List of publications 8 - Information and Library Network ...shodhganga.inflibnet.ac.in/bitstream/10603/6299/17/17_chapter 8.pdf · List of publications (a) ... your family needs and

Publications

TERI University Ph.D. Thesis, 2006

List of publications

(a) In Journals

Dutta V, Chander S and Srivastava L. 2005

Public Support for Water Supply Improvements: Empirical Evidence

from Unplanned Settlements of Delhi, India

Journal of Environment and Development 14 (4): 439–462

Sage Publications, University of California, San Diego, USA

Dutta V and Tiwari A P. 2005

Cost of Services and Willingness to Pay for Reliable Urban Water

Supply: A Study from Delhi, India

Journal of Water Science and Technology: Water Supply 5 (6): 135–144

IWA Publishing, London


Pricing water - Reflections on the Increasing Block Pricing policy of

Delhi’s water utility

Journal of Indian Buildings Congress 12 (1): 233–243

Dutta V. 2004

A conceptual inquiry into sustainability criteria for urban water systems

Urban India, Journal of National Institute of Urban Affairs 24 (2): 89–131

(b) In Conference Proceedings

Dutta V. 2006

Public support for water supply reform in unplanned sector: Empirical

evidence from an urban water utility

6th Annual Global Development Conference

Global Development Network, January 17–24, St. Petersburg, Russia

8

Publications

TERI University Ph.D. Thesis, 2006

Dutta V. 2005

Preference heterogeneity, public choice and willingness to pay for water

supply improvements in planned and unplanned areas of Delhi, India

International Congress on Environmental Planning and Management

Brasilia, Brazil, September 11 – 15, 2005. ISBN: 85-905036-2-3, Pp. 24


Water as an economic good – a framework for valuing environmental

externalities for the urban water supply and use

IWA International Conference on Water Economics, Statistics, and Finance

Rethymno, Greece, July 8 – 10, 2005. ISBN 960-88711-1-5, 9–14


Sector Reforms, Regulation and the Challenges of sustainability:

Demand side Analysis for Urban Water Utility of Delhi, India

XII World water Congress of IWRA – Water for Sustainable Development –

Towards Innovative Solutions, 22–25 November 2005, New Delhi, India

Dutta V and Chander S. 2003

Art and Science of urban water sustainability: a vision for sustainability

Proceedings of National Seminar on Water for National Capital Region

Central Water Commission, Ministry of Water Resources, Govt. of India, 30 April,

2003, 1–23

Questionnaire: Willingness to pay for water of differential quality (Set 1: Ascending)

RECORD No.: SURVEYED BY:

WARD NAME: VERIFIED BY:

COLONY: CHECKED BY:

PROPERTY CATEGORY: DATE OF SURVEY:

Name of the respondent:

Address:

Gender: Male [ ] Female [ ] Age:

1. Respondent is the head of the household: Yes/No

2. Qualification:

(a) Postgraduate [ ] (d) Primary education [ ]

(b) Graduate [ ] (e) No schooling [ ]

(c) Secondary/higher secondary [ ] (f) Others, pls. Specify [ ]

3. Occupation:

(a) Personal Business (d) PSU

(b) Government service/Retired (e) Housewife

(c) Private job (f) Others, pls. Specify

4. Housing category:

(a) Bungalows with garden/lawn (e) Traditional Houses

(b) Bungalows without garden/lawn (f) Government Qtrs.

(c) DDA Flats: HIG/MIG/LIG (g) Private house on a plotted land

(d) Group Housing Society (h) Slums/JJ Clusters

5. Tenurial status of the house: (a) Owned [ ] (b) Rented [ ] (c) Govt. [ ]

6. If it is a rented house, what is the rent per month? Rs____________

7. No. of rooms:

8. No. of person residing:

(a) No. of adults (> 16 yrs.)

(b) No. of minors (< 16 yrs.)

9. What is your household’s average monthly income? Rs___________

(a) Less than 2500

(b) 2500 – 5000

(c) 5000 – 8000

(d) 8000 – 12000

(e) 12000 – 15000

(f) 15000 – 20000

(g) More than 20000

10. What is your household’s average monthly expenditure?

(a) less than 2500

(b) 2500 – 5000

(c) 5000 – 8000

(d) 8000 – 12000

Rs___________

(e) 12000 – 15000

(f) 15000 – 20000

(g) More than 20000

11. Most important environmental problem in your area, according to you is: (please tick one)

(a) Air pollution (f) Other, pls. specify

(b) Garbage disposal

(c) Water quality/pollution

(d) Traffic congestion

(e) Sewerage congestion/overflow


12. Are you aware of the water supply problems in your area? (Please tick one)

(a) Fully aware (c) Somewhat aware

(b) Highly aware (d) Ignorant

13. Do you think the overall quality of piped water (taste, smell and colour) supplied by the municipality is appropriate?

(Tick as applicable)

(a) Gives poor taste [ ] (b) has turbidity [ ] (c) smells bad [ ]

OR, (d) Overall quality is appropriate

Additional comments about the water quality, if any:

14. Do you have a municipal water supply/connection? (a) Yes [ ] (b) No [ ]

If no, please state the source of water supply:

(a) Personal groundwater borewell

(b) DJB tanker

(c) Private water tanker/vendor

(d) Bottled water

(e) Public hand pump

(f) Public tap

15. Do you feel the quantity of water supplied is adequate for your needs? (a) Yes [ ] (b) No [ ]

If no, how many additional hours per day of water supply will be required to meet all your needs? ______

16. Do you prefer round the clock 24-hours water supply? (a) Yes [ ] (b) No [ ]

17. What is your household’s main source of water for bathing and washing cloths?

(a) Municipal water [ ] (b) Groundwater [ ] (c) Public handpump [ ]

(d) Other, pls. specify

18. Do you treat water before using it for potable purposes? (a) Yes [ ] (b) No, We drink directly [ ]

If yes, how do you treat water?

(a) Boil (b) Boil + Filter

(c) Cloth Filter (straining) (d) Zero B

(e) Ceramic candle filter (f) UV (e.g. Aquaguards) filter

(g) Reverse Osmosis filter (h) Other_________________

19. What is the water supply situation in your house?

(a) No. of times in 24 hrs.

(b) Duration (in hrs/min.):

Morning: Daytime: Evening:

20. How do you rate the water pressure: (a) High [ ] (b) Average [ ] (c) Weak [ ]

21. For how many hours/minutes do you run the booster pump to increase the pressure of municipal supply?

______________

22. Do you have any preference for the water supplier?

(a) Govt. – owned

(b) Privatised

(c) Govt. – private joint

(d) Any other (specify)/indifferent

23. What do you prefer a (a) Fixed charge (b) metered bill

24. How do you pay for water?

(a) We pay by flat and fixed monthly tariff

(b) We pay by metered tariff on actual consumption

(c) We don’t pay because our supply is unmetered

25. If metered, does this meter work? (a) Yes [ ] (b) No

26. Last monthly bill: Rs_______


27. How much did you pay for water last year? Rs ________for_______months

28. Total no. of toilets in your house:

(a) Indian style with flush [ ]

(b) Traditional dry pit type [ ]

(c) Modern flush type (Western style) [ ]

(d) Others, Pls. specify_____________________________

29. Total no. of bathrooms in your house:

30. Do you use groundwater in your premises? (a) Yes [ ] (b) No [ ]

If yes, for how many hours/min. you run the motor everyday?

Pls. specify the capacity of the motor:________HP

31. Water supply arrangement (please tick as appropriate):

(a) Water goes direct to overhead tank with the help of booster pump because pressure is not sufficient

(b) Water goes direct to overhead tank without booster pump because pressure is high

(c) We first store at the under ground tank and then pump to over head tank

(d) We don’t have over head tank and use running water

(e) Any other (Describe arrangement)

32. What type of storage do you have; what is the total volume of your storage and if you know how much was the

installation cost, please write in Rs.?

Type Capacity (liters or gallons)

(1) Overhead tank

(2) Underground tank

(3) Drum

(4) Bucket/vessel

(5) Others

(6) None

33. Total consumption of water per day for your family for different purposes: (If you don’t know the exact amount in litres,

answer in buckets) If you don’t know, you can write total consumption of water per day based on frequency of tanks-filling

Drinking Floor washing

Bathing Flushing

Cooking Gardening

Washing clothes Car washing

Water coolers Others

TOTAL =

34. How do you balance shortfall in water supply?

(a) We have installed tubewell in our premises

(b) We rely on bottled water

(c) We get water from tankers

(d) Public handpump

(e) Public tap

(f) Other pls. Specify


35. How much water you draw from these alternate

source(s) on a daily basis: (in liters/ or in buckets)

[Litres/Buckets]

How much money do you spend per month on drawing

water from these alternate sources?

[Rs]

(a) Tubewell

(b) Bottled water

(c) Tanker

(d) Handpump

36. Do you think that you or your family members get some health impacts/ waterborne illnesses due to bad quality

water? (a) Yes [ ] (b) No [ ]

37. If Yes, please give some impacts that you have experienced during last six months in your family or existence of any

chronic illness due to the consumption of unsafe water?

Disease yes/no no of persons affected sickness days money spent on treatment

Diarrhoea

Dysentery

Jaundice

Typhoid

Cholera

Vomiting

Others, pls specify

38. In a year, how many such episodes occur in your family?

39. Expenses on medical care/treatment – if the sick person got treatment, how much was the total medical cost?

40. Any additional comments in convenience, quantity, quality and condition of water supply that you would like to

make?


41. Do you think that water supply in your area needs improvement? If you think so, which water supply options according to your family needs and your preferences would be the best in terms of quality, quantity and reliability for your family? (Pls. tick one)

[Note: Enumerators should explain risk and reliability for each of these options to the respondent for making an informed

choice]

1. DUAL SUPPLY (POTABLE + NON POTABLE)

Go to Scenario 1

2. EXISTING SUPPLY WITH IMPROVED QUALITY AND QUANTITY

Go to Scenario 2

3. EXISTING SUPPLY WITH MY PERSONAL PURIFIER OR BOTTLED WATER

Go to Scenario 3

(a) WITH BOTTLED WATER (Packaged Drinking Water)

(b) WITH WATER PURIFIER

4. NO CHANGE IN WATER SUPPLY, I AM HAPPY WITH THE PRESENT SYSTEM (Status Quo)

STOP


Valuation section: Scenario 1 [DUAL SUPPLY: POTABLE + NON POTABLE]

As things stand currently, you pay a certain unreliability cost (cost of unserved/underserved water) as well as pay for

any damage due to bad quality, unhygienic water. You may also pay for stand-by source of water, such as bottled

water (Rs. 600 – 800 per 1000 litres), water tankers (Rs. 100 per 1000 litres) – which can be far more expensive per

kilolitre than municipality supply. For example on an average you indirectly pay Rs. 2000 every year on reducing

unreliability and Rs. 3000 – 3500 on treating water borne sickness. This may be 10 times more than what you pay to

the municipality for annual water consumption. An improved water supply may improve your family’s health and you

can avoid considerable cost on waterborne diseases such as jaundice, gastroenteritis, diarrhoea, etc1. These will

also avoid any cost on bottled water, home treatment (aqua guards etc...), and result in increased life of pipelines and

household fixtures such as water heaters.

In the light of above, the city water agency plans to provide the appropriate quality of reliable drinking water to

consumers in an environmentally sensitive manner. The two main objectives could be defined as:

(1) Improvement of supplied water quality to appropriate standards equivalent to WHO norms

(2) Assured supply in terms of better reliability gradually going on a 24 x 7 scheme

In order to do this it will require significant new investments. This new investment is expensive and the cost would

have to be recovered through the water tariff. The water will be reliable all day so long as you pay for it.

1. We do not yet know exactly how much this investment would cost but I would like to know, if you agree in

principle with paying for an improved water supply. If this project is implemented and if piped water quantity

for water is sufficiently increased gradually to 24 hours supply per day, it would result in higher tariff rates.

Would you want this project to be implemented?

(a) Yes (b) No

2. If ‘No’ give reasons:

(a) Cannot afford to pay

(b) Present supply is adequate to our need and demand is

already met from our installations.

(c) I believe the current tariff is already high.

(f) Other (please specify)

3. Now I am going to read you some costs per kilo liter (1000 liters) of potable water that the proposed

water supply might cost you. Please tell me if you would be willing to pay this amount. (Start bidding

game) [Note: 1000 litres can be equated to 50 buckets of water]

(a)

I would be willing to pay Rs 40 for 1 kilolitre of reliable potable water that meets WHO norms.

If no then proceed to (c). If yes, proceed to (4).

(b)

I would be willing to pay Rs 30 for 1 kilolitre of reliable potable water that meets WHO norms.

If no then proceed to (d). If yes, proceed to (4).

(c) I would be willing to pay Rs 20 for 1 kilolitre of reliable potable water that meets WHO norms?

If no then proceed to (e). If yes, proceed to (4).

1 Cost of illness for a representative household in Delhi due to diarrhoeal diseases is Rs.1094.31 per annum (Dasgupta, 2001)


(d) I would be willing to pay Rs 10 for 1 kilolitre of reliable potable water that meets WHO norms?

(4) What is the maximum number of rupees per month as WATER BILL you would be willing to pay for reliable

potable water that meets WHO norms? Rs__________

5. Now I am going to read you some costs per kilo liter (1000 liters) of non-potable water that the

proposed water supply might cost you. How much are you willing to pay for non-potable water that is

supplied from the Water Utility?

a) I would be willing to pay Rs 12 for 1 kilolitre of non-potable water? If no then proceed to (b). If yes,

proceed to (6).

b) I would be willing to pay Rs 8 for 1 kilolitre of non-potable water? If no then proceed to (c). If yes,

proceed to (6).

c) I would be willing to pay Rs 4 for 1 kilolitre of non-potable water? If no then proceed to (d). If yes,

proceed to (6).

d) I would be willing to pay Rs 2 for 1 kilolitre of non-potable water? If no then proceed to (e). If yes,

proceed to (6).

6. What is the maximum number of rupees per month as WATER BILL you would be willing to pay for non-

potable water? Rs_____________

7. Would you be willing to pay a one-time connection fee? Yes/No

8. If yes, how much you are willing to pay one time connection fee? Rs___________

9. Which of the following statements best describes the way you felt when I described the option of an improved

supply to you: (please tick one only):

(a) “The introduction of new investment in to the water sector in this city has a good chance of improving the

water supply”

(b) “This new investment may improve the water supply but it will take so long to produce benefits it is not

worth much to me”

(c) “I just do not believe that this system for improved supply you described could be introduced or made to

work in our city”

To understand dual-piping system, kindly follow this: A decentralized dual-piping system is being planned – one

for potable and other for non-potable purposes. Water of appropriate quality would be provided for potable purposes

at sufficient pressure with appreciable reliability in the kitchen and bathrooms. In addition to that, recycled non-

potable water would be provided for toilet flushing and other non-potable requirements such as gardening, floor-

washing etc. It would also be used for recharging the aquifers (groundwater basins), which may help you and the

water agencies to manage water supplies and minimize the severity of rationing during dry months.


Scenario 2 [EXISTING SINGLE SUPPLY WITH IMPROVED QUALITY

POTABLE WATER]

As you are aware, people in the city face problems of water supply. In the light of above, the city water agency plans

to provide the appropriate quality and quantity of reliable drinking water to consumers. The two main objectives are:

(1) Improvement of supplied water quality to appropriate standards equivalent to WHO norms

(2) Assured supply in terms of better reliability gradually going on a 24 x 7 scheme

An improved water supply may improve your family’s health and you can avoid considerable cost on waterborne

diseases such as jaundice, gastroenteritis, typhoid, etc. These will also avoid any cost on bottled water, home

treatment (aquaguards etc...), and result in increased life of pipelines and household fixtures such as water heaters.

In order to do this it will require significant new investments. This new investment is expensive and the cost would

have to be recovered through the water tariff.

1. We do not yet know exactly how much this investment would cost but I would like to know, if you agree in

principle with paying for an improved water supply. If this project is implemented and if piped water quantity

for water is sufficiently increased gradually to 24 hours supply per day, it would result in higher tariff rates.

Would you want this project to be implemented?

(a) Yes (b) No

2. If ‘No’ give reasons:

(a) Cannot afford to pay

(b) Present supply is adequate to our need and demand is

already met from our installations.

(c) I believe the current tariff is already high.

(a) Other (please specify)

3. Now I am going to read you some costs per kilo liter (or 1000 liters) that the proposed water supply might cost

you. Please tell me if you would be willing to pay this amount for the improved services. (Start bidding game)

(a)

I would be willing to pay Rs 20 for 1 kilolitre of reliable water that meets WHO norms?

If no then proceed to (c). If yes, proceed to (4).

(b)

I would be willing to pay Rs 16 for 1 kilolitre of reliable water that meets WHO norms?

If no then proceed to (d). If yes, proceed to (4).

(c) I would be willing to pay Rs 12 for 1 kilolitre of reliable water that meets WHO norms?

If no then proceed to (e). If yes, proceed to (4).

(d) I would be willing to pay Rs 8 for 1 kilolitre of reliable water that meets WHO norms?

If no then state how much you are willing to pay_Rs____________________and proceed to (4).

4. What is the maximum number of rupees per month as WATER BILL you would be willing to pay for reliable

potable water that meets WHO norms? Rs__________


Scenario 3 [SINGLE SUPPLY NON POTABLE WITH HOME TREATMENT]

Suppose the Water Utility supplies only non-potable water that would be acceptable up to bathing standards and for

all purposes other than drinking and cooking. For drinking and cooking, you would need to treat it on your own by

installing home-treatment appliances. Generally, such home-treatment appliances could be either Ultraviolet-based

(UV) or based on reverse osmosis (RO). The cost of RO purifiers range from Rs. 15,000 – 20,000 and UV purifiers

are available anywhere between Rs. 6000 – 8000. The cost of water from a UV treatment system works out to be

around 2 paise per litre (or Rs. 20 per kL), whereas it would be around 60 paise per litre (or Rs. 600 per 1000 litres)

from that of reverse osmosis unit. The RO-based purifier uses pressure to force water molecules through a

membrane making water chemically and microbiologically potable as well as removing suspended particles,

pesticides, heavy metal contaminants and other brackish water problems. The UV purifier provides bacteria-free

drinking water. UV, by itself, does not remove any particulate matter or turbidity or heavy metal contaminants. It does

not remove volatile organic compounds such as pesticides or insecticides. No form of carbon filter removes bacteria.

