 230 0.12 Coastal 503 0.27 550 0.29 600 0.32 330 0.17 Plateau 348 0.18 381 0.20 600 0.32 228 0.12 National 447 0.24 489 0.26 600 0.32 293 0.16 Sources: Compiled using data from National Institute of Statistics 1 Minimum wage for garment factory workers (US$50 per month) 2 Estimated by Economic Institute of Cambodia A1.6 Premature death cost estimation Cost of premature death is calculated by multiplying the number of deaths by the unit �nancial and economic value of a death. Data on the number of premature deaths from health information systems are unreliable, due to underreporting as well as misdiagnosis. For example, the annual number of reported deaths in Cambodia using HIS for under �ves is only 954, while the predicted numbers of deaths based on under-�ve child mortality rates and birth rate is about 30,600. Clearly, the �gures appear to grossly underestimate the actual death rate. Therefore, Table A19 below is the estimated death due to poor sanitation and hygiene by disease. It should be noted that the �gures presented in the table for under-�ve children are estimated based on WHO data and other studies described in Annex A1.3. The diarrheal mortality for population over �ve years are estimated based on the Economic Impacts of Sanitation in Cambodia 80 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES WHO’s global burden of disease regional data, where about 10% of the total diarrheal deaths are among over-�ve population. Moreover, the number of deaths by age group among over-�ve population is allocated by using the WHO data for WPR-B for diarrheal case fatality rate which are 0.0002207 for 5-14, 0.0000949 for 15-59, and 0.000948 for adults 60 years and over. According to the data in Table A19, the number of under-�ve children deaths from poor sanitation and hygiene accounts for about 30% of the total child mortality in Cambodia. This �gure may be much higher if other indirect diseases are included (i.e., 37% according to Table A11). Table A19. Estimated number of annual deaths from poor sanitation & hygiene Age grouping Disease 0-4 5-14 15+ Total Diarrheal diseases 5,808 448 344 6,600 Skin diseases N/A N/A N/A N/A Malnutrition N/A N/A N/A N/A ALRI1 1,786 N/A N/A 1,786 Measles1 420 N/A N/A 420 Malaria1 1,033 N/A N/A 1,033 1 Estimated based on WHO and other studies (See method in Annex 1.3) N/A – Not available or estimated Premature death affects society in a number of ways, and has proven to be difficult to value with any degree of precision. As a result, economists have employed a range of methods for valuing premature loss of human life [37]. The most tangible economic impact of premature death is the loss of a member of the workforce, with implications for the economic outputs generated. Hence, this approach, what has been termed the ‘human capital approach’ (HCA) approximates the welfare loss by estimating the future discounted income stream from a productive person, from the time of death until the end of (what would have been) their productive life. However, this technique has been criticized for the fact that it values human life exclusively for its productive potential. Empirical evidence indeed proves that life has a value beyond the productive worth of a human, which both society as a whole and individuals are willing to pay for in order to safeguard [38, 39]. Various other methods are available to estimate this broader economic as well as inherent worth of human life. Two major methods are used to value life: (1) observations about actual market and individual behavior with respect to what they pay to reduce the risk of death (e.g. safety measures) or what they are willing to accept for an increase in the risk of death (e.g. wage premium for risky jobs). This approach is known as ‘hedonic pricing’. (2) Stated preference from individuals exposed to risk, using interview technique. This approach is known as ‘contingent valuation’. Both these approaches estimate directly the willingness to pay of individuals, or society, for the reduction in the risk of death, and hence are more closely associated with actual welfare loss compared with the human capital approach. The problem in valuing life is that the alternative methods can give very different estimates of the value of life, and applications of the same techniques to different contexts can also reveal very different implicit values in reducing the risk of death. For example, willingness to pay studies generally show greater value of life than the human capital approach. These variations and differences will affect the credibility of economic studies when used for policy decisions, and hence considerable care is needed in estimating and presenting the economic impact of premature loss of life to policy makers. Therefore, in order to sound more plausible to policy makers, this present study uses the more conservative human capital approach, described below. Sensitivity analysis explores the implications of alternative values for loss of human life using the willingness to pay approach. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 81 ANNEXES Human capital approach The human capital approach summates the future years of income at the average age of death. Given lack of data on exact age of death, three time points of death were used: 2 years of age for the 0-4 age group; 9 years of age for the 5-14 age group; and 40 years of age in the 15+ age group. The discount rate applied was 3%, reflecting the social rate of time preference approximated by the long-term real interest rate. Also, given that per capita income grows over time, a presumed long-term per capita income growth of 2% was applied to future incomes. Average income was taken from the average compensation of employees, and adjusted to sub-national level by applying GRP per capita ratios. For the younger age groups who will not be in the work force for several years, the net present value of future earnings are further discounted to take this into account. Values are shown in Table A20. Financial costs of premature death were approximated using the human capital approach by assuming a coping period following the loss of an adult member of the family. The coping period could be the period after which the income of the lost adult is expected to be replaced. A period of one year is conservatively used in this study. Therefore, the average compensation of employees for a single year is applied to the number of adult deaths to estimate the �nancial impact of premature death. In this case, the average compensation of employees used is US$489 for Cambodia. Willingness to pay approach Given the lack of estimates of willingness to pay for avoiding death in developing countries, and Southeast Asian countries in particular, the bene�ts-transfer method was applied for the willingness-to-pay method. This essentially involves taking value-of-statistical-life (VOSL) values from a meta-analysis of studies in developed countries and transferring the value directly using an adjustment for differences in income. While this approach has many weaknesses [40], the absence of data from developing countries justi�ed the bene�t transfer approach. The VOSL reported in North American and European studies was highly variable, ranging from around US$1 million to more than US$10 million [38, 41-45]. A meta-analysis of 40 VOSL studies reported by Bellavance et al in 2007 reported average VOSL of US$9.5 million and median VOSL of US$6.6 million [46] , similar to the mean estimate of US$5.4 million found by Kochi et al (2006) [47]. Developing country studies are few. A study of the Indian labor market found VOSL varying from roughly US$0.14 to US$0.38 million [48]. Given the large number of studies from OECD countries, an adjusted bene�t transfer is justi�ed, using a highly conservative VOSL estimate of US$2 million. This value is signi�cantly lower than the values presented in the meta-analyses conducted by Bellavance (2007) [46] and Kochi (2006) [47], but consistent with the mid-range in the meta-analysis conducted by Mrozek and Taylor (2002) [49]. This value also reflects the lower end of the US$2 million to US$4 million recommended by Abelson for public policy [38]. The VOSL of US$2 million is transferred to Cambodia by adjusting downwards by the ratio of GDP per capita in Cambodia to GDP per capita in the USA. The calculation is made at both official exchange rates (more conservative) as well as at purchasing power differences (less conservative), and assuming an income elasticity of 1.0. Direct exchange from higher to lower income countries implies an income elasticity assumption of 1.0, which may not be true in practice. Therefore, the bene�ts transfer from OECD studies was also made at income elasticity of 0.8 and 0.6. Values are shown in Table A20. Economic Impacts of Sanitation in Cambodia 82 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES Table A20. Unit values for cost of a premature death Values (US$, year 2005) Variables Low Base case High Human capital approach (economic)1 0-4 years 12,098 17,223 36,721 5-14 years 14,913 20,328 40,492 15+ years 9,569 10,795 13,895 Human capital approach (�nancial) 489 489 489 Willingness to pay using bene�ts transfer method2 VOSL Income elasticity 1.0 at OER 10,219 20,439 40,878 VOSL Income elasticity 0.8 at OER 25,559 51,118 102,236 VOSL Income elasticity 0.6 at OER 63,923 127,846 255,692 Input values Value of a Statistical Life (VSL) in USA 1,000,000 2,000,000 4,000,000 GDP p.c. in Cambodia (using OER) 447 447 447 GDP p.c. in USA (using OER) 43,740 43,740 43,740 GDP p.c. in Cambodia (using PPP) 2,929 2,929 2,929 GDP p.c. in USA (using PPP) 41,950 41,950 41,950 1 Low and high values are produced by using income per capita growth of 1% and 4% (base case 2%) 2 Low and high values are produced by using US$1 and US$4 million as VOSL (base US$2 million) A2 Water resources The 2003 United Nations Report “Water for people, Water for life� states that many rivers, lakes and groundwater resources are becoming increasingly polluted, and that human excreta is one of the most frequent sources of pollution [50]. In Southeast Asian countries, a signi�cant proportion of human excreta is flushed directly into water resources due to low coverage of sewage treatment for piped sewerage, or else human excreta eventually �nds its way into water resources through open defecation, leaking septic tanks or seepage from pit latrines. As a result, levels of suspended solids in rivers in Asia have risen by a factor of four over the last three decades and Asian rivers have a higher biological oxygen demand and bacterial content than the global average [50]. The results of polluted water on human activity are many: previously safe drinking water sources are rendered unusable, and water becomes less productive or less usable for agricultural purposes including �sh production, or for industrial and domestic uses. According to the Asian Development Bank, the threat to �sh production is especially important, given the economic importance, subsistence value as well as nutritional value of �sh in the Southeast Asian region [51]. Domestic sources contribute importantly to water pollution in most developing countries, where the majority of households do not have their sewage or wastewater safely disposed of or treated. However, the presence of other sources of water pollution means that overall economic impact of polluted water cannot be attributed to poor sanitation alone. Pollutants which affect water-related economic activity include microorganisms, organics, chemicals, solids, gases and heat [52]. Pollution originates from a variety of sources: • Households (sewage and grey water from bathing, laundry, cooking) • Small industries (garments, washing, brewery) • Manufacturing industries (production or processing) • Other offices, hospitals, etc. • Chemical fertilizers, pesticides, and treatment of acid-sulfate soils • Animal waste • Silt release following build-up behind dams • Salinity intrusion from coastal areas Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 83 ANNEXES Major categories of water use include drinking water, domestic uses, crop and �sh production, energy production, industry, recreation and transport. For some of these activities, good quality water is important – such as for drinking – while for other uses water quality standards are not so strict such as for agricultural and some industrial uses. Therefore, only selected impacts of polluted water are examined in this present study, with selection of uses of water where there is a strong proven association between poor sanitation and the associated costs. Annex B3 provides full algorithms for estimation of economic impacts of water resource. A2.1 Water quality measurement Inland water quality is affected by many variables, the two main ones being the quantity of polluting substances released and the overall quantity of water resources for absorption of the pollution load. Hence, water quality indicators will need to be interpreted based on these two variables, as well as the multitude of factors that determine these variables. Furthermore, the economic impact of polluted water depends on what productive and non-productive uses the different water resources have, or could have assuming improved water quality. Water quality monitoring is limited in Cambodia. Different organizations or agencies are interested in different uses of water, and hence measure different water quality indicators. In Cambodia, Phnom Penh Water Supply Authority (PPWSA) conducts water quality tests for the purposes of supplying water to Phnom Penh residents, covering 3 locations close to the city: Chroy Changva on the Mekong, Phum Prek on the Tonlé Sap river, and Chamkamon on the Bassac river. The Mekong River Commission (MRC) conducts water quality tests in many locations in Cambodia which are mainly on the Mekong river (upstream, midstream, and downstream locations) and the Tonlé Sap Lake. Table A21 below shows selected drinking water quality standards, comparing the �gure based on WHO guideline with Cambodia standards. Some of the main implications for drinking water are pollution from infectious pathogens (microbial agents) such as E Coli, dangers from heavy metals, bad odor due to organics, turbidity caused by solids, and bad taste due to low pH and solids. Table A21. Selected drinking water quality standards Indicator Unit CNDWQS1 WHO2 Color Pt/Co 5 15 pH value Unit 6.5 - 8.5 6.5 – 8.5 Suspended solids mg/l - 1 Turbidity NTU 5 5 Total dissolved solids mg/l 800 1000 Dissolved oxygen mg/l - <10 Total coliform cfu/100ml 0 0 Thermotolerant coliform cfu/100ml 0 0 N-Ammonia (NH3-N) mg/l - 0.05-0.50 Source: Phnom Penh Water Supply Authority, 2006 1 Cambodian National Drinking Water Quality Standard 2 The �gures are adopted by PPWSA based on WHO guideline. These �gures were used by PPWSA before the existence of CNDWQS. A2.2 Contribution of poor sanitation to water pollution Water pollution from domestic sources can be estimated from the annual release or eventual seepage of untreated