i

Document of

The World Bank

Report No: ICR00001845

IMPLEMENTATION COMPLETION AND RESULTS REPORT

(TF-58243 TF-58244 TF-58245)

ON A

Grant

IN THE AMOUNT OF

(US$ 0.85 MILLION EQUIVALENT)

TO THE

Naandi Foundation

FOR A

Andhra Pradesh Rural Water Scheme Project

September 19, 2011

GPOBA

India

SAR

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

ii

CURRENCY EQUIVALENTS

(Exchange Rate Effective March 28, 2011)

Currency Unit = Indian rupee (INR)

US$ 1.00 = 44.8397

FISCAL YEAR

ABBREVIATIONS AND ACRONYMS

Vice President: Isabel M. Guerrero

Country Director: Roberto Zagha

Sector Manager: Ming Zhang

Project Team Leader: Cledan Mandri-Perrott

ICR Team Leader: Josses Mugabi

iii

COUNTRY

Project Name

CONTENTS

Data Sheet

A. Basic Information

B. Key Dates

C. Ratings Summary

D. Sector and Theme Codes

E. Bank Staff

F. Results Framework Analysis

G. Ratings of Project Performance in ISRs

H. Restructuring

I. Disbursement Graph

1. Project Context, Development Objectives and Design ............................................... 1

2. Key Factors Affecting Implementation and Outcomes .............................................. 1

3. Assessment of Outcomes ............................................................................................ 2

4. Assessment of Risk to Development Outcome ........................................................... 2

5. Assessment of Bank and Borrower Performance ....................................................... 3

6. Lessons Learned ......................................................................................................... 3

7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners ............ 3

Annex 1. Project Costs and Financing ............................................................................ 4

Annex 2. Outputs by Component ................................................................................... 5

Annex 3. Economic and Financial Analysis ................................................................... 6

Annex 4. Bank Lending and Implementation Support/Supervision Processes .............. 7

Annex 5. Beneficiary Survey Results ............................................................................. 8

Annex 6. Stakeholder Workshop Report and Results ..................................................... 9

Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR ..................... 10

Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders ....................... 11

Annex 9. List of Supporting Documents ...................................................................... 12

MAP

iv

I N S E R T

D A T A S H E E T

H E R E

AFTER APPROVAL BY COUNTRY DIRECTOR

AN UPDATED DATA SHEET SHOULD BE INSERTED

MANUALLY IN HARD COPY

BEFORE SENDING A FINAL ICR TO THE PRINT SHOP.

NOTE: The Data Sheet is generated by the system

using the information entered in the Operations Portal

each time you use “Send Draft”, “Print” or “Submit Final” functions.

5

1. Project Context, Development Objectives and Design

1.1 Context at Appraisal

In 2005, India was a low-income country with a gross national income (GNI) per capita of $720

compared to the world average of $6,9871 at that time. The population of India in 2005 was estimated

at 1,132 million, of which 71% or 766 million resided in rural areas and the rest in urban areas. An

estimated 21% of communicable diseases in the country were water related, and the number of deaths

due to diarrhea alone was 700,000 in a single year. Although 86 % of the population had access to

improved water sources, only 18% of the rural households had access to an improved water source.

The State of Andhra Pradesh (AP) where the pilot project was implemented is considered a leading

reform state in India, with a clear long-term strategy toward infrastructure development laid down in

its Vision 2020 document. It is the fifth most populous state in the country with approximately 80

million people, which in 2005, accounted for nearly 8% of the country’s total population. In addition,

the GNI per capita in AP was below India’s average. Disparities between rural and urban

development indicators in AP was noticeable, given that the rural population accounted for 73% of the

state’s total and was relatively poorer.

In 2005, AP had made considerable progress in its development agenda. Back in the late 1990s, the

state was relatively poor and in the midst of a fiscal crisis. In 1999/2000, AP’s GNI per-capita was

around US$385, nearly 22% of its population was below the poverty line, and the state’s fiscal deficit

to gross state development product (GSDP) ratio was reaching the 5% mark. Since then, the state

made impressive progress in all dimensions of development. Per-capita income rose to nearly

US$700 in 2005, although this was still slightly below the national average at that time2. During the

same period, water supply coverage increased from 3% to 65% of the total population3.

Despite these improvements, approximately 17 million people or over 22% of the total population in

the State had experienced bacteriological contamination of water at the time of project appraisal.

Rural households in the coastal districts of AP were the most affected. Most of the villages in these

coastal districts were dependent on irrigation canals/ponds for drinking water, which was being

treated through slow sand filters and distributed through individual household taps and public taps.

But due to the poor operation and maintenance of the slow sand filters and the distribution network,

the quality of water available to rural households was extremely poor. For example, it was estimated

that over 500,000 people lived with severe fecal contamination level of >2,000 Most Probable

Number per milliliter (MPN/ml), whereas the World Health Organization (WHO) permissible limit

for fecal contamination is <100 MPN/ml.

Thus, at the time of project appraisal, there was a clear need for increased investment in improving

access to safe drinking water for the rural poor in AP, together with creating awareness among

1 World Development Indicators Database, World Bank, 1 July 2006. 2 India - Third Andhra Pradesh Economic Reform Credit/Loan Development Policy Lending, Program Document, 11 December 2006. 3 Environmental Health in India-Priorities in Andhra Pradesh, World Bank, 2001.

6

households on the health risks posed by the consumption of unsafe water. The project was consistent

with the Bank Group Country Assistance Strategy (CAS) for India, and the Government of India's

Eleventh Five Year Plan (2007-2012), both of which recognize that provision of adequate

infrastructure, including improved and safe drinking water, is critical to sustaining economic growth.

1.2 Original Project Development Objectives (PDO) and Key Indicators

The objective of the project was to provide safe drinking water to 75,000 inhabitants (earning less

than US$20 per month) in the coastal areas of AP. Achieving this objective was expected to result in:

Health benefits from reduced exposure to risks posed by unsafe water

Economic benefits from reduction in medical expenses to treat water-borne diseases, increased

productivity and capacity to work due to fewer water-borne disease related to days off from

work (especially for mothers of children under five who are prone to diarrhea), and lower

expenditure related to boiling water.

Social benefits from equitable access to safe drinking water for villagers from different

religions and social backgrounds

Improved financial viability/sustainability of the existing water supply systems due to full

participation and ownership of the beneficiary community.

Transparent use of funds in the water sector

Replication of affordable village water treatment technology

Key performance indicators to measure achievements included the following:

number of households making upfront cash or in kind contribution for the project against

number of households planned to be mobilized;

number of households registered to the community safe water scheme (CSWS) represented by

paid subscription fees4;

number of consumers actually using the CSWS represented by number of sold water coupons5;

number of below-poverty level (BPL) households reached with safe drinking water;

number of households discontinuing use of unsafe water for drinking purpose;

percentage decrease in incidences of water-borne diseases and deaths; and

percentage reduction in medical/health expenditure.

1.3 Revised PDO and Key Indicators, and Reasons/Justification

There was no revision to the PDO and indicators.

4This means a user having paid a subscription fee which entitles the user to buy water from the CSWS. 5 Water Coupons were to be numbered and each household given a unique identifying code. The plant operator would then verify the

code to ensure that the Water Coupon is used only by the family it was sold to.