In fact under quite normal operating condition all carbon forms can and do become perfect breeding grounds for

bacteria, including pathogenic bacteria. Given that, the Water Company will provide only non-potable water and

consequently you will have to treat it on-site for potable purposes, do you think you would accept non-potable water

and install water purifiers?

Yes [ ] No [ ]

1. If ‘Yes’ how much do you think you would be willing to pay for installing such home treatment systems?

(a) Rs. 6000 – 8000

(b) Rs. 8000 – 12,000

(c) Rs. 12,000 – 15,000

(d) Rs. 15,000 – 20,000

(e) Specify----------------

2. Now I am going to read you some costs per kilo liter (1000 liters) of non-potable water that the

proposed water supply might cost you. How much are you willing to pay for non-potable water that is

supplied from the Water Utility?

a) I would be willing to pay Rs 12 for 1 kilolitre of non-potable water? If no then proceed to (b). If yes,

proceed to (3).

b) I would be willing to pay Rs 8 for 1 kilolitre of non-potable water? If no then proceed to (c). If yes,

proceed to (3).

c) I would be willing to pay Rs 4 for 1 kilolitre of non-potable water? If no then proceed to (d). If yes,

proceed to (3).

d) I would be willing to pay Rs 2 for 1 kilolitre of non-potable water? If no then proceed to (3)

3. What is the maximum number of rupees per month you would be willing to pay as WATER BILL for non-

potable water? Rs__________

10.1177/1070496505281841THE JOURNALOF ENVIRONMENT & DEVELOPMENTDutta et al. / SUPPORT FOR WATER SUPPLY IMPROVEMENTS

Public Support for Water Supply Improvements:Empirical Evidence From

Unplanned Settlements of Delhi, India

VENKATESH DUTTA

SUBHASH CHANDER

LEENA SRIVASTAVA

The unplanned sector of India’s capital city has an enormous backlog in theprovision of reliable water supplies to its population, which is further exacer-bated by the growing number of informal urban settlements. In this context,contingent valuation method (CVM) is applied to evaluate a policy of provid-ing better water supply with improved quality and reliability in unplannedsettlements of Delhi. Willingness to pay (WTP) questions are used to value aspecific outcome of a policy intended to assure a reliable water supply that hasno health risk of contamination. The estimation from linear utility modelsassert that the proposed changes would provide positive net benefits to custom-ers who are otherwise incurring considerable amounts of “coping cost” in theabsence of a reliable water supply. The findings have important policy implica-tions for gauging public support for water supply improvements in infra-structurally disadvantaged households.

Keywords: urban development; water supply; unplanned settlements; copingcost; willingness to pay

In most South Asian cities today, public sector agencies are strugglingto provide adequate water supplies to their customers. Municipal bod-ies, supposedly in charge of service delivery, are caught in a downwardspiral of disrepair and appear to be in no position to meet increasingdemand or maintain the standards. Inefficiency, low investment levels, alack of financial viability, and an absence of customer orientation meanthat customers face an inadequate and unreliable supply of low-qualitywater. This results in huge coping costs and poor customer satisfaction,resulting in low willingness to pay (WTP). Policy makers frequently

439

AUTHORS’ NOTE: The fieldwork for this study was financially supported by the Waterand Sanitation Program (WSP)—South Asia. The first author gratefully acknowledgesUniversity Grants Commission (UGC), Government of India for the award of research fel-lowship under JRF scheme to carry out his doctoral work at the Faculty of Policy and Plan-ning, TERI School of Advanced Studies, New Delhi, India. Thanks are due to Dr. PuneetChitkara for his critical review and guidance. Comments received from three anonymousreferees helped to improve the article considerably.

The Journal of Environment & Development, Vol. 14, No. 4, December 2005 439-462DOI: 10.1177/1070496505281841© 2005 Sage Publications

refrain from raising water tariffs, fearing that people living in infra-structurally disadvantaged unplanned settlements will not pay. Credi-ble estimates of WTP and other means of assessing demand could beused to demonstrate that they are already paying much more than theofficial tariff rate through informal channels and coping strategies andthat they would be willing to pay even more to secure better services(Zerah, 2000, p. 168).

This article examines how much money people in unplanned areasare willing to pay to support a policy that provides them with a betterand reliable water supply. The alternate scenario emphasizes twoaspects of the proposed improved services: (a) longer hours of servicethat would gradually move from intermittent to continuous scheme and(b) good quality water with no health risk of contamination. Policyimplications are subsequently discussed, keeping in mind the cost ofprovision of water supplies. Also included is the assessment of the costof unreliable supply (coping cost), which households have to bear in theabsence of a reliable water supply.

The results of the study provide evidence from unplanned settle-ments that households’ willingness to pay for improved water servicesis much higher than their current water bills. The findings have signifi-cant policy implications for gauging public support for the ongoing per-formance improvement plan, which aims at universal, clean, potablewater for an increased number of hours, gradually shifting toward con-tinuous and more equitable water supply.

The Context:Unplanned Settlements

The National Capital Territory of Delhi (NCTD) is the fastest-growingIndian metropolis with a population of 14 million and a decadal growthrate of 46.31% between 1991 and 2001 against a national average of21.34% (Office of the Registrar General of India, 2001). Of the many chal-lenges facing the city, the most important is that of spiraling andunplanned population growth, which has brought to the forefront arelated problem of provision of reliable water supply. About half the cityis still outside the planning framework. At present, three fourths of thetotal population lives in conditions similar to that of slums. About 1.4million people live in unauthorized colonies, another 1.3 million inJhuggi-Jhopari (JJ)1 clusters, nearly 1.2 million in resettlement colonies,0.15 million in urban villages, and nearly 0.5 million in rural villages.

440 THE JOURNAL OF ENVIRONMENT & DEVELOPMENT

1. Jhuggi Jhopari (JJ) clusters are squatter settlements (slums), mostly on governmentland. Unauthorized colonies comprise settlements that come up on land initially not meantfor human habitation. As such, these do not have any service provision from the municipal

Every year about 0.3 million migrants settle into slums and resettlementcolonies. This type of upsurge in population exerts tremendous pressureon the city’s infrastructure and urban utilities such as water supply. Thenew Master Plan—2021 (MPD 2021) seeks to make Delhi free from slumsand unplanned colonies by bringing them at the center stage of theurban development process. However, without the provision of in-house piped water supply in such areas, such a vision would not becomea reality.

Dutta et al. / SUPPORT FOR WATER SUPPLY IMPROVEMENTS 441

Table 1A Glimpse of Water Service Provisions in Delhi (2003 to 2004 Level)

No. of % ofAccess to Water Sources Households (HHs) HHs

Piped water supply 1,924,140 75.33Hand pumps/tubewells 559,518 21.91Well 1,019 0.04Other sources (river/canal/tanks) 69,472 2.72

Service Provision in the No. of % ofUnplanned Settlements Colonies Colonies

Urban villages covered with pipedwater supply 135 100.00

Urban villages covered with sewer facility 93 68.88Regularized-unauthorized colonies covered

with piped water supply 557 98.23Regularized-unauthorized colonies covered

with sewer facilities 458 80.77Resettlement colonies covered with piped

water supply 44 100.00Resettlement colonies covered with sewer

facilities 44 100.00Unauthorized colonies covered with piped

water supply 229 21.38In-house toilet facility available (no. of

households) 1,991,209 77.96

Source: Planning Department, Government of Delhi (2004).

agencies. However, because of administrative and political reasons, these colonies are con-sidered for periodic in situ regularization, keeping in view parameters like environmentand land use, which become eligible for individual level water and sewer connections. Vil-lages where agricultural land is taken up for expansion of the city and thereby become apart of urban Delhi are known as “urban villages.” Households located in these areas areeligible for municipal services. There are 1,700-odd unauthorized colonies and 1,100 slumsettlements; however, their numbers are not static and keep varying. According to the newmaster plan, despite about 130,000 households and Jhuggis having already been relocated,there are still around 300,000 eligible squatter families who need to be relocated.

Piped supplies in unplanned settlements are made available eitherthrough individual household connections or public hydrants (see Table1). Tankers (portable tankers) are also used to supply water by the publicwater utility called the Delhi Jal Board (DJB). The DJB meets the waterdemands of 14 million people with 1.5 million connections, having atotal pipeline network of close to 9,500 kilometers. Water supplied by theDJB is supposed to be safe under normal circumstances; however, con-tamination is common where joints/valves lie close to sewers. Problemsalso arise in areas where water pipes have become old and corroded. Alarge number of inhabitants living in slums, unauthorized colonies, andsquatters depend on shallow hand pumps’ water, which in most cases, iscontaminated and thus unfit for drinking. Discontinuous water supplywith quality limitations gives rise to several deficiencies, resulting in anunwillingness to pay for substandard services.

Several residents are outside the piped network. Without connectionsto the water supply system, the urban poor typically have to buy water,often of dubious quality, from water vendors that charge anywhere from5% to 2,500% more per liter than what a customer connected to the mainsupply would pay. Often, this water comes after hours of waiting,arrives at inconvenient hours, and at designated places from which thehousehold supply needs to be carried home. Because of the generallypoor quality of water, the urban poor suffer the debilitating effects ofwater-borne diseases like diarrhea, gastroenteritis, and cholera, thusrequiring them to spend on medical treatment, causing their children tomiss school or causing the adults to lose income through missed work-ing days. The consequences of unreliable water supply (in terms of lostproductivity and negative health impacts) result in huge direct and indi-rect costs. This potential source of revenue is not captured by the formalproviders, but instead it is paid directly to unregulated, small-scale pri-vate sectors such as informal water markets, water vendors, and the like.

Problems also arise because of the legal land tenure status. Many resi-dents lack legal land tenure, and the DJB generally refrains from supply-ing services to settlements that lack legal land tenure mainly because it isseen as the political acceptance of the settlement as a bona fide place ofresidence. This is the case even where policies exist, which state thatwater and sanitation should be made available to all citizens (MPD2021).

Review of Literature

The increased use of contingent valuation method (CVM) in develop-ing countries has led to a growing confidence in its use as a tool for theassessment of effective demand for improved water supplies. It must be


recognized at the outset that there is no one best method for evaluatingthe consequences of a particular policy choice or alternative on humanwell-being. However, in conducting demand assessment studies in adeveloping country context, including India, this method continues tobe extensively used by researchers (Ahmad, Goldar, Misra, & Jakariya,2003; Bateman & Willis, 1999; Choe, Varley, & Bijlani, 1996; Dasgupta &Dasgupta, 2004; Goldblatt, 1999; Griffin, Briscoe, Singh, Ramasubban, &Bhatia, 1995; Ready, Malzubris, & Senkane, 2002; Whittington &Swarna, 1994). This technique is considered to be particularly suitablefor assessing consumer preferences for nonmarket commodities, wheremarkets are absent, imperfect, or incomplete. Imperfections generallyexist in the market for household water services and, as a result, marketprices do not reflect the true values that households place on water.

Worldwide, several studies have been carried out in a number of low-income countries to determine affordability and WTP for water and san-itation services (Ahmad, Goldar, & Misra, 2005; Altaf, Whittington,Jamal, & Smith, 1993; Briscoe, 1993; Crane, 1994; Davis, 2004; Griffinet al., 1995; McPhail, 1993; Murty, James, & Misra, 1998; Ntengwe, 2004;Whittington & Lauria, 1991; Whittington, Pattanayak, Jui-Chen Yang, &Bal Kumar, 2002; Whittington, Xinming, & Robert, 1990).2 These studiessuccessfully implemented CVM to assess WTP for improved supplyconditions and the efficacy of government policies on water supply. Thestudies affirm that despite the poverty prevailing in the low-incomecountries, many households are able and willing to pay for water andsanitation services if the public water utilities are run along commerciallines. These studies also show that demand-side information abouthousehold preferences and priorities, which have traditionally beenneglected, can provide a valuable input into the planning process.

Across India, a number of studies, coupled with practical experienceon the ground, have shown that many urban and rural communities arewilling to pay for water and sanitation services. A 1993 study in ruralKerala (Singh et al., 1993) used CVM to test the sensitivity of householdsto the monthly tariff for water from a yard tap compared to the house-hold connection with improved quality. The study found that the realconstraint to providing household connections was not the high cost butthe limited availability of local credit. A study by Vaidya (1995) inBaroda found that about 85% of households without household connec-tion expressed WTP for improved standpost service. Among house-holds with individual connections, about 63% were found to be willingto pay for better pressure and 11% for better quality. Nearly 80% of thosewith household connections were willing to pay more even if servicewas not improved, as much as 3 times the current municipal rate.Approximately 58% of households were willing to pay a one-time con-


2. See Merrett (2002) for detailed critique of the literature that has applied contingentvaluation to the WTP for water services in low-income countries.

nection charge. However, the share of income that households were will-ing to pay for water declined as income levels increased, with WTP forthe highest-income group being only 60% higher than that of the lowest-income group. This suggested limited scope for cross-subsidizationacross income groups. Similarly, a study in Dehradun in 1996 showedthat a high percentage of households were willing to pay more than thecurrent tariff for an improved water service (Choe et al., 1996). InMumbai, one of the metropolitan cities in India, Raje, Dhobe, andDeshpande (2002) described the impact of various factors on WTPthrough a logistic regression analysis. Nearly 50% of the respondentswere ready to pay partially more than their current bill amounts. Thus,all four studies revealed that households were willing to pay amountssignificantly higher than the current tariffs for improved water services.

Theoretical Construct and Random Utility ModelsUsed for the Study

The starting point for the theoretical construct on modeling utilityfrom improved water supply can be succinctly stated: True economicvalues are unobservable. However, when some economic parameterchanges, we ask how much compensation, paid or received, wouldmake economic agents indifferent about the change. The prechange orpostchange level of welfare is used as a baseline for valuation.3


Table 2Explanatory Variables Used in Estimation of

Willingness to Pay for Improved Water Supply

Mean StandardVariables Value Deviation

Education: Respondent’s years of education 10.88 4.36HH Size: Size of the household 6.26 2.96Awareness: Households’ awareness about water

supply and reliabilitya 2.81 0.77Quantity: Dummy for quantity of supplied waterb 0.30 0.46GWater: Dummy for private groundwater well as

alternate sourcec 0.28 0.44Storage: Actual volume of daily storage in liters 584.21 640.74Mon_Bill: Household’s current water bill in

Rs per month 55.75 38.85Pay_Inc: Increment to water bill in Rs per month 104.81 78.10

a. On a scale from 4 (fully aware) to 1 (ignorant).b. 1 = adequate, 0 = otherwise.c. 1 = yes, 0 = otherwise.

3. For details on random utility models, please see basic econometrics textbooks such asGreen (2000), Gujarati (1995), and Maddala (1982).

When an individual faces a change in a nonmarket or environmentalgood of a measurable attribute—for example, water quality indicator qfrom q0 to q1 (with q1 > q0), the indirect utility function u after the changebecomes higher than the status quo. u0j = ui(yj zj q0 ε0j) represents the util-ity for the status quo and u1j = ui(yj, zj, q1, ε1j) represents utility in the finalstate. Here, yj is the jth respondent’s discretionary income, zj is an m-dimensional vector of socioeconomic variables and attributes of choice,and εj is the error component known to the individual respondent butnot observed by the researcher.

Based on this utility function, respondent j answers yes to a requiredpayment of tj if the utility with the improved scenario exceeds utility ofthe status quo.

u y t z u y zi j j j j j j j( , , ) ( , , )− >ε ε1 0 0 . (1)

However, researchers do not know the random part of preferencesand can only make probability statements about yes and no. The proba-bility of a yes response is the probability that the respondent thinks he isbetter off in the proposed scenario, even with the required payment, sothat ui > u0. For respondent j, this probability is

[ ] [ ]Pr( ) Pr ( , , ) ( , , )yes u y t z u y zj i j j j j j j j= − >ε ε1 0 0 . (2)

This probability statement provides an intuitive basis for analyzingresponses and can be used as the starting point for estimating the wel-fare from an alternate scenario of supplying the customers with water ofbetter quality and reliability. But it is too general for parametric estima-tion of WTP. Two modeling decisions are needed. First, the functionalform of utility ui(yj, zj, εij) must be correctly chosen. Second, the distribu-tion of the εij must be specified. Virtually all approaches begin by speci-fying the utility function as additively separable in deterministic andstochastic preferences:

u y z u y zi j j ij i i j ij( , , ) ( , )ε ε= + . (3)

The linear utility function results when the deterministic part of thepreference function is linear in income and covariates:

u y z yij j i j i j( ) ( )= +α β (4)


where yj is discretionary income, zj is an m-dimensional vector of vari-ables related to respondent j and αi an m-dimensional vector of parame-

ters, so that α αi j i k jkk

mz z=

=∑

1

. A CV question induces the respondent to

choose between the proposed conditions at the required payment t, andthe current state. The deterministic utility for the proposed alternate sce-nario is

u y t z y tij j j j j j( ) ( )− = + −α β1 1 (5)

where t j is the price offered to the jth respondent. The status quo utility is

u y z yj j j j0 0 0( ) = +α β . (6)

The change in deterministic utility is

u u z y t yij j j j j j− = − + − −0 1 0 1 0( ) ( )α α β β . (7)

A reasonable assumption is that the marginal utility of income is con-stant between the two CV states, unless the proposed CV scenario pro-vides a substantial change (Haab & McConnell, 2002). Hence, β1 = β0 andthe utility difference becomes

u u z tij j j j− = +0 α β (8)

whereα α α= −1 0 andα αz zj k jkk

m=

=∑

1

. With the deterministic part of pref-

erences specified, the probability of responding yes becomes

Pr( ) Pr( )yes z tj j j j= − + >α β ε 0 . (10)

Once utility is specified as the sum of random and deterministic com-ponents, the differences in the random components between the statusquo and the CV scenario cannot be identified, and so there is no reasonnot to write the random term as εj = ε1j – ε0j. To proceed in estimating theparameters of the utility difference, it is necessary to specify the nature ofthe random terms. The assumptions that εij are independently and iden-tically distributed (IID) with mean zero describes most distributionused. Two widely used distributions are the normal (probit) and logistic


(logit). Both these distributions are symmetric, which facilitates modelestimation from econometric package like LIMDEP.