feces, urine and gray water into inland water bodies. It is estimated by applying the number of population with unimproved sanitation, the proportion of sewage released to water bodies, and average human (and animal) waste production per year. Table A22 presents the �gures and assumptions behind the release of human waste to water Economic Impacts of Sanitation in Cambodia 84 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES bodies. Pollution load from human waste is based on an assumption of an average 0.15 kg feces production and 1.5 liter urine production per person per day. Table A22 shows the assumptions used in this study for estimating the amount of untreated sewage that is discharged to water bodies in Cambodia. According to the Municipality of Phnom Penh (MPP), it is estimated that more than 80% of household waste water in the capital is discharged to the wetland nearby Phnom Penh. The rest goes directly to the ground and water bodies. It is stated that the sewage water treated by wetland is considered safe to discharge to the water body21. Based on the data from the Phnom Penh Department of Public Works and Transport (DPWT), the wetland can reduce the pollutants in sewage water such as suspended solids by 72%, BOD by 83%, and COD by 86%. According to Cambodian standards, this reduction of pollutants makes the waste water discharged from the wetland not overly harmful to the environment, or speci�cally to the receiving water body. On average the pollutants can be reduced by 80%. So for the purpose of this study, 20% is assumed for the sewerage system ‘leaking’ to water bodies from the zone of Phnom Penh. For other zones, however, all sewerage system (100%) are assumed ‘leak’. In Cambodia, the majority of open defecation is to land (99%) and only 1% directly in water bodies22. For septic tank, it is assumed that 90% of them are not properly managed or poorly designed without conforming to engineering standards. In addition, 40% of the effluents �nd their way to ground water. For pit latrines, 50% of the effluents are assumed to leak to the ground water bodies. Table A22. Proportion of untreated sewage discharged to water bodies % septic tanks not % sewage % Sewerage managed properly % leaking pit Region discharged directly systems % of which latrine to ground- into water body leaking % Total to ground- water1 water1 Phnom Penh 20 20 90 40 50 Plains 100 100 90 40 50 Tonlé Sap 100 100 90 40 50 Coastal 100 100 90 40 50 Plateau 100 100 90 40 50 Total 84 84 90 40 50 Source: Discussion with MRD and DPWT: leaking sewage, septic, and pit latrine 1 A crude assumption is made regarding the proportion of leak of septic tank and pit latrine that �nd their way to the ground water. It is assumed to be 40% for septic tank and 50% for pit latrine. Table A23 shows the assumptions on polluting substances discharged per day, reflecting urban households with piped water connection. Rural households without piped water connection are conservatively assumed to have zero gray water, but the same amount of sewage as urban households. According to interviews with an official of the Ministry of Environment (MoE), it is assumed that domestic waste accounts for somewhere between 60% and 70% of total water pollution in Cambodia. In this case, the attributable fraction of pollution to poor sanitation is therefore assumed to be 65%. In addition, approximately 75% of water uses are transformed into household waste water in the form of both gray water and sewage. In general, 50-80% of the total residential waste water is gray water23. For the purpose of this study, however, 65% is assumed. 21 An interview with a lecturer of Royal University of Agriculture 22 Demand assessment for sanitary latrines in rural and urban areas of Cambodia, 2007 23 http://en.wikipedia.org/wiki/Gray_water, retrieved on October 29, 2007 Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 85 ANNEXES Table A23. Waste load production per capita per day, subdivided by gray water and sewage, for urban households with pipe connection Biological Oxygen Chemical Oxygen Total Suspended Source Nitrogen (grams) Demand (grams) Demand (grams) Solids (grams) Gray water 15 40 2 48 Sewage 35 35 7 20 Total 50 75 9 68 Source: [53] A2.3 Cost implications of water pollution for drinking water supply Both water consumers and water providers treat water because water sources are not clean. Some households, particularly the more wealthy, purchase more expensive bottled water which is either chemically treated or from a protected (mineral) source. The more polluted the water source, the more likely the household will take some form of precautionary measures, which leads to higher unit cost of treatment. In some cases, households will not haul water from more polluted water sources if less polluted sources are available, but this may lead to additional time or �nancial costs. Given that drinking water sources are polluted from several sources and not just from poor sanitary practices, by removing the human (and animal) waste component of polluted water, the need to treat water is not altogether removed. However, the removal of human and animal waste content from water sources may reduce the necessity for treatment and/or it lowers the unit cost of treatment. This present study compares selected indicators from the water quality guidelines presented in Table A21 with available water quality measures to conclude how polluted water is for drinking purposes. Some of the main indicators which will cause households to purchase, treat or walk further to access cleaner water are perceived or actual presence of infectious pathogens (microbial agents) and heavy metals, bad odour due to organics, turbidity caused by solids, and bad taste due to low pH and solids. For the purposes of cost estimation, household drinking water sources are sub-divided into three categories (see Table A24 below): 1. Households receive piped water supply, sourced either from water treatment companies or from open community water sources. 2. Households purchase water from other non-piped suppliers, such as tanker truck, water by the bucket, or bottled water. 3. Households collect water from free or low cost community or public sources. Table A24 presents national level data with rural/urban breakdown. From the table, it is observed that about 282,000 households in Cambodia have piped water as their source of drinking water, while nearly 2.5 million households haul water from various sources. It is also noted that about 1.8 million households treat their water prior to consuming by using different methods. Economic Impacts of Sanitation in Cambodia 86 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES Table A24. Numbers of households by main source of drinking water Other purchased water Households Piped water (‘000) Hauled (‘000) treating water Location water Treatment Other Bottled themselves Vendor (‘000) plant source water (‘000) Rural 112.4 - 72.6 14.1 2,142.7 1,491.7 Urban 169.6 - 20.6 28.7 218.9 352.0 Total 282.0 - 93.2 42.7 2,361.5 1,843.6 Source: CDHS (2005) According to national data, households receiving their water supply from all sources tend to treat their water for drinking purposes, even when the water sources have been protected or properly treated. Several methods exist for household water treatment, including chemical treatment, �ltration, boiling, and solar disinfection. However, national surveys indicate that the main treatment method in Cambodia is boiling (CDHS, CSES). In addition to households, some industries may also have to treat water that does not conform to the required properties for certain industrial processes. However, these costs are excluded in this present study due to the lack of information. Various methods are available to estimate costs to avoid drinking polluted water. The lower bound on �nancial cost could be reflected by identifying speci�c actions by water providers and households to remove bacteria, such as chemical treatment for piped water, open wells, and household treatment. The upper bound on �nancial cost of water pollution can be reflected by apportioning to poor sanitation a fraction of the total cost of water treatment and purchase. However, it is noted that households choose more convenient but more costly water sources for a variety of reasons, which include (but are not limited to) water pollution, convenience of access, time savings and no other available water supply. Hence the tariffs paid by households are adjusted downwards by a further 50% to account for these other bene�ts of piped water supply. This adjustment factor is referred to as Factor A. The economic costs of accessing (clean) drinking water includes not only the �nancial cost, but also the efforts made by households to access cleaner drinking water sources, such as walking further to protected wells or boreholes, and the time taken to treat water in the home. For accessing rainwater, however, the study assumes zero access cost. Algorithms used to calculate �nancial and economic costs of avertive behavior to avoid consumption of unclean water for drinking purposes are provided in Annex B3. Algorithms were applied at sub-national (regional) level, using data available on drinking water sources, prices, and household water treatment practices The minimum drinking water per capita per day was assumed to be 4 liters, taken from WHO sources [54]. For hauled water, the proportion of households traveling further to access cleaner water was crudely assumed to be 10% to reflect the proportion of populations living close to polluted water sources which are unusable for drinking water purposes (lakes, rivers and polluted groundwater); while additional journey time was taken from surveys of time for collecting drinking water i.e. from CDHS 2005. The value of time reflecting the welfare loss for those traveling to fetch water for drinking is assumed as 30% of an adult’s income. The attribution to poor sanitation of the overall costs of sourcing clean water was made differently for the different water sources: • Overall attributable water pollution to poor sanitation (Factor B): a judgment is made about this fraction based on available data on releases from industry, agriculture and domestic sources. It is assumed that this attributable fraction is 65%. • Attributable fraction of water purchased due to poor sanitation (Factor C): given concerns of households about bacteria in water available from open sources, this fraction may be higher than the fraction above. Other characteristics of water available from local sources (taste, color, cloudiness) are also taken into account, which often account for switching sources. It is assumed that this fraction is 70%. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 87 ANNEXES • Attributable fraction of water treated (Factor D) at the household level due to poor sanitation is assumed to be 70%. In addition, due to the scarcity of data on costs that are used for the estimation of economic costs of water resource, some assumptions are made to enable estimation. The assumptions are, however, based on discussion with stakeholders, local citizens, and experts. Table A25 shows the assumed unit cost of water and water treatment (boiling). According to the table, piped water tariff is US$0.07 per m3 in urban areas, while it is more than US$0.3 per m3 in rural area. This can be explained by the fact that water supply in urban areas is more efficient than in rural areas. The water tariff for purchased water from vendor is roughly 15 times more than piped water tariff in rural areas, and 35 times that in urban areas. The bottle water is even more expensive with average cost of US$43 per m3 calculated from a standard 20-liter bottle water. The water treatment cost is calculated for boiling method as it is the main method used in Cambodia. According to the table, the cost of boiling water in urban area is roughly US$16.5 per m3 while it is only US$8 per m3 in rural areas due to abundant fuel wood in rural areas which can be used for boiling water. Table A25. Unit cost of water Type of water use Unit cost (US$/m3) Piped water tariff Rural 0.34 Urban 0.07 Purchased water (vendor) Rural 4.94 Urban 2.47 Purchased water (bottle) Rural 43.21 Urban 43.21 Water treatment cost Rural 8.23 Urban 16.46 Source: Author’s estimate based on actual market price A2.4 Water quality and domestic uses of water In addition to the uses of surface and groundwater sources for domestic use, industrial use, �sheries and agriculture, water is an essential ingredient to many other human and non-human activities [55]. In the present study, it is not possible to conduct an exhaustive analysis of all the different uses of water. However, the following three categories were assessed for relevance: • Non-commercial household (domestic) activities • Leisure activities [56] • Wildlife, covering flora and fauna and animal species. Non-commercial activities are concentrated at the household level, and include water for cooking purposes, washing clothes and kitchenware, and personal hygiene. Also, some traditional customs as well as leisure activities are closely related to water. Again, some of these require good quality water given it will be ingested, while others do not require quality water. Economic Impacts of Sanitation in Cambodia 88 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES Activities affected by below standard water quality were assessed in the following way: • Proportion of households and population that use untreated or unprotected surface or ground water for cooking, washing and bathing. Even though, the quality of water for non-drinking purpose may be lower than that for drinking, the water quality for domestic used needs to be at an acceptable level. The polluted water can contaminate the human body through these activities. However, in many areas in Cambodia, people usually use well, pond or open river for washing and bathing. • Proportion of households and population that switch water source due to the preference to have clean water for domestic activities. For example, for laundry and bathing, purchased water (via pipe or vendor) may be used rather than using local water bodies. Although this case may exist, the fact that people choose other sources is not mainly because the local water source is polluted by poor sanitation and hygiene. The main reason for this is that the local water bodies in some regions of the country contain some minerals creating unfamiliar smell to users and coloring clothes after laundry24. • Extent of other cultural and leisure activities related to water, and that require water of a minimum quality standard, including swimming. Wildlife, large and small, is dependent on water resources. For plants and trees, much of the water need is met from rainfall, and hence water pollution is not an issue. Some plants and trees, most animals and all aquatic life depend on surface water and therefore can be affected by water polluted from different anthropogenic sources. Despite the importance of water use in the three categories above, in this study only the costs of accessing domestic water are evaluated quantitatively. For the purpose of estimation, the study adopts the domestic water use of 28 liters per day25 (including drinking water of 4 liters), which is based on minimum requirements de�ned by WHO. It is, however, assumed that purchased water users use water twice less than those using piped water, i.e. 50% of water used by piped water users. A2.5 Water quality and �sh production value Fisheries in Cambodia Fisheries play a very important role in Cambodia in providing employment, income and food security to many people. The vast majority of people in Cambodia depend upon �sh as a central part of their diet from which they obtain protein and vital nutrients and vitamins. It is estimated that 90% of animal protein consumed by the people in central Cambodia is from �sh [57]. On average, however, the protein from �sh accounts for 75% of the total animal protein [58]. It is estimated that 11% of the households in the �shing-dependent communes around the great lake and along the Mekong and Bassac Rivers are engaged in full-time �shing and related activities, while other 35% are engaged in part-time [58]. Currently, the �sheries sector employs around 259,000 people and contributes 7.3% to the country’s GDP in 2006. As of 2006, Cambodia exported 12,000 tons of fresh water �sh, 7,500 tons of marine water �sh, and 6,835 tons of other seafood26. Pollution and �sh production Pollution and river diversion have driven freshwater �sheries into collapse worldwide, and the extinction of freshwater species far outpaces the extinction of mammals and birds [59-61]. Fish populations are affected by a multitude of changes taking place due to human interventions, such as hydroelectric dams, water diversions for agriculture, flood control levees, dredging, water pollution, and habitat degradation such as logging. According to the Food and Agriculture Organization (FAO), the long-term productivity of �sh stocks is related to the carrying capacity of their environment, which alters as a result of natural variability and of changes induced by human activity, such as coastal habitat degradation, destructive �shing methods and pollution.