7

1.4 Main Beneficiaries

The project was for the development of 25 individual Community Safe Water Schemes (CSWS)

targeting 75,000 people in 25 villages. Each CSWS involved the community (represented by the

village council or “Panchayat”), Naandi Foundation, an NGO as grant recipient and project manager,

and Water Health India (WHIN), a disinfection technology provider and operator of community

village water treatment facilities, in a public-private partnership. The scheme provided a one-off

subsidy linked to the provision of clean water, with subsidy payments made largely after the delivery

of agreed outputs. These ‘outputs’ included construction and installation of the CSWS; registration of

at least 500 households to the scheme, and three months of billed user fee consumption.

The project targeted coastal districts in AP as the water contamination situation is more pronounced

than in other districts in the State. Naandi Foundation, the implementing agency, together with the

Panchayat authorities took the lead in determining which villages were chosen.

1.5 Original Components

Component 1 – Construction, Installing, Commissioning and Operating Community Safe Water

Schemes

This component involved constructing, installing and commissioning community safe water schemes

(CSWSs) consisting of ultraviolet (UV) filter water plants and storage tanks, registering households to

the CSWSs, operating CSWSs and verifying usage of CSWSs, in approximately 25 villages in

Krishna, Guntur and West Godavari districts of Andhra Pradesh.

Component 2 – Promotion of Awareness on Safe Drinking Water

This component involved undertaking campaigns and developing and implementing information,

education and communication (IEC) strategies for the purpose of educating rural communities in

Andhra Pradesh on the benefits of safe drinking water and promoting the use of treated water from the

CSWSs.

Component 3 – Independent Verification, Monitoring and Evaluation

This involved engaging the services of consultants for carrying out technical verification, evaluation

and monitoring of activities and outputs for components 1 and 2 of the project.

1.6 Revised Components

There was no revision to the project components.

1.7 Other Significant Changes

There were no significant changes to the project.

8

2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry

The project involved the development of community safe water schemes (CSWS) in response to the

lack of potable drinking water in the rural coastal areas of AP. Each CSWS involved the construction,

installation, commissioning and operation of a standalone UV water purification plant connected to a

water distribution point from which household purchase water in jerry cans. The key factors during

project preparation or issues related to quality at entry that affected implementation and outcomes can

be summarized as follows:

The project built on the positive experiences of “fee-for-service” community based water

schemes in the State. The design reflected the lessons learned and the positive experiences of

the Naandi Foundation (the lead implementing agency) in implementing similar community

based water schemes on a “fee-for-service” model working at the grass root level with full

community engagement. The project was also based on a fairly sound analysis of the demand

and willingness to pay for improved water quality in the target rural areas.

The project design was pragmatic. For years in the rural coastal areas of AP, there were no

alternatives to the highly contaminated tap water from government-run piped water systems,

which were poorly operated and maintained. In some areas, there was no piped system at all

and people relied on contaminated rivers/canals as their source of drinking water. Therefore, a

project to set up small standalone UV water purification plants as point sources for drinking

water was pragmatic and fit well within the realities from both a technical and bacteriological

perspective. Although the project did not involve domestic connections, it was a first step in

providing safe and affordable drinking water in areas where typically the existing piped water

schemes had failed largely due to broader failures in the sector, including lack of O&M

capacity and incentives at the local level, and lack of a sustainable tariff structure. On the

surface, this project may appear as a short-term fix since small purification plants have

obvious scale limitations - i.e. their ability to provide sufficient quantities of water to a

growing rural population is limited. It could be argued that a better option would have been to

support improvements /rehabilitation of existing government-run water treatment systems and

distribution network. However, the harsh reality in most of rural India is that water distributed

to individual households through piped networks is unlikely to be potable for the foreseeable

future given the massive investment required and the lack of institutional capacity. The

solution is not simply fixing plants and pipes. Significant reforms are also needed to fix the

highly dysfunctional institutional environment which is mostly responsible for the poor O&M

of existing systems. Fixing institutions is a long term undertaking, and traditionally, in the

Indian water sector, it is also a significant challenge that could not be addressed within the

limited scope and timeframe of the project.

9

The project objective was clear and realistic. The aim was simply to provide safe drinking

water to 75,000 inhabitants (earning less than US$20 per month) in the coastal areas of AP.

The anticipated outcomes at the household level (i.e. health, economic and social) were

consistent with the project scope, duration, resources and approach, and thus the project could

reasonably be held accountable for their achievement. The likelihood of achieving project

outcomes was enhanced by promoting awareness on safe drinking water use, in addition to

providing the actual safe water source.

The project design and implementation arrangements were structured around core OBA

principles – i.e. transparency and explicit use of subsidies; accountability for results; private

sector participation; innovation and efficiency; monitoring of results; and sustainability – and

aligned to the twin objectives of improving access to safe drinking water and creating

awareness among rural households on the health risks posed by consumption of unsafe water.

The project adopted a community-based public-private partnership model with strong

community participation. The schemes involved the village council or “Panchayat,” Naandi

Foundation, and WHIN in a public-private partnership which was operationalized through a

Tripartite Agreement. The Panchayat provided a regular source of water, secure land, the

community financial contribution, and electricity at a pre-agreed tariff for the CSWS. Naandi

acted as the project manager. Naandi also mobilized the community towards raising the

community contribution and developing a communication and awareness campaign, working

with key stakeholders such as schools, community self-help groups, and health workers to

change the behavioral practices of poor families regarding water. WHIN was the technology

partner/operator. The company built and installed the ultraviolet (UV) filter water purification

plants and the basic infrastructure needed to operate them (i.e., pumping system, storage tank,

and shelter) in the 25 villages; hired and trained a plant operator for each village; and shared

project implementation risks by providing Naandi with operational performance guarantees.

Community participation was ensured through baseline surveys/pre-launch awareness

campaigns; community contribution; community training, and use of community-based plant

operators.

The project incorporated sustainability into the project design. The core of the design is a

performance-based subsidy linked to the delivery of pre-agreed outputs which included three

months of billed water services. The output-based approach requires that tariffs paid by users

for consumption cover the costs of operation and maintenance, including education and

communication activities. Institutional sustainability was enhanced through the Tripartite

Agreement that set out the roles and responsibilities of all actors, the education campaign that

accompanies each project (and which is one of the verifiable indicators for OBA subsidy

disbursement) and the mechanism to ensure mobilization/collection of user charges.

Furthermore, sustainability is ensured through the continued access to drinking water by at

least 50% of total households in each village. Finally, the design incorporated accountability

mechanisms through output verification by an independent third party and regular reporting to

10

Panchayat and feedback to appropriate government agencies at both the District and State

levels.

2.2 Implementation

The project was not restructured and was never at risk. The use of output-based aid facilitated regular

output reporting and independent verification that outputs had been delivered. Payments /

reimbursements to the implementing agency were made on a per village basis as follows: 20 percent

on plant commissioning (i.e. after a technical test has been undertaken and the plant produces water to

the agreed standard); 60 percent upon verification that 500 households have been registered and paid

subscription fees, and the remaining 20 percent upon verification of three months of billed user fee

consumption. This disbursement profile provided a strong incentive for the implementing agency to

build plants only where there is demonstrated demand and willingness to pay.

The implementing agency however faced a few minor challenges which it resolved without Bank

intervention. For example, although the water was extremely affordable, at Rs. 3 ($0.06) for a 20 liter

can, beneficiaries were initially reluctant to pay, and some felt that the fee being charged should

include free delivery to their door. Clearly, they were not used to paying for water, not even from the

existing government-run systems. This prompted the implementing agency to increase outreach and

communication efforts. They also changed the mode of payment from water coupons (which were

being forged/shared by some users hence affecting the revenues) to cash payment at the water

collection point.