In this article, we use both normal and logistic distributions with lin-ear utility function. Here, in estimating the people’s WTP, dichotomouschoice data are statistically modeled by fitting a probability function andthen integrating for the computation of the welfare measures repre-sented by the mean and the median WTP. The dependent variable is adiscrete variable taking the value of 1 if households were willing to paythe operation and management (O & M) cost of infrastructure provisionrelated to continuous water supply with better quality, or 0 if they werenot willing to pay. To better understand the determinants of respon-dents’ preferences for improved water supply services and to seewhether these determinants are consistent with economic demand the-ory, a series of multivariate regression analyses were performed with thesurvey data. Theory and intuition suggest that preferences for improvedwater supply and WTP would differ across population groups with dif-ferent sociodemographic characteristics, existing water situations, andopinions about water quality and public policy. Accordingly, in themodel, respondent’s awareness, education, household size, storage, andgroundwater usage are taken as independent variables (see Table 2).Income is not taken as explanatory variable in the respondent’s utilityfunction as marginal utility of income is constant between the two CVscenario sets.

Thus the probability of yes for respondent j can be estimated as

Pr( ) Pr( ( ) )

Pr( ( )

α β ε α β ε

α β

z t z

z tj j j j t j j

j j

− + > = − − <

= − − −

0

1 >

= < −

ε

ε α βj

j j jz t

)

Pr( ).

(11)

The last equality exploits the symmetry of the distribution. For sym-metric distributions F(χ) = 1 – F(–χ). Suppose that ε σj N~ ( , )0 2 . It is nec-

essary to convert ε σj N~ ( , )0 2 to a standard normal N(0,1) variable. Let

θ ε σ= / , then θ ~ ( , )N 0 1 and

Pr( ) Pr( )ε α β θα

σβσ

α

σβσ

j j jj

j

jj

z tZ

t

zt

< − = < −

= −⎛

⎝⎜

⎞

⎠⎟Φ

(12)

where Φ(χ) is the cumulative standard normal, i.e., the probability that aunit normal variate is less than or equal to x. This is the probit model. The


probit model is based on a latent regression model in which the distur-bances are assumed to have a normal distribution. The normal distribu-tion for the difference ε = ε1 – ε0 would result if ε1 and ε0 are each inde-pendent normal. If εj is assumed to be logistically distributed, it hasmean zero and unknown variance π σ2 2 3L / . Normalizing byσ L creates a

logistic variable with mean zero and variance π 2 3/ . The probability thata variate distributed as a standard logit is less than or equal to χ equals( exp) ( ))1 1+ − −x . Then the probability that respondent j answers yes is

Pr( ) [ exp( ( / / ))]yes z tj j L j L= + − − −1 1α σ β σ . (13)

This is the logit model. This can be constructed as a random utilitymodel in which it is assumed that the random parts of the utility func-tions are distributed as independent extreme value. The logistic can bederived as the difference of two extreme value distributions. Maximumlikelihood estimation method is used to estimate the parameters. Sup-pose the sample size is T and let Ij = 1 if respondent j answers yes. Thelikelihood function becomes

L y z tZ t Zj j

j

TI j

j( , / , , )α βα

σ

β

σ

α

σ= −

⎛

⎝⎜

⎞

⎠⎟

⎡

⎣⎢

⎤

⎦⎥ − −

=∏ Φ Φ

1

1β

σ

t jI j⎛

⎝⎜

⎞

⎠⎟

⎡

⎣⎢

⎤

⎦⎥

−1 (14)

for the probit and when the logit is estimated

[ exp( ( / / ))]1 1+ − − −α σ β σz tjj L L (15)

is substituted forα

σ

β

σ

Z tj j−⎛

⎝⎜

⎞

⎠⎟. Maximum likelihood routines use the log

of the likelihood function to calculate the maximum likelihoodestimates.

THE CALCULATION OF MEAN AND MEDIAN WTP

For the linear random utility model defined above, WTP can be de-fined as

α β ε α β ε1 1 0 0z y WTP z yj j j j j j j+ − + = + +( ) . (16)


Solving this equation for WTP yields

E WTP zj jε α β( ) /= . (17)

To solve for the median WTP, we find the WTP that solves the expres-sion that the probability that u1 > u0 = 0.5.

[ ]Pr ( ( )) .

Pr[ (

α β ε α β εε

ε

1 1 0 0 0 5z y M WTP z y

Md Wj j j j j j j+ − + > + + =

TP zj j j) / / ] . .> + =α β ε β 0 5

(18)

Because ε is symmetric with mean zero, this expression yields

Md WTP zj jε α β( ) /= . (19)

In the case of a linear utility function and a symmetric mean zeroerror, the mean and median WTP with respect to random preferences areequal. We introduce the notation Εe and Mde for the mean and the medianof the random preference term to emphasize the different sources of theuncertainty.

Sample Selection and Design

Three separate urban local governments with nonoverlapping juris-dictions serve the city of Delhi, covering a total area of 1,486 sq km. Thelargest of these is the Municipal Corporation of Delhi (MCD), which hasmore than 94% of the area (1,400 sq km) under its jurisdiction. The NewDelhi Municipal Council (NDMC) and the Delhi Cantonment Board(DCB) control about 86 sq km. The MCD is among the largest municipalbodies in the world, providing civic services to more than 13.78 millioncitizens in the capital city. It is next only to Tokyo in terms of area. Withinits jurisdiction are some of the most densely populated areas in theworld. The entire MCD area is divided into nine districts having 12zones. These 12 zones are further divided into 134 wards (see Figure 1).Each ward has residential colonies/enclaves that come under severalcategories such as rural and urban villages, resettlement colonies, regu-larized unauthorized colonies, squatter settlements, slums, and so on.Municipal Valuation Committee (MVC) of MCD released category-wiseclassification of residential colonies into eight classes in April 2004 forproperty tax calculation under which households typically belong.These classes represent fairly well the level of affluence as well as the


level of planning. As we go down from A to H, the level of affluence,quality-of-life, and planning levels decreases. Water supply situation isalso characteristically divergent in these areas. Class A is highest in thestrata, with high level of planning having households under high-income group followed by B, C, and D. The unplanned dwelling unitsmostly come under E, F, G, and H classes, which may be furtherclassified into JJ clusters, squatter settlements, rural and urban villages,unauthorized colonies, and so on.

For the present study, unplanned residential areas that come under E,F, G, and H classes inside MCD jurisdiction are taken for sampling. Torepresent the entire population multistage stratified random sampling isfollowed. In the first stage, the Census and Economic Survey reports areused to identify the entire population as distributed by census or plan-ning blocks. In the second stage, to geographically spread the samplingunits encompassing different socioeconomic backgrounds and watersupply situations, E, F, G, and H classes are taken as the sampling framefrom which unplanned colonies are selected. Colonies in respective


National Capital Territory of Delhi (1,486 sq km)

Municipal Corporation of Delhi (MCD) 1,400 sq km

New Delhi Municipal Corporation (NDMC) 44 sq km

Delhi Cantonment Board (DCB) 42 sq km

9 Districts

12 Zones

134 Wards

Planned Units

Unplanned Units

Planned-Private

Planned-DDA

J J Clusters

Slum Designated Areas

Unauthorised Colonies

Resettlement Colonies

Regularised Unauthorised Colonies

Urban/Rural Villages

A

B

C

D

E

F

G

H

Figure 1: Classification of Residential Units Under MCD Jurisdiction Based on Typeof Settlements

strata are selected randomly depending on their property class, plan-ning status, and level of water stress. During the survey in a particularresidential enclave, numbers of households are divided into four catego-ries of water situations—good, average, below average, and bad basedon information available with the local area councilors and ResidenceWelfare Associations (RWAs). Secondary literature mainly availablewith the municipal water authority is also consulted to delineate trou-ble areas within a colony prior to actual sampling. The distribution of thesampled households in a particular colony followed the same ratio. Fordeciding about the stratum weight, their area-extent was taken as theirproportions, which may pose some unavoidable difficulties in project-ing the survey as being typical of the unplanned areas of the whole city.4

Despite this difficulty, utmost care is taken in selection of the colonies toencompass the ingrained heterogeneity typical of the metropolitan city.If a household refused to answer questions in a particular category, thenext unit in the same category in that particular block replaced it.Because probability-proportional-to-size sampling depends on the sizeof the population, some wards had more than one cluster in the finalsample.

Questionnaire Design andHousehold Survey

The design of the household questionnaire was based on researchpapers in contingent valuation studies (Whittington, McClelland, &Davis, 1998) and available guidelines on the subject such as Wedgwoodand Sanson (2003), MacDonald and Young (2002), the World Bank’s Liv-ing Standard Measurement Survey manual (Grosh & Glewwe, 2000),DLTR’s (Department for Transport, Local Government and the Regions)guidelines on WTP survey (Pearce, Ozdemiroglu, et al., 2002), the ADBtraining manual on water sector, and the Indian Water Works Associa-tion (2002) Manual on Water Demand Assessment for Urban Water SupplyProjects. The final questionnaire for the present study consisted of thefollowing four sections: (a) a section on socioeconomic characteristics,(b) a preliminary attitudinal section, (c) a section on water supply situa-tion, and (d) a valuation section.

The first section dealt with basic income-expenditure, demographic,educational, and occupational data for the respondents. The second sec-tion included some preliminary attitudinal questions keeping in mindthe hypothesis that attitude and awareness of the residents influenced


4. The number of dwelling units in each of the stratum cannot be arrived accurately asMCD population data is available at the ward level and a single ward may include morethan two strata.

their water use behavior and hence willingness to pay. The third sectionconsisted of questions on prevailing water supply services, respon-dents’ consumption pattern, money spent on availing reliable watersupply services, and so forth. The information gathered from this sectiongave us an idea of the wide range of averting expenditures incurred bythe consumers. And, finally, the fourth section dealt with valuationquestions eliciting respondents’ willingness to pay for improved watersupply.

The purpose of the valuation section was to obtain an estimate of thevalue of urban water supply improvements. In particular, the enumera-tor emphasized two aspects of the proposed improved services: longerhours of service that would gradually move from intermittent to contin-uous scheme and good quality water with no health risk of contamina-tion. The enumerator described the demerits of the status quo situationand clearly explained benefits of reform project. To avoid any hypotheti-cal bias about the provision of continuous supply, the statements usedwere

It is not that as soon as the work starts, water supply would be there con-tinuously, but it is that the efficiency level and quality of services wouldimprove gradually, and over a period of time clean potable water would beavailable for increased number of hours. The objective of the performanceimprovement program is to improve water availability in your area and agradual shift toward continuous and more equitable water supply thor-ough rehabilitation of existing infrastructure and establishment of newsupply lines.

Having presented a scenario and policy, the respondent was then facedby a set of valuation questions. First, the respondent was asked a closed-ended, yes-no type question to support a policy change as per the choicegiven and then faced by a WTP question, provided the response to thiswas yes. The elicitation format used in this study was a split biddinggame, with different subgroups facing starting point in increasing ordecreasing order—the starting bid being 3 times higher or lower than theactual cost. The bidding game was followed by a final open-ended ques-tion on households’ maximum willingness to pay. The valuation sectionwas finally concluded by the respondent’s belief in the likelihood ofimproved water supply attributes and reasons for not willing to pay,if any.

Prior to designing the survey, a series of exploratory, qualitativegroup discussions were conducted with residential customers and sur-vey enumerators, each exploring customer’s perceptions and experi-ences of water services. The focus group was comprised of researchers,statisticians, students, and local representatives agreed on a sample sizeof about 100 to properly reflect the distribution of households in differ-


ent income brackets and geographic locations. Four focus groups wereconducted. The information that was obtained during the focus groupswas used to design the questionnaire, including which service attributeswere included in the experiments, how the attributes were described,and the levels that each attribute could take. It was decided in the begin-ning that the only way reasonable WTP information for improved waterquality could be obtained at the household level was through a per-sonal, door-to-door interview. A total of 650 interviews were completed,spread over various unplanned settlements. Few of them were rejectedbecause of nonresponse and missing data.

The survey administration included background briefing of respon-dents on the implications of allowing the status quo to continue and theimportance of continuous reliable supply. People were asked about theirknowledge of the current water supply situations and their current billand expenses on averting supply unreliability. Different subsamples ofconsumers were confronted with different initial bids and were askedwhether they would pay the nominated increase in the bill as the price topay for guaranteed quality.

Estimation Results

In general, continuous water supply with good quality water is per-ceived as a high priority to customers in unplanned settlements. In addi-tion to the average monthly water bill of Rs 56, a domestic customerin unplanned areas is willing to pay Rs 101 (average of logit and probitestimates) extra for a reliable water supply. This is consistent with theresults of a previous research on the valuation of a reliable water supply(MacDonald, Barnes, Bennett, Morrison, & Young, in press). The loglikelihood ratio (pseudo R2) of 0.46 (probit) and 0.50 (logit) indicates thatthe main effects model is a good fit (see Table 3). Education and aware-ness of the respondent is positively correlated with his WTP, implyingthat there is a positive relationship between a respondent’s WTP and hisnumber of education years as well as his level of awareness. Respon-dents with personal groundwater borewells are less likely to pay morethan the existing amount for an improved water supply. This is becausethey have already invested a significant amount in securing a reliablewater supply. The coefficient of current monthly water bill is negativelysigned—respondents getting higher water bills appear to be less likelyto increase their payments for improvements in water supply. Alterna-tively, respondents who are getting lower monthly bills are more likelyto choose the nonstatus-quo option. The negative coefficients for house-holds who have developed a large water storage capacity shows thatthey have already invested a significant amount of money in storing


water to avoid unreliable water supply; therefore, their additional WTPfor an improved water supply system is actually lower. Respondent’shousehold size is a significant variable, with a negative coefficientmeaning larger household sizes would probably result in less likelihoodof paying more for an improved water supply. Similarly, the coefficientfor quantity of daily supply is negatively signed, but it is not signifi-cantly higher than the 90% confidence limit. In this context, a largemajority of the customers receive water between 3 to 4 hours daily, and itcan be inferred that the level of satisfaction among customers in un-planned areas is very low.


Table 3Results From the Estimation of Willingness to Pay Through Probit and

Logit Model Using Maximum Likelihood Estimates

Pr _θα α αα αα

<+ + +

+ +0 1 2

3 4

5

EDU GWATER

MON BILL STORAGE

HHSIZE QUAN AWR t

a

+ + −

⎛

⎝

⎜⎜⎜

⎞

⎠

⎟⎟⎟

⎡

⎣

⎢⎢⎢

⎤

⎦

⎥⎥⎥α α β

σ

6 7

/

Probit Model Logit ModelParameter Parameter

Variables Est. t Ratio Est. t Ratio

Constant –1.7844 –5.04 –3.9838 –5.79Education 0.0450 2.57 0.0497 1.52b

GWater –0.6216 –3.72 –1.1702 –3.71Mon_Bill –0.0003 –1.88b –0.0005 –1.63b

Storage –0.0001 –1.91 0.0572 –2.81HH Size –0.1891 –6.66 –0.0002 –6.28Quantity –0.2675 –1.60b –0.3457 –1.49b

Awareness 0.0918 1.03b –0.4620 1.34b

Inc_Billc –0.0245 12.56 –0.05723 10.96Log likelihood (LogL) –226.50 –209.07Restricted log likelihood (LogL0) –418.65 –418.65Estrella (1–(L/L0)^(–2L0/n) 0.57 0.61McFadden (pseudo R2) 0.46 0.50Mean WTP (increment to

monthly bill) Rs 104.53 Rs 98.46

Note: See Table 2 for a definition of each variable.a. Loomis, Kent, Strange, Fausch, and Covich (2000) have adopted a similar modelingapproach from a CV study in Colorado for the water use in the South Platte River. It is alsodiscussed in Haab and McConnell (2002, chapter 2, p. 31).b. Denotes acceptability below 95% confidence limits.c. Note that this is the parameter on –t.

Descriptive Statistics

Access to piped water supply. In the sampled unplanned colonies,28.92% of the total households had no direct piped water supply, andthey relied on alternate sources for their daily water requirements. Thisfigure is surprisingly higher, as an earlier report (Delhi Urban Environ-ment and Infrastructure Improvement Project, 2001) that explored ineq-uity in water supply observed that 10% of Delhi’s population had nopiped water supply at all and 30% had grossly inadequate access. Butthis report was based on the entire population of Delhi, covering bothplanned and unplanned areas. Statistics of coverage of about 90% andbare figures of quantity of water supplied in the cities as claimed byauthorities tend to hide several realities regarding both the operations ofthe system and the experience of customers. The coverage figure mayrelate to installed capacity. What is relevant for the customers is theactual operating capacity of the water supply system, or the averageactual supply through in-house connections during a sustained period.The figure of 28.92% could be more realistic as population in the un-planned areas has increased rapidly in the past 5 years without the pro-vision of in-house piped water supply. Absence of safe and reliablewater supply in such areas could be a serious bottleneck, which mayfurther hinder the urban growth and in situ development.

Level of consumption. According to the survey average, daily con-sumption in the unplanned areas was 420 liter per household or 67 litersper capita per day (lpcd). An earlier study (Barah, Sipahimalani, & Dhar,1998) reported average per capita consumption in Delhi from 313 lpcd inaffluent households to 140 lpcd in relatively less-well-off householdsand a meager 16 lpcd for the slum households. A more recent study(Wolf & Kraft, 2002) reported widespread variation in per capita waterconsumption that ranged from 17 lpcd in slums to 646 lpcd in affluentareas. These studies show that the per-capita average consumption inunplanned areas is much less. The availability of less water to them islargely because of discontinuous water supply and limited availabilityof in-house piped water connections.