� (page 47, [62]). Environmental degradation has been cited as one of the key threats to inland �sh producers in countries of the lower Mekong basin [63]. Of 24 An interview with Dr. Chea Samnang, Ministry of Rural Development 25 State of the Environment report 2004, MoE (p.77) 26 Ministry of Agriculture, Forestry and Fisheries Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 89 ANNEXES particular concern for water quality for �sh production in Southeast Asia are suspended solids, dissolved oxygen, heavy metals and pesticides [64]. However, as one of the few publications on water quality and �sh production in Asia notes (writing in 1986) “data on the effect of water quality on Asian species of �sh are not readily accessible� (page 15, [64]). Furthermore, it is difficult to predict the exact impact of water pollution on �sh production given variations between �sh species and the multiple other determinants of �sh production such as food availability, water depth, flow, and temperature. Domestic sources contribute importantly to water pollution, largely through the biological oxygen demand (BOD) exerted by organic matter which reduces dissolved oxygen levels [65]27. Fish living below a sewage treatment plant have a signi�cantly higher mortality rate than �sh upstream [66-68]. Pharmaceutical discharge in urine can affect �sh health directly. The scienti�c literature testi�es, albeit incompletely, to the adverse effects of sewage release on �sh reproduction and �sh growth. One key determinant of �sh health which has received attention from scientists is the level of dissolved oxygen [66- 76]. For example, experiments undertaken in Canada on native �sh and benthic macroinvertebrate species showed that exposure to low dissolved oxygen and low temperatures caused delays in hatching of eggs, reduced mass of �sh post-hatch, and depressed feeding rates and lowered survival of �sh [72]. Additionally, micro-organisms contained in human and animal waste such as parasites and bacteria have a number of implications for �sh health [63, 70, 71, 75, 76], as well as safety of �sh for human consumption [66, 69, 77-80]. Common illnesses from contaminated �sh and shell �sh include typhoid, salmonellosis, gastroenteritis, infectious hepatitis, Vibrio parahaemolyticus and Vibrio vulni�cus infections, paralytic shell�sh poisoning (PSP), and amnesic shell�sh poisoning (ASP). A further consideration that needs to be addressed is the fact that, in many contexts, the nutrients from sewage act as a source of food for �sh, and hence positively affect the production of �sh. This happens both intentionally, when sewage is fed to farmed �sh in a regulated way, and unintentionally when �sh in open water bodies are exposed to untreated sewage disposed upstream. Hence, in recognizing the bene�ts of sewage for �sh production, the impact analysis addresses only unregulated, unintentional, pollution of water with sewage. It should be noted, though, that sewage-fed farmed �sh may not be optimally managed, and negative health effects may result. Methods for modeling the relationship between sewage release and �sh production Given the lack of empirical evidence linking water quality and �sh production in Cambodia, this study uses innovative methods to examine the likely importance of sewage release for �sh production. While the following three key links are identi�ed, only the �rst is assessed quantitatively in this study: • The proven link between sewage and dissolved oxygen levels, and the resulting impact of lowered dissolved oxygen levels on �sh production28. 27 A major determinant of �sh reproduction, growth and survivability is dissolved oxygen (DO). When an organic waste is discharged into an aquatic system, a biological oxygen demand (BOD) is created. BOD is a measure of the oxygen required to break down organic compounds, and high BOD levels signi�cantly deplete the amount of dissolved oxygen in surface water. Consequently high BOD levels have a detrimental effect on the health of aquatic species that require elevated levels of DO. From human waste, damages result from direct biological oxygen demand, as well as increased growth of algae from nitrates and phosphorous contained in human waste. The algae biodegrade the nutrients, thus reducing the amount of DO available. 28 Dissolved oxygen was selected as the key water quality parameter because aquatic organisms require oxygen in speci�ed concentration ranges for respiration and efficient metabolism, and because dissolved oxygen concentration changes above or below this range can have adverse physiological effects. Even short-lived anoxic and hypoxic events can cause high mortality rates of aquatic organisms. Exposure to low oxygen concentrations can have an immune suppression effect on �sh which can elevate their susceptibility to diseases for several years. Moreover, the toxicity of many toxicants (lead, zinc, copper, cyanide, ammonia, hydrogen sul�de and pentachlorophenol) can double when DO is reduced from 10 to 5 mg/L. The amount of oxygen available in the water also decreases with temperature and when plants die. Oxygen requirements increase at a higher temperature (e.g. an increase in water temperature from 10 to 20°C at least doubles the oxygen demand). The presence of other pollutants such as nitrogen and marine life overcrowding reduce DO levels. In cloudy conditions, plants use up more of the available DO. Plants proliferate with the presence of nitrate and phosphates from agricultural run-off, sewage and excess �sh feed. Economic Impacts of Sanitation in Cambodia 90 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES • The proven link between micro-biological contents of water and �sh disease, and hence survival. • The link between micro-biological contents of water inhabited by �sh and the transmission of disease to humans via �sh consumption, due to inadequate de-contamination of �sh prior to consumption. This study assesses the water quality indicators available for different freshwater locations where �sh are (or used to be) farmed or caught, and assesses the various issues related to �sh reproduction, �sh populations, and overall �sh health, and attributes estimated economic impact to poor sanitation (sewage and grey water release) as one of several sources of water pollution in those water bodies. The focus of this study is on freshwater �sh, given that dissolved oxygen is more affected in water bodies where oxygen depletion is more acute, resulting from release of untreated sewage into freshwater. It is recognized that the impact of poor sanitation on �sh stock, �sh growth and eventual �sh catch is extremely difficult to quantify. Coefficients linking water body pollution and yield reduction have not been developed. For a crude quanti�cation of the possible loss in �sh value due to water pollution, a modeled relationship based on assumptions is used, represented in Figure 6. The Figure shows the estimated reduction in volume of �sh caught at lower levels of dissolved oxygen, for an average �sh species in Cambodia. The amount of oxygen needed for the survival of �sh varies with time of year and species. Oxygen needs vary even with the life stage of a species. Young species tend to be more sensitive to low oxygen conditions than adults. Also important is the duration of periods with low oxygen. Most species can survive short periods of reduced oxygen, but suffer during longer periods. According to Meck [74] and others, the minimum limiting oxygen concentrations for a �sh is dependent upon its species, physical state, level of activity, long term acclimation, and stress tolerance. A research study from the USA examined the lowest DO at which different �sh species survived for 24 hours, varying from 6.0 mg/L down to 3.3 mg/L [73]. Usually larger �sh are affected by low DO before smaller �sh. Given the lack of published studies on the empirical relationship between these two variables, the following assumptions were made based on a mixture of available scienti�c literature, internet sources, and expert opinion. A range is assessed in sensitivity analysis, shown by dotted line in Figure 6. • Water with an oxygen concentration of less than 3.0 mg/l will generally not support �sh. When concentrations fall to about 3.0-4.0 mg/L, �sh start gasping for air at the surface or huddle around water falls or higher concentration points. • Numerous scienti�c studies suggest that 4.0-5.0 parts per million (ppm) of DO is the minimum amount that will support a �sh population for short periods of 12-24 hours. • Above 5.0 mg/l, almost all aquatic organisms can survive inde�nitely, provided other environmental parameters are within allowable limits. When there are too many bacteria or other aquatic organisms in the area, they may overpopulate, using DO in great amounts [67]. • Levels of 6.0 mg/L and above support spawning, and above 7.0 mg/L support growth and activity [72, 73]. • The DO level in good �shing waters generally averages about 9.0 parts per million (ppm). In order to assess likely impacts of polluted water on �sh production, geographical locations of the principal �sh catches and water quality indicators are matched for major selected inland water bodies. Based on the observed DO levels in these water bodies, the function in Figure 6 is applied that estimates the loss of �sh catch due to lower than optimal levels of dissolved oxygen. The current �sh production levels are adjusted upwards to predict what the �sh catch would be in the presence of optimal DO levels, using algorithms provided in Annex B3. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 91 ANNEXES Figure 6. Modeled relationship between dissolved oxygen levels and �sh production (with lower and upper range) 10 9 8 7 Dissolved oxygen (mg/l) 6 5 4 3 2 1 0 0 10 20 30 40 50 60 70 80 90 100 110 Fish production (% optimal level) Source: Authors own estimates Notes: the upper line represents the maximum effect of reduced DO levels on �sh production volume, with linear reduction from 8mg/L to 4mg/L. In the base case, the linear reduction from 7.5mg/L to 3mg/L, while for the least effect, the linear reduction from 6mg/L to 2mg/L. Impact on �sh reproduction and growth due to water pollution is assessed by spatially comparing actual yields under current pollution levels and potential yields under a situation of good water quality, based on a water quality – �sh production function. Where �sh yields have dropped close to zero, historical records were sought for when �sh were previously caught in the given water body. The focus of the initial analysis is on �sh production that is officially recorded in national statistics. Where �sh catch values are available, these are recorded; where not available, �nancial value is estimated by applying current market prices to the average type of �sh. The economic impact of low DO levels for non-recorded �sh catch is not assessed in the present study. Likewise for the economic estimates, the �sh catch from subsistence �shing was not estimated; and the nutritional importance is not discussed despite the fact that �sh has been very important in providing food and nutrition to the people. Given the shortage of data, seasonal patterns are not assessed either although the water quality by season/month is strongly related to the �sh catch values by season. Moreover, since the �sh catch region is based on the geographic location, it is hard to match the �sh catch with the speci�c water bodies. To estimate attributable impact to poor sanitation, a proportion of this loss is assigned to sewage and domestic gray water, as compared to other sources of water pollution (industry, agriculture, silt/natural erosion). Normally, this can be done by estimating the proportion of BOD from these different sources. Due to the lack of reliable information on proportion of BOD from poor sanitation, it is assumed that poor sanitation contributes 65% of BOD to the water body in Cambodia. A3 Environment The release of waste into the environment has other effects besides water pollution, given the unpleasant smells emanating from feces, urine, and other waste products [81, 82]. In countries where open defecation and unofficial dumping of waste are common, the quality of land is affected, rendering it unattractive and unusable for productive use. Even in Asian countries where municipalities are responsible for collecting solid waste, solid waste collection is Economic Impacts of Sanitation in Cambodia 92 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES not commonly done, or it is inadequately done. Unregulated waste dumping presents a threat to those disposing of waste, those living in the vicinity of the dumping area, as well as the poorest of the poor who often live off the waste (e.g. recycling activities). Waste grounds are also inhabited by