Another issue encountered was customer complaints regarding the taste of the water. This was

common in high salinity areas where the UV technology alone could not work effectively in removing

taste. The issue was resolved by retrofitting the plants with a combined UV and Reverse Osmosis

(RO) treatment system to deal with the salinity.

There were also instances where installation of plants was significantly delayed due to lack of

political support and delays in collecting community contributions. This however was a fact that was

recognized at the design stage and therefore no disbursements were to be made until such time when

all pre-agreed outputs were achieved.

Overall, all project stakeholders interviewed agree that the project was implemented fairly efficiently

compared to other community-based water schemes. It was simple in design, and there was a strong

commitment to the objectives by all participants. The output-based approach allowed for a flexible

implementation process which was constantly adapted to respond to what the implementing agency

saw as the main challenges as the project progressed.

Other key factors that influenced the implementation process can be summarized as follows:

The project adopted rigorous selection criteria for the target villages to maximize impact on

the poor. The project used various techniques to target the subsidy to the poorest households.

The three project districts were chosen because of their high poverty rate and lack of access to

11

quality water services. Within this geographical area, villages were selected based on the

presence of a water source that can be purified by ultraviolet technology, and the willingness

and ability of the village to adopt a fee-for-service scheme. To target individual beneficiaries

in the villages, the project used the government’s ‘white ration card,’ a system that entitles

low-income individuals to obtain basic commodities (e.g. rice, flour) at a reduced price. Other

indicators considered include family size, construction of a house from low-cost material,

limited or no possession of durable goods such as a TV, and time spent by women and

children of a given household in fetching water. Naandi Foundation carried out the targeting

in close collaboration with the village Panchayat. The process was verified by the independent

verification agent.

Use of innovative social-marketing techniques. This project had a significant awareness and

behavioral change component aimed at educating rural communities regarding the benefits of

safe drinking water and promoting the use of treated water. Decades of experience in the rural

water sector in India and elsewhere have shown that providing a safe water source alone does

not guarantee that communities will use it. Therefore, in addition to traditional health

education methods, the implementing agency developed innovative ways of promoting the

new drinking water sources through social marketing techniques such as branding (see Box 1)

which proved to be effective in increasing the likelihood of achieving project outcomes.

Delays in disbursing project funds. The project experienced delays in the disbursement of

funds. The main issue was the selection of four new villages to replace those previously

selected villages where the community could not afford the user contribution fee (the village

contribution was a key component in the overall project design). As a result there were delays

in achieving the original target of 25 plants, as these new villages came on board close to the

grant closing date. Furthermore, the project engaged the services of an independent

Box 1: The “Dr. Water” Brand

“..For years, in the heart of rural Andhra Pradesh, there were no alternatives to contaminated pond water. Then,

two years ago, the Naandi Foundation, an NGO with a strong presence in the state, approached the GPOBA with

a proposal to set up purification plants with the technical assistance of WHIN. Capitalizing on the immense

respect that doctors command in rural communities, Naandi and WHIN created the Dr. Water brand to give the

water purification plants funded by GPOBA an identity. Today, for more than 16,000 families across 25 villages,

Dr. Water is the water that keeps them free from disease resulting in a huge improvement in their quality of life.

Shyamala, a health educator in West Godavari district, Andhra Pradesh, India, had tried for months to convince families in the region to pay a small fee for clean drinking water. “I work with women’s self help groups in 10

villages. Nothing worked anywhere– flip charts, posters, skits, and role plays. Women were not convinced about

the need to use or pay for clean drinking water.”

Faced with an impasse, Shyamala then hit upon a novel idea. “We connected a microscope to a projector and

asked women in the village of Jagannadhapuram to apply untreated pond water samples to the glass slides. They

were shocked by the microorganisms that they could see on the screen. Some women actually wept on seeing the

germs that their families were routinely being subjected to.”

The practical demonstration instantly changed behavior: the number of women using the clean water facility in

this village grew from 20 to 390 members..”

Source: GPOBA (2010). Project Feature Story, September 16, 2010

Source

12

verification agent (IVA) to verify and audit the expected outputs before the grant recipient can

claim any reimbursements. Unfortunately, the IVA took longer to validate the results than

originally anticipated. However, despite this setback, the project successfully achieved the 25

treatment plants as per the original project objectives. Given the delays in disbursements, the

Bank agreed to a grant extension request submitted by Naandi to allow for a three month

period of operation validation for the plants completed close to grant closing, and thus allow

for the full reimbursement to the Naandi Foundation. Another major reason for the hold-up

was that more than half of the disbursements were linked to registration of households – which

required a prolonged period of community mobilization.

2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization

a) M&E design

The grant recipient and project implementer (Naandi Foundation) had the primary responsibility for

monitoring project progress, and for conducting baseline, mid-term and post-implementation

evaluations to measure the impact of the scheme. Naandi retained the services of an Independent

Verification Agent (IVA) paid for by the project. The IVA verified both the achievement of outputs

and the eligibility of beneficiaries benefiting from the project, through a combination of physical

observations, semi-structured interviews, in-depth interviews and focus group discussions with

households, village elders, village administration, health workers and the plant operators. In addition,

it is understood that WHIN also collected information on the performance of the plants in terms of

water quality and reliability, and financial information such as water sales and operation and

maintenance costs. However, this information was never systematically reported as part of the project

monitoring.

b) M&E implementation

The verification process involved the following:

Baseline survey – at the beginning of the project

Physical verification of plant – one month after Baseline Survey

Verification of ‘Below Poverty Line (BPL)’ Household registration as per the criteria – one

month after the plant verification

Verification of 3 months’ billing / consumption as per the criteria – one month after checking

the BPL criteria

The IVA prepared Output Verification Reports (OVR) at each stage of the verification process. In

total three OVRs were prepared for each village/plant.

13

c) M&E utilization

The above M&E system was being utilized mainly for disbursement purposes. However, the OVRs

provided more than just verification. They also contained a wealth of information from the household

survey which helped in tracking project outcome indicators relating to health, social and economic

benefits. The survey methodology was fairly sound and the resulting data and evidence could be relied

upon to come to a reasonable conclusion on the extent of achievement of project outcomes (see

Chapter 3). However, the sustainability of these M&E arrangements beyond the project is

questionable, as the main driver was to verify outputs for disbursement purposes and not necessarily

to track outcomes.

2.4 Safeguard and Fiduciary Compliance

Naandi Foundation complied satisfactorily with Bank fiduciary policy, safeguards and standard

requirements. The project management staff was skilled in its tasks. Project statements of expenditure

and disbursement requests were generally submitted in a timely manner, as were the audit reports,

except for a few audit reports towards the end of the project which took some time due to staff

changes. The procurement of consultant services and goods were conducted in full compliance with

the applicable Bank procurement guidelines and principles. No safeguard issues arose during project

implementation.

2.5 Post-completion Operation/Next Phase

The estimated service life of the CSWS is 15 years. The CSWS requires low maintenance and WHIN

has full responsibility for operation and maintenance for the initial 8 years. Thereafter, the Village

Panchayat may either choose to renew the contract with WHIN or assume responsibility for continued

operation and maintenance. WHIN will build local capacity for managing the scheme by recruiting

and training local plant operators. These will be available to the Village Panchayat beyond the

duration of the O&M contract.