Metering and billing information. Metering on actual consumption inunplanned areas is very low. Fewer than 50% of the households reporteduse of water meters within their premises, out of which 31.65% haveworking water meters and billing is based on actual consumption. As awhole, billing hardly reflects actual water consumption—8.27% ofhouseholds don’t pay at all and 31.47% of the population pay by “flatand fixed tariff” irrespective of the consumption (minimum charge = Rs30 per month). To realize the customers’ WTP, increased level of water


metering is essential. If water utilities are to attain sustainability in thelong term, they will have to embark on developing and applying theright water tariff based on volumetric consumption through increasedlevel of water metering both in planned and unplanned areas.

Cost of unreliable supply. Because of the water agency’s inability to pro-vide efficient and reliable supply, customers spend a significant amountof money to make the supply reliable in terms of both quality and quan-


31.47 31.65

8.27

28.21

0

5

10

15

20

25

30

35

Flat and Fixed Tariff Metered Tariff Don't Pay No municipalconnection

Per

cent

age

Figure 2: Extent of Billing in Unplanned Sector

49.9 51.1

18.53

31.46

0

10

20

30

40

50

60

MeteredConnections

UnmeterdConnections

Meters not working Meters working

Figure 3: Extent of Metering in Unplanned Sector

tity (see Table 4).5 An increasing proportion of urban customers arealready making their own investments to simulate 24-hour, 7-day-a-week water supply at the household level—borewells, surface and over-head storage tanks, booster pumps, tankers suppliers, and so forth.These investments are supplemented with water purification methodssuch as filtration and boiling. Poor people bear a disproportionate shareof the impact of inefficient water. Customer inconvenience results in lossof household income or productive time as at least one family memberhas to cope with securing reliable water on a daily basis. Typically, theresidents need to supplement public supplies with water obtained fromprivate sources, and this is usually much more expensive. The extent towhich this is done—the proportion of water procured privately againstthe quantity supplied is determined by individual households demandand socioeconomic characteristics. Several households use multipleaverting measures, for example, private tankers and communitystandposts for sourcing water and boiling for purification. Thus, the lackof access to a reliable, adequate, and safe water supply impacts directly


Table 4Cost of Unreliable Supply Borne by People in Unplanned Settlements

% of Annualized Recurring/Averting Measures Households Capital Cost O & M Cost

Storage in buckets/drums 88.23 Rs 180 Rs 25Overhead tanks 51.47 Rs 0.75/liter Rs 50Use of private tankers 4.90 NA Rs 200Use of public hand pumps 17.15 NA NAUse of community standposts 14.70 NA NAUse of bottled water 3.26 NA NAUse of ceramic filters 16.99 Rs 141.50 Rs 200Use of UV filters 12.25 Rs 1061 Rs 400Use of RO filters 0.32 Rs 2743 Rs 1500Boiling 12.74 NA Rs 532Use of boosting pumps 53.10 Rs 813.50 Rs 1015Use of borewell pumps 27.77 Rs 976 Rs 533Sickness because of diarrheal diseases 19.60 NA Rs 704Average annual coping cost Rs 3112

Note: For calculating cost on boiling, consumption of cooking gas for 25 minutes per day istaken. Considering gas consumption of 180 gm/hr and price of cooking gas cylinder (14.5kg) at Rs 282, the cost on boiling comes out to be Rs 1.46/day. O & M = operation and man-agement; NA = not applicable.

5. Unreliability cost is calculated as an annualized sum of money spent on drawingwater from alternate sources, groundwater pumpage cost, household water treatmentcost, and cost on treating waterborne illness mainly because of diarrheal diseases.

on the livelihoods and incomes of the urban poor—their ability toengage in income-generating activities, the types of livelihood activitiesthey can engage in, their incomes from these activities, and their overallcost of living.

Preference for service provider. Preference for privatized water utilityamong households is very low (8.9%), and a large number of respon-dents prefer the government-owned water agency (65.3%). However,20.4% of the respondents are indifferent to their service provider as longas sufficient quantity and quality is maintained. People fear completeprivatization, and affordability is one of the prime concerns apart fromabrupt tariff increases. Many people feel that water services are a basicright and should be provided to them regardless of whether they can paythe “full price.” Partnership between communities, local government,and service providers is important in creating models for development(such as joint public and private provision) that not only work but couldalso be sustainable in the long run.

Conclusion

The provision of efficient water systems in unplanned settlementsforms a vital underpinning to urban development and economicgrowth. There are five main common difficulties that have to be over-


8.945.20

65.37

20.49

0.00

10.00

20.00

30.00

40.00

50.00

60.00

70.00

Private serviceprovider

Joint public andprivate

Government owned Indifferent to serviceprovider

%of

HH

s

Figure 4: Household’s Preference for Service Provider—Preference for Private UtilityIs Very Low Among Unplanned Residents

come for urban communities in unplanned settlements to access betterand reliable water services. These are (a) the legal position of the resi-dents with respect to land tenure as several areas come under “unautho-rized” jurisdiction; (b) the political position—willingness of local gov-ernment to bring them in the mainstream of urban development process;(c) the locality of the settlement in which the communities are living,including distance and accessibility, because of narrow access roads anddense population (the lack of planning also limits the technology thatcan be used and also the level of service); (d) the cost recovery (construc-tion, operation, maintenance, servicing, etc.) of accessing services; and(e) political will to charge higher and efficient rates by local governments.

The fourth point (i.e., the cost of accessing services) is very much tiedup with the four other barriers outlined above. Both legality and localityof settlement will have an impact both on the cost of installing servicesand on the willingness and ability of poor communities to pay for basicservices. It has been well documented that poor households are not onlypaying more in absolute terms, but they also pay more in social costsbecause of the degraded environments in which they live, through poorhealth and lost employment. Generally, their income poverty is used asan excuse for not providing them with basic services in the belief thatthey will not be able to pay and/or not be willing to pay. However, theresults indicate that reliability of water supply with no health risk of con-tamination is of value to residential customers even in unplanned areas,and they are willing to pay higher than what they are currently payingfor better services. Currently fewer than 50% of them are willing to payhigher than the O & M cost even though the indirect cost of unreliablesupply to customers is Rs 259 per month. Reasons for this could be lowbelief in the service provider whereby respondents did not fully believethat the new system would work reliably. This also shows protestresponses whereby customers feel that they should not have to pay forimproved water. To recover a significant portion of the costs, the waterutility should begin to recognize the value of including unplanned com-munities in the network and provide them with reliable water serviceswhen it is apparent that this will increase revenue through metering andbilling.

Finally, lack of political will to commercialize water utilities andcharge a cost-based tariff is a real issue still to be addressed adequately indeveloping countries context. Because water supply is considered asocial and moral obligation, the onus of providing water has tradition-ally been thrust on the government, made available at heavily subsi-dized rates. Political interference is considered as being an obstacle tocost recovery and tariff increase even when customers are willing to payhigher rates. It shows itself particularly in “unwillingness to charge” forwater services without guaranteeing proper financing from othersources but also in other unsustainable water policies. For example, the


local government might proclaim that it should provide its people with“free” water when in fact the funding, institutional arrangements, andcapacities to provide that water service do not exist. If policy makers canestablish what customers are willing to pay, they should be able to revisetariffs to their willingness to pay, plan future investment keeping inmind what customers really want, and move toward financialsustainability and independence.

References

Ahmad, J. K., Goldar, B. N., & Misra, S. (2005). Value of arsenic-free drinking water to ruralhouseholds in Bangladesh. Journal of Environmental Management, 74, 173-185.

Ahmad, J. K., Goldar, B. N., Misra, S., & Jakariya, M. (2003). Willingness to pay for arsenic-free,safe drinking water in Bangladesh. New Delhi, India: Water and Sanitation Programme–South Asia, The World Bank.

Altaf, M.A., Whittington, D., Jamal, H., & Smith, K. (1993). Rethinking rural water supplypolicy in Punjab, Pakistan, Water Resources Research, 29, 1943-1954.

Barah, B. C., Sipahimalani, V., & Dhar, P. (1998). Urban water supply and sanitation. InU. Sankar & O. P. Mathur (Eds.), Economic instruments for environmental sustainability(pp. 53-65). New Delhi, India: National Institute of Public Finance and Policy.

Bateman, I., & Willis, K. G. (Eds.) (1999). Valuing environmental preferences: Theory and prac-tice of the contingent valuation method in the U.S., EU and developing countries. London:Oxford University Press.

Briscoe, J. (1993). When the cup is half full: Improving water and sanitation services in thedeveloping world, Environment, 35(4), 7-37.

Choe, K., Varley, R. C. G., & Bijlani, H. U. (1996). Coping with intermittent water supply: Prob-lems and prospects, Dehradun, India (Activity Report No. 26). Washington, DC: Environ-ment Health Project, USAID.

Crane, R. (1994). Water markets, market reforms and the urban poor: Results from Jakarta,Indonesia. World Development, 22(2), 71-83.

Dasgupta, P., & Dasgupta, R. (2004). Economic value of safe water for the infrastructurallydisadvantaged urban household: Acase study in Delhi, India. Water Resources Research,40(11).

Davis, J. (2004). Assessing community preferences for development projects: Arewillingness-to-pay studies robust to mode effects? World Development, 32(4), 655-672.

Delhi Urban Environment and Infrastructure Improvement Project. (2001). Detailed projectreport. Delhi: Ministry of Environment and Forest, Government of India and PlanningDepartment, Government of National Capital Territory of Delhi.

Goldblatt, M. (1999). Assessing the effective demand for improved water supplies in infor-mal settlements: A willingness to pay survey in Vlakfontein and Finetown, Johannes-burg. Geoforum, 30, 27-41.

Green, W. H. (2000). Econometric analysis (4th ed.). New York: Macmillan.Griffin, C. C., Briscoe, J., Singh, B., Ramasubban, R., & Bhatia, R. (1995). Contingent valua-

tion and actual behaviour: Predicting connection to new water systems in the state ofKerala, India. World Bank Economic Review, 9, 373-395.

Grosh, M., & Glewwe, P. (2000). Designing household survey questionnaires for developingcountries: Lessons for 15 years of the Living Standards Measurement Study. Washington, DC:World Bank.

Gujarati, D. (1995). Basic econometrics (4th ed.). New York: McGraw-Hill.


Haab, T., & McConnell, K. E. (2002). Valuing environmental and natural resources: The econo-metrics of non-market valuation. Cheltenham, UK: Edward Elgar.

Indian Water Works Association. (2002). Manual on water demand assessment for urban watersupply projects. Mumbai, India: Author.

Loomis, J. B., Kent, P., Strange, L., Fausch, K., & Covich, A. (2000). Measuring the total eco-nomic value of restoring ecosystem services in an impaired river basin. Ecological Eco-nomics, 33, 103-117.

MacDonald, D. H., Barnes, A., Bennett, J., Morrison, M., & Young, M. D. (in press). Whatconsumers value regarding water supply disruptions: A discrete choice analysis. Jour-nal of American Water Association.

MacDonald, D. H., & Young, M. (2002). Valuing customer service preferences: A methodology(CSIRO Project Report). Collingwood, Australia: Commonwealth Scientific and Indus-trial Research Organisation (CSIRO).

Maddala, G. (1982). Limited dependent and qualitative variables in econometrics. Cambridge,UK: Cambridge University Press.

McPhail, A. A. (1993). The “five percent rule” for improved water service: Can householdsafford more? World Development, 21(6), 963-973.

Merrett, S. (2002). Deconstructing household’s willingness to pay for water in low-incomecountries, Water Policy, 4, 157-172.

Murty, M. N., James, A. J., & Misra, S. (1998). Economics of industrial pollution abatement: The-ory and empirical evidence from the Indian experience. Oxford, UK: Oxford UniversityPress.

Ntengwe, F. W. (2004). The impact of consumer awareness of water sector issues on will-ingness to pay and cost recovery in Zambia (Special Issue 15-18). Physics and Chemistryof the Earth, 29, 1301-1308.

Office of the Registrar General of India, Government of India. (2001). Census of India. NewDelhi: Author.

Pearce, D., Ozdemiroglu, E., Bateman, I., Carson, R. T., Day, B., Hanemann, M., et al. (2002).Economic valuation with stated preference techniques—summary guide. London: Depart-ment for Transport, Local Government and the Regions.

Planning Department, Government of Delhi. (2004). Socio-economic profile of Delhi, 2003-2004. New Delhi, India: Author.

Raje, D. V., Dhobe, P. S., & Deshpande, A. W. (2002). Consumer’s willingness to pay formore municipal supplied water: A case study. Ecological Economic, 42, 391-400.

Ready, R. C., Malzubris, J., & Senkane, S. (2002, February). The relationship between envi-ronmental values and income in a transition economy: Surface water quality in Latvia.Environment and Development Economics, 7(1), pp. 147-156.

Singh, B., Ramasubban, R., Bhatia, R., Briscoe, J., Griffin, C., & Kim, C. (1993). Rural watersupply in Kerala, India: How to emerge from a low-level equilibrium trap. WaterResources Research, 29(7), 1931-1942.

Vaidya, C. (1995). Willingness to pay for water supply and sewerage in Baroda. New Delhi, India:Human Settlements Management Institute.

Wedgwood, A., & Sanson, K. (2003). Willingness to pay surveys—a streamlined approach,Guidance notes for small town water services. Leicestershire, UK: Loughborough Univer-sity, Water, Engineering, and Development Centre.

Whittington, D., & Lauria, X. (1991). A study of water vending and willingness to pay forwater in Onitsha, Nigeria. World Development, 19(2), 179-198.

Whittington, D., McClelland, E., & Davis, J. (1998). Implementing and demand drivenapproach to community water supply planning: Acase study of Lugazi, Uganda, WaterInternational, 23, 134-145.

Whittington, D., Pattanayak, S. K., Yang, J.-C., & Bal Kumar, K. C. (2002). Householddemand for improved piped water services: Evidence from Kathmandu, Nepal, WaterPolicy, 4(6), 531-556.


Whittington, D., & Swarna, V. (1994). The economic benefits of potable water supply projects tohouseholds in developing countries (Economic Staff Paper No. 53). Manila, the Philippines:Economics and Development Resource Centre.

Whittington, D., Xinming, M., & Robert, R. (1990). Calculating the value of time spent col-lecting water—some estimates for Ukanda, Kenya. World Development, 18(2), 226-280.

Wolf, T., & Kraft, T. (2002). Health for all—water for all? The megacity Delhi (Status Report).New Delhi, India: GEOMED.

Zerah, M.-H. (2000). Water: Unreliable supply in Delhi. New Delhi, India: Centre de SciencesHumaines, Manohar Publishers & Distributors.

Venkatesh Dutta received his B.Sc. in environmental sciences in 1999 from University of Delhi andM.Sc. in environmental management in 2001 from Indraprastha University, Delhi. He is a Ph.D.candidate in regulatory and policy research at Faculty of Policy and Planning, TERI School ofAdvanced Studies, The Energy and Resources Institute, New Delhi since 2001. Currently he is alecturer (environmental economics and policy) in the Department of Environmental Sciences,B.B.A. Central University, Lucknow, teaching postgraduate students. His research interests are inintegrated urban and regional environmental management, environmental economics, and policyanalysis.

Subhash Chander received a B.Tech. in civil engineering, an M.Tech. in water, power, and dam con-struction, and a Ph.D. in hydrology. He taught water resources engineering at Indian Institute ofTechnology, Delhi at postgraduate and doctoral levels from 1961 to 1997 as emeritus fellow. Cur-rently he is serving as an advisor at TERI School of Advanced Studies, New Delhi and InterstateWater Resources, Government of Andhra Pradesh Hyderabad. His research interests are in inte-grated water resources management, water policy analysis, and hydrological modeling.

Leena Srivastava has a Ph.D. in energy economics from the Indian Institute of Science, Bangalore,India. She is currently the executive director, TERI, New Delhi, an independent not-for-profitresearch institution working in the areas of energy, environment, and sustainable development. Shehas worked on a range of issues covering regulatory policy/planning and economics of energy devel-opment pathways in India. Since June 2000 she has also been dean of Faculty of Policy and Planningat TERI School of Advanced Studies, where she is teaching doctoral courses on energy policy andplanning and infrastructure economics.


135

COST OF SERVICES AND WILLINGNESS TO PAY FOR RELIABLE URBAN WATER SUPPLY: A STUDY FROM DELHI, INDIA V. Dutta1 and A.P. Tiwari2

1 Centre for Regulatory & Policy Research, Department of Policy and Planning, TERI School of Advanced Studies, India Habitat Centre, Lodi Road, New Delhi 110 003, India, [email protected].

2 Housing & Urban Development Corporation (HUDCO), Ministry of Urban Development & Poverty Alleviation, India Habitat Centre, Lodi Road, New Delhi 110 003, India, [email protected].

Abstract The provision of safe and universal water supply in an equitable and efficient manner is extremely important for urban water reform programs currently being carried out in developing countries. The sector reform not only requires a significant amount of working capital, but also people’s willingness to pay for the improved infrastructure. This paper serves two purposes – first, it explains the meaning of ‘full-cost of water services’ in urban areas and attempts to provide a framework to value economic and environmental externalities for the urban water supply and use through a case study of India’s capital city – Delhi. The second part uses contingent valuation method to establish people’s willingness to pay from a survey of 1100 households for water supply with better quality and reliability. Policy implications are subsequently discussed, keeping in mind cost of provision of water supplies. Also included is the assessment of the cost of unreliable supply (coping cost), which otherwise households are spending in the absence of a reliable supply. The paper shows several instances of reciprocal externality wherein the residents themselves absorb the cost of over-extraction, in terms of declining water tables, and cost of salinity in terms of decentralised treatment cost.

Keywords: Urban water supply; willingness to pay; externalities; pricing.