stray dogs or other animals, which are often diseased and pose a threat to human health. Even where there is a private or public agency taking care of disposal, it is often not performed according to plan. In cities, waste carts stay on the streets for many days, with resulting smell and unsightliness for local inhabitants and tourists. These aesthetic aspects of sanitation cause a loss of welfare for those coming into contact with the waste. However, given the lack of available data on these aspects, there is considerable uncertainty on the overall importance of these impacts. A3.1 Aesthetics Aesthetics are not strongly related to productivity or income. Economic evaluation studies do not usually quantify aesthetics such as smell and sight in economic terms. Instead these aspects are described as a potential additional bene�t provided by sanitation programs. Studies assessing welfare impacts of sanitation options, including willingness to pay studies, tend to limit the focus to the physical boundaries of the household and not assess impacts on the broader environment [83, 84]. Hence, �ndings on welfare aspects of aesthetics in this study are presented mainly in qualitative terms. In Cambodia, due to lack of published information, most information was collected based on interviews with experts who are working closely to the �eld of environment. In this study, ‘aesthetics’ refers to the impacts of waste on air quality (smell) and visual appearance caused by inappropriate waste disposal. Since the problem of solid waste is seen as a major environmental concern, the main focus is on the inadequate disposal of solid waste by households and waste disposal services. The practice of solid waste handling and management in Cambodia is briefly described, and its impact is qualitatively assessed based on experts’ opinion. The key informants in this regard are the Phnom Penh Waste Management (PPWM), Ministry of Environment (MoE), and Municipality of Phnom Penh (MPP). A3.2 Land quality When it has alternative uses, land is a tradable commodity. In this regard, land that is used for improper, unofficial disposal of waste will be unusable for other more productive uses, and hence will reflect an economic loss to society. This present study assesses the waste disposal practices and resulting effects on land availability and land quality descriptively. Due to the lack of previous studies on this particular issue, the information is obtained from interviews with key resource people. A4 Other welfare The type of sanitation facility a household has will have a range of impacts on population welfare. An important but difficult to quantify aspect is the welfare impact on individuals and families who use a sub-standard, uncomfortable latrine or who have no latrine at all. Except for the disease impact (see section A1), these less tangible aspects of human welfare have limited direct �nancial implications, and cannot easily be captured by market values. However, in the broader assessment of welfare impact on households and individuals, these aspects are potentially important [85]. Intangible aspects include user preferences for sanitation options. For women and girls especially, there can be physical dangers of using distant toilets or open spaces, especially at night; this practice may also cause damage to a person’s status. More tangible impacts of sub-standard latrine or no facilities is the time spent journeying to open defecation sites or public latrines, or waiting due to insufficient shared or public latrines per head of population [86]. Time savings can be used for other productive or leisure activities, and thus have an economic value. Also, life decisions such as schooling or choice of employment, and absences from school and the workplace, can be linked to the presence of sub-standard or no latrines [87]. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 93 ANNEXES A4.1 Intangible user preferences User preferences which could be described as ‘intangible’ – or difficult to quantify – include: • Comfort & acceptability– the acceptability of the squatting or seating position; the ease to perform personal hygiene functions; the freedom from rushing to complete toilet-going due to unhygienic latrine conditions, flies and foul smelling air. • Privacy and convenience – the bene�ts of not being seen using the toilet; or being seen walking to access toilet facilities (women) [88]. • Security – the location of the latrine within or near to the home means that excursions to the outdoor do not need to be made for toilet-going needs, in particular at night, where there may be dangers (theft, attack, rape, and other injuries sustained from dangerous animals or snakes). • Conflict – on-plot sanitation can avoid conflict with neighbours or the community, where tensions exist on the shared facilities, or �elds and rivers for open defecation. • Status and prestige – when visitors come to the house, it gives prestige to the household to be able to offer their guests a clean and convenient toilet to use. Families may hold more social events at their house as a result of a clean latrine. According to Table A26 below, there are nearly 10 million people who do not have a proper latrine; so they still practice open defecation. Of those people, up to 93% are living in rural areas, while 7% are in urban areas. Since open defecation is a practice which does not provide comfort and privacy for toilet going, it can be argued that the majority of Cambodian people still experience discomfort and lack of privacy. In addition, as many as 245,000 people use uncovered latrines which are open to flies causing health hazard and create bad smell affecting them in intangible ways. It is also observed that about 0.6 million people are using shared latrine facilities which are inconvenient to them in terms of time lost. Table A26. Lack of latrine – indicators of defecation conditions Unimproved latrine No latrine: open Open pit: no Shared facilities Area which open to flies defecation (‘000) covering (‘000) (loss time) (‘000) etc. (‘000) Number % Number % Number % Number % Rural 9,032.4 78.1 104.1 0.9 185.0 1.6 474.2 4.1 Urban 724.1 32.3 15.7 0.7 60.5 2.7 159.2 7.1 Total 9,756.5 71.4 119.8 0.9 245.6 1.8 633.3 4.5 Source: CDHS (2005), Population Projection 1998-2020 A4.2 Access time Welfare loss from increased access time due to unimproved sanitation can be due to journey time for open defecation or waiting time for shared or public latrine. The algorithms for the calculation are provided in Annex B4. The welfare loss due to journey time and waiting time has some economic values. As with health time productivity loss, the adults’ time value reflecting welfare loss is assumed as 30% of their income while children’s time value is worth 50% of those adults’ time value. For shared toilet, it is assumed that a shared latrine is normally shared by 2-3 households which are equivalent to roughly 12.5 people. In this case, it is assumed that the average waiting time for shared latrine is 3 minutes per person per day. According to local citizens, journey time to and from open defecation site is roughly 10 minutes. So, assuming conservatively that a person visits toilet only once a day (excluding urination), Table A27 gives the average access time for toilet-going. Economic Impacts of Sanitation in Cambodia 94 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES Table A27. Toilet access, by geographical location sub-type Population experiencing Average time access per day for sub-optimal access in terms Population with those with sub-optimal access of proximity & waiting time Location access time already (minutes) (‘000) minimized (‘000) Shared Open Shared Open latrines defecation latrines defecation Rural 1,816 474 9,032 3 10 Urban 1,258 159 724 3 10 Total 3,073 633 9,756 - - Source: Compiled using CDHS (2005); and Population Projection 1998-2020, NIS A4.3 Impact on life decisions and behavior Running water supply and sanitary latrines in schools are a luxury in most of the developing world, and in many workplaces latrines are unhygienic, poorly maintained, and do not cater for special needs of women. The presence of hygienic and private sanitation facilities in schools has been shown to affect enrolment and attendance, especially for girls [87, 89]. Good latrine access at the workplace has implications for women participation at traditionally male- dominated employment areas. Furthermore, sanitary and adequate latrines in schools and at workplaces not only affect participation rates, but they also improve welfare of all pupils and employees. Given the complex web of causative factors and eventual life decisions, and the many factors determining absenteeism from school or the workplace, it becomes difficult to quantify the exact relationship between poor sanitation conditions, education and work decisions, and eventual economic outcomes. However, it can be easily seen how sanitation is really important in education and work decisions, especially among girls and women. It has been recognized that cleanliness and privacy related to toilet have a strong impact on girls and women attitudes in participating in education29. Most students perceive that a good toilet at school makes them at ease when they need to use it during their study. If the toilet and the environment are not good enough, the students tend not to be willing to come to school and, in some cases may result in an absence from school. It has been argued that the frequent absenteeism has made the students unable to catch up the lesson leading to high dropout rate. Cambodian women and girls are shy and have difficulty in using the toilet that does not have water supply especially while they are in menstrual period. With a good and clean latrine at school, female students will have more privacy as they can avoid from being seen by others when using toilet. Besides the privacy issue, schools having poor sanitation may strongly affect students’ health and feeling. These effects will in turn have impacts on their study performance. Based on the observation of an Ministry of Education Youth and Sports (MoEYS) official, school with proper toilets is likely to perform better than those without toilets because both students and teachers feel more at ease about toilet-going. In the case where there are not adequate toilet facilities at schools, teachers and students usually need to wait uncomfortably for the study session to end to rush to the toilet at home. For the school equipped with toilets facilities, it can be observed that toilets for girls are usually separated from those for boys. A news article highlighted that the shortage of latrines at school in Cambodia is currently keeping girls out of school. It was quoted that when school girls get older, toilet is increasingly important30. It is recognized that lack of education has largely constrained employment opportunities for Cambodian women. Regarding workplace, it is generally stated that workplaces such as factories in Cambodia have toilets being in good 29 Interview with MoYES official who is working in education health 30 The Cambodia Daily, Volume 36 Issue 39, February 20, 2007 Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 95 ANNEXES condition and enough for workers to use31. Nevertheless, from perspective of some workers, the toilets in most factories are not clean enough which may in turn have impacts on their health and working condition. Table A28 below shows the number of schools by region having toilet facility. From the table, there are nearly 2,000 primary schools in Cambodia lacking of latrines, while nearly 2,500 do not have water supply. In addition, more than 200 and nearly 400 secondary schools are lacking of toilet and water supply respectively. However, it is important to note that not all toilets facilities at school are adequate. According to a MoEYS official, in Phnom Penh there are only about 30 schools having adequate latrines, while in the areas outside Phnom Penh, the number is very uncertain, but it may turn out to be very low. It is also observable that some schools have separate toilets for teacher and students, where in most cases latrines for teachers are in better condition than latrines for students. Table A28. Water and sanitation coverage in schools (2005-2006) Toilet access Water supply access Establishment With Without With Without Primary school Phnom Penh 102 11 71 42 Plains 1,661 640 1,577 724 Tonlé Sap 1,581 665 1,274 972 Coastal 334 116 248 202 Plateau/Mountain 676 491 614 553 Total 4,354 1,923 3,784 2,493 Secondary school Phnom Penh 31 - 26 5 Plains 301 88 237 152 Tonlé Sap 210 80 157 133 Coastal 61 25 41 45 Plateau/Mountain 91 35 70 56 Total 694 228 531 391 Source: Ministry of Education Youth and Sports Menstruation may be one of the obstacles causing girls not to come to school which do not have water supply and clean toilets. Besides, this is not the only reason keeping the girls out of school. It is recognized that the dropout of girls from school is known to be caused by several reasons: • Some girls keep out of school because they are busy to help family’s living as some go to work in the garment factory. • Some poor families decide not to send their daughters to school, as they cannot afford for their children education. • Some female students also drop out of schools because of the distance and safety to commute their schools. Based on Table A29 below, the number of boys enrolling in primary school is higher than girls of about 8.3%. While more than 1.3 million boys enrolling in primary school, there are only 1.2 million female students have done so for the school year 2005-2006. It is also noticed that the number of male enrollment is even much higher than female students in secondary school. Among 0.8 million secondary school students in 2005-2006, there are about 0.5 million boys enrolling in school comparing to 0.3 million girls. Although the difference is not signi�cant, this �gure highlights the low participation rate of girls in education, where, among others, one of the possible reasons 31 An interview with MoLVT official Economic Impacts of Sanitation in Cambodia 96 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES is the lack of sanitation as stated earlier. Regarding the number of teachers at school, among 97,831 teachers, about 39% of teachers are women comparing to 60% male teachers. In terms of employment, it should be noted from the table that the number of female employment is more than that of male of 7%. Here, it is not because women have more job opportunities than men, but because most women involve in unskilled employment with very limited future prospects. Table A29. Male / female participation rates in school and workplace (2005-2006) Variables Female Male Total Primary school Enrolment (million) 1.2 1.3 2.6 Completion rate (%) 89.8 90.4 90.1 Drop-out rate (%)1 11.9 11.3 11.6 Secondary school Enrolment (million) 0.3 0.5 0.8 Completion rate (%) 23.5 33.4 28.3 Drop-out rate (%) 22.7 19.7 21.0 Teacher (‘000) 36.2 61.6 97.8 Employment in 2001 (million) 3.23 3.02 6.24 Source: Ministry of Education Youth and Sports, and Statistical yearbook, 2006 1 Number of drop-out divided by number of enrolment Based on the above data sources, the expected impacts of poor sanitation in schools and workplaces are assessed by estimating school days and workdays lost due to poor sanitation: 1. School enrolment and completion: a proportion of non-enrolled pupils and absentee days are apportioned to lack of sanitation. However, the attributable fraction of non-enrollment and absenteeism to poor sanitation is not known in Cambodia. 