Summary of the targets from the paper
Water and Sanitation
||Benefit for every dollar spent
||Eliminate open defecation (rural only)
||Universal access to basic drinking water at home
||Universal access to basic sanitation at home
UN-Water has developed an integrated and broad Water Goal proposal, which has the contribution and buy-in of many governments and sector partners.
On drinking-water, sanitation and hygiene (WASH) specifically, a highly consultative process has been managed by the WHO/UNICEF Joint Monitoring Program (JMP) since 2011, leading to a series of proposed WASH targets and indicators for the post-2015 period. These go beyond the MDG target on improved drinking water and basic sanitation: the targets also include hand washing, WASH outside the household, more advanced water and sanitation services, and accelerated coverage for poor and disadvantaged groups until the target year 2030. An interim target includes ending open defecation by the year 2025.
This paper provides an evidence base with which to compare different WASH targets and world regions by benefit-cost ratio.
This paper is part of a larger study being conducted by the World Bank on WASH targets proposed by WHO and UNICEF. The proposals include the universal coverage of households with basic WASH services by 2030, with faster acceleration of access for the population groups currently with lowest access. Once the underlying coverage data sets are available, targets that provide a greater proportion of the overall population with more advanced WASH services will be analyzed. The present paper provides benefit-cost ratios for basic WASH services.
The estimation model
The basis of all the calculations are two key statistics, one on population numbers over the study period and the other for WASH service coverage in the year 2015 under different service definitions. The model moves populations from lower to higher service levels, calculating the costs and benefits of doing so. This is done for each wealth quintile separately, accelerating coverage at a faster rate to those populations with lower coverage.
The quantitative model is run at country level, and the results aggregated to give the regional and global totals or averages, weighted by country population size. High-income countries are excluded from the study, except Equatorial Guinea, which has below 50% sanitation coverage and Russia, which has closer to 90% sanitation coverage but still has a significant number of child deaths attributed to poor WASH.
Population sizes for rural and urban areas were sourced from UN Statistics for the latest year (2012) and UN projected estimates to 2030. The countries included represent 6.12 billion (84%) of the world's 7.3 billion population in 2015, and 7.15 billion (85%) of the projected 8.4 billion population in 2030. !n 2015, 43% will live in urban areas, expected to rise to 56% in 2030. For the health impact analysis, populations are disaggregated into three age groups (0-4 years, 5-14 years and 15+ years) due to the differential information available for these groups on disease incidence.
The targets proposed for household WASH services are:
1.Eliminate open defecation
2.Achieve universal access to basic drinking water, sanitation and hygiene for households, schools and health facilities;
3.Halve the proportion of the population without access at home to safely managed drinking water and sanitation services; and
4.Progressively eliminate inequalities in access
Each of these needs a concrete definition in order to conduct an economic analysis and enable consistent monitoring over time.
Eliminating open defecation is a necessary milestone on the way to everyone having basic sanitation. Note that sewage flushed from a toilet to a drain that leads directly to a canal, river or other open water without treatment is currently classified as ‘improved’ sanitation, even though it is effectively open defecation from an environmental standpoint. The indicator used for the current study is the percentage of population practicing open defecation. Calculations assume that the lowest cost options are used, including a private or shared traditional latrine in rural areas or communal toilets in urban areas. The apparent cost advantage of latrines with the lowest capital cost may be reduced when renovation and replacement costs are taken into account.
Basic drinking water at home. Drinking water is water used for drinking, cooking, food preparation, personal hygiene or similar purposes. Households are considered to have a ‘basic’ drinking water service when they use water from a household piped water supply, a protected community source such as a well, spring or borehole, or collected rainwater. The indicator used is the percentage of population using a protected community source or piped water with a total collection time of 30 minutes or less for a roundtrip, including queuing. For costing purposes, half of the unserved population is assumed to be supplied by a protected community borehole, with the other half supplied by a dug well.
Basic sanitation at home. To be counted as ‘basic’ sanitation, facilities must effectively separate excreta from human contact, and ensure that excreta do not re-enter the immediate environment. The indicator is the percentage of population using basic sanitation facilities shared among no more than five families that know each other. For the costing exercise, the mix of basic facilities assumed to be used by households includes a pour-flush pit latrine (50% of unserved) and a dry pit latrine (50% of unserved) in rural areas, and a flush toilet to septic tank (50% of unserved) and any type of pit latrine (50% of unserved) in urban areas. In the case of shared facilities, the average number of households sharing is assumed to be 2.5.