INTRODUCTION Reliable water supply with higher levels of quality is attainable through higher costs and hence higher prices to customers. To determine the appropriate level of service improvements, information is needed on full-cost of water provision, what customers want and how much they are willing to pay. This information is important for making explicit decision about appropriate mix of service quality and price (Whittington and Swarna, 1994). However, there is some confusion about the exact meaning of some of the articulated principles of “full-cost pricing”, like environmental damage cost. In particular, it is not clear at many places how to value environmental externalities for the urban water supply and use. Though economists have tried to calculate this cost (Zhang, 2003; Bowers and Young, 2000; Bhatia et al., 1994; Briscoe, 1996; Munasinghe, 1990), there appears to be some bias in the methodology adopted like over estimation of the damages. A review of the existing literature on marginal cost pricing for urban water showed that this area has not been fully examined and hence, further research is needed. Ebarvia (1997) suggested setting the optimum price equal to the marginal opportunity cost (MOC). The principle of MOC pricing outlines three major components for the supply of a natural resource: (a) marginal production costs (MPC), (b) marginal user costs (MUC), and (c) marginal environmental costs (MEC). According to the researchers,

Journal of Water Science and Technology: Water Supply, Vol 5(6) 2005, IWA Publication, UK

136

the socially optimal price should equal the sum of these components. The measurement of each of them is briefly discussed below from a case study of Delhi’s water utility. Subsequently, the paper examines how much people are willing to pay to support a policy of providing them a reliable water supply that meets World Health Organization (WHO) standards. The alternate scenario emphasizes two aspects of the proposed improved services: (i) longer hours of service that would gradually move towards continuous supply, and (ii) good quality water with no health risk of contamination. MARGINAL PRODUCTION COST Marginal value reflects the economic value of water but it is difficult to implement it. Difficulties in implementation arise because it is difficult to define and estimate marginal cost in quantitative terms needed to determine appropriate user charges. Because of capital indivisibility problem of a typical water supply system, costs will be marginal at certain times and non-marginal at other times. This results in significant fluctuations in price. Studies by Saunders and Warford (1977) indicate that when the problem of capital indivisibility exists, computing the marginal cost as the average unit cost of incremental output becomes more appropriate. This necessitates the calculation of average incremental cost (AIC) as given below. The numerator in the formula is the present value of the least-cost investment stream plus the incremental operating and maintenance costs while the denominator is the present worth of the incremental volume of water produced over the period considered (Warford, 1994):

∑

∑

=

=

+−

+−+= T

t

tt

T

t

ttt

rQQ

rMMIAIC

10

10

)1()(

)1()( (1)

where, It: is the investment cost in year t, Mt – M0: is the operations and maintenance cost in year t due to incremental consumption of water in year t or Qt – Q0, and r: is the discount rate.

Using this, the AIC for per m³ of water comes out to be 13.18 Rs (0.30 US$) (Figure 1).

211

344399

5456014.63

7.538.76

11.9613.18

0

100

200

300

400

500

600

700

Total revenue O & M Cost O & M +Depreciation

O & M + Interest Total Expenses

Cost component

Rs i

n cr

ores

0

2

4

6

8

10

12

14

Rs/

m³

Figure 1. Total production cost and revenue for Delhi’s water utility - total revenue per m³ is 4.63 Rs while total

expense is 13.18 Rs (2003 – 2004 data; 1 Rs = 0.023 US $ at 2005 price level, 1 Crore = 10 million Rs) MARGINAL USERS COST

The cost of future use foregone due to the depletion of a resource may be estimated as the cost of replacing the depleted asset at some future date (if substitutes are available). The MUC can, therefore, be

137

estimated by getting the difference between the present value of the MPC of the substitute or replacement technology and the present value of the MPC of existing technology.

Tb

r

CPMUC)1(

)(

+

−= (2)

where, Pb: is the price of replacement technology, C: is the price of existing technology, r: is the discount rate, and T: is the time at which the replacement technology comes in or the switch to the backstop occurs. Following strategies are considered for Delhi’s water supply augmentation: (a) supply augmentation through leakage reduction; (b) surface water augmentation through construction of dams; and (c) augmentation through groundwater development. Currently 42% (1279 million liters a day, MLD) of treated water is lost due to leakages. The present level of 42% system losses could be reduced to 25% in the short term (by 2011) with a long-term target of 15% (by 2021) through upgradation and replacement of distribution network. The total investment for this is set at 32 million US$ (2005 price level) for a total augmentation of about 819 MLD including 187 MLD from recovery of water at the 4 water treatment plants. For surface water augmentation from dams, Tehri dam (728 MLD at 161 million US$), Kishau dam (1683 MLD at 255 million US$), Renuka dam (1251 MLD at 185 million US$) and Lakhwar Vyasi dam (614 MLD 140 million US$) are considered totaling 741 million US$ for augmenting 4277 MLD of water. Plans for dams at Kishau, Renuka and Lakhwar Vyasi are at a very preliminary stage only, and these dams would take at least a decade and half to be realised. The third option of augmenting through groundwater development is not considered for future, as there is hardly any scope for further groundwater development due to falling water table. Only a small fraction of the water demand is being met by the groundwater and these are already overexploited in the current scenario. Using the above figures at a discount rate of 12% and 20 years timeframe, the MUC comes out to be 14.31 Rs/m³ of water (0.32 US$/m³).

MARGINAL EXTERNALITY COST Economic externalities This consists of estimating the cost of bad quality unhygienic and unreliable public water supply termed as the “unreliability cost”. Due to water agency’s inability to provide efficient and reliable supply, customers spend significant amount of money to make the supply reliable in terms of both quality and quantity (see Table 1). The indirect cost of unreliable supply to customers in Delhi is found to be 259 Rs (2.97 US$) per month per household - 4.6 times the average monthly water bill paid to the public water utility. An increasing proportion of urban customers are already making their own investments to simulate ‘24×7’ water supply at the household level – borewells, surface and overhead storage tanks, booster pumps, tankers suppliers, etc. These investments are supplemented with water purification methods such as filtration and boiling. Table 1. Cost of unreliable supplya borne by people in unplanned settlements in Delhi as measure of economic externalities.

Averting measures % of households Annualised capital cost Recurring/O&M cost Storage in buckets/drums 88.23 180 Rs 25 Rs Overhead tanks 51.47 0.75 Rs /Litre 50 Rs Use of private tankers 4.90 NA 200 Rs Use of public handpumps 17.15 NA NA Use of community standposts 14.70 NA NA Use of bottled water 3.26 NA Use of ceramic filters 16.99 141.50 Rs 200 Rs

138

Use of UV filters 12.25 1061 Rs 400 Rs Use of RO filters 0.32 2743 Rs 1500 Rs Boiling b 12.74 NA 532 Rs Use of boosting pumps 53.10 813.50 Rs 1015 Rs Use of borewell pumps 27.77 976 Rs 533 Rs

Cost of sickness Sickness due to diaorrheal diseases 19.60 NA 704 Rs Average annual coping cost per household 3,112 Rs (72 US $) a Unreliability cost is calculated as annualized sum of money spent on drawing water from alternate sources, groundwater

pumpage cost, household water treatment cost and cost on treating waterborne illness mainly due to diaorrheal diseases. b For calculating cost on boiling, consumption of cooking gas for 25 min/d is taken. Considering gas consumption of 180 gm/hr

and price of cooking gas cylinder (14.5 kg) at 282 Rs, the cost on boiling comes out to be 1.46 Rs/d. Data collected from 650 residential households.

Environmental externalities Examples of environmental externalities in urban water systems include the decline in natural water quality and quantity. When a wastewater stream pollutes a river or contaminates a groundwater aquifer, the condition of the water resource changes. The impact of such changed environmental conditions may decrease the society’s welfare as a whole. The environmental externalities cost can be calculated as indicated below (a, b and c).

a. Due to river getting polluted by wastewater and effluents discharge: Lost scenic boat trips along the river side = estimate of total visitor days in a year × gross daily

net revenue Lost fisheries = (Carrying capacity of river for fisheries at the current level – carrying capacity

of river for fisheries at the target level) × unit market value of the species concerned × sustainable off-take percentage (50%)

Lost wildlife and biodiversity = additional costs of achieving minimum target river water quality and flow level = cost of upgrading sewage treatment plants + cost of improving stormwater control + cost of importing water to the catchment

Care should be taken to ensure that there is genuinely some lost value. Minor impacts are included in lost wildlife and biodiversity opportunities. Where wildlife and biodiversity objectives are poorly defined, the work may need to be underpinned by a series of studies to clarify the nature of public preferences for wildlife and biodiversity conservation. Alternatively, the impact of wastewater disposal can be indirectly measured in terms of the impaired ability of the river to produce potable water because of wastewater disposal. The investment and operating costs of wastewater collection and treatment up to the river’s receiving water quality can be taken as a proxy of the environmental cost of wastewater disposal (Warford et al., 1994). With respect to sewerage services in Delhi, around 73% of the population is connected to the sewer network. There are 17 wastewater treatment plants with nominal capacity of 2330 MLD. This is more than adequate for an estimated present day volume of 1847 MLD of wastewater for treatment. However, to meet the future demand, as well as reduce unsafe disposal of wastewater in the environment, 20% additional capacity would be required for year 2021 with a total investment of 70 million US$. The current cost of treatment is 2.80 Rs/m³. Using the formula for AIC, at a discount rate of 12% for 20 years annualized stream, the cost per m³ when additional capacity is built would be 8.50 Rs/m³ (0.19 US$).

b. Due to groundwater aquifer getting saline (external costs due to salt water intrusion): Treatment cost to desalinise saline water: This can be measured as the cost to desalinise water

to the potable standards. Reduced life span and productivity of wells: When the productivity of the wells declines, the

operation and maintenance cost goes up. In some situations, one has to abandon the existing well and look for alternate sources of water.

In our study area, it was observed that many group-housing societies have installed Reverse Osmosis (RO) plant for treating the saline water in the absence of a reliable municipal supply. The cost of treating saline groundwater comes out to be 78 Rs (1.8 US$) to 48 Rs (1.1 US$) per m³ of water treated depending upon the membrane life and capacity of the plant (Figure 2). In the absence of a reliable data,

139

this cost estimate can be taken as the cost of groundwater getting saline due to excessive withdrawal. This is an example of reciprocal externality wherein the residents themselves absorb the cost of over-extraction, in terms of declining water tables, and cost of salinity in terms of RO treatment cost.

c. Due to groundwater aquifer getting depleted (external costs due to over extraction): This is calculated as cost of over extraction in terms of declining water table, i.e., increasing pumpage cost. The basic equation relating pumpage to water table is (Gisser, 1983):

WWRt

HSA n −+=Δ

Δ α·· (3)

where, A: is the area of the aquifer, S: is the storativity coefficient, H: is the depth of water table, Rn: is the natural recharge, alpha is the return-flow coefficient, and W: is the volume of water pumped from the aquifer.

During the period tΔ , the water table has been lowered by HΔ . The vertical distance between the water table, H, and the pumping elevation, PE, termed the lift (PE - H), increases over time as the water level declines. Correspondingly, the pumping efficiency of wells declines as the lift increases. Thus, the falling water levels result in rising marginal cost of pumping. Let c denote the marginal cost of pumping per cubic meter per meter of lift (i.e., cost of energy to lift water), then the operation cost (OC) can be expressed as:

)( HPcOC E −= (4)

In general, in the study area, it was found that on an average for every one meter decline in water level, there is additional burden of 0.48 kwh of electricity on households.

16.8 16.8

5.6 5.6

33.2 33.2

33.2 33.2

3.6 3.6

3.6 3.6

19.7

6.3

6.6

4.6

4.6

0.9 0.9

0.3 0.32.11.5

1.5

0

10

20

30

40

50

60

70

80

30 m³/d, Membrane = 5 years




RO

trea

tmen

t cos

t (R

s/m

³)

Maintenance (Contractror, ROmembrane and filter cleaning)

Spare equipment (ROmembrane+ACF+Cartridgefilter)

Labour

Chemical consumption (Acid,antiscalant)

Electricity consumption

Investment

Figure 2. Cost of treating the saline groundwater in a West-Delhi housing society through RO plant (90 m³/d

capacity for 90 families). WILLINGNESS TO PAY FOR A RELIABLE WATER SUPPLY

Preferences for change versus status quo and consumer’s willingness to pay are estimated through discrete choice multinomial logit (MNL) and nested logit (NL) model with a linear utility function using

140

maximum likelihood estimation technique (Haab and McConnell, 2002; Ahmad et al., 2005; MacDonald et al., 2005). In total 1100 household survey was completed spread over various planned and unplanned settlements through multistage stratified random sampling. Few of them were rejected due to non-response and missing data. Specifically, having presented a scenario and policy the respondents were asked their willingness to pay for improved ‘alternate scenario’ through a set of split bidding game valuation questions. The valuation section was finally concluded by respondent’s belief in the likelihood of improved water supply attributes and reasons for not willing to pay if any. The estimated parameters of the choice model define the utility functions for each alternative. Thus, the dependent variable is choice of water supply system. In making a choice, individuals are assumed to evaluate ‘alternatives’ on the basis of their ‘attribute profile’ and then choose the alternative that maximizes their utility. Theory and intuition suggest that preferences for improved water supply and WTP would differ across population groups with different socio-demographic characteristics, planning status, existing water situations, and opinions about water quality and public policy. Accordingly, in the model, respondent’s awareness, education, household size, etc. were taken as independent variables. Quality was defined in terms of single improved quality meeting WHO standards or dual quality supply with separate provision of potable and non-potable water. Reliability was defined in terms of increased supply hours (over and above current supply hours) that meet the customer’s end use demand. The upper bound level, representing potential reliability improvements that are considered to be technically feasible is continuous supply. Thus three principal alternative specific factors influencing an individual’s utility are the quality, supply hours, and the bid amount. The mean WTP is calculated as:

E (WTP) )*exp(1ln[1k

knnk

kkn

nn

yZZSASCASC ∑∑∑ ′++++= δβγ

β

(5)

where,

yβ : gives the marginal utility of income and is the coefficient of the cost attribute or bid levels, nS : represents socio-economic or environmental attitudinal variables for the nth individual, kZ : is a vector of choice attributes,

β : is a vector of coefficients of explanatory variables. The effects of attributes in the scenario sets are captured by the Z variables, while the ASC captures any systematic variations in choice observations that are associated with an alternative that are not explained either by the attribute variation or respondent’s observed socio-economic characteristics (Train, 2003). It is possible to include socio-economic as well as attitudinal variables into the utility functions by estimating the variables interactively, either with the ASC or with any of the attributes from the choice set. Table 2. List of explanatory variables and their description used in the discrete choice model

Variable Description ASC Alternate Specific Constant for Status Quo Option PLANNING Level of planning from A to H as classified by property class EXTR_DUR Total duration of supply demanded in hours PAYMENT Increment to monthly water bill in Rs (Bid levels) AWARE Respondent’s level of awareness AGE Respondent’s age in years HEAD Dummy for head of the households; 1 if yes, 0 otherwise ENV Dummy for water quality as major environmental problem; 1 if yes, 0 otherwise EDU Respondent’s total years of education

141

From the discrete choice model, it is concluded that customers are willing to pay 295 Rs per month (6.78 US$) for dual quality supply and 189 Rs per month (4.35 US$) for single quality reliable supply. However, this cost is above the operation and maintenance cost of service provision, but do not cover the user’s cost and environmental externalities. This shows that customers are not willing to pay the “full-cost” of water primarily because (i) the full-cost of water is very high due to environmental and resource costs; and (ii) the commonly held view that the provision of basic water and sanitation services is the job of the government, and the customers have a right to access such services at a low price, irrespective of their ability to pay for them. Based upon the model, customers are willing to pay higher amounts with increasing service hours, his level of awareness, and education. Therefore, the public institutions that currently deliver water should increase their service standards to fully receive the customer’s willingness to pay and restructure the institutions to bring the cost at an economically efficient level. Table 3. Estimation of WTP for two alternate scenarios using nested structure under multinomial logit model

Single Supply (Scenario 1) Dual Supply (Scenario 2) Variable Coefficient t-ratio Mean Coeff.×Mean Coefficient t-ratio Mean Coeff.×MeanASC -3.88049 -3.758 1 -3.88049 -2.76757 -3.527 1 -2.76757EXTRA_DUR 0.165237 11.876 8.21 1.356596 0.165237 11.876 5.58 0.922022PAYMENT -.00413 -1.760* – – -.00413 -1.760* – –ASC×PLANNING -0.33562 -4.406 3.05 -1.02365 0.223538 3.221 5.29 1.182516ASC×AWARE 0.56249 3.723 3.03 1.704345 0.749801 5.684 2.11 1.58208ASC×AGE -0.03886 -3.700 39.6 -1.53903 -0.03292 -3.708 42.01 -1.38299ASC×HEAD 0.414153 1.390* 0.43 0.178086 0.607752 2.414 0.61 0.370729ASC×ENV 1.04699 3.197 0.24 0.251916 1.10176 3.826 0.246 0.271033ASC×EDU 0.266067 5.441 14.36 3.820722 -0.00066 -.026* 10.78 -0.00716Log likelihood function -642.353 Chi-squared 730.065 Pseudo R-square .44996 No of observations 1063 E (WTP) 295.05 Rs 189.32 RsCorresponding amount at 95% confidence interval

304.04 – 286.05 Rs

198.31 – 180.32 Rs

*Denotes acceptability below 95% confidence limits

SHOULD EXTERNALITY COST BE INCLUDED IN PUBLIC PRICING POLICY? Typically, the supply costs (incurred in financing and operating the abstraction, transmission, treatment and distribution systems) are considered to be relatively higher, while the opportunity costs (imposed on others as a result of use of the water) lower than the supply cost. Accordingly, the priority issue for the economic management of urban water supplies relates primarily to the supply cost. However, researchers argue that adequate recovery of costs of water services has to be implemented into the pricing policy taking sustainable social, economic and environmental impacts into account. Nonetheless, it is ambiguous at many places how to internalise such impacts for the urban water supply and use. Therefore, designing an effective public externality pricing policy is challenging due to uncertainty involved in the measurement of externality cost. While the theoretically best approach to incorporate external costs in water price (i.e., charging each user the cost of external damage they generate) is not practical; another suitable approach that gives the right signals can be pursued. There are several instances of reciprocal externality wherein the residents themselves absorb the cost of over-extraction, in terms of declining water tables, and cost of salinity in terms of decentralised treatment cost. Environmental externality costs should only be included where they are actually incurred and paid by the water utility. It is not fair to charge for externalities that are incurred due to public water utility’s

142

inefficiency. Water prices may be set so that revenues from water sales cover all operating costs, on-going maintenance costs, capital expenses necessary for ongoing operation, and costs of water use to the environment. The costs of water use to the environment may be included in the water bill by a “sewerage surcharge” that is required to treat the wastewater up to the stream’s receiving quality. In summary, to charge each water user for the exact external cost that their actions impose on others is unworkable. This would require estimates of the environmental damage in monetary terms that additional increments of water use cause for each water user and development of a set of differential charges reflecting damage cost. This task is very complicated and purely site-specific. Quantifying appropriate charges to “internalise” the externality is thus a major research challenge.