2. From Table A29 above, the employment rate among women and men are very much similar. In Cambodia, the information on days lost from work for women in workplaces is very limited, and hence no link is possible with the adequacy or inadequacy of latrines in the workplace. Nonetheless, if the days lost incurs, there would be some economic cost in that as it impacts on the productivity of affected workers. A5 Tourism Tourists are sensitive to their environment, and are less likely to choose destinations which are dirty or where the risk of disease is high. In Cambodia, while tourism sector is one of the main contributors to the economy, it is important that tourism impact of poor sanitation is addressed in this study. It is recognized that a country may be losing tourist revenues due to the degraded environment and high infectious disease rates among the general population, as well as actual or perceived health risks to tourists. Hence any initiative to attract more tourists to a country will need to consider the part sanitation plays in this [90-92]. A5.1 Introduction to tourism Poor sanitation can have several impacts beyond those traditionally assessed such as health and environmental impacts. However, the linkages are sometimes difficult to make, due to lack of data as well as difficulties in isolating speci�c cause-effect relationships. One such impact that this study attempts to quantify is that of tourism. Unarguably, the quantity of tourists choosing a country for their holiday is partially related to the general sanitary conditions of Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 97 ANNEXES the country, as well as whether the country has suffered speci�c events such as a cholera epidemic. Better sanitary conditions can lead to attracting ‘high value’ tourists – tourist who are willing to pay more for their holiday. The attractiveness of a country for tourists is related to several aspects of sanitation – quality of water resources (either for enjoyment or for use), quality of the environment (smell, sightlines), food safety (hygiene in food preparation), general availability of toilets offering comfort and privacy in restaurants, bus stations, etc, and health risks. In Cambodia, there are a lot of tourist places such as Angkor Wat in Siem Reap that attract more than 440 thousands tourists in 2005 which accounts for 31% of total tourist arrivals to Cambodia. With the increasing in numbers of tourists, the authority builds and develops the facilities in this area to �t the tourist requirement. Hundreds of hotels, restaurants which provide convenient services have been established. Moreover, clean toilets nearby temple areas are set up for all tourists who need to use during their traveling. However, it is argued that a number of tourists have had diarrhea during their stay in Cambodia which may be caused by poor sanitation and hygiene in the areas around tourism spot32. The economic impact of tourism assessed in this study is the sub-optimal exploitation of tourist potential in the country. A5.2 Tourism and sanitation Tourism is a booming industry, and continues to experience double-digit growth in many developing countries around the world [93], fuelled by cheapening airfare costs coupled with the realization of developing country governments and private sector of the potential economic bene�ts of tourism. Tourism is playing an important role in boosting the revenues of governments as well as contributing to much-needed economic growth in the developing world. Countries of Southeast Asia have been among those enjoying year-on-year increases in tourist numbers and income. In 2006, about 1.7 million visitors coming from different regions visited Cambodia which is 20% higher than in 2005. In addition, the number of holiday tourists is very high compared to business visitors and visitors who come for other purposes. More than 97% of tourists come for holiday purposes as opposed to 0.32% for business. The national income from tourism sector accounts for more than 1 billion US$, which is 26% higher than in 2005, showing the increasing importance of tourism sector in Cambodian economy. In the Asian region, some countries have done better than others at exploiting the growth in tourism. Tourist preferences clearly play a key role in this: there are clearly many factors that determine tourists’ choice of destination, and sanitation will be one among many. Tourist growth depends on what the country can offer such as tourist transport infrastructure, quality of accommodation and restaurants, clean and adequate sanitation facilities, type of experience offered (culture, climate, culinary, relaxation), and safety. Also, the price of tourist services determines the relative attractiveness of a country for foreign tourists, which is partially determined by the stability and level of the local currency. Tourists are often heavily influenced in their choice of destination by the availability of information (positive media) on a destination, the offer of package tours or package deals in their home country, and/or the ease of booking flights and hotels on the internet or by phone. The availability of package deals and the ease of bookings are themselves a function of the level of development of a tourist destination. Hence there is a self-reinforcing loop, which can – over time – lead to large resort complexes and tourist destinations such as the various coasts and islands of Southern Europe and the Caribbean, coastal areas of Thailand and Malaysia, and well known tropical islands (Zanzibar, Maldives), among others. What role does sanitation play in a country’s attractiveness for tourists? The environment is one of the key attracting elements of a tourist destination – as a popular refrain goes: “sun, sea and sand� – which are recognised factors for attracting tourists. But if the sea is brown from the pollution released by the country’s rivers, if the sand or 32 An interview with Association of Travel Agent Economic Impacts of Sanitation in Cambodia 98 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES roadsides are soiled with the excreta of the local inhabitants, and if food preparation standards are low, then the tourism potential of a location is clearly limited [94]. Tourist perceptions about the sanitary conditions of a potential tourist destination are gathered from their own research and experience, as well as from the stories and perceptions circulating via travel agencies and social networks. Important aspects for sanitary conditions to tourists include, but are not limited to: • Aesthetics of the local environment (sight, smell) • Cleanness of water for swimming or sightseeing • Availability of clean toilets and water, soaps and towels for personal hygiene, in accommodation, restaurants, bus stations etc. • Expectations of getting sick either from food poisoning or environmental factors It is also noted that the minister of tourism has recently recognized the importance of hygienic restrooms in attracting tourists to Cambodia, and has expressed the intention to improve toilets in the tourism sites and in the whole country. A major step that Ministry of Tourism (MoT) will do is that restaurants and hotels in Cambodia will be rated and classi�ed according to their toilet facilities. It is believed that this action will provide incentive for those establishments to improve their restrooms adequately33. A5.3 Estimation of tourist losses due to poor sanitation While it is accepted that the standards of tourist facilities in Cambodia are improving over time, this present study assumes, based on evidence, that the sanitary standards remain sub-optimal. Hence, it is hypothesized that Cambodia could attract more tourists now and in the future; one of the aspects that must improve for that to happen is hygiene and sanitation. Given the limited options for countries to boost tourist numbers and hotel occupancy rates from improved sanitation in the near-term, the tourist losses are not estimates as a �nancial cost. However, in the longer term it is assumed that Cambodia can not only increase the hotel occupancy rates under the existing capacity constraints of tourist infrastructure (airport, hotels, internal transport, restaurants), but also to expand the tourist infrastructure as well as making tourist destinations more attractive for tourists to accommodate signi�cantly increased foreign tourist arrivals. Infrastructure requirements were based on mid- to long-term government targets for tourist growth and total numbers; where these were not available, realistic assumptions are made of the tourist growth achievable over a �ve year period until 2010. A target occupancy rate of 80% is assumed for the present study although it is largely season-speci�c. The formula is shown in algorithm in Annex B5, and the values used presented in Table A30. According to Ministry of Tourism (MoT), the occupancy rate of tourism in Cambodia in 2006 is nearly 55%. Moreover, given the importance of tourism sector in the country and possible sensitivity of sanitation on tourism growth, it is assumed that the attributable fraction of tourism loss to poor sanitation is 10%. This attribution can be justi�ed by the increasing interests of MoT in improving sanitation to attract tourists to the country. 33 The Cambodia Daily, Volume 37 Issue 67, August 17, 2007 Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 99 ANNEXES Table A30. Inputs for calculating the �nancial losses in tourist receipts 1 Parameter Symbol Cambodia Actual occupancy (%) OCA 55 Potential occupancy (%) OCO 80 Contribution of sanitation to tourist losses (%) φ 10 Actual number of tourists (millions) ta 1.7 Average expenditure per tourist (US$/day) E 95 Average length of stay (days) - 6.5 Source: Ministry of Tourism, 2006 1 Table reflects values used for calculation of losses due to lower than optimal foreign tourists. Economic losses are reflected by the gap between the current tourist revenues, and the tourist revenues that would be possible at signi�cantly higher tourist visit numbers such as those experienced by neighboring countries with higher visit rates. At present, the potential number of tourists at an assumed optimal occupancy rate of 80% is 2.5 million, which is 46% higher than the tourist number in current occupancy rate. In a potential future scenario, the prospect of an annual tourist increase of 20% is adopted to estimate the potential number of tourists, based the tourism vision of the Ministry of Tourism [95]. Hence, based on 20% increase of tourist capacity at 80% occupancy rate, the potential number of tourists to Cambodia is about 3 million per year. Then the economic costs can be estimated using algorithm in Annex B5. It is worth to note that the economic losses are estimated at the average annual loss over the mid-term period (5 years) based on the tourist capacity and the target occupancy rates. A5.4 Tourist sickness Once the tourist is on-site, they may experience a reduction in enjoyment of their holiday experience by becoming sick due to a disease related to poor sanitation and hygiene. While having an illness episode is not only a bad experience in itself, it also eats into holiday time, and may incur some expenses related to treatment. In the worst case, the return journey of the tourist is affected or they need to get emergency transport. However, the number of sick tourists and the disease episode that may be necessary in the calculation of economic loss in Cambodia is not well-documented. Nonetheless, it can be argued that, for the economic cost, the welfare loss of the tourist from having days off sick is added to the �nancial cost of sickness treatment. However, due to lack of data, the estimation of economic cost is not included in the present study. It is, however, noted that although these losses may incur among tourists, they may need to be excluded from national estimates of economic impact, as the majority of them are experienced by foreigners. Nevertheless, the tourist sickness may have negative implications in the long-term development of tourism sector in Cambodia. The frequent occurrence of sickness among tourists may, to some extent, slow down the growth of tourist arrivals in the future, which badly impacts on the country’s economic growth. A6 Impact mitigation associated with improved sanitation and hygiene Having estimated the costs of poor sanitation, from a policy viewpoint it is important to know by how much these �nancial and economic costs can be reduced by implementing improved sanitation options. While there are many types and con�gurations of sanitation improvement available, this present study aims to estimate potential bene�ts obtainable for a small number of generic categories of sanitation improvement. This aids interpretability by policy makers. Indeed, it is the aim of the second study of ESI to estimate the costs and bene�ts of speci�c sanitation options, which are the most relevant policy options in each country context. Economic Impacts of Sanitation in Cambodia 100 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES A6.1 Health impacts mitigated The �nancial and economic gains from improved sanitation and hygiene will be a proportion of the total losses estimated for diseases associated with poor sanitation and hygiene. The proportion of costs avertable will depend on the expected effectiveness of the interventions employed to prevent disease. No health intervention, as implemented in practice, will be 100% effective in reducing the overall loss. However, sanitation and hygiene interventions have been proven to be effective in a number of �eld trials [96, 97]. Given that good quality epidemiological studies are limited in number, and have already been reviewed in previous meta-analyses, no additional country-level studies were used to estimate disease cases prevented. Hence the estimates of intervention effectiveness are based on the international literature, which includes the most up-to-date reviews on effectiveness [96-99]. The latest and most authoritative review by Fewtrell et al (2005) presented summaries of effectiveness from a meta- analysis of �eld trials on water, sanitation and hygiene separately, as well as together [97]. The reader is referred to the paper for details of individual studies. Table A31 below shows the summary of the meta-analysis. Table A31. Summary of meta-analysis results on WSH intervention efficacy for diarrheal disease