Progressive elimination of inequalities in access. Future indicators will be disaggregated on four dimensions, but this study considers just wealth quintiles and urban versus rural areas.
The total population of the 140 countries included in this study is predicted to grow from 6.12 billion in 2015 to 7.15 billion in 2030. Therefore, a coverage assumption is needed for this additional global population of 1 billion. We assume that these people will be accommodated in new dwellings. Since the cost of providing WASH facilities is difficult to disaggregate from the cost of the new dwelling itself, the same unit costs are used as for adding WASH facilities to existing dwellings.
The total intervention cost consists of all resources required to put in place, operate and maintain a WASH service. The main components of this are investment costs (CapEx), major maintenance costs (CapManEx) and regular, recurrent costs (OpEx). For this study, emptying of septic tanks and latrines is considered as capital maintenance, as it is more likely to happen every few years as opposed to every year. Due to lack of unit cost data on some cost components, management and support costs for initial program delivery including behavior change are added as 10% of the CapEx, and CapManEx is estimated at 30% of the CapEx every five years for hardware maintenance, while for safe excreta management the emptying and treatment of septic tanks and pit latrines is considered an additional cost.
There is also a distinction to be made between the incremental costs of extending access to basic and safely managed WASH to those not currently having access, and the costs of maintaining, renovating and replacing WASH services for all populations served.
Cost data were obtained through an extensive search of both published and grey literature. In addition, available cost data were sent to experts in the 40 countries with the largest unserved populations for verification and request to provide latest country-based estimates. Costs are expressed in US dollars in the baseline year.
Given that cost data between different studies, even in the same country, can be highly variable, with the major data source being agency reports as opposed to peer reviewed journals, the results of a global costing exercise are highly uncertain. The quality and representativeness of the cost data sets are themselves uncertain, and the level of service the unit cost refers to is often also unclear. Besides costs, other uncertainties include those in lifespan of technologies, in the appropriate discount rate, and unforeseen future events like population growth, migration and climate change.
A large range of economic and social benefits can result from the provision of improved WASH services, but in many cases there is insufficient evidence to enable a credible global assessment.
Over recent decades, compelling evidence has been gathered that significant and beneficial health impacts are associated with improvements in access to safe drinking-water, basic sanitation and handwashing facilities. The present study focuses on water-borne and water-washed diseases, which are the major ones at the household level. Infectious diarrhea is the most significant of these diseases; this category includes cholera, salmonellosis, shigellosis, amoebiasis, and other protozoal and viral intestinal infections. Also, since environmental factors are estimated to account for 50% of undernutrition in the developing world, we include diseases with a higher incidence or case fatality rate due to malnutrition. Specifically, a proportion of the total burden of respiratory infection and malaria in under-fives is attributed to poor water supply and sanitation, on a regionally –determined basis.
The economic benefits are threefold. There are savings related to less need to seek and provide healthcare, a reduction in the loss of productive time due to disease, and a reduction in premature mortality. The time of adults too sick to work is valued at 30% of average GDP per capita, on an hourly basis. For school-age children and adults caring for under-fives, time is valued at 15% of GDP per capita. Mortality is valued using the human capital approach. Health benefits are presented by income quintile for two variables; WASH coverage and under-five deaths.
There are also time benefits to be gained. Gaining access to basic improved water supply reduces roundtrip times from 40 to 20 minutes in urban areas and from 60 to 20 minutes in rural areas. The time saving is a combination of closer access and higher number of water points, leading to less queuing time. For sanitation, in the baseline, only one trip per day is assumed for defecation.
Sensitivity and scenario analysis
The baseline and all scenarios are presented at discount rates of 3 and 5% while a one-way sensitivity analysis was performed on the value of death, substituting a cost per DALY of US$ 1,000 and US$ 5,000 in all countries instead of the value of life using the human capital approach.
Based on the new indicator definitions for basic water and sanitation, 2.3 billion people will need to be covered with basic water and 3 billion will need to be covered with basic sanitation. For water supply, over 900 million of the unserved reside in sub-Saharan Africa, while for sanitation over 1 billion of the unserved reside in each of sub-Saharan Africa and South Asia.
The benefit-cost ratio of ending open defecation is 5.8-7.3, depending on the assumptions used. The benefits vary from US$ 81 billion to US$ 99 billion per year, at DALY values of US$ 1,000 and US$ 5,000, respectively. The annual cost of ending open defecation over a 15 year period from 2015-2030 is US$ 12-14 billion
The benefit-cost ratio of providing basic water is between 3.3. and 4.4 depending on the discount rate and DALY value. The benefits of basic water vary from US$ 50 billion to US$ 66 billion per year, at DALY values of US$ 1,000 and US$ 5,000, respectively. The annual cost of providing basic water supply over a 15 year period from 2015-2030 is approximately US$ 13-15 billion.