ACKNOWLEDGEMENTS

The authors gratefully acknowledge Dr Leena Srivastava, Prof Subhash Chander and Dr Puneet Chitkara for the academic guidance provided during their doctoral research at Faculty of Policy and Planning, TERI School of Advanced Studies, New Delhi. Research Fellowship provided by the University Grants Commission (UGC), Government of India to the first author is sincerely acknowledged.

REFERENCES Ahmad J.K., Goldar B.N., Misra S. (2005). Value of arsenic-free drinking water to rural households in

Bangladesh. Journal of Environmental Management, 74(2005), 173 – 185. Bhatia R., Cestti R. and Winpenny J. (1994). Policies for Water Conservation and Re-allocation: Good

Practice Cases in Improving Efficiency and Equity. World Bank Technical Paper, World Bank, Washington, DC.

Bowers J., and Young M. (2000). Valuing Externalities: A Methodology for Urban Water Use, CSIRO Urban Water Program.

Briscoe J. (1996). Financing water and sanitation services: the old and new challenges. Water Supply, 14(3/4), 1-17.

Ebarvia M.C. (1997). Pricing for Groundwater Use of Industries in Metro Manila, Philippines, Research report, International Development Research Centre, Ottawa, Canada.

Gisser M. (1983). Groundwater: focusing on the real issue, Journal of Political Economy, Vol. 91, No.6. Haab T., and McConnell K.E. (2002). Valuing Environmental and Natural Resources: the econometrics

of non-market valuation, Edward Elgar, Cheltenham, UK. MacDonald D. H., Barnes A., Bennett J., Morrison M., and Young M.D. (2005). What consumers value

regarding water supply disruptions: a discrete choice analysis, Journal of American Water Association, Forthcoming.

Munasinghe M. (1990). Managing water resources to avoid environmental degradation: policy analysis and application, Environment Working Paper, No. 41, World Bank, Washington D.C.

Saunders R.J. and Warford J.J. (1977). Alternative Concepts of Marginal Cost for Public Utility Pricing: Problems of Application in the Water Supply Sector, World Bank Staff Working Paper, No. 259.

Train, K. (2003). Discrete Choice Methods with Simulation, Cambridge University Press. Warford J. (1994). Marginal Opportunity Cost Pricing for Municipal Water Supply. EEPSEA Discussion

Paper. Paper prepared for the Economy and Environment Programme for Southeast Asia, August 1994.

Warford J., Hansen S., Panayotou T. and Spofford W. (1994). Marginal Opportunity Cost Pricing: Water, Coal and Forest Resources in China.

Whittington, D. and Swarna V. (1994). The Economic Benefits of Potable Water Supply Projects to Households in Developing Countries, ADB Economic Staff Paper No. 53. Manila: Asian Development Bank.

Zhang F. (2003). Marginal Opportunity Cost Pricing for Wastewater Disposal: A Case Study of Wuxi, China, Research Report, Economy and Environment Program for Southeast Asia, International Development Research Centre, Canada.

Sector reforms, regulation and the challenges of sustainability: Demand side analysis for urban water utility of Delhi, India

VENKATESH DUTTA Centre for Regulatory & Policy Research, The Energy and Resources Institute, India

ANAND PRAKASH TIWARI Housing & Urban Development Corporation, New Delhi, India

Abstract Even though, the provision of safe and reliable water supply to mankind is considered crucial for the urban and economic development, yet, for a long time, this sector was relegated to the policy backburner by the urban local bodies (ULBs). In most South Asian cities today, public sector agencies are struggling to provide water supply and sanitation services satisfactory to customers. Municipal bodies, supposedly in charge of service delivery, are caught in a downward spiral of disrepair and in difficult position to meet the increasing demand or even maintain the standards. To make decisions that lead to focused, efficient and sustainable solutions, it is important to have a clear picture of the situation, so that problems can be identified. Keeping this objective in mind, the main purpose of this paper is to get a good understanding of the baseline conditions regarding water supply services for Delhi’s residential customers, including coping strategies and costs, demand and preferences for services, and contributions affordable to each group in the light of current reform process. This is done primarily to explore regulatory policy and institutional context for urban water reforms and set a rational policy framework for sustainable urban water supply and use decision-making. The situational analysis is done through demand and supply side analysis, which allows to measure the urban water sustainability criteria. Keywords : Urban water supply, Demand analysis, Reform.

INTRODUCTION

As water resources dwindle rapidly while demand escalates, there is a growing awareness that water development and supply management practices have to improve. India cannot afford to ignore the warning signs. Water supply in large to medium-sized urban centers in India is confronted with several problems such as – many people, especially the resource poor do not have access to safe drinking water; unacceptably large quantities of water are lost due to leakages; water supplies are intermittent and of inadequate quality; efficiency and productivity are low; repair and replacement costs are excessive as a result of inadequate preventive maintenance; technical options chosen are not suited to the technical, social or financial environment in which they operate; poor cost recovery, high operating costs and poor financial management (Dutta, 2004; Saleth and Dinar, 2004; Raghupati, 2002; Basu and Main, 2001; Zerah 2001). These problems become more challenging with heterogeneous planning levels and differing age of water infrastructure. Due to poor customer’s willingness to pay in the status

quo situation, the accumulated deficit in revenue base has limited the ULBs both to improve the current service in those areas with water scarcity problems, and to invest the necessary resources to guarantee

XII World Water Congress “Water for Sustainable Development - Towards Innovative Solutions”

22-25 November 2005, New Delhi, India

Venkatesh Dutta et al. 4-14

sustainability of supply. The enormous volumes of water and extensive infrastructure required to fulfill urban water demand have frequently exceeded the ability of government to provide secure supplies, and have also created severe environmental problems. To make decisions that lead to focussed, efficient and sustainable solutions, it is important to have a clear picture of the situation, so that problems can be identified. Keeping this objective in mind, the main purpose of this paper is to get a good understanding of the baseline conditions regarding water supply services for Delhi’s residential customers, including coping strategies and costs, demand and preferences for services, and contributions affordable to each group in the light of current reform process. This is done primarily to explore regulatory policy and institutional context for urban water reforms and set a rational policy framework for sustainable urban water supply and use decision-making. The study has attempted to develop a decision-support framework (DSF) to identify customer preferences; their willingness to pay to achieve these preferences (or willingness to accept compensation where a reduced standard is contemplated) and the cost of achieving their preferences, which allows cost-benefit analysis to be carried out. In short the DSF has following 5 components (Fig. 1):

• Aspect (supply) – Impact (demand) analysis: gives the identification of the decision problem, i.e., change in the provision of the public goods through demand and supply side analysis

• Selection of “attributes” and their “labels” – 3 attributes “quality (marginal vs high quality)”, “quantity (sufficient to meet differential end uses)” and “reliability (intermittent vs continuous)”

• Utility Choice/Preference Model to measure welfare (a) benefits: compensating surplus and equivalent surplus and (b) marginal rate of substitution between “attributes”

• Cost of achieving the desired “attributes”

• Policy analysis

The first component reviews the current performance of the utility through identified criteria of urban water sustainability. This gives the existing situation of demand and supply and outlines the need for a change in water supply provisions. The second component highlights three important attributes – quality, quantity and reliability that are considered fundamental for improving the performance. The third component measures the benefits of service improvements through random utility models. The fourth component attempts to estimate cost of achieving these performance improvements and finally policy analysis is done based on customers’ willingness to pay (WTP) and the cost of services. Due to the limited scope of the paper, we will highlight the demand analysis in the first component of the DSF. Before this, a brief outline of Delhi’s water supply system is provided in the context of changing socio-economics and urban growth. CITIES AND URBAN WATER SUPPLY SYSTEMS : CHANGING PATH OF EXPANSION

Urban water supply systems and practices have evolved in direct relation to the physical, social, economical and institutional environment of the cities. The course of the cities in time displays different patterns – cities in most of the developing countries have followed different paths of spatial and demographic growth. Water supply systems have also a number of differences reflecting the diversity of urban contexts and the different tracks of the urban growth and expansion. Water distribution networks have typically followed the emerging patterns of urban development with little overall and long-term planning, even in the cases of cities with rigorously programmed urban plans. The evolution of the metro cities as agglomerations of initially separated residential nuclei results in

Sector reforms, regulation and the challenges of sustainability: Demand side

analysis for urban water utility of Delhi, India 4-15

Fig. 1 Phases in the decision-support framework developed for the study

problematic network structures due to gradually merging the networks of differing ages and due to, equalizing pressures through a retrofitted system. For example, City of Delhi contains populations marked by huge economic disparities both in terms of income and infrastructure facilities. Existing water service conditions are highly heterogeneous across zones, and that these conditions are significantly associated with household planning status in those areas. The problem of determining appropriate pricing strategies for such city is highly complex, due to the cities’ heterogeneity in terms of service conditions and socio-economic characteristics. Although in some zones the water supply network is sufficient, the service standards can often vary drastically from one area to another as a result of engineering, geographic or urbanization problems. In socio-economic terms, practically all the zones present a remarkable diversity, with a relatively clear distinction between different planning


levels and property classes (Dutta et al., 2005). One of the biggest bottlenecks that could hinder the growth of economy and urban development could be a slow response to urban water systems as inadequate coverage, poor quality, as well as unreliable and unsustainable supply of water have an adverse impact on cities’ socio-economic development. The capital city’s water utility – the Delhi Jal Board (DJB) is the primary provider of piped water supply and sewerage services. It serves a total population of nearly 14 million through 1.47 million water connections. Three separate urban local governments with non-overlapping jurisdictions serve the National Capital Territory of Delhi (NCTD) covering a total area of 1486 sq km. The largest of these is the Municipal Corporation of Delhi (MCD), which has over 94% of the NCTD area (1400 sq km) under its jurisdiction. The New Delhi Municipal Council (NDMC) and the Delhi Cantonment Board (DCB) control about 86 sq. km. The MCD is among the largest municipal bodies in the world providing civic services to more than estimated population of 13.78 million citizens in the capital city of India. It is next only to Tokyo in terms of area. Within its jurisdiction are some of the most densely populated areas in the world. The entire MCD area is divided into 9 districts having 12 Zones. These 12 zones are further divided into 134 wards (Fig. 2). Each ward has residential colonies/enclaves that come under several categories such as rural and urban villages, resettlement colonies, regularised unauthorised colonies, Jhuggi Jhopri (JJ) clusters, slums etc. JJ clusters are squatter settlements mostly on government land. Unauthorized colonies comprise settlements that come up on land initially not meant for human habitation. As such these do not have any service provision from the municipal agencies. However, due to administrative and political reasons, these colonies are considered for periodic in-situ regularization, keeping in view parameters like environment and land use, following which they become eligible for individual level water and sewer connections. Villages where agricultural land is taken up for expansion of the city and thereby become a part of urban Delhi are known as ‘urban villages’. Households located in these areas are eligible for municipal services. According to MCD, there are 1700-odd unauthorized colonies and 1100 slum settlements in Delhi, however, their numbers are not static and keeps varying. According to the new Master Plan (MPD – 2021, Draft Report), despite about 1.30-lakh households and Jhuggis having already been relocated, there are still around 3-lakh eligible squatter families who need to be relocated. With the emergence of various types of settlements, particularly unplanned settlements, the urban scenario of Delhi has become very complex and difficult for management by local bodies. To ensure appropriate allocation of land and development of all public utilities/physical infrastructures, MPD 2001 made provisions of 9 land-use categories with further 37 use zones. Unfortunately, the objectives of the Master Plan could not be achieved due to failure in making available adequate developed land for physical infrastructure and public utilities like water supply to the concerned agencies in time on the one hand, and the lack of adequate developed land at affordable rates to the public for housing. The capital city’s water utility cannot afford to ignore the warning signs. In the recent past, things seem to have taken a turn for the better. The capital’s primary provider of piped water supply and sewerage services initiated a major step towards reform and possible restructuring of the utility in 2004 (DJB, 2004). The Board envisages a complete overhaul of this system through its reform plan, which is titled the Delhi Water Supply and Sewerage Project. The project, which spans 10 years and is scheduled for completion in 2015, has six different components. The first phase involves improvement in the distribution system to attain continuous supply. In mid-February 2005, the Board invited pre-qualification bids from private operators for the distribution and management of water supply in two zones of South Delhi. The primary objective of the contract is to ensure continuous supply of water in these zones. In the second component, bulk water supply and sewerage infrastructure will be enhanced for citywide improvement in these services. The third component involves consultancy studies for the rationalization of trunk systems for changeover to continuous supply regimes. Under the fourth component, organizational strengthening measures would be introduced to improve DJB’s internal efficiency and consumer focus and responsiveness. The fifth component involves targeted interventions, including pilot projects, for improved services to the poor. The final component comprises a rollout programme for scaling up services and replication across the entire city.



Fig. 2 Classification of residential units under MCD jurisdiction based on type of settlements (the bar on right shows the different class under which they typically belong)

DEMAND VERSUS SUPPLY SIDE ANALYSIS

Demand side analysis is a utility concept referring to measures targeting user’s attitude (by information, pricing, etc.) as well as interventions that improve the efficiency of distribution and use. The supply side analysis on the other hand primarily concerns technical and engineering aspects of water supply and service provision. Though inadequate provision of water supply due to resource constraints may play a part, the major reason for the present ‘crisis’ is the misdirected emphasis on investing in physical infrastructure without ascertaining its utility or ensuring its maintenance and sustainability. Current reform process is still in short in comparison with the magnitude of the ongoing supply problems. From an economic perspective supply side interventions only address one part of the problem. Economic prescriptions for tackling sustainability issues also require information concerning the demand side of the water market. The demand side analysis constitutes a vital input to the economic decision making process which must underpin any long term sustainable policy for reform of the water supply of the city. In doing situational analysis, we have considered following five aspects (Fig. 3):

• Infrastructure functioning and use: quantity, quality, and reliability of services


Fig. 3 Sustainability of urban water systems as mapped by 5 major aspects and their criteria

• Institutional aspects: effectiveness of institutions in service delivery and management, assessing the degree of satisfaction of service objectives

• Financial aspects: adequacy of cost recovery for scheme operation

• Social aspects: affordability, equitable access and participation in benefits and management by urban poor and socially disadvantaged groups

• Environmental aspects: assessing whether the impacts of the system on the natural environment are taken into account

Infrastructure Functioning and Use

Supply is intermittent: Although the water supply system is operated on the basis of continuous 24-hr water production and transmission, the transmission system transports water to the various underground terminal reservoirs intermittently from where it is put into the distribution systems through pumping

Sustainability of urban water system s

Institutional aspects

Financial aspects

Social aspects

Environmental aspects

Infrastructure func tioning and use

Supply reliability

Service delivery

Consumer satisfaction

Willingness to pay

Equitable access

Participation

Cost recovery

Tariff design/Pricing

Revenue water

Non - revenue water

Energy conserv ation

Resource utilisation

Reuse and recycling



stations. A few areas of the city have distribution systems that are connected directly to the transmission system and such areas obviously enjoy a near-continuous supply of water. For distribution, the water is supplied from the terminal reservoir on an intermittent basis, depending on the volume of water available from the transmission system and the size of the distribution system. To ration the available water, distributions are often sub-zoned for supply by manipulation of valves. Until January 2003, the distribution areas were divided into 12 zones; after that date, the number of zones was increased to 21. Under both systems, the zones operate as maintenance and revenue collection zones, the boundaries of which do not coincide with the physical supply boundaries. Water from the WTP is supplied continuously to the distribution zones via numerous transmission mains supplying various UGRs each with its own BPSs for onward delivery to the distribution zones. There are three main in-line booster-pumping stations in the transmission system. The total length of the transmission main is 568 km, the dia of which ranges from 450 mm to 1500 mm. The Bhagirathi system ranges from 1100 mm to 1200 mm; the Haiderpur system ranges from 1000 mm to 1500 mm; the Wazirabad system from 600 mm to 1500 mm, and the Chandrawal system from 450 mm to 1200 mm. There are main reservoirs at 61 locations where underground tanks and associated booster pumping stations supply the various distribution systems. These UGRs have a total storage of 731,673 m3 (160 million gallons). Water distribution is undertaken in 21 revenue zones. These revenue zones or maintenance zones do not match the physical supply zones - the precise no of which is not known. However, it can be assumed that there are 61 major zones (one for each underground reservoir) and more than 400 minor zones, as there are 492 distribution booster-pumping stations. The total length of distribution system is around 9000 km. there are some distribution zones that receive water from direct tapings to the transmission system. There is large variation of duration of supply across zones. Most of the zones have 4 – 6 hours of daily supply, on an average 4.5 hours of supply per day; though a few small areas do enjoy a 24-hour supply because they are directly connected to the transmission system (Tables 1 & 2). A further manifestation is the prevalence of small private booster pumps to enable customers to abstract more water at a higher pressure than the system presently permits. This is against the DJB by-laws. The supply duration is lowest in the West zone.

Demand supply gap: A substantial part of total water demand is made up of residential customers, although there are huge variations between households. Within the category of residential customers, there are important differences between per capita use in single-family establishments and multi-family establishments (flats and apartments). Considering different settlements types and their per capita water requirements based on norms of CPHEEO & MoUD, the total domestic water requirements are given in Table 3. In the DJB’s 10th Five Year Plan, the total water demand per capita is calculated at 60 gallons per capita per day or 274 lpcd, broken down as – Domestic: 172 lpcd; Industrial, commercial and institutional: 47 lpcd; Fire protection: 3 lpcd; and Floating population and special use: 52 lpcd. Using the DJB 10th Five Year Plan estimated demand, the 2006 water demand for the total predicted population of 155 million would be 4521 thousand m3/day or 995 MGD. If the total demand is estimated using the DJB’s norms of per capita supply and 15% losses, the total demand is estimated to be 800 MGD. Thus the current shortfall (gap between water demand and supply) is estimated to be 200 MGD or 33% of the present level of supply. The supply of water in the service area is not even: Though half the revenue zones have an average supply per connection of between 25 and 35 cubic metres per month (m3/month), six zones have more (up to 75 m3/month) while eight have less as low as 2 m3/month. To provide water to the areas not covered by the distribution system, DJB supplies water by tanker services. There are three different norms of per capita water supply referred to in Delhi. The ISI standards prescribe 135 liters per capita per day (lpcd). The DJB uses a norm of 50 gpcd (227 lpcd) whereas the Delhi Master Plan uses a norm of 363 lpcd. Based on the above figures of water production and bulk water sales, it is estimated that whereas the per capita water supply in MCD areas is 184 lpcd, it is significantly higher in NDMC and CB areas which receive 283 lpcd and 350 lpcd respectively.