WHIN charges an affordable fee of $0.06 for every 20 litre can. This rate was set to cover operation

and maintenance and the operator’s initial share of investment (net of subsidy). The socio-economic

study conducted as part of project preparation showed that willingness to pay for improved water

quality was higher than this tariff rate. Therefore, it is likely that the O&M expenses and other

associated costs of the CSWS will always be well within the user fee charged thus enhancing financial

sustainability.

Anecdotal evidence suggests that the awareness campaign implemented by Naandi has played a

critical role in creating demand for improved drinking water quality, to the extent that the community

now fully appreciates the need to pay for the services (the same was also confirmed through the focus

group discussions conducted during the baseline and OVR3 as a part of IVA’s verification process).

Unfortunately, WHIN has not provided revenue and O&M cost data for all the plants supported by the

14

grant. However, interviews with plant operators and field-based sales representatives suggest that

about a quarter of the plants are currently breaking even.

It is critical that the technical and financial sustainability of the plants be monitored and evaluated in

the future as there have been concerns regarding the high capital costs of the system, and its scale

limitations in terms of the ability to provide sufficient quantities of potable water to a rapidly growing

rural population that is demanding higher levels of services for which they are willing to pay.

Naandi was requested to report on the status of the project at the first and second anniversary of the

closing of the project. The first anniversary report is still pending. Both Naandi and WHIN are

interested in continuing the project, but need to identify additional sources of grant funding. The

methodology developed for each CSWS can be readily applied to other villages in AP. In fact, the

project has partially helped to stimulate a new sub-sector of "distributed off-grid water supply" in

rural AP. This market did not escape the attention of the Rural Water Supply and Sanitation

Department (RWSSD) of the GoAP, which decided to award BOOT contracts for water purification

plants in a few villages, but which did not really take off due to various reasons, most notably the

inability of private operators to access finance.

RWSSD is also keen to explore options for adapting the OBA concept in improving the effectiveness

and results-orientation of O&M functions for existing rural piped water schemes, with some form of

private sector participation if feasible. In other words, RWSSD is more interested in scaling up

the model piloted in the project, rather than the technology. But they acknowledge that there may be

cases where UV/RO water purification plants are the only technically feasible option (e.g. in fluoride

affected areas).

3. Assessment of Outcomes

3.1 Relevance of Objectives, Design and Implementation

The relevance of the project development objective is high. It is as relevant to the country’s rural

water sector development needs today as it were at appraisal. The twin focus on providing a safe

water source and at the same time creating community awareness of the health risks of drinking

contaminated water was highly effective in increasing the likelihood of sustainable outcomes.

However, there is currently a vigorous debate on the cost-effectiveness of the UV/RO water

purification plants and the sustainability of the off-grid water supply model. At one hand, the private

sector and NGOs who promote these plants contend that such systems are the only way to guarantee

safe drinking water to rural communities, given the dysfunctional O&M structure of existing

government-run piped water schemes. They argue for a ‘dual-system’ sector structure in which

existing piped systems are used for purposes requiring lower quality water (such as washing and

bathing), while the UV/RO plants (which ironically also tap from the same raw water source as the

existing piped systems) are used for drinking water only. The hope is that the cost would reduce as

more plants are rolled out in the communities.

15

State government agencies on the other hand maintain that the sector vision is to provide potable,

continuous (24x7) water supply to rural communities via a household connection, with a minimum of

40 liters/capita/day. UV/RO plants and the off-grid model in general, are important in filling the gap

in the short to medium term, but they are not substitutes to single-village or multi-village piped water

schemes.

There is also a technology-choice issue at play in this debate. Slow Sand Filters (SSF) – the most

commonly used treatment technology in rural drinking water supply in AP – pose maintenance

challenges, especially in a poor institutional environment. In a well-managed SSF, the efficiency of

pathogen removal can be as high as 99.5 percent. However, the maintenance of SSF is cumbersome as

cleaning of sand must be done every three months. This period comes down to 30 days in high

turbidity regions or during the monsoon period. In fact, most of the SSFs do not function during the

monsoon time.

This recurring problem with the SSF is one of the reasons for the rapidly growing market of UV/RO

plants and off-grid water supply. The RWSSD is currently exploring alternative water treatment

technologies, with a view of replacing the SSF in future schemes. In addition, there is interest in

adapting the OBA concept in improving the effectiveness and results-orientation of O&M functions

for existing rural piped water schemes, with some form of private sector participation if feasible. This

is clearly an acknowledgement that the issue is more than just a technology fix.

In summary, while the project was a modest pilot intended simply to provide safe drinking water to

75,000 poor rural inhabitants (who would have otherwise continued consuming contaminated water),

the way it was designed and implemented, has partially helped to ignite a wider sector discussion on

issues ranging from technology choice, O&M arrangements and accountability. This is evidence of

the wider relevance of the project and its achievements, which are further elaborated in the following

section.

3.2 Achievement of Project Development Objectives

The objective of the project was to provide safe drinking water to 75,000 poor inhabitants (earning

less than US$20 per month) in 25 coastal villages of AP. The project has delivered results on the

ground by providing safe water access through construction and installation of 25 UV water

purification plants in 25 villages in Guntur, Krishna and West Godavari districts. By grant closure, all

water plants were fully operational and serving a total of 16,104 poor households (or 77,878 people),

which is 29 percent higher than the original target of 12,500 BPL households. Household surveys

conducted after grant closure found that 98 percent of the households reached by the project still

continue to use water from the new plants for drinking purposes (i.e. they have not reverted back to

existing contaminated sources). This implies a high community awareness of the health risks of

contaminated water, and evidence of the effectiveness of the awareness campaign which

complemented the hardware component.

Evidence from household surveys also suggests that the project has led to health and economic

outcomes (see Table 3.1), which can be reasonably and directly linked to the project outputs.

16

Table 3.1 – Evidence of Health and Economic Benefits from Household Surveys

Before the project After the project

Average time spent collecting water was

estimated at 61 minutes

Incidence of water borne diseases in children and

adults was 68 percent and 84 percent

respectively

On an average a family was spending around

Rs.300-400 for treatment on these diseases for a

single visit

Average time spent collecting water

estimated at 20-22 minutes which shows

that there was a time saving of around

40 minutes per trip of fetching of water.

Overall, the incidence of water borne

diseases has dropped by 85 percent

(figure collaborated by health workers

in the village health centers)

Household savings on medical expenses

alone was Rs 650-750 a year.

Source: TNS India (2011). Assessment of the Community Safe Water Scheme in Andhra Pradesh. Presentation to the World bank Team, February 2011

3.3 Efficiency

The total project cost was $970,000 compared to $1.25 million estimated at appraisal. This was

because the actual unit cost of installing the UV plant turned out to be 16 percent lower than what was

estimated at appraisal.

An ex-post economic analysis of the project was carried out. Based on the key benefits (derived and

expected) from the project, at a project cost of $970,000, the analysis at grant closing yielded a

positive and significant NPV of $4.4 million and an overall IRR of 71 percent (against an IRR of 73

percent at appraisal). The economic benefit of the revenue generated from water sold alone at the

established tariff rate is also significant with NPV of $2.5 million and an IRR of 38 percent (against

31 percent at appraisal). When the net benefits (over and above their willingness to pay) of time cost

saved to fetch water are added, the IRR increases to 49 percent and with added net benefits from

improved health, the overall IRR reaches to 71 percent.

The ex post financial analysis reveals that the operators will be able to cover its operating costs and

with a profit margin only starting year 2012. With available data on revenue generation and O&M

costs as of 2010, the measures being taken to improve consumption and revenue, as well as other

relevant information (see Annex 7), it is expected that the a total of 23 schemes under operation by the

operator will generate enough revenue to cover the cost with a profit margin starting 2012 and in the

15 year of its useful life. The FRR is estimated at 23% (against 64% at appraisal) when taking into

account the subsidy from the grant; and it is estimated at 3.5% (against 6% at appraisal) without the

subsidy.