The benefit-cost ratio of providing basic sanitation is between 2.9-3.3. The benefits of basic sanitation vary from US$ 94 billion to US$ 107 billion per year, at DALY values of US$ 1,000 and US$ 5,000, respectively. The annual cost of providing basic sanitation supply over a 15 year period from 2015-2030 is US$ 23-38 billion.
In urban areas, the benefit-cost ratio for basic water varies between regions, from 2.2 (South Asia) to 5.4 (Eastern Asia), with a global ratio of 3.4. At a 5% discount rate, the global benefit-cost ratio reduces from 3.4 to 3.1. In rural areas, the benefit-cost ratio for basic water varies from 4.5 (South Asia) to 16 (Eastern Asia), with a global ratio of 6.8. At a 5% discount rate, the global benefit-cost ratio reduces from 6.8to 5.7.
For the provision of basic sanitation, the benefit-cost ratio in urban areas is between 1.2 (sub-Saharan Africa) and 5.7 (Oceania), with a global ratio of 2.5. At a 5% discount rate, the global benefit-cost ratio reduces from 2.5 to 2.3. In rural areas, the benefit-cost ratio for basic sanitation varies from 3.8 (sub-Saharan Africa) to 47 (Oceania), with a global ratio of 5.2. At a 5% discount rate, the global benefit-cost ratio reduces from 5.2 to 4.8.
To eliminate open defecation, simpler sanitation options are feasible. However, these options have a shorter lifespan and require continued support to motivate communities to remain free of open defecation and repair or replace or their latrine when it stops functioning. When the lifespan of a simple or traditional latrine is assumed to be one year only, the benefit-cost ratio is 6.0 globally, varying from 3.9 (sub-Saharan Africa) to 33 (Oceania). The results are highly sensitive to the assumptions on how long the hardware and support package last for – if these are increased to 2 years then the benefit-cost ratios are doubled. At a 5% discount rate, the global benefit-cost ratio stays the same at 6.0.
In all cases, benefit-cost ratios are higher for poorer populations.
The general findings for water and sanitation provision are:
1.Benefit-cost ratios are higher in lower income quintiles, accounted for by the higher health impacts in these populations. This is a compelling reason to serve the poorest first.
2.Benefit-cost ratios are higher in rural areas, accounted for by lower unit costs and higher capacity to benefit from health and access time savings.
3.Benefit-cost ratios are higher for sanitation than for water supply – with a greater difference in rural areas. This is accounted for by greater access time savings per person for sanitation than for water, especially in rural areas.
There are significant decreases in benefit-cost ratio when using a higher discount rate (of 5% instead of 3%). Using an even higher discount rate (e.g. 8% or 10%, as many countries as well as development banks do use), the benefit cost ratios can be reduced by half. The lower discount rates partly explain the difference in ratios between this study and the most recent previous global analysis, which used an 8% discount rate.
An analysis was done with DALYs being valued at either $1,000 or $5,000. Globally the benefit-cost ratios are slightly lower than the basecase when valued at US$1,000 and slightly higher when valued at US$5,000. The overall benefit-cost ratios are closer to the urban value, since costs are estimated to be twice as much in urban areas compared to rural ones. While the DALY varies by five times, the effect on benefit-cost ratios is not that significant. This is because overall mortality contributes a small share of the overall economic benefit and the major benefit comes from time savings.
Provision of basic water and sanitation facilities to the billions of people currently unserved would be a good investment in economic terms. The gains are greatest in rural areas and BCRs are higher for lower income quintiles. Economic returns varied between different regions of the world due to different price levels of water and sanitation services, different capacity to benefit.
The benefit-cost ratios for water presented in this study are twice as high as the most recent global study. This is partly due to higher health benefits, resulting from an updated health impact of basic water supply (34% instead of 18% reduction in diarrheal disease). It is also due to an updated unit cost database, which has lower unit costs than used in previous studies. A third reason is that in this study some populations receive a lower cost technology (divided between borehole and dug well) than in the previous study which assumed only borehole.
The benefit-cost ratios are also marginally lower for sanitation, comparing the global BCR of 5.5 in the previous study with 3.4 (urban) and 6.8 (rural) in the present one.
CCC ... Copenhagen Consensus Center
The text being discussed is available at