reduction Number of Estimate of effect (relative risk)2 Intervention studies included1 Low Mid High Household treatment of water 8 0.46 0.61 0.81 Water supply 6 0.62 0.75 0.91 Sanitation 2 0.53 0.68 0.87 Hygiene 8 0.40 0.55 0.75 Multiple interventions 5 0.59 0.67 0.76 Source: Fewtrell e al (2005) [97] 1 Includes only studies of good quality, as de�ned by Fewtrell et al 2 Relative risk of disease when intervention tested against baseline of no intervention (relative risk of 1.0) These relative risk reductions are used to estimate expected rates of diarrhea under a situation of basic improved sanitation and hygiene practices, and carried through to estimation of health care cost, productivity and income, and premature deaths. Hence, based on the literature, the following reductions of disease incidence are predicted: • Sanitation: incidence reduced = 32% (range 13% to 47%) • Hygiene: incidence reduced = 45% (range 25% to 60%) Note however that hygiene and sanitation interventions implemented together will not have the sum of the individual effects. The literature does not provide evidence for the proposition that two interventions are more effective than one. This point needs to be taken into account in interpreting the estimations of economic loss avoided from health interventions. A6.2 Other economic losses due to poor sanitation Given that the attributed costs of poor household sanitation are the object of the study (described in sections A1 to A5 above), the effect of improving sanitation will – in theory – be to mitigate the full estimated losses. This assumes that the interventions are fully effective in isolating human excreta (at least in its harmful form) from the environment, and it assumes a linear reduction in economic impact at different levels of attribution. In other words, it is assumed that by removing x% of the pollution source, a similar x% of economic losses would be averted. However, such linearity cannot necessarily be assumed. Also, for some environmental effects where the environment has been degraded considerably over time, there will also need to be expenditure on a clean-up operation to bring the land and water resources back to usable or fully productive condition. These costs are not estimated in this present study. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 101 ANNEXES A6.3 Market for sanitation inputs Economic impacts and with effects up the supply chain (multiplier effect), will be for small local entrepreneurs as well as larger, non-local companies. There is also a potential for improving livelihoods of poor people through sanitation programs, largely through health improvement and employment generation [88]. Given the needs of sanitation programs for human labor and materials, sanitation programs will have a number of economic effects, whether it be for small local entrepreneurs or larger companies. Table A32 presents the unit costs of different sanitation options. These unit costs reflect the cost per household associated with each improvement option. Financial costs are made up of purchased services (labor, materials, equipment), while economic costs also include non-purchased inputs to the sanitation option (such as household and community-provided labor). Sums also include the expected sale of hygiene products (soap, cleaning materials) for the hygiene interventions. These are multiplied by the expected coverage with different sanitation options to estimate total potential market values. However, since there is scarce information for calculation of economic costs, the present study attempts only to estimate the �nancial costs. While on the one hand the costs reflect government and household spending, it represents an economic gain for those involved in providing the services and will have broader economic effects. Table A32 also gives the number of households to receive different sanitation options. Allocation of each sanitation option is based on CMDG target for latrine improvement, and on National Biodigester Program (NBP)34 target for EcoSan. The number of households to be receiving each improvement is based on the CMDG target of the sanitation coverage in 2015 which is 72% for urban areas and 30% for rural areas. It is estimated that to reach this CMDG target, more than 404 thousands households need to be equipped with an improved latrine, which is equivalent to 40 thousands households each year. However, since there are various sanitation improvement options available, the assumed number of households adopting each option needs to be allocated. In this case, as pit latrine is the least cost choice for most households in Cambodia, particularly in rural areas, it can be assumed that 40% of households will adopt simple pit latrine while the rest of other options are equally shared (ventilated pit latrine, septic tank, and sewer connection). For biodigester plants, however, the estimation of the number of plants is based on the NBP target to have 17,500 plants constructed between 2006 and 2009. Currently, since there are 600 plants35 that have been constructed, the total remaining input number of biodigester plants is 16,900. So, the remaining number of households adopting this EcoSan needs to be allocated annually over four-year period, which gives 4,225 per year. Table A32. Unit prices of sanitation improvement options (per household) Septic tank Simple pit Simple Piped sewer Variable VIP with sludge latrine EcoSan connection removal Households to receive each 16,171 8,085 8,085 4,225 8,085 improvement (annually) Unit prices (US$) Super structure 14 14 19 - 19 Slab 3.5 3.5 10 - 10 Underground 3 13.5 16 - 48 Total unit price (US$) 20.5 31 45 297 77 Source: Informed choice manual on rural household latrine selection of MRD [100], National Bio-digester Program Cambodia [101]. 34 NBP is a national program of MAFF involving the reuse of animal dung and human waste for biogas and fertilizers. 35 Information Folder, National Biodigester Program Cambodia, 2006 Economic Impacts of Sanitation in Cambodia 102 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES A6.4 Market for sanitation outputs Where human excreta is used as fertilizer, the availability of nutrients from human excreta can lead to the replacement of chemical fertilizer, which saves costs [102]. Furthermore, where fertilizer was not being used optimally before, the nutritional content and economic value of crops may increase. Also, there are long-term bene�ts of reducing the use of chemical and mineral fertilizers, especially taking into account the fact that some fossil resources are in increasingly short supply (e.g. phosphorous). Alternatively, families with livestock may instead invest in a biogas reactor, which provides biofuel for cooking and can even be used for lighting where other improved sources (electricity) are not available [103]. The reuse of human waste for fertilizer or biogas production cannot be assumed to be population-wide, given cultural attitudes towards handling and re-use of human waste, and low practical feasibility in many locations. Success often depends on local perceptions of the expected returns on re-use of human waste, whether it be for biogas or fertilizer. On the contrary, the re-use of animal waste tends to be adopted more widely among the rural population. In Cambodia, the NBP which is commissioned by the Ministry of Agriculture Forestry and Fisheries (MAFF), has largely involved in the re-use of both animal dung and human waste for biogas and fertilizer in rural households. However, the program focuses more on animal rather than human waste due to the large volume of waste required. This program mainly relates to the own-use of biogas and fertilizer. According to NBP, there are potentially 25% of rural households which are technically feasible for construction of biodigester plant considering geographically atmospheric condition. As of May 2007, only about 600 households have the biodigester plants, the main use being for biogas, while 72% also use them for fertilizer. According to the recent survey by NBP, the cash saving by using biogas for cooking and lighting is about US$11 per month, equivalent to US$132 per year [104]. The average annual sanitation output is estimated based on the annual allocation of the NBP target (17,500 households building biodigester plants) over 4 years from 2006 to 2009. In this sense, the number of households assumed to adopt EcoSan annually is estimated as 17,500 divided by 4 which is equal to 4,375. It is noted that Table A33 gives the number of households adopting EcoSan (biodigester plants) and the unit value gains from the output of biogas. For fertilizer, however, the value has not been estimated by NBP. Table A33. Input values for estimation of returns to re-use of human waste Fertilizer Biogas Application to own Own use Variable Sales of land Sales of gas fertilizer Annual and sludge Annual Current Current average average Number of rural households adopting - 432 3,150 - 600 4,375 EcoSan (biodigester) Unit value per household - N/A - 132 (US$/year) Source: National Biodigester Program Cambodia [101] A7 Uncertainty analysis Tables A34, A35 and A36 provide alternative input values to reflect three main types of data uncertainty in the present study: • Uncertainty in the estimation of overall impacts, such as in the epidemiological and economic variables (Table A34). • Uncertainty in the attribution of the overall impact to poor sanitation (Table A35). • Uncertainty in the actual size of impact mitigation achievable (Table A36). Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 103 ANNEXES Table A34 presents the main uncertain economic variables, and the alternative – low and high values – used in the one-way sensitivity analysis36. Table A34. Alternative assumptions and values used in one-way sensitivity analysis Variables selected Low estimate of impact Base case estimate High estimate of impact Health Diarrheal incidence DHS data for under �ves; DHS data for under WHO regional data for WHO regional data for over �ves; WHO regional data all age groups; 88% of �ves; 70% of diarrheal cases for over �ves; 88% of diarrheal cases attributed attributed to poor S&H diarrheal cases attributed to poor S&H to poor S&H Hourly value of non- 30% of GDP per capita 30% of compensation of Minimum wage income earning time employees – economic only Hourly value of Children given value of zero Children given 50% value Children given same value productive time for as adults as adults children Premature death Human capital approach, Human capital approach, VOSL bene�t transfer of using 2% growth and GDP using 2% growth and US$2 million, using 0.6 per capita compensation of income elasticity employees Water Fish production and Lower range used (�sh less Mid range used Higher range used (�sh DO relationship1 affected by low DO) more affected by low DO) Other welfare Time access 5 10 15 (minutes/day) Value of time See under ‘health’ above Tourism Future growth in 10% increase per year; to 20% increase per year; to 30% increase per year; to tourist numbers 60% occupancy 80% occupancy 80% occupancy 1 The �gure on �sh production is based on the graph showing relationship between DO and �sh production. For water pollution, it is assumed in the base case that 65% of pollution is originated from poor sanitation. Nevertheless, in the situation where other sources of pollution such as industry and agriculture become prominent, the low estimate case assumes poor sanitation contributes only 50% to water pollution. On the contrary, in the case where the release or leakage of human waste to water bodies remains increasingly signi�cant it is assumed that 70% of water pollution is from poor sanitation. For tourism sector, 10% of tourist number impact attributed to poor sanitation is assumed for the base case, 5% for low estimate case, and 15% for high estimate case. Table A35 shows the possible linkage between poor sanitation and impacts using low estimate, base case estimate and high estimate. It is seen from the table that, in the base case, 88% of the diarrheal diseases are attributed to poor sanitation and hygiene, while 70% is used in the low estimate case and 95% for the high estimate case. While the low estimate refers to the fact that poor sanitation and hygiene do not have great impacts on diarrheal diseases, the high estimate assumes that nearly all diarrheal diseases are attributed to poor sanitation and hygiene, which is a rather extreme case. 36 In a one-way analysis, only one parameter at a time is varied. Economic Impacts of Sanitation in Cambodia 104 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES Table A35. Alternative assumptions for links between poor sanitation and impacts Variables selected Low estimate of Base case estimate of High estimate of impact impact impact Health Disease incidence attributed Lower value for disease Mid value for disease Higher value for disease to poor sanitation and hygiene incidence: 70% incidence: 88% incidence: 95% (diarrhea cases) Water Water pollution attributed to 50% 65% 70% poor sanitation Tourism Tourist numbers impact 5% 10% 15% attributed to poor sanitation Table A36 below shows the achievable impacts of mitigation measure at different estimates: low case, base case, and optimistic case. The low case refers to the fact that although poor sanitation is mitigated, the achievable impact is not as high as the base case. For health impact, it is noted that while in the base case the improvement of sanitation and hygiene can reduce disease rate of 32-45%, the conservative case estimates that the mitigation can only reduce about 13-25%, while the optimistic case expects 47-60% reduction of diseases. For water pollution and tourism, while the impacts by poor sanitation can be 100% mitigated in the base case, in the conservative case, it is assumed that only 70% of impacts can be mitigated through the improvement of sanitation. Table A36. Alternative assumptions for impact mitigation Low (conservative) High (optimistic) Variables selected Base case estimate estimate estimate Health Sanitation-related diseases mitigated 13% 32% 47% Hygiene-related diseases mitigated 25% 45% 60% Water Sanitation-related drinking water 70% 100% Not tested1 pollution costs mitigated Sanitation-related �sh production 70% 100% Not tested1 costs mitigated Tourism Sanitation-related tourist losses 70% 100% Not tested1 mitigated Sanitation markets Sanitation output coverage 1,000 4,375 5,000 (households using biodigester) 1 Value cannot be > 100% Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 105 ANNEXES Annex B. Algorithms B1. Aggregating equations Total costs of sanitation and hygiene C = CH + CW + CU + CT (1) Health related costs of poor sanitation and hygiene CH = CH_HC + CH_P + CH_D (1) Water related costs of poor sanitation and hygiene CW = CW_Drink + CW_Domestic + CW_Fish (2) Other welfare losses of poor sanitation and hygiene CU = CU_T (3) Tourism losses from poor sanitation CT = CT_RL (4) B2. Health costs related to poor sanitation and hygiene Total health care