17

3.4 Justification of Overall Outcome Rating

Rating: Highly Satisfactory

The overall outcome of the project is rated as highly satisfactory on the basis of the high relevance of

the project, the level of achievement of the PDO, the efficiency of investments made, and household’s

continued use of the water sources for drinking purposes as revealed by the household surveys

conducted before the ICR mission.

The project has delivered more than the expected results through an innovative partnership between

Naandi, WHIN and the community, and through a financing mechanism that fosters transparency and

accountability for results, innovation and efficiency and sustainability. The implementation approach

also promoted participatory community involvement and building capacity in the villages for

managing the services.

The project shows how potable water schemes can be designed in a sustainable way, with real and

direct impact for the community. It enabled over 75,000 poor people to have a safe water source, as

without the water purification plants the alternative would have been to drink water from

contaminated rivers or from the existing piped water systems whose treatment facilities

(and distribution networks in general) are dysfunctional, and therefore provide unsafe water fit only

for other purposes such as washing. The project also demonstrated that rural households are indeed

willing to pay more for clean drinking water.

3.5 Overarching Themes, Other Outcomes and Impacts

(a) Poverty Impacts, Gender Aspects, and Social Development

The project targeted mainly BPL households (at least 500 households per plant), as this is the main

reason for the OBA capital subsidy grant. However, this does not mean that other better off

households could not access the water plants. The project retained an independent verification agent

to ensure that the targeting criteria of the project were met for the initial 500 households registered.

Thus, the project has reached the poor population at the grassroots level, people who otherwise would

not have been reached through public sector agencies.

The poverty impact of improving access to safe drinking water within a reasonable distance is well

documented. Access to potable drinking water helps to reduce both the medical expenditure

associated with water borne diseases and the time spent in collecting water (see for example Table

3.1). While the former has an immediate impact on disposable income, the latter frees time for more

productive activities or, in cases where children are responsible for collecting water, for education.

(b) Institutional Change/Strengthening

18

The project relied heavily on the proven track record of the grant recipient (Naandi) and the

implementing agency (WHIN) in planning and implementing community-based water schemes.

There was no involvement of State government institutions/agencies. Naandi and WHIN worked with

the village Panchayat in a tripartite agreement, which helped to strengthen the role of the village

Panchayat in service delivery and its capacity to engage with the private sector. Some of the village

Panchayats delegated their responsibilities to functional committees from among its members and also

co-opted new ones to not only generate demand for safe water, but also in promoting hygiene

practices at personal, home and community level. Regular monthly meetings of Functional

Committee/User groups were convened to discuss and make decisions on a wide range including

protection of raw water source and its catchments, and setting water charges.

(c) Other Unintended Outcomes and Impacts (positive or negative)

The project has contributed to the development of a new sub-sector of "distributed off-grid water

supply" in rural AP. As a result of this project and other NGO/private sector initiatives, the State

government has taken interest in this mode of supply, especially in areas with very high fluoride

levels in the ground water. The State government has experimented with BOOT contracts for UV/RO

water purification plants, with investments entirely raised and made by the private sector. Five 10-

year contracts were awarded to five different firms (including WHIN) covering two to three districts

per contract. However, some of the firms backed out of the contracts due to failure to raise the

required capital or because some of the villages contracted were too small to be viable.

The project has also partially contributed to stimulating a wider sector discussion on issues related to

appropriate water treatment technology for rural water supply, institutional arrangements for O&M,

and improving accountability and transparency in service delivery.

4. Assessment of Risk to Development Outcome

Rating: Moderate

The risk that the development outcomes will not be maintained is moderate. Some of the key

measures taken to enhance the likelihood of sustainable outcomes included an extensive community

awareness campaign on the health risks of drinking contaminated water, strong community

participation, and a modest user fee to cover the costs of operation and maintenance of the plant and

other associated expenditures.

Both WHIN and Naandi have already adopted similar initiatives across the State, and there is

sufficient commitment to scale up. In fact, Naandi Foundation has recently incorporated a private

company (Naandi Water) to help drive the roll-out of similar CSWS schemes in other districts in AP,

as well as other States. WHIN is also currently engaged in an expansion drive. Over the past few

years the company has instituted a number of changes including changes to its business model,

blending technologies, increasing its education and awareness campaigns and a series of changes into

19

its product offerings which has brought in substantial cost savings. In addition, investors in the

company are long term investors, including IFC, who are prepared to provide the patient capital

required for this new sector.

However, the off-grid water supply model is largely unregulated in India. The village Panchayats set

the price for the water purification services provided by the system operators or NGOs, and in some

cases provides the land and subsidized electricity to run the plants. But their capacity to monitor the

performance of the service is limited. Moreover, despite the rapid growth of the off-grid water market

in the last few years, uncertainties still remain regarding the long-term sustainability of the model, in

light of the rapidly growing rural populations who continue to demand access to higher levels of

service for which in many cases, they are willing and able to pay for.

5. Assessment of Bank and Borrower Performance

5.1 Bank Performance

(a) Bank Performance in Ensuring Quality at Entry

Rating: Satisfactory

The project preparation ensured a high likelihood of meeting the project development objective. The

selected grant recipient and implementing agency had demonstrated an excellent track record and a

proven capacity to deliver the outputs. The selection of competent implementing agencies was

instrumental to ensuring the project’s success. As the output-based approach resulted in funding only

being disbursed after independent verification of output delivery, the risk to the Bank was at all times

low.

The output-based nature of the project shifted the demand risk and other operational risks to the

recipient and implementing agencies. This ensured a very strong ownership and helped to ensure that

the project development objective was reached. The use of an independent verification agent ensured

that progress was constantly monitored and the Bank was regularly updated on the number of outputs

delivered.

During project preparation the task team assessed the recipient’s procurement and financial

management capacity. As this capacity was found satisfactory the project relied on the recipient’s and

implementing agencies’ procurement and FM arrangements, resulting in a more efficient project

implementation. Using existing capacity of the service provider to procure inputs for the project

helped to reduce transaction cost and to speed up implementation. The fixed reimbursement on

verified outputs effectively protected the Bank against procurement and FM-related risks.

(b) Quality of Supervision

Rating: Satisfactory

20

The fact that grant disbursements were conditional on the delivery of agreed outputs ensured that

project implementation focused on results rather than on inputs. A high quality of performance

reporting was assured, as disbursements were conditional on the Bank receiving output verification

reports (OVRs) prepared by an independent verification agent. The use of the recipient’s procurement

and FM systems, ensured through regular audits by a reputable audit firm, is considered best practice.

With this set-up and the small size of the project, the Bank team did not have a significant presence in

the field in form of implementation support/supervision missions. India-based staff of the Water and

Sanitation Program (WSP) – a partnership between the World Bank and the United Nations

Development Program – provided supervision support with regards to the selection of the IVA and

during the preparation of this ICR. In addition, WSP’s AP State co-coordinator provided coordination,

logistical and advisory support during project identification, design and implementation.