Table 1 Duration of supply hours in the zones

Zone Total hrs. of supply at end points No. of times per day

Central 6 2

City & SP 6 2

Karol Bagh 4 2

Mehrauli 4 2

Najafgarh - -

Rohini 8 2

RWS-N 4 2

South 4 2

West 1-3 2

Civil Lines 24 Near continuous

Shah/N 24 Near continuous

Shah/S 4-6 2

Table 2 Water availability in the zones

Zone Source Average per capita supply (lpcd)

Estimated demand-supply gap reported (%)

Central - - - City & SP Wazirbad & Chandrawal 146 37% Karol Bagh Haiderpur & Chandrawal 100 45% Mehrauli Deer Park & GK-I 60 40% Najafgarh - - - Rohini Wazirabad & Haiderpur 175 15% RWS-N Haiderpur 60 - South Bhagirathi & Haiderpur 140 40% West Wazirabad & Haiderpur 125 30% Civil Lines Wazirabad & Chandrawal 135 28% Shahadara(N) Bhagirathi 160 - Shahadara(S) Bhagirathi 105 -

Table 3 Total water demand considering different settlements type

Type of settlement Approx. population in million (2000 level)

% of population

Litres per capita per day demand (based on norms of CPHEEO & MoUD)

Total domestic water demand, 1000 m3/day (2004 level)

Slum clusters 2.072 14.8 70 160.1 Slum designated areas 2.664 19.1 70 207.0 Unauthorised colonies 0.740 5.3 70 57.5 Resettlement colonies 1.776 12.7 150 295.7 Rural villages 0.740 5.3 150 123.2 Regularised unauthorised colonies

1.776 12.7 168 331.2

Urban villages 0.888 6.4 168 165.6 Planned colonies 3.308 23.7 200 734.7 Total of NCT Delhi 11.964 100.0 2076



Part of network is no longer adequate to serve an area with much increased population density: There are several areas, where the network is not sufficient to provide water to the dense population. For example, Kalkaji Extn. , Govindpuri, Sunlight Colony, Hari Nagar Ashram, Amritpuri. Urban villages mostly come under this category. Due to lower land prices and close vicinity to neighbourhood urban colonies, these urban villages attract many private builders for housing, construction and commercial activities without caring for water infrastructure. Urban villages in the heart of city do not show any character of ‘village’ and they become economic centers of informal activities through provision of services such as food & vegetables, domestic services, automobiles workshops and household maintenance as well as trade and recycling. The consequent increase in demand for basic housing and services for urban populations, as well as skewed distribution of investment towards affluent suburban developments, has resulted in the rapid expansion of illegal or unplanned and unserviced settlements in such areas, with unhealthy living conditions and extreme overcrowding. The numbers of people living in these settlements are expanding so rapidly that urban local bodies are unable to keep up with the necessary infrastructure development . The problem is further exacerbated by haphazardly built and high-density housing making provision of water through conventional means extremely difficult. It is not unusual to see one water connection being shared by more than five households in such areas resulting in severely affecting quality and quantity of water supply. Congested and unplanned settlements with narrow roads and by lanes create massive problem in retrofitting the current piped network. Water losses in the piped systems, due to age, poor construction, leakage and inadequate planning for operation and maintenance are adversely affecting the economic sustainability of existing systems in such areas.

Level of consumption is very low: The average water consumption in Delhi is generally cited at being 240 lpcd, the highest in the country, but, according to the survey, the average consumption per household in the unplanned areas was 420 liter or 67 liters per capita per day and in the planned areas it was 165 lpcd (Fig. 4). On an average people need 36 lpcd for potable purposes and 125 lpcd for domestic purposes. The mean consumption in planned areas was found to be 19.8 kl/month and 12.6 kl/month in unplanned areas. An earlier study (Barah et al., 1998) reported per capita consumption in Delhi ranging from 313 lpcd in affluent households to 140 lpcd in relatively less well off households and, a meagre 16 lpcd for the slum households. A more recent study (Wolf and Kraft 2002) reported that the range varies from 17 lpcd to 646 lpcd. Whatever the statistics be, the average consumption in unplanned areas is very less and continuous water supply remains a pipe dream to them. Discontinuous and inadequate water supply with quality limitations poses serious risks to customer’s health.

In many areas water is not supplied to the network: There are several residential areas in which water supply is not regularly supplied despite the presence of piped networks. JJ colonies, unauthorized colonies and urban villages mostly come under this category. There are several reasons for this, for example, (a) the urban local bodies often regard the provision of water services as conferring a degree of legitimacy on the settlements and are therefore reluctant to supply them even when there exist a piped network; (b) connectivity of the distribution system is not properly maintained or poor piped network within the area; (c) limited availability of water at the booster pumping stations (d) water is rationed to the neighbouring colonies; (e) tube well water is mixed with the limited municipal supply locally (this was case in several areas in South and Central zones, the tube wells are installed by the area councilor, mixed water is often of non-potable quality due to high salinity and is invariably not tested for quality). Institutional Aspects

Coping cost to avert unreliable water supply is significantly high: Coping cost is calculated as annualised sum of money spent on drawing water from alternate sources, groundwater pumpage cost, household water treatment cost and cost on treating waterborne illness mainly due to diarrhoeal diseases. While it is difficult to estimate the full economic (including environmental and health) costs to these household-averting measures, they are economically inefficient (as they reduce savings which


could be invested more productively), are unaffordable to the poor who suffer disproportionately, and are not sustainable. To cope with the unreliable water supply (which has quality implications too) households resort to several coping strategies such as storage in the tanks, installing water purifiers, boiling, purchasing packaged water from water vendors etc. The annual cost of unreliability differs significantly in different property classes. It can be said that with increase in quality of life and planning status, households spend higher amount of money to avert unreliability – it is Rs 2525 per annum in lower stratum property class G and goes up to Rs 6000 for class A. Planned and unplanned areas put together, the average cost of unreliability is Rs 4533 per annum per household in the MCD areas. The average coping cost in unplanned areas is Rs 3486 per annum per household (Table 4).

Fig. 4 Average share of water for different usage for a household (Liters/household)

Table 4 Coping cost or cost of unreliable supply borne by people in unplanned settlements

Averting measures % of households Annualised capital cost Recurring/O&M cost

Storage in buckets/drums 88.23 Rs 180 Rs 25

Overhead tanks 51.47 Rs 0.75/Liter Rs 50

Use of private tankers 4.90 NA Rs 200

Use of public handpumps 17.15 NA NA

Use of community standposts 14.70 NA NA

Use of bottled water 3.26 NA NA

Use of ceramic filters 16.99 Rs 141.50 Rs 200

Use of UV filters 12.25 Rs 1061 Rs 400

Use of RO filters 0.32 Rs 2743 Rs 1500

Boiling 12.74 NA Rs 532

Use of boosting pumps 53.10 Rs 813.50 Rs 1015

Use of borewell pumps 27.77 Rs 976 Rs 533

Sickness due to diaorrheal diseases 19.60 NA Rs 704

Average annual coping cost Rs 3486

Share of water use category per household

120

85

75

65.5

32

16.520.5

10 12.5

0

20

40

60

80

100

120

140

Bathing Used in

toilet

Washing

clothes

Washing

utensils

Cleaning

house

Drinking

water

Cooking

water

Gardening Others

Lit

ers

pe

r h

ou

se

ho

ld



Poor people bear a disproportionate share of the impact of inefficient water. Customers inconvenience results in loss of household income or productive time as at least one family member has to cope with securing reliable water on a daily basis. Typically, the residents need to supplement public supplies with water obtained from private sources, and this is usually much more expensive. The extent to which this is done – the proportion of water procured privately against the quantity supplied is determined by individual households demand and socio-economic characteristics. Several households use multiple averting measures, for example, private tankers and community standposts for sourcing water and boiling for purification. Thus the lack of access to a reliable, adequate and safe water supply impacts directly on the livelihoods and incomes of the urban poor – their ability to engage in income-generating activities, the types of livelihood activities they can engage in, their incomes from these activities, and their overall cost of living.

Water quality is not adequate/appropriate: Even though the results of quality check by DJB indicates water of a potable quality being delivered to customers, however, it is highly likely that some contamination of the water takes place in the distribution system due to the intermittent supply situation. Poor water quality may be also attributable to leaking pipes, proximity of water lines and sewer lines, user of boosters etc. Even though, the tests conducted reveal that quality of treated and distributed water meets the requisite standards, however the customer is often dissatisfied with the quality of water. The deficient sampling methods, timing, testing and location of procedures being followed may explain this apparent disparity. Unplanned settlements have numerous problems associated with water and sanitation provision and unsafe hygiene. Water supplied by the DJB is supposed to be safe under normal circumstances; however, contamination is common where joints/valves lie close to sewers. Problems also arise in areas where water pipes have become old and corroded. Several households reported the current levels of water quality as inadequate with respect to their desired expectations. Due to improper drainage and sewerage system several low lying unplanned areas get flooded during rains. The sewage enters drinking water pipeline and contaminates the water. A large number of inhabitants living in slums, unauthorized colonies, and squatters depend on shallow handpumps’ water, which in most cases, is contaminated and thus unfit for drinking. Salinity and over exploitation has contributed to depletion and drastically affected the availability of water in different parts of the city. 22.58% households in planned areas rated poor quality of municipal water as the major environmental problem in their areas. The usage of water purifiers is very high in planned areas – while 11.2% drink tap water directly without any form of home treatment, 88.8% households use some form of in-house water treatments. The penetration of UV type purifier ‘Aquaguard’ is highest – 44.19% people in planned areas use this. Financial Aspects

Poorly managed metering, billing and collection: Non-metering of water implies that pricing has no effect on water use. Metering on actual consumption in unplanned areas is very low. Less than 50% of the household reported use of water meters within their premises out of which 31.65% have working water meters and billing is based on actual consumption. As a whole, billing hardly reflects actual water consumption – 8.27% of households don’t pay at all and 31.47% of the population pay by ‘flat and fixed tariff’ irrespective of the consumption (minimum charge @ Rs 32.50 per month). To realize the customers’ WTP, increased level of water metering is essential. If water utilities are to attain sustainability over the long-term, they will have to embark on developing and applying the right water tariff based on volumetric consumption through increased level of water metering both in planned and unplanned areas. DJB records showed a total of 1,33,833 customer connections in December 2001, of which about 95% were registered as domestic connections. In principle, all these connections should be metered. However, the records showed that 301,733 (23% of the total) connections had no meter and the meters of 572,853 (43%) were out of order. It follows that only 459,247 (34%) of connections had a


meter in working order. The overwhelming majority of connections with no meter are domestic connections. Reasons for meter going out of service include, tampering, blockage, weer and corrosion. In addition, meters cannot be expected to operate with any accuracy in an intermittent supply situation. The overall metering situation in DJB is therefore very poor. Water production is not metered; it is estimated by factoring pump operation hours with duty point outputs. Similarly for consumption, a customer with a working meter is billed either for actual consumption or for 24 m3/month whichever is greater. For non-working meters, billing is based on either average past consumption (when the meter was working) or a flat rate equivalent to (for domestic customer) 24 m3/month depending on the circumstances. The year 2000 – 01, BDB billing records indicate that less than 6% bills were based on actual consumption. Cost recovery and pricing: The price of water and the charging system are important determinants of the level (vis-à-vis conservation) and affordability of water use. Since 1958, customers had been paying minimal charges for the water they used. The tariff structure was first revised to change in 1958 – 59 but for most of the customers, it remained just the same – 11 paise per kilolitre. The slabs continued till 1962 – 63. The most elaborate tariff was introduced in 1963 when the entire capital was divided into five categories. The water utility carried on with this structure with minor rate changes till 1970 – 71 when a three-category system was introduced again. With the growing population, the need for a separate body was felt and finally in 1998, the DJB was formed with a revised tariff structure. Prior to the hike in December 2004, the lowest slab started from 35 paise per kilolitre (kl) and went up to Rs 3 per kl above consumption of more than 30 kl with 50% surcharge on wastewater treatment. This tariff structure continued till December 2004 when the DJB rationalised the tariff to facilitate the sectoral reforms aimed at achieving better level of services and customer orientation. The new monthly bill is calculated on the basis of a formula that comprises of a fixed monthly access charge, a variable component and a sewer surcharge of 1.5. The fixed access charge is given on the basis of the type of accommodation. The first 6 kl is free for domestic customers, but they have to pay fixed access charges, whether they use this amount or not. The variable component is the product of the actual consumption in kl and a penalty quotient for consuming more. The penalty quotient goes up with each consumption slab: it is 0 for those consuming upto 6 kl; 2 for 6-to-20kl slab; 5 for 20-to40-kl slab and 10 for those consuming more than 40 kl. In the current scenario, to recover the total annual expenses of Rs 601 crore, the price of water has to be raised to Rs 13.18 per kl (based on DJB data, 2003 – 2004, (Fig. 5) The lack of data on consumer demand structure is a real problem for making rational consumption block structure in case of increasing block tariff design (Dutta & Tiwari, 2005). Considering the new pricing policy adopted by Delhi’s water utility, if a household is living in a cooperative group housing society (CGHS) flat, the new IBT tariff will make him pay for wasteful neighbours and put the entire society on a higher billing category because the billing is done on the basis of bulk metering. With the new tariff, the water bills have doubled for residents of CGHS flat with just the fixed charges being levied. For example, “a bill of Rs 7,000 per month has gone up to as much as Rs 20,000 even though consumption is the same” (In spate: Inflated bills for water users, Times of India, March 25, 2005). In a society, water is supplied to a group of flats and a common bulk meter is installed. The society then distributes the water and the bill is shared equally. Most societies do not have individual connections. Because water is usually supplied to them in bulk, they are pushed to the maximum consumption category and are charged quite a lot. Similarly, if many households in unplanned areas were sharing one meter, the IBT would penalize for collective consumption, which would fall in higher consumption slab (Fig. 6). Further, if they were to install individual meters, their consumption might be below 6 kl of lifeline-slab which would further tend towards underpricing and would be like giving water free of cost to the entire society. A group of poor families might thus have collective consumption that pushes them into the highest-priced blocks of the IBT, leaving only relatively wealthy households in the lower (subsidized) blocks.



Fig. 5 Cost of services for DJB, to recover the total annual expenses of Rs 601 crore, the price of water has to be raised to Rs 13.18 per kl (based on DJB data, 2003 – 2004)

If a tenant has a separate water connection, he and the landlord will get two separate water bills. Now, with the new tariff structure in place, tenants have been billed for the entire plot area even if he lives on half the area or even lesser. For instance, a person living in a bungalow on a 155-sq meter plot has to pay a fixed access charge of Rs 150. If he has let out first floor of the bungalow to a tenant, the plot area being utilised by the tenant is much less than 155 sq meters. If it is 100 sq meter, it would easily put the tenant in a lower tax bracket of Rs 100 as access charge. DJB has conveniently billed both the landlord and tenant for the same plot area.

211

344399

5456014.63

7.53

8.76

11.96

13.18

0

100

200

300

400

500

600

700

Total revenue O & M Cost O & M +

Depreciation

O & M + Interest Total Expenses

cost component

Rs in

cro

res

0

2

4

6

8

10

12

14

Rs p

er

KL

Rate per kilolitre (kl) at different water consumption level

6.6

7

8.3

3

12.5

0

16.6

7

25.0

0

5.6

7

6.3

3

8.0

0

9.6

7

13.0

0

9.1

0

9.5

0

10.5

0

11.5

0

13.5

0

8.6

4

8.9

3

9.6

4

10.3

6

11.7

9

15.8

9

16.1

1

16.6

7

17.2

2

18.3

3

3.00

7.00

11.00

15.00

19.00

23.00

LIG Type II HIG Plots of 100 - 150

sq m

Plots above 150

sq m

Rs

pe

r k

ilo

litr

e (

kl)

6 kl 15 kl 25 kl 35 kl 45 kl

Fig. 6 Rate per kilolitre at different water consumption level for different classes, the horizontal dotted line shows DJB’s cost of production


High non-revenue water (NRW): According to DJB sources, there were 13 lakh total water connections in Delhi (MCD) as on December 2000. Out of those 10 lakh connections were metered and 3 lakh were unmetered. However, it is also noted that 35.5% of the meters were defective. Thus effectively the extent of metering in Delhi is about 50%. The DJB has provided about 13.47 lacs water connections in the city till April 2001. Nearly 74% of these are metered connections. According to DJB the percentage of UFW is as high as 50% (Table 5). The real losses comprising transmission and distribution losses are 37%. This figure is further checked with the Finance department of Delhi Jal Board. Total water billed is estimated to be 300 MGD while the total production of filtered water for supply is 650 MGD and the rate of recovery on revenue demanded is about 57%. Whereas the domestic sector accounts for 80% of total water consumption, its share in total revenue is about 12% of consumption and 45% of revenue, and the industrial sector accounts for about 8% of consumption and 155 of revenue. Table 5 Components of non-revenue water (NRW)

A B C D E F

Authorised

consumption 60%

Billed metered consumption

(including water exported in

bulk)

14% Billed authorised

consumption 50%

Billed unmetered consumption 36%

Revenue

water

50%

Unbilled metered consumption 1% Unbilled authorised

consumption 10% Unbilled unmetered consumption 9%

Unauthorised consumption 2.5% Apparent losses 3%

Metering inaccuracies 0.5%

Leakage on transmission mains 14%

Leakage and overflow at utility’s

storage tanks

0%

System

input

volume

Water Losses

40% Real losses 37%

(transmission and

distribution losses)

Leakage on distribution mains

and service connections up to

point of customer metering

23%

Non-

Revenue

Water

(NRW)

50%

Inadequate preventive management and repair and replacement costs are high: Operation is inefficient – and maintenance is at far lower levels than is needed - with energy and establishment costs accounting for 42% and 45% of the O&M costs (excluding debt charges) respectively. Maintenance at water treatment plants, booster pumping stations and for pipeline consists of mostly corrective or breakdown measures. In mitigation of this situation, it should be mentioned that most water treatment plants are hydraulically overloaded causing adverse operating conditions. However there is very little regard paid to knowledge and awareness of operations at the treatment units. Maintenance is performed with limited tools, resources and information. Minimal preventive maintenance is carried out. The result is a crisis management approach resulting in frequent outages, high repair and rehabilitation costs adding to the poor level of water supply services. This also results in high operating costs (for chemicals, energy, manpower, defective equipments, spare parts and total inventory procedures).