costs (5) CH _ HC = ∑ CH _ HCi i Health care cost per disease (6) CH _ HCi = α i ⋅ pop ⋅ βi ⋅ ∑ χ ih ⋅ vih ⋅ phealthih h Total productivity costs (7) CH _ P = ∑ CH _ Pi i Productivity cost of disease type i (8) CH _ Pi = α i ⋅ pop ⋅ βi ⋅ dhi ⋅ ptime Total cost of premature death (9) CH _ D = ∑ CH _ Di i Cost of premature death per disease (10) CH _ Di = ∑ deathia ⋅ γ ia ⋅ pdeatha a B3. Water related costs associated with poor sanitation and hygiene Total cost associated with accessing clean drinking water (11) CW _ Drink = ∑ CW _ Drinkm m Economic Impacts of Sanitation in Cambodia 106 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES Cost of accessing clean drinking water per source/treatment method (12) CW _ Drinkm = hm ⋅ wdrinkm ⋅ pwaterm ⋅ δ ⋅ π m Total domestic water access cost (excl. drinking water) (13) CW _ Domestic = ∑ CW _ Domesticm m Domestic water access cost by source/method (14) CW _ Domesticm = hm ⋅ wdomm ⋅ pwaterm ⋅ δ ⋅ θ m Fisheries loss (15) CW _ Fish = AFP − PFP Potential �sh production level (16) AFP PFP = ε B4. Other welfare impacts Time access cost for unimproved latrine (18) CU _ T = pop _ u ⋅ taccess ⋅ ptime ⋅ 365 B5. Tourism losses Lost revenues (19) ⎛ oc ⎞ CT _ RL = ϕ ⋅ ⎜ O − 1⎟ ⋅ ta ⋅ et � oc A ⎠ B6. Variable de�nition summary Tables B1 to B3 present the subscripts, variables and parameters used in the algorithms in Sections B1 to B5 above. Table B1. Subscripts used in algorithms Code Description Elements a Age group 0-4 years, 5-14 years, over 15 years i Disease types Diarrhea, Cholera, Typhoid, Malnutrition related diseases, etc h Health care provider Public hospital, private hospital, informal care, self-treatment m Treatment method Piped water, non-piped water, home-treated, hauled water Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 107 ANNEXES Table B2. Variables used in algorithms Symbol Description AFP Actual �sh production value C Total cost of poor sanitation and hygiene CHC Health costs of poor sanitation and hygiene CH_HC Health care costs of all diseases CH_HCi Health care cost of disease type i CH_P Productivity costs of diseases CH_Pi Productivity cost of disease type i CH_D Premature death costs of diseases CT Tourism losses associated with poor sanitation and hygiene CT_RL Revenue losses CU Other welfare losses associated with poor sanitation and hygiene CU_T Time access cost for unimproved latrine CW Water related costs of poor sanitation and hygiene CW_Drink Clean water drinking access costs CW_Drinkm Clean water drinking access cost for method m CW_Domestic Domestic water access costs CW_Domesticm Domestic water access cost for method m CW_Fish Fisheries production loss deathia Number of premature deaths, by disease type i and age group a dhi Number of days taken off work or daily activities due to disease i egirls Number of adolescent girls enrolled in school et Expenditure per tourist (US$) hm Number of households using water source or treatment method oca Actual occupancy rate (%) oco Optimal occupancy rate (%) pahc Average health care cost per case pdeatha Value of premature death for age group a PFP Potential �sh production value phealthih Unit price of care (per visit or day) for disease type i at health facility h ptime Daily value of time pwaterm Water price or time value per m3 of water pop Population pop_u Population with unimproved access to sanitation ta Actual number of tourists taccess Average access time (journey or waiting) per day vih Visits to or days for disease type i at health facility h wdrinkm Drinking water consumption/HH (m3) from water source/treatment method m wdomm Consumption/HH of domestic purpose (m3) from water source/treatment method m Economic Impacts of Sanitation in Cambodia 108 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES Table B3. Parameters used in algorithms Symbol Description Incidence rate per person of disease type i αi Proportion of episodes attributed to poor sanitation for disease type i βi Proportion of cases seeking care for disease type i and provider h χ ih Proportion of deaths attributable to poor sanitation, by disease type i and age group a γ ia δ Attributable water pollution to poor sanitation ε Ratio of the �sh production at the current DO level to �sh production at the optimal DO level µ Proportion of diseases related to sanitation Importance of averting drinking polluted water in relation to overall bene�ts of piped water πm supply; where π m = 1 for m ≠ piped-water Importance of averting using polluted water in domestic activities in relation to overall bene�ts θm of piped water supply; where θ m = 1 for m ≠ piped-water φ Attributable tourism loss to poor sanitation Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 109 ANNEXES Annex C. Recommendations for research Given the exploratory and wide-ranging nature of this study, a lot of searches have been made for data and information, stakeholders have been interviewed, and data sources have been evaluated to enable application of the methodology in order to calculate sanitation impacts. For some of the impacts estimated in this study, the �gures reflect �rst estimates which are tentative in nature. Some estimates are not supported by sufficient scienti�c evidence to be taken at face value. Hence, there are a number of proposals for further research, both scienti�c and operational in nature. 1. More routine measurement of variables is needed at in Cambodia for monitoring sanitation impact, as well as more in-depth analysis, compilation and comparison between existing data sets collected by different agencies. In some cases, sampling of certain areas and small-scale studies can be used rather than nationwide surveys. Areas of particular importance include, but are not limited to: disease rates and mortality; water quality; �sh stocks; sewage discharge; and gender. 2. The studies on attribution to poor sanitation of the impacts estimated above are needed to gain insights of the linkage of poor sanitation and each impact. As noted in the study, only the attribution of diarrhea to poor sanitation and hygiene is drawn on scienti�c study at the regional and global level, while others are based on assumptions and experts’ opinions. In this case, the knowledge of the linkage will lead to better accuracy in the estimation of impacts. Some important attributable fractions to poor sanitation that deserve attention are: water pollution, tourist reduction, sanitation and hygiene-related disease incidence. 3. Scienti�c research is required to understand sanitation impacts, linkages and relationships better, such as: determinants of disease, and impact of water pollution on different �sh species. 4. Existing household (health, budget), employment, school and tourist surveys could be expanded to collect more sanitation-speci�c data, such as: sanitation access time; expenditure on sanitation and hygiene products; tourist rate of disease, tourist perceptions of environment and sanitation; and contribution of �sh catch to local income generation, and role in the household economy. 5. The present study is limited by lack of studies examining economic values associated with sanitation services, and by weaknesses in existing estimates of economic value. Further economic research is required to inform Cambodian policy makers about the (economic) value of sanitation-related intangible impacts to different population groups, and the value of life (e.g. willingness to pay to avoid premature death). 6. The present study was a national study, aggregated from estimates made at the �rst level of sub-national division. In order to convince local decision makers such as city mayors or district officers, local level studies especially in the areas where lack of sanitation is predominant are needed to evaluate speci�c impacts in those settings. Given the variations in sanitation impacts depending on very variable disease rates, and differences in importance of other impacts, such local studies would be more credible in convincing local decision makers to invest in sanitation. 7. This study has shown the negative economic impacts of poor sanitation and hygiene in Cambodia, and estimated which costs could be averted if sanitation and hygiene were to be improved. However, these data do not quantify the actual efficiency of the different measures to improve sanitation and hygiene. Given that advocates for the sanitation sector are competing with advocates for other sectors, arguments for higher investments in sanitation must be supported by evidence that returns on sanitation investments are higher than investments in other sectors. In this sense, as the complement to the global cost-bene�t analysis of sanitation [86], the country level study shall be conducted to inform national decision makers how to effectively and efficiently invest in sanitation and hygiene. Economic Impacts of Sanitation in Cambodia 110 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) Annex D. Sanitation coverage by region and rural/urban grouping Improved sanitation (%) Unimproved sanitation (%) Region Type House Septic Pit Public Pit Total Open Other Total Connection tank latrine1 toilet latrine2 Rural 28.5 43.3 1.5 73.3 - 0.0 22.7 3.9 26.7 Phnom Penh Urban 90.0 3.9 0.6 94.4 - 0.0 2.8 2.8 5.6 Total 60.6 22.8 1.0 84.3 - 0.0 12.3 3.3 15.7 Rural 0.1 16.6 2.4 19.2 - 0.9 78.5 1.7 81.0 Plains Urban 3.8 51.9 0.8 56.6 - 0.0 41.2 2.2 43.4 Total 0.5 19.9 2.3 22.7 - 0.8 75.0 1.5 77.3 Rural 0.3 14.6 1.9 16.9 - 0.6 81.1 1.4 83.1 Tonlé Sap Urban 5.5 43.1 3.1 51.7 - 0.2 43.0 4.9 48.1 Total 1.4 20.4 2.2 23.9 - 0.6 73.4 2.1 76.1 Rural 0.1 10.6 4.1 14.8 - 2.6 75.4 6.8 84.7 Coastal Urban 7.7 33.3 0.8 41.8 - 1.2 49.7 7.3 58.2 Total 2.8 18.6 2.9 24.3 - 2.1 66.3 7.2 75.7 Rural 0.1 7.9 2.0 10.0 - 1.0 88.5 0.6 90.0 Plateau/Mountains Urban 1.6 28.0 3.7 33.3 - 3.4 62.1 1.0 66.6 Total 0.5 13.1 2.5 16.0 - 1.6 81.7 0.7 84.0 Rural 1.1 12.6 2.0 15.7 - 0.9 78.1 5.2 84.2 Total Urban 28.9 25.8 1.4 56.1 - 0.7 32.3 10.8 43.8 Total 5.2 14.5 1.9 21.6 - 0.9 71.4 5.9 78.2 Source: CDHS (2005) 1 Improved pit latrine = flush pit latrine, VIP, simple pit latrine with slab 2 Unimproved pit latrine = traditional pit latrine without slab A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) Economic Impacts of Sanitation in Cambodia ANNEXES 111 ANNEXES Annex E. List of Contributing Stakeholders No Name Institution Scope of work 1 Mr. Mak Sœun FAO Water resource impact 2 Mr. Sin Somuny MediCam Health impact 3 Mr. Nang Phirun MoE Water resource impact 4 Mr. Phin Rady MoE Water resource impact 5 Mr. Saron Sambo MoE Environment impact 6 Mr. Pen Saroeun MoEYS Welfare impacts 7 Mrs. Kuy Phalla MoEYS Sanitation at school 8 Ms. Khoun Engmuny MoH Health impact 9 Mr. Sok Touch MoH Health impact 10 Mr. Veng Thay MoH Health & environmental health 11 Dr. San Sary MoH Hospital 12 Mr. Pok Vanthat MoLV Welfare impacts (workplace) 13 Ms. Hou Neamita MoWA Gender 14 Mr. Mao Hak MoWRM (MRC) Water quality and river flow 15 Mr. Chiep Sivorn MPP Waste management 16 Dr. Chea Samnang MRD Rural sanitation 17 Mr. Chreay Pom MRD Rural sanitation 18 Mr. Mao Bunsoth NIPH Health impact 19 Dr. Ou Kevanna NMCH Malnutrition 20 Ms. Chy Kimhuy NPH Malnutrition 21 Mr. Oun Syvibola Plan International Rural water supply and sanitation 22 Mr. Sao Kun Chhon PPWM Waste management 23 Mr. Ros Kim Leang PPWSA Water supply 24 Mr. Khuth Vuthearith PPWSA Water supply 25 Mr. Keo Heng PPWSA Water supply 26 Mr. Jan Lam NBP Small-scale biogas 27 Ms. Lam Saoleng NBP Small-scale biogas 28 Ms. Sieng Leakna UNDP Gender specialist 29 Ms. Hilda Winarta UNICEF Sanitation Economic Impacts of Sanitation in Cambodia 112 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES Annex F. References 1. WHO/UNICEF/JMP, Meeting the MDG Drinking Water and Sanitation target: the urban and rural challenge of the decade. 2006, Joint Monitoring Programme. World Health Organization, Geneva; and the United Nations Children’s Fund, New York. 2. MoP, National Strategic Development Plan 2006-2010. 2005. 3. Rosenboom, J.W., J. Ockelford, and A. Robinson, Water supply and sanitation in Cambodia: Poor access for poor people. 2006, World Bank Water and Sanitation Program. 4. Joint Monitoring Programme, Global Water Supply and Sanitation Assessment Report. 2000, World Health Organization, United Nations Children’s Fund, Water Supply and Sanitation Collaborative Council: Geneva and New York. 5. Hutton, G., Evaluation of the Economic Impacts of Sanitation: Methodology and Guideline. 2007, Unpublished document. World Bank, Water and Sanitation Program. 103 pages. 6. Hutton, G. and L. Haller, Manual for Conducting Economic Evaluation of Water and Sanitation Interventions in China. 2006, World Health Organization. 7. Roberts, M. and A. Long, Demand Assessment for Sanitary Latrines in Rural and Urban Areas of Cambodia. 2007, The World Bank Water and Sanitation Program: Phnom Penh. p. 28. 8. Neou, S., et al., Cambodia Economic Watch. Vol. 6. April 2007, Phnom Penh: Economic Institute of Cambodia. 9. Prüss, A., et al., Estimating the global burden of disease from water, sanitation, and hygiene at the global level. Environmental Health Perspectives, 2002. 110(5): p. 537-542. 10. Prüss-Üstün, A., et al., Unsafe water, sanitation and hygiene, in Comparative Quanti�cation of Health Risks: Global and Regional Burden of Disease due to Selected Major Risk Factors, M. Ezzati, et al., Editors. 2004, World Health Organization: Geneva. 11. Wagner, E. and J. Lanoix, Excreta disposal for rural areas and small communities 1958, World Health Organization. Monograph series No.39: Geneva. 12. Dorfman, M., N. Stoner, and M. Merkel, Swimming in sewage: the growing problem of sewage pollution and how the Bush administration is putting our health and environment at risk. 2004, National Resources Defense Council; Environmental Integrity Project. 13. Strickland, G.T., ed. Hunter’s Tropical Medicine and Emerging Infectious Diseases. Eighth Edition ed. 2000, W.B. Saunders Company. 1192. 14. Bairagi, R., et al., The association between malnutrition and diarrhoea in rural Bangladesh. International Journal of Epidemiology, 1987. 16(3): p. 477-81. 15. Becker, S., R. Black, and K. Brown, Relative effects of diarrhea, fever, and dietary energy intake on weight gain in rural Bangladeshi children. American Journal of Clinical Nutrition, 1991. 53: p. 1499-1503. 16. Black, R., K. Brown, and S. Becker, Effects of diarrhea associated with speci�c enteropathogens on the growth of children in rural Bangladesh. Pediatrics, 1984. 73: p. 799-805. 17. Rowland, M., T. Cole, and R. Whitehead, A quantitative study into the role of infection in determining nutritional status in Gambian village children. British Journal of Nutrition, 1977. 37: p. 441-50. 18. Rowland, M., S. Rowland, and T. Cole, Impact of infection on the growth of children from 0 to 2 years in an urban West African community. American Journal of Clinical Nutrition, 1988. 47: p. 134-8. 19. Martorell, R., et al., Acute morbidity and physical growth in rural Guatemalan children. Pediatrics, 1975. 129(1296-1301). 