Whereas this was a small pilot project implemented by an NGO and a private operator, there was

merit in continuously engaging with State agencies (especially RWSSD) regarding the benefits of the

approach in general and the potential for wider application. The Bank team working on preparation of

the Andhra Pradesh Rural Water Supply and Sanitation (AP RWSS) project (which became effective

in 2009) had detailed discussions with GoAP on the OBA pilot and the possibility of scale-up of the

model in water quality affected areas as part of the AP RWSS project. However, for reasons already

explained in section 3.1, these discussions did not go far. Nevertheless, GoAP went ahead (without

Bank support) and experimented with BOOT contracts for UV/RO water purification plants which, as

already explained earlier, did not work.

(c) Justification of Rating for Overall Bank Performance

Rating: Satisfactory

Overall, Bank performance is rated as satisfactory on the basis of satisfactory performance in ensuring

quality at entry and quality of supervision, as justified above.

5.2 Borrower Performance

(a) Government Performance

Rating: Satisfactory

Government involvement in this project was only at the local village council or Panchayat. The State

Government was not involved in this project, although they endorsed Naandi’s grant application to

GPOBA. The performance of the village councils is rated as satisfactory on the basis of its strong

supporting role during project preparation and implementation in each of the villages targeted. Their

buy-in was essential, since they provided access to land, the raw water source and electricity at a pre-

21

agreed tariff for the treatment plant and mobilized the community to support the project. However,

their involvement in monitoring the performance of the operators has been limited, largely due to lack

of capacity.

(b) Implementing Agencies Performance

Rating: Satisfactory

The two implementing agencies were Naandi Foundation (the grant recipient and project manager)

and WHIN (the operator). The two agencies were selected for their capacity and proven track record

in implementing similar community-based water schemes. The performance of both agencies is rated

as satisfactory. Both demonstrated a strong commitment and knowledge of working at the grass roots

level and developing a sense of community ownership necessary to enhance the sustainability of the

project.

Both Naandi’s and WHIN’s efforts in creating awareness of the health risks of drinking contaminated

water, through innovative social marketing tools is considered best practice, and further demonstrated

the commitment to go an extra mile to achieve the project development objectives.

Both agencies however fell short of adequately complying with the grant agreement provisions related

to timely submission of final audit reports and project closing reports. There have also been

significant delays on the part of WHIN in providing the required information on the financial and

technical performance of the schemes, with the result that it was not possible to carry out a

meaningful financial analysis ex post.

(c) Justification of Rating for Overall Borrower Performance

Rating: Satisfactory

Overall, the performance of the village councils, grant recipient and implementing agency is rated as

satisfactory on the basis of the justification provided above. At the time of grant closing, the grant

recipient had complied with all the provisions of the grant agreement, except the submission of a

project closing report.

6. Lessons Learned

The design and implementation of the project offers the following key lessons that might be useful in

shaping future projects of this nature:

Rural households are willing and able to pay for clean drinking water. This has been a

consistent lesson emerging from many Bank-supported rural water supply and sanitation

projects in India. However, in most cases this demand has to be created through community

awareness campaigns. In this project, paying a cost recovery tariff was a new concept for the

beneficiaries, which for decades have been accustomed to expecting free service delivery from

22

the government. The shift in payment behavior was made possible by the community

awareness and social marketing campaigns carried out by the implementing agencies, and a

high level of transparency and accountability ensured through the OBA approach.

A win-win strategic partnership between the grant recipient and the technology

provider/operator is critical to success. The role of the Naandi Foundation as grant recipient,

administering the project and interfacing with the implementing agency and the Panchayat

through a tripartite agreement, has been instrumental for the success of the project. Naandi

provided support in selecting communities and in targeting beneficiaries. Throughout the

project Naandi conducted visits to beneficiary communities, trained and facilitated

community-based health promoters, managed awareness campaigns and helped clarify issues

relating to project implementation with the implementing agency. The partnership resulted in

a win-win situation whereby WHIN benefited from increased demand generated by Naandi’s

safe-water education campaigns, and Naandi benefited from linking its traditional health

education work to new and reliable clean water supplies. However, as operators increase their

capacity in using the community approach, the need for tripartite agreements between the

operator, an NGO and the village Panchayat becomes less necessary. In such a case a direct

contractual agreement between the Panchayat and the operator is likely to be more efficient.

Small grant-financed pilot projects can stimulate discussion on wider sector issues, but

Bank teams must be flexible enough to respond quickly and appropriately. This project

partially contributed to stimulating a wider sector discussion on issues related to appropriate

water treatment technology for rural water supply, institutional arrangements for O&M, and

improving accountability and transparency in service delivery. However, the Bank team

missed an opportunity to engage with the line State government agencies (such as the

RWSSD) regarding the potential benefits of the OBA approach in general and the potential for

wider application.

Appropriate application of the Bank’s fiduciary procedures to an output-based arrangement

minimized the fiduciary risk to the Bank without encroaching on the implementers’ freedom

to use its own systems to deliver outputs. The project has confirmed that paying on outputs

effectively transfers procurement and FM-related risks to service providers.

7. Comments on Issues Raised by Grantee/Implementing Agencies

(a) Grantee/Implementing agencies

The Grantee (Naandi Foundation) did not have substantive comments on the ICR

(c) Other partners and stakeholders (e.g. NGOs/private sector/civil society)

23

The private operator (WHIN) did not have substantive comments on the ICR

24

Annex 1. Project Costs and Financing

(a) Project Cost by Component (in USD Million equivalent)

Components Appraisal Estimate

(USD millions)

Actual/Latest

Estimate (USD

millions)

Percentage of

Appraisal

Components 1 and 2 – construction,

Installing, commissioning and

operating community safe water

Schemes and promotion of awareness

on safe drinking water

1.25 0.971 78%

Independent Verification 0.05 0.034 69%

Total Baseline Cost 1.30 1.005 77%

Physical Contingencies

0.00 0.00

N/A

Price Contingencies

0.00 0.00

N/A

Total Project Costs 1.30 1.005 77%

Project Preparation Costs 0.075 0.064 85%

Total Financing Required 1.375 1.085 79%

(b) Financing

Source of Funds Type of Co-

financing

Appraisal

Estimate

(USD millions)

Actual/Latest

Estimate

(USD millions)

Percentage of

Appraisal

Trust Funds 0.00 0.00 N/A

Global Partnership on Output-based Aid 0.85 0.834 98%

Community Contribution 0.25 0.17 68%

Commercial borrowing 0.20 0.19 95%

25

Annex 2. Outputs by Component

Component 1 – Construction, Installing, Commissioning and Operating Community Safe Water

Schemes

Component 2 – Promotion of Awareness on Safe Drinking Water

This component involved undertaking campaigns and developing and implementing information,

education and communication (IEC) strategies for the purpose of educating the targeted rural

communities on the benefits of safe drinking water and promoting the use of treated water from the

CSWSs. As awareness creation was not an OBA output, data on number of campaigns carried out

and number of people reached was not systematically tracked and therefore was not available at the

time of the ICR.

Component 3 – Independent Verification, Monitoring and Evaluation

The Independent Verification Agent produced 3 reports per village (i.e. OVR 1, II and III). This

translates into a total 75 reports for all the 25 villages.