Social Aspects

Many areas do not have access to piped drinking water supply: All planned settlements are connected with DJB supply. But, the figure is different in unplanned settlements. Of the 135 urban villages, all have been supplied with piped water. Of the 567 unauthorized colonies, 560 were supplied by March 2003, including 6029 standposts. The exact number of JJ Colonies is not known as it keeps on varying, but according to Planning Department statistics, 820 JJ colonies have already been provided with piped water out of 1082 (Socio-economic survey: 2003 – 2004). There are 11,533 standposts within the entire system and these are estimated to provide about 49 MGD of water. In the sampled unplanned colonies, 28.92 % of the total households had no direct piped water supply and they relied on alternate sources for their daily water requirements. This figure is surprisingly higher as an earlier report (DUEIP 2001) that explored inequity in water supply observed that 10% of Delhi’s population had no piped water supply at all and 30% had grossly inadequate access. But this report was based on the entire population of Delhi covering both planned and unplanned areas. Statistics of ‘coverage’ of around 90 per cent, and bare figures of quantity of water supplied, tend to hide several realities regarding both the operations of the system, and the experience of customers. The ‘coverage’ figure may relate to installed capacity. What is relevant for the customers is the actual operating capacity of the water supply system, or the average actual supply through in-house connections over a sustained period. The figure of 28.92% could be more realistic as population in the unplanned areas has increased rapidly in the last 5 years without the provision of in-house piped water supply. Absence of safe and reliable water supply in such areas could be a serious bottleneck, which may further hinder the urban growth and in-situ development. The situation is very poor in JJ clusters. According to the statistics, out of 1080 slum clusters at the end of 2003 having population of 30 lakhs, only 75.78% are provided with in-house piped water supply (Socio-economic Survey, 2003 – 2004). Providing them with in-house water supply could be a great challenge to the water utility. Many residents lack legal land tenure and this has a detrimental impact on their ability to engage with the formal sector, whether to raise capital for business, access services, find a job or resist evictions. The DJB generally refrain from supplying services to settlements that lack legal tenure because it could be seen as political acceptance of the settlement as a bona fide place of residence. This can be the case even where policies exist which state that water and sanitation should be made available to all citizens (New Master Plan – 2021, Draft Report). Reasons given for the lack of services are that the land is unsuitable for habitation, inaccessibility, overcrowding or a perception that poor communities are not able to pay for services. Environmental Aspects

Reliance on groundwater amongst both planned and unplanned dwellers is high: Despite the existence of a piped water system at least 36% of the planned population meets 90% of its water need from personal tubewells. Reliance on groundwater becomes almost 100% in many society flats in CGHS category who resort to deep drilling to withdraw groundwater. They have their own water distribution systems that often provide 24×7 water supply. The large-scale extraction of groundwater is a result of widening gap between the demand and supply of water. According to a report released by the Central Ground Water Board (GCWB), Delhi’s ground-water level has gone down on an average by about eight meters in the last 20 years at the rate of about a foot a year. The present levels of groundwater withdrawals exceed the estimated availability. So the present rate of consumption is already unsustainable. The groundwater table is Delhi has depleted to 20 –30 metres in various areas across the city. Compared to a level of 30 – 40 feet at the time of Independence, the water table has dropped to 350 feet at certain places. It is said to be falling at 10 feet per year on an average. Groundwater levels have depleted by 2 – 6m in Alipur and Kanjhwla blocks, 10m in the Najafgarh block, and about 20m in Mehrauli block. In the unplanned areas, very high proportion of rural villages are outside the piped network and in such areas groundwater forms the main source of drinking water. The households dig shallow wells


fitted with handpumps or install motors and draw subsoil water. Owing to this situation of escalating population without a commensurate increase in the availability of raw water, the groundwater in many blocks has been over exploited. This has disturbed the hydrological balance leading to decline in the productivity of wells, increasing pumping costs and more energy requirement. Utilisation of groundwater in Najafgarh and Kanjhawala block is very common which means that people consider this resource as potentially important. Groundwater quality is deteriorating: The quantity of fresh water is very marginal and most of the groundwater reserves consist of brackish or saline water. Therefore, people using groundwater complained that soap did not lather properly or that their skin remained sticky after bath, dry hair and unpleasant smell and salty taste. The quality of underground water is deteriorating and in several places it is unfit for human consumption. Salinity and over exploitation has contributed to depletion and drastically affected the availability of quality water in different parts of the city. In Alipur block saline water aquifer occurs at shallow depth. In groundwater Fluoride content ranges between 1.70 to 4.23 ppm whereas the permissible limit is less than 1.5 ppm. In most part of the Kanjhawla block, chemical quality of shallow groundwater is brackish with electrical conductivity more than 3000 ms/cm at 25°C except north eastern part i.e., all along western Yamuna canal and its tributaries where shallow groundwater is fresh with electrical conductivity in between 1000 ms/cm at 25°C to 3000 ms/cm at 25°C. Fluoride content ranges between 1.80 to 3.0 ppm and the nitrate content is between 67 to 1000 ppm, where as the permissible limit is less than 45 ppm. The increase in the nitrate content is due to discharge for cesspools or due to leakage of sewage pipes. Groundwater of Najafgarh block is saline except along Najafgarh drain and at some places around Samalkah, Baqargarh and Ojwah areas. At Shikharpur and Kanganheri areas the hand pump water at 30 m depth, though marginally saline can be classified within tolerable limits for drinking and irrigation purposes. The saline groundwater covers an area 32 sq km and marginally saline over the area of 129 sq km. A study done by the NEERI for MoEF revealed high nitrate and fluoride concentrations. Fluoride content ranges between 1.52 to 3.25 ppm and the nitrate content is between 48 to 272 ppm. High metallic content, particularly manganese and iron have also been observed in the samples collected. The iron concentration was found to be verying from 4.05 mg/l to 0.337 mg/L. In the city block the quality of groundwater is generally better than the other blocks. The Fluoride content ranges between 1.70 to 2.60 ppm and the nitrate content is between 54 to 850 ppm. Groundwater in maximum part of the South district is fresh and potable with electrical conductivity ranging from 320 to 4130 micro-mhos/cm at 25` C. Electrical conductivity values more than the permissible limit are observed at Deragaon, Molarbund and at Gadaipur. High values of nitrate are found at four locations i.e. at Gadaipur, at Rajokri and at Jaunapur. In north Ghitorni and Andheri More, Fluoride is more than permissible limits. Except chromium concentrations at one locality, ground water is devoid of pollution by heavy metals. A comparison of ground water quality from 1983 to 2000 shows quality deterioration in the central part of the Chhattarpur basin and in the areas around Nizamuddin. In central part of the Chattarpur basin, quality deterioration is mainly because of over-development of ground water resources. Wastewater is not properly conveyed and treated: Around 90 % of the wastewater produced is collected in the sewer system but of this amount, only 60 % is conveyed to treatment works. Around 40 % (600 million litres/day) of the collected wastewater is discharged as raw wastewater into drains and Nallah’s, flowing ultimately in the Yamuna river. This results in severe pollution and nuisance. Currently the effluent quality at most treatment works does not consistently meet the current 20 mg/litre BOD and 30 mg/litre SS effluent standards, with limited coliform reduction. Currently trapping of flows in the drains at their outfall into Yamuna River and directing the flows to the treatment works is the approach. This solution does not resolve the problem of heavily polluted drains and Nallah’s. As the drains also carry treated wastewater and drainage water, the strength of the drain water is low, which hampers the functioning of the treatment processes.



CONCLUSIONS

From the situational analysis of Delhi’s water utility, it becomes clear that the current service level in the sector suffers from serious deficiencies. However, there are some significant points that that need to be addressed urgently. These are:

(a) Quantity of daily supply must meet the customer’s demand for both potable and non-potable end uses.

(b) Quality of water should be adequate to standards. Where there is demand for differentiated quality water and it is sustainable to do so, plan to make such arrangements should be taken on site-specific basis.

(c) Ensuring water availability to unaccessed areas through decentralized ‘end of use’ systems such as water tankers, water kiosks etc. in the short run with comprehensive plan to provide piped water supply in such areas in future. The urban poor often live in informal settlements treated as illegal or unauthorized by MCD and DDA. The lack of legal recognition of these settlements and the corresponding lack of tenure rights of inhabitants can be a major hurdle to securing access to improved water sources. Solutions may require innovative alternatives on the part of municipal governments as well as residents of the informal settlements. Public authorities can grant de facto recognition to informal communities, extending basic water services to them, well before the time-consuming procedures of legal formalization have been completed. Improved access to alternative water sources, such as community taps or water kiosks operated by community micro entrepreneurs, can be provided without waiting for resolution of land ownership claims.

(d) There is an urgent need to rethink the pricing design currently adopted. Willingness to pay can be used to create equity-based policy of water tariffs reflecting property class under which households typically belong. Every claimed advantage of an increasing block tariff (IBT) can be achieved with a simpler and more efficient tariff design with uniform price coupled with a rebate (UPR) for unplanned areas. With a two-part tariff based on a uniform volumetric price coupled with a rebate, the customer’s water bill is based on the sum of two calculations: (1) a fixed charge with rebate for unplanned settlements, and (2) a uniform volumetric tariff set at or near marginal cost of water/or customer’s willingness to pay.

(e) Sub-division of water supply networks into District Metering Units (DMU’s) and “block mapping” programs to detect illegal connections, theft and leakages: When water input to a zone and its usage within the zone is not known, it become difficult to measure the extent of non-revenue water. Zones may either be areas supplied by individual service reservoirs or discrete well-defined areas supplied by a particular branch off the trunk main.

(f) Involvement of customers/residential housing societies or RWAs in monitoring services, installing water meters and detecting leakages, wastages etc. For communities to deliver these roles effectively several conditions must be met: the strategy must be firmly based on people’s needs which can be identified by assessing the threats and opportunities in the community environment; the strategy must be appropriate for the community to undertake, that is it must be in accordance with the community’s capacity; technical, financial and educational inputs to the community must be appropriate or compatible with the needs and community’s capacity; and finally the strategy must belong to the people, including the technical financial and educational inputs. “Champions” for Water Demand Management amongst residential welfare groups should be identified and their role of promoting demand management within the community and within the Municipality should be defined. Specific roles need to be defined for these champions, such as assisting during metering and billing, appearing at various functions, lobbying other departments to conserve water, and many others.


(g) There is an urgent need to generate awareness among customers about the need for reforming the sector. Customers should be made aware of the ‘full-cost’ of water and it should be printed on the backside of water bill. The importance of increasing investment in new infrastructure, as well as for the operation and maintenance of the current system should be disseminated through print and electronic media.

Acknowledgements The authors gratefully acknowledge the support and guidance provided by Dr. Leena Srivastava and Prof. Subhash Chander at Faculty of Policy and Planning, TERI School of Advanced Studies, New Delhi. REFERENCES

Barah, B.C., Sipahimalani,V., and Dhar, P. (1998) Urban water supply and sanitation in Sankar, U. and Mathur, O.P. (eds.) “Economic Instruments for Environmental sustainability”, National Institute of Public Finance and Policy, New Delhi. 53 – 65.

Basu, S.R. and Main, H.A.C. (2001) Calcutta’s water supply: demand, governance and environmental change, Applied Geography, Volume 21, Issue 1, January 2001, Pp. 23 – 44.

DJB (2004) Workshop on vision for Delhi water supply and sewerage sector, Organised by Delhi Jal Board and Water & Sanitation Program, South Asia, March 12 & 13, 2004.

Delhi Urban Environment and Infrastructure Improvement Project (DUEIP)(2001) Detailed Project Report, Ministry of Environment and Forest, Government of India and Planning Department, Government of National Capital Territory of Delhi.

Dutta, V. (2005) Preference Heterogeneity, Public Choice and Willingness to pay for Water Supply Improvements in Planned and Unplanned areas of Delhi, India, International Congress on Environmental Planning and Management, Brasilia, Brazil, September 11 – 15, 2005. ISBN: 85-905036-2-3, p. 24.

Dutta, V., and Tiwari, A. P. (2005) Pricing water - Reflections on the increasing block pricing policy of Delhi’s water utility, Journal of Indian Buildings Congress, New Delhi, Vol. XII, No. 1 (2005), Pp. 233 – 243.

Dutta, V., Chander, S., and Srivastava L. (2005) Public Support for Water Supply Improvements: Empirical Evidence from Unplanned Settlements of Delhi, India, International Journal of Environment and

Development, Vol. 14, No. 4, P. 24, University of California, San Diego, CA.

Dutta, V. (2004) A conceptual inquiry into sustainability criteria for urban water systems, Urban India, Journal of National Institute of Urban Affairs, New Delhi, XXIV, No. 2 (2004), Pp. 89 – 131.

MPD – 2021, Draft Report, Delhi Development Authority, Delhi.

Raghupati, U. P. and Foster, V. (2002) Water: A Scorecard for India, Water Tariffs and Subsidies in South Asia Paper 2, Water and Sanitation Program-South Asia.

Saleth, R. and Dinar, A. (2004) The institutional economics of water: A cross-country analysis of institutions and performance, Cheltenham, UK: Edward Elgar Publishing Limited.

Socio-economic Profile of Delhi (2003 – 2004) Government of National Capital Territory of Delhi, India.

Wolf, T., and Kraft, T. (2002) Health for all – water for all ? The megacity Delhi, Status Report, GEOMED.

Zerah Marie-Helene (2000) Water: unreliable supply in Delhi, Centre de Sciences Humaines, Manohar Publishers & Distributors, New Delhi, pp. 168.



AUTHORS’ BIODATA Venkatesh Dutta obtained B.Sc. (Environmental Sciences) in 1999 from University of Delhi standing first in the order of merit and M.Sc. (Environmental Management) in 2001 with distinction from Indraprastha University, Delhi. He received UGC-Research Fellowship for his doctoral research at TERI School of Advanced Studies, New Delhi. He is the recipient of Bryan Watts Memorial Fund, USA (2005); Vice-Chancellor’s Green Vision Award (2001) and Academic Excellence Award (1996–99). He has travelled to Italy, Greece, France & Brazil for presenting his research work. His major research interest areas are Integrated Urban & Regional Environmental Planning & Management, EIA, Environmental Economics and Policy. He is the member of International Society for Ecological Economics (ISEE), Maryland and International Association of Hydrological Sciences (IAHS), UK. He has 6 research papers and several articles to his credit. Currently he is serving as Lecturer in the Department of Environmental Sciences, B.B. Ambedkar University (a central university), Lucknow teaching postgraduate students.

Anand Prakash Tiwari has a B.E. in Environmental Engineering and Masters in Environmental Planning from CEPT, Ahmedabad. Currently he is a Ph.D. candidate at TERI School of Advanced Studies, New Delhi. His doctoral work is on urban water reform focusing on institutions and delivery mechanisms. He is serving as Senior Infrastructure Officer at Housing & Urban Development Corporation (HUDCO), Delhi. His major research interest areas are Environmental Planning & Policy, Water Resources Management and Institutional Economics. He has widely presented and published his research work both in India and abroad. He is a member of International Society for Ecological Economics (ISEE), Maryland.

Medal for Research on Development (First place)

Theme 2: Institutions, policies, and long-run growth

Urban water institutions are being pulled along the public-private continuum by different

societal forces and actors in emerging and yet-to-emerge socio-political economies in developing

world. Interlinkages between institutions, social welfare and long-term growth of urban economy

are complex but identifiable. The emerging paradigm shift in urban development puts water

institutions in a challenging position because of the imperative citywide needs for service

expansion, enhanced capacity building and internal efficiency improvements. One of the biggest

bottlenecks that could hinder the growth of urban economy and development could be a slow

response to water institutions as inadequate coverage, poor quality, as well as unreliable and

unsustainable supply of water have an adverse impact on cities’ socio-economic development. It

is the issues of organization and institutional development, finance, and human resource

development that are the crucial ones in ensuring future service delivery and long-term growth

of urban water utilities. There is a clear window of opportunities for further change, and, for

providing effective policy guidance. The paper addresses growing perspectives on provision of

continuous water supply with better quality in infrastructurally disadvantaged unplanned areas

addressing concrete policy issues in the context of institutional and sector reform, long-term

growth of water utilities and public policy decision-making. It argues that a lack of political will –

resulting from a traditional belief that the government has a moral obligation to provide clean

water at highly subsidiesd rates or for free – stands in the way of commercializing water utilities

for the benefit of such households, who may now remain underserved. The findings have

important policy implications for gauging public support for water supply improvements in

infrastructurally disadvantaged households including institutional reforms conducive to pro-

poor growth.

Public Support for Water Supply Reforms Public Support for Water Supply Reforms

in Unplanned Sector: Empirical Evidence in Unplanned Sector: Empirical Evidence

from an Urban Water Utilityfrom an Urban Water Utility

Institutions and Development: At the Nexus of Global Change

St. Petersburg -- Russia

January 19-21, 2006

SEVENTH ANNUAL GLOBAL DEVELOPMENT CONFERENCE

Venkatesh Dutta

Documents

List of publications 8 - Information and Library Network ...shodhganga.inflibnet.ac.in/bitstream/10603/6299/17/17_chapter 8.pdf · List of publications (a) ... your family needs and