20. Molbak, K., et al., Cryptosporidium infection in infancy as a cause of malnutrition: a community study from Guinea-Bissau, West Africa. Am J Clin Nutr, 1997. 65(1): p. 149-52. 21. Kolsteren, P., J. Kusin, and S. Kardjati, Morbidity and growth performance of infants in Madura, Indonesia. Annales Tropical Pediatrics, 1997. 17(3): p. 201-8. 22. Condon-Paoloni, D., et al., Morbidity and growth of infants and young children in a rural Mexican village. American Journal of Public Health, 1977. 67: p. 651-6. 23. Checkley, W., et al., Asymptomatic and symptomatic cryptosporidiosis: Their acute effect on weight gain in Peruvian Children. American Journal of Epidemiology, 1997. 145: p. 156-163. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 113 ANNEXES 24. Adair, L., et al., Growth dynamics during the �rst two years of life: a prospective study in the Philippines. European Journal of Clinical Nutrition, 1993. 47(1): p. 42-51. 25. Zumrawi, F., H. Dimond, and W. JC, Effects of infection on growth in Sudanese children. Human Nutrition and Clinical Nutrition, 1987. 41: p. 453-61. 26. Villamor, E., et al., Human immunode�ciency virus infection, diarrheal disease and sociodemographic predictors of child growth. 93, 2004. 3(372-9). 27. Fishman, S., et al., Childhood and Maternal Underweight, in Comparative Quanti�cation of Health Risks: Global and Regional Burden of Disease due to Selected Major Risk Factors, M. Ezzati, et al., Editors. 2004. 28. WHO, Country estimates of cause-speci�c mortality. 2004, World Health Organization: World Health Report 2004 and Discussion Paper 54 29. WHO, Global burden of disease regional estimates. 2004, World Health Organization: Geneva. 30. Korenromp, E., Malaria incidence estimates at country level for the year 2004 – proposed estimates and draft report. 2005, World Health Organization. Roll Back Malaria: Geneva. 31. Begoña, A., N. Hanley, and R. Barberán, The value of leisure time: a contingent rating approach. Journal of Environmental Planning and Management, 2001. 44(5): p. 681-99. 32. Isley, P. and R. Rosenman, Using revealed preference to evaluate the reservation wage and the value of leisure. Litigation Economics Digest, 1998. 3(1): p. 61-7. 33. Lee, K. and I. Kim, Estimating the value of leisure time in Korea. Applied Economics Letters, 2005. 12: p. 639- 41. 34. Shaw, M., Value of leisure time based on individuals’ mode choice behavior. Journal of Advanced Transportation, 2004. 38(2): p. 147-62. 35. Organisation for Economic Co-operation and Development, Economic Valuation of Environmental Health Risks to Children. 2006, OECD, Paris. 36. Hutton, G., L. Haller, and J. Bartram, Global cost-bene�t analysis of water supply and sanitation interventions. Journal of Water and Health, 2007. 5(4): p. 481-502. 37. Brannon, I., What is a life worth? Regulation, 2004. Winter 2004-2005(http://www2.gsu.edu/~ecolot/docs/ appendix.pdf ). 38. Abelson, P., The value of life and health for public policy. The Economic Record, 2003. 79(Special Issue 1): p. s2-s13. 39. Jones-Lee, M., ed. The value of life and safety. 1982, North-Holland Publishing Co.: Amsterdam. 40. Bateman, I.J., et al., Bene�ts transfer in theory and practice: a review. 2000, CSERGE Working Paper GEC 2000- 25; Centre for Social and Economic Research on the Global Environment, University of East Anglia http://www.uea.ac.uk/env/cserge/pub/wp/gec/gec_2000_25.pdf. 41. Persson, U., The value of statistical life in transport: �ndings from a new contingent valuation study in Sweden. Journal of Risk and Uncertainty, 2001. 23(2): p. 121-34. 42. Smith, S.V., Jury verdicts and the dollar value of human life. Journal of Forensic Economics, 2000. 13(2): p. 169- 88. 43. Georges, D. and P. Lanoie, How to make a public choice about the value of a statistical life: the case of road safety. 2002, Working Paper 02-02. École des Hautes Études Commerciales (HEC) Montréal. 44. Baranzini, A. and G. Ferro Luzzi, The economic value of risks to life: evidence from the Swiss labour market. Swiss Journal of Economics and Statistics, 2001. 137(2): p. 149-70. 45. Perreira, K.M. and F.A. Sloan, Living healthy and living long: valuing the nonpecuniary loss from disability and death. Journal of Risk and Uncertainty, 2002. 24(1): p. 5-29. 46. Bellavance, F., G. Dionne, and M. Lebeau, The value of a statistical life: a meta-analysis with a mixed effects regression model. 2007, Canada Research Chair in Risk Management; Working Paper 06-12. 47. Kochi, I., B. Hubbell, and R. Kramer, An empirical bayes approach to combining and comparing estimates of the value of a statistical life for environmental policy analysis. Environmental and Resource Economics, 2006. 34: p. 385-406. 48. Madheswaran, S. and K.R. Shanmugam, Impact of trade unions on the compensation for job risks: evidence from the Indian labor market. International Review of Economics and Business, 2006. 50(1): p. 121-41. 49. Mrozek, J. and L. Taylor, What determines the value of life? A Meta Analysis. Journal of Policy Analysis and Management, 2002. 21(2): p. 253-70. Economic Impacts of Sanitation in Cambodia 114 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES 50. United Nations, Water for people, water for life. 2003: World Water Development Report. UN World Water Assessment Programme. Published by UNESCO and Berghahn Books. 51. Asian Development Bank, An evaluation of small-scale freshwater rural aquaculture development for poverty reduction. 2005: Operations Evaluation Department, ADB. 52. Various, Philippines sanitation sourcebook and decision aid. 2005, Government of the Philippines, WSP-EAP, GTZ, Government of Australia. 53. Anon. Proceedings of the 6th International Conference on Diffuse Pollution, Amsterdam, Sept 30 - Oct 4 2002. Organized by the International Water Association (IWA) in cooperation with the Netherlands Association on Water Management (NWA) and Aquatech. 2002. 54. WHO, Minumum water quantity needed for domestic uses. 2005, WHO/SEARO Technical Notes for Emergencies; Technical Note No. 9; WHO Regional Office for Southeast Asia. 55. Emerton, L. and E. Bos, Value: counting ecosystems as water infrastructure. 2004: IUCN The World Conservation Union; Water and Nature Initiative. 56. Machado, F. and S. Mourato, Improving the assessment of water reated health impacts: Evidence from coastal waters in Portugal. CSERGE Working Paper GEC 99-09, 1999. 57. DOF, Cambodia post-harvest �sheries overview. 2005, Ministry of Agriculture Forestry and Fisheries: Phnom Penh. 58. Torell, M., A.M. Salamanca, and B.D. Ratner, eds. Wetlands Management in Cambodia: Socioeconomic, Ecological, and Policy Perspectives. 2004, WorldFish Center Technical Report 64: Malaysia. 59. Abromovitz, J., Imperiled waters, impoverished future: the decline of freshwater ecosystems. 1996, Worldwatch Paper No. 128. 80 pages. Worldwatch Institute. 60. Calamari, D. and H. Naeve, eds. Review of pollution in the African aquatic environment. 1994, Committee for Inland Fisheries of Africa Technical Paper. No. 25. 118 pages. FAO: Rome http://www.fao.org/docrep/008/ v3640e/V3640E00.HTM. 61. Gupta, M., Challenges in sustaining and increasing �sh production to combat hunger and poverty in Asia. 2006, NAGA, WorldFish Center Quarterly Vol. 29 No. 1 & 2 Jan-Jun 2006 http://www.inspection.gc.ca/english/ anima/�spoi/manman/cssppccsm/chap2e.shtml. 62. FAO, An evaluation of small-scale rural aquaculture development for poverty reduction. 2005, Food and Agriculture Organization: Technical Guidelines for Responsible Fisheries, No. 10. FAO: Rome. 63. Virdin, J., et al., Saving �sh and �shers: toward sustainable and equitable governance of the global �shing sector. 2004, The World Bank. Agriculture and Rural Development Department. Report No. 29090-GLB. 109 pages. 64. Alabaster, J., Review of the state of water pollution affecting inland �sheries in Southeast Asia. 1986, Food and Agriculture Organization Fisheries Technical Papers, No. 260. p. 25 pages. 65. Williamson, R. and J. Klamut, Transport and fate of contaminants in surface water. 2001, University of California Agricultural Extension Service & California Department of Health Services. 66. Escher, M., et al., The effect of sewage plant effluent on brown trout (Salmo trutta fario): a cage experiment. Aquatic Sciences, 1999. 61: p. 93-110. 67. Tsai, C., Effects of chlorinated sewage effluents on �shes in Upper Patuxent River, Maryland. Chesapeake Science, 1968. 9(2): p. 83-93. 68. Tsai, C., Changes in �sh populations and migration in relation to increased sewage pollution in Little Patuxent River, Maryland. Chesapeake Science, 1970. 11(1): p. 34-41. 69. Al-Muzaini, S., et al., The quality of marine water around a sewage outfall. Water Science and Technology, 1999. 40(7): p. 11-15. 70. Bucher, F. and R. Hofer, The effects of treated domestic sewage on three organs (gills, kidney, liver) of brown trout (Salmo trutta). Water Research, 1993. 27(2): p. 255-261. 71. Burkhardt-Holm, P. and K. Scheurer, Application of the weight-of-evidence approach to assess the decline of brown trout (Salmo trutta) in Swiss rivers. Aquatic Sciences, 2007. 69: p. 51-70. 72. Chambers, P., et al., Dissolved oxygen decline in ice-covered rivers of northern Alberta and its effects on aquatic biota. Journal of Aquatic Ecosystem Stress and Recovery, 2000. 8(1): p. 27-38. 73. Environmental Quality Center, Dissolved oxygen. website, Environmental Engineering and Earth Sciences Department, Wilkes University, Pennsylvania, US http://www.water-research.net/Watershed/ dissolvedoxygen.htm. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 115 ANNEXES 74. Meck, N., Pond water chemistry. 1996, Koi Club of San Diego http://www.vcnet.com/koi_net/H2Oquality. html. 75. Mellergaardl, S. and E. Nielsen, Impact of oxygen de�ciency on the disease status of common dab Limanda limanda. Diseases of Aquatic Organisms, 1995. 22(101-114). 76. Svobodová, Z., et al., Water quality and �sh health. 1993, Food and Agriculture Organization (FAO); EIFAC Technical Paper No. 54. FAO: Rome. 59 pages ftp://ftp.fao.org/docrep/fao/009/t1623e/t1623e.pdf. 77. Pastorok, R. and G. Bilyard, Effects of sewage pollution on coral-reef communities. Marine Ecology, 1985. 21: p. 175-189. 78. Geary, P. and C. Davies, Bacterial source tracking and shell�sh contamination in a coastal catchment. Water Science Technology, 2003. 47(7-8): p. 95-100. 79. Burkhardt, W. and K. Calci, Selective accumulation may account for shell�sh-associated viral illness. Applied Environmental Microbiology, 2000. 66(4): p. 1375-8. 80. Dommison, C. and C. Ross, Animal and human fecal pollution in New Zealand rivers. New Zealand Journal of Marine and Freshwater Research, 1999. 33: p. 119-28. 81. Brisson, I. and D. Pearce, Bene�ts transfer for disamenity from waste disposal. CSERGE Working Paper, 1995. WM 95-06. 82. Powell, J.C. and I. Brisson, The assessment of social costs and bene�ts of waste disposal. CSERGE Working Paper, 1994. WM 94-06. 83. Mukherjee, N., Learning What Works for Sanitation: Revisiting Sanitation Successes in Cambodia. July 2002, Water and Sanitation Program East Asia and the Paci�c (WSP-EAP). 84. Whittington, D., et al., Household demand for improved sanitation services in Kamasi, Ghana: A contingent valuation study. Water Resources Research, 1993. 29(6):1539-1560. 85. WSP, Demand assessment for sanitary latrines in rural and urban areas of Cambodia. 2007, Water and Sanitation Program, World Bank. 86. Hutton, G. and L. Haller, Evaluation of the non-health costs and bene�ts of water and sanitation improvements at global level. 2004, World Health Organization 87. United Nations Childrens’ Fund and International Red Cross. Water, Sanitation and Hygiene Education for Schools. Roundtable Meeting. 2005. Oxford, UK 23-25 January 2005. UNICEF/IRC. 88. Borba, M.-L., J. Smet, and C. Sijbesma, Enhancing livelihoods through sanitation. 2007, Thetic Overview Paper 19. IRC International Water and Sanitation Centre. 89. Doyle, B., Increasing education and other opportunities for girls and women with water, sanitation and hygiene. 1995, UNICEF, New York. 90. Giannico, L., Prevention of the most frequent communicable diseases in tourists, in Tourist Health: A New Branch of Public Health, W. Pasini, Editor. 1988, Volume I, Rimini, Italy; IATM/WHO/WTO. 91. Zampieri, A., Epidemiology of infections in tourists, in Tourist Health: A New Branch of Public Health, W. Pasini, Editor. 1988, Volume I, Rimini, Italy; IATM/WHO/WTO. 92. Frechtling, D., Annoted Bibliography on Health and Tourism Issues. 1992: Washington, DC: International Institute of Tourism Studies/The George Washington University. 93. World Tarvel and Tourist Association and Accenture, Travel and tourism traveling to new heights. Executive summary. 2006, http://www.wttc.org/2006TSA/pdf/World.pdf. 94. U.S. Natural Resources Defense Council, Health risks and economic impacts of beach pollution. 2006. 95. MoT, Tourism statistical report: First semester 2007. 2007, Statistics and Tourism Information Department, Ministry of Tourism: Phnom Penh. 96. Esrey, S., R. Reachem, and J. Hughes, Interventions for the control of diarrhoeal diseases among young children: Improving water supplies and excreta disposal facilities. Bulletin of the World Health Organization, 1985. 63: p. 757-772. 97. Fewtrell, L., et al., Water, sanitation, and hygiene interventions to reduce diarrhoea in less developed countries: a systematic review and meta-analysis. Lancet Infectious Diseases, 2005. 5(1): p. 42-52. 98. Clasen, T., et al., Interventions to improve water quality for preventing diarrhoea. 2006, In: The Cochrane Library, Issue 3, 2006. Oxford: Update Software. 99. Cairncross, S. and V. Valdmanis, Water supply, sanitation and hygiene promotion, in Disease Control Priorities in Economic Impacts of Sanitation in Cambodia 116 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) ANNEXES Developing Countries, D. Jamison, et al., Editors. 2006, 2nd Edition. New York: Oxford University Press. 100. DRHC, Informed choice manual on rural household latrine selection. 2006, Department of Rural Health Care, Ministry of Rural Development. 101. MAFF and SNV, National Biodigester Programme Cambodia: Information Folder. November 2006, National Biodigester Programme Cambodia. 102. Werner, C., et al. Reasons for and principles of ecological sanitation. in 2nd International Symposium on Ecological Sanitation, April 2003. 2003. GTZ and IWA: http://www.gtz.de/de/dokumente/en-ecosan-reasons- and-principles-2004.pdf. 103. Renwick, M., P. Sagar Subedi, and G. Hutton, A cost-bene�t analysis of national and regional integrated biogas and sanitation programs in sub-Saharan Africa. 2007, Prepared by Winrock International for the Dutch Ministry of Foreign Affairs. 104. Chandararot, K. and L. Dannet, Biodigester User Survey Report. May 2007, National Biodigester Programme. Economic Impacts of Sanitation in Cambodia A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI) 117 ANNEXES Economic Impacts of Sanitation in Cambodia 118 A five-country study conducted in Cambodia, Indonesia, Lao PDR, the Philippines and Vietnam under the Economics of Sanitation Initiative (ESI)