Village Mandal District Projected Actual Percentage ppl/hh # of Beneficiaries

Paulipadu Gurajala Guntur 500 780 156% 4 3,120

Machavaram Machavarm Guntur 500 710 142% 4 2,840

Obulesinepalli Durgi Guntur 500 690 138% 5 3,450

Ravipadu Narsaraopeta Guntur 500 539 108% 6 3,234

Nujendla Nuzendla Guntur 500 560 112% 5 2,800

Nuthakki Mangalari Guntur 500 610 122% 4 2,440

Ikkarru Narsaraopeta Guntur 500 627 125% 4 2,508

Pedakalapalle Mopidevi Krishna 500 578 116% 4 2,312

Kowthavaram Gudlavelleru Krishna 500 543 109% 5 2,715

Tellaprolu Unguturu Krishna 500 757 151% 5 3,785

Pedathumidi Bantumilli Krishna 500 640 128% 4 2,560

Kaza Movva Krishna 500 605 121% 5 3,025

Thadinada Kalidindi Krishna 500 740 148% 5 3,700

Sanarudravaram Kalidindi Krishna 500 640 128% 5 3,200

Adakikolanu Nidamarru West Godavari 500 586 117% 4 2,344

A. Vermavaram Achanta West Godavari 500 542 108% 6 3,252

Tetali Tanuku West Godavari 500 670 134% 6 4,020

Marteru Marteru West Godavari 500 732 146% 6 4,392

Chataparru Eluru Rural West Godavari 500 685 137% 6 4,110

Kodamanchili Achanta West Godavari 500 629 126% 4 2,516

Pothunuru Denduluru West Godavari 500 662 132% 5 3,310

Pasivedala Kovvuru West Godavari 500 597 119% 5 2,985

Valluru Achanta West Godavari 500 640 128% 5 3,200

Jagannathapuram Tadepalligudem West Godavari 500 692 138% 5 3,460

Gollavanithippa Bhimavaram West Godavari 500 650 130% 4 2,600

12,500 16,104 77,878

26

Annex 3. Economic and Financial Analysis

Economic Analysis

The project at completion provided safe drinking water to some 16,104 (against the original target of

12,500) poor households (or 77,878 people) in 25 coastal villages of Andhara Pradesh (AP). The

project provided significant short and long-term benefits with many positive spillover social and

economic impacts at the individual and community levels. Health and environmental benefits and

cost saving from medical and health related expenses are the largest benefits of any water supply and

sanitation project, but they are difficult to quantify in entirety.

A benefit-cost framework using a “with” and “without” project methodology has been used to

calculate the Economic Rate of Return and the Net Present Value (NPV) of the project, ex post. The

economic analysis covers a project life of 15 years. Cash flow is discounted at 12%, which is the

estimated cost of capital. A conservative inflation rate of 6% per annum is assumed for cash flow

analysis. Without project situation is what would have been prevailed without the project vis-a-vis

factors such as population increase, continued exposure to contaminated water, and economic growth.

The quantifiable economic benefits include: (i) water sold at flat tariff rate established by the operator

which comprise of economic value of incremental water consumptions by households with access to

safe water; (ii) health benefit of the project for the beneficiaries; and (iii) potential time cost saving for

collecting water from distance. The economic value of cost of time saved is estimated using the

opportunity cost of labor or the income foregone in other income generating activities; which is

considered benefits to the households and the society.

The ex post economic analysis yield a positive and significant NPV of Rs.199 million (or US$4.4

million) and an overall IRR of 71% (against IRR of 73% for ex ante economic analysis at appraisal).

The economic benefit of the revenue generated from water sold alone at the established tariff rate is

also significant with NPV of R.113.9 million (or US$2.5 million) and an IRR of 38% (against 31% at

appraisal). When the net benefits (over and above their willingness to pay) of time cost saved to

fetch water are added, the IRR increases to 49% and with added net benefits from improved health,

the overall IRR reaches to 71%.6

Financial Analysis

The ex ante financial analysis at appraisal revealed that project was financially sustainable from the

operator’s perspective. The ex post financial analysis reveals that the operators will be able to cover

its operating costs and with a profit margin only starting year 2012. The reasons are: (i) of the 25

6 The gains from time cost saved to fetch water and improved health are significantly over and above

the beneficiaries’ willingness to pay; therefore, their marginal benefits (net of willingness to pay)

were used in the NPV calculation.

27

schemes constructed 23 are operated and maintained by the operator (2 are operated by the

communities); (ii) of the 23 schemes in operations at the project closing in 2010, only 11 covered

direct O&M costs and remaining sites did not generate enough revenue to cover operating costs; (iii)

plant constructed were in operation only 8 hours a day; (iii) consumption in 9 sites was low due to bad

taste of water; and (iv) in 3 sites community is small and therefore insufficient volume of water

produced and consumed. All these issues are being addressed, schemes are being upgraded to be

operated more efficiently with better quality water, and operational hours of the schemes are being

increased to cater increasing demand. With available data on revenue generation and O&M costs as

of 2010, the measures being taken to improve consumption and revenue, as well as other relevant

information, all 23 schemes are expected to generate enough revenue to cover the cost with a profit

margin starting 2012 and in the 15 year of its useful life. The FRR is estimated at 23% (against 64%

at appraisal) when taking into account the subsidy from the grant; and it is estimated at 3.5% (against

6% at appraisal) without the subsidy.

28

Annex 4. Grant Preparation and Implementation Support/Supervision Processes

(a) Task Team members

Names Title Unit Responsibility/

Specialty

Lending/Grant Preparation

Cledan Mandri-Perrott Sr. Infrastructure Specialist FEUFS Task Team Leader

Mustafa Zakir Hussain Infrastructure Specialist FEUFS Infrastructure Specialist

Juri Sekiguchi Consultant FEUFS Consultant

Luis Tineo Sr. Procurement Specialist GPOBA Sr. Procurement

Specialist

Priyanka Sood Consultant GPOBA Consultant

Kumaraswamy Sankaravadivelu Procurement Analyst SARPS Procurement Analyst

Ranjan Samantaray Consultant (Environment) SASES Consultant

(Environment)

Suryanarayan Satish Senior Social Development

Specialist

SASES Senior Social

Development

Specialist

Atul Deshpande Financial Management

Specialist

SARFM Financial Management

Specialist

Shellka Arora Legal Associate SARIM Legal Associate

ICR

Josses Mugabi Water & Sanitation Specialist SASDU ICR Team Leader

Daniel Coila Information Officer GPOBA Information/Data retrieval

Shideh Hadian Sr Infrastructure Economist SASDU Economic/Financial

Analysis

(b) Staff Time and Cost

Stage of Project Cycle

Staff Time and Cost (Bank Budget Only)

No. of staff weeks USD Thousands (including

travel and consultant costs)

Lending

FY06 5.36 23.70918

FY07 6.88 39.92773

Total: 12.24 63.63691

Supervision/ICR

FY09 2.28 11.35776

FY10 6.93 23.52481

FY11 9.38 66.68742

FY12 3.7 15.33612

Total: 22.29 116.90611

29

Annex 5. Beneficiary Survey Results (if any)

Not applicable

30

Annex 6. Stakeholder Workshop Report and Results (if any)

Not applicable

31

Annex 7. Summary of Grantee's ICR and/or Comments on Draft ICR

The Grantee did not submit their own ICR and did not provide substantive comments on the Bank

ICR. However, the private operator (WHIN) provided s short report in response to a number of

question posed by the Bank team during the ICR mission. The full original text of the report is

provided below.

**********

Background: • WHIN entered into contract with Naandi in 2007 to construct and operate 25 Community Water

systems in the state of Andhra Pradesh.

• The projects were to be funded through a mix of community contribution, GPOBA funding, bank

Loan / WHIN Contribution.

• The 25 plants are spread in 3 districts of the state, namely, Krishna, Guntur and West Godavari.

• All the 25 CWS plants (the sites are listed in Annexure) were operational in 2010 and WHIN

received the agreed GPOBA Contribution from Naandi (last tranche received in Nov 2010).

• WHIN has deployed UVW technology constructed all the 25 sites. In 2009-10, WHIN has used its

own funds to retrofit 5 sites (Machavaram, Obulesunipalle, Nuzendla, Nutakki and Ikkuru) with

RO technology due to high TDS content in the raw water.

• The prevailing price points for water are 15 paise / litre for UV plants and 20 paise / liter for RO

plants.

• Of the 25 CWS constructed, 22 were under WHIN operation at the end of 2010. Community has

taken over operations of two CWS (Tetali & Telaprolu). The Community had taken over a third

plant (Marteru), which it has handed back to WHIN in Q4 2010. WHIN has recently started

operating the Materu facility as of the beginning of 2011.

Responses to questions in mail dated 25th Feb 2011 1. Actual amount of community contribution realized from the 8 villages (from which you collected

community contribution)

Per our records, we have not received any contribution from the communities of the 8 villages.

Naandi discontinued collecting from the local communities in those 8 villages, and WHIN was

unsuccessful in collecting the dues from the community, partly because the facility was already built

and partly because of confusion with Naandi’s departure (e.g. certain of the communities claimed that

they had paid Naandi, a claim that WHIN could not verify). Therefore, WHIN ended up funding the

community shortfall in those 8 communities.

2. The amount of private capital contributed to the project

Please refer to the table below for the breakup of the funding for the 25 plants.

32

In addition, WHIN has also provided funding of approx. Rs [1.2 Mn] for RO augmentation.

3. Average yearly O&M costs for each plant, and for all the 25 plants, 2007- 2010 data, where

applicable.

WHIN groups into O&M costs of a CWS into 4 categories as follows:

• Direct O&M expenses: Wages of operator, utility expenses, consumables, repair &

maintenance and insurance

• Interest costs : interest payable on the bank loan allocated to a site

• Allocated O&M expenses: includes allocated wages of WHIN staff who manage the plants

(sales, service, quality teams)

• Corporate costs: Allocated WHIN corporate costs [and interest on WHIN contribution]

Please refer to the Table 1 in the Annexure for direct O&M expenses of the plants for period of 2007

– 2010.

4. Average yearly water sales (in both m3 and $$) for each plant, and for all the 25 plants

33

The portfolio performance during 2007-2010 is summarized as chart above. Please refer to the Tables

2 & 3 in the Annexure for site-wise data.

As can be seen in the chart, 11 of the 22 sites that WHIN has been operating are covering their direct

O&M expenses. The remaining 11 sites are not covering their direct O&M expenses for the following

key reasons:

In Pedakallepally, water is not available throughout the year, thus hampering operations

In Obulesnapalli, Sanarudravara and Ravipadu NTR respectively, the community is too small

and therefore sufficient volumes are not processed through the WHC to ensure profitability

In Khaza, Pasivedula, Pulipadu, Valluru, Kodamanchali, Tadinada and Nuzendla the

community does not like the taste of water (due to high TDS level) and therefore uses other

sources of water

5. No. of plants already breaking even out of the 25 plants

34

The portfolio of 22 plants has started meeting all direct O&M expenses in 2010. Please refer to the

chart above.

Of the 22 plants under WHIN operation in 2010, 11 sites have covered the direct O&M expenses. The

other milestones for profitability are coverage of interest costs, allocated overheads and WHIN

corporate costs, which these 11 sites are expected to attain in [2012].

Of the 11 sites where Revenues in 2010 were insufficient to meet the direct O&M expenses, 9 sites

are planned to be upgraded to RO technology which will improve the taste of the water (which is a big

driver for community adoption) and drive consumption. The RO technology up-gradation is expected

to be completed by the end of 2011. The remaining 2 sites - Obulesnapalli and Sanarudravaram - will

not be upgraded as they require other interventions, which may include handover of the CWS to the

community or relocation of CWS to a different village.

6. Summary table of key performance indicators for all the plants (reliability, quality, production,

hours of operation, water losses, etc)

Plants are operational for at least [8] hours each day – 4 in the morning and 4 in the evening.. Where

the demand is higher, operational hours are increased to cater to the demand of the community.

35

* WHIN tests the quality of water produced at each plant twice a month at 15 day interval – one at the

plant using a field test kit which yields the proximate analysis and the other at its lab in Vijayawada

where the water is tested for [20] parameters. The water tested for compliance with WHO standards

and IS 10500 (Potable drinking water standards are prescribed by Bureau of India Standards

** Out of the total universe of samples tested for all the GPOBA funded WHCs, only 10 samples

were defective in 2008, 0 were defective in 2009 and 7 were defective in 2010. All sites, which tested

negative, were re-sanitized for issue redress

Annexures:

Table 1: Site wise annual direct O&M expenses for 2007 – 2010 in Indian Rupees

• The sites are listed in the order in which they commenced operations.

• Three sites were taken over by the local community in early 2010, one of which

has been handed back to WHIN for operation

o Marteru was taken over by Community in and was handed back to WHIN

in Q42010.

o Telaprolu and Tetali were taken over by community in [early 2010] and

are under community operation. WHIN continues to be in discussion with

the local community to retake operations of the facility

o In the instance where the community took over the plant, WHIN continued

to incur O&M. This is because the plant was taken in March. Thus, for the

months of January, February and March, O&M costs were incurred

o

• Average per site O&M expenses have reduced by 30% between 2008 and 2010

due to (i) better procurement strategies for consumables and (ii) optimization of

consumables, Repairs & maintenance processes and (iii) optimization of the

operator pool.

Table 2: Site wise annual Revenue (User Fees) for 2007 – 2010 in Indian Rupees

• The sites are listed in the order in which they commenced operations.

• Marteru was taken over by Community in [early 2010] and was handed back to

WHIN in Q42010.

• Telaprolu and Tetali were taken over by community in [early 2010] and are under

community operation.

• Avg revenue per site has increased 4 times between 2007 and 2010.

• Machavaram was fitted with RO in [second half of 2010] hence a significant

improvement in revenue over 2009.

Table 3: Site wise annual volume dispensed for 2007 – 2010 in litres

• The sites are listed in the order in which they commenced operations.

• Marteru was taken over by Community in [early 2010] and was handed back to

WHIN in Q42010.

• Telaprolu and Tetali were taken over by community in [early 2010] and are under

community operation.

• Overall Volumes dropped after 2008 due to change in user fee from 7.5 paise /

Liter to 15 paise / Liter and increased competition offering RO and Ozonized

water.

Annex 8. Comments of Co-financiers and Other Partners/Stakeholders

Not applicable

Annex 9. List of Supporting Documents

1. Mandri-Perrott (2008). Output-based Aid in India: Community Water project in

Andhra Pradesh. OBApproaches Note 21 (October 2008). GPOBA/World Bank,

Washington DC.

2. World Bank (2009). Andhra Pradesh Rural Water Supply and Sanitation Project.

Project Appraisal Document. World Bank, Washington DC

3. World Bank (2004). Country Assistance Strategy for India. World Bank,

Washington DC

MAP

I N S E R T

M A P

H E R E

AFTER APPROVAL BY COUNTRY DIRECTOR

AN ORIGINAL MAP OBTAINED FROM GSD MAP DESIGN UNIT

SHOULD BE INSERTED

MANUALLY IN HARD COPY

BEFORE SENDING A FINAL ICR TO THE PRINT SHOP.

NOTE: To obtain a map, please contact

the GSD Map Design Unit (Ext. 31482)

A minimum of a one week turnaround is required