Social Benefits of Potable
Water , Sanitation and Hygiene Education in Coastal
Kerala
Evaluation criteria for 'Centralized
vs. Decentralized' public provision*
K. Pushpangadan**
1. Introduction
Drinking water and sanitation are
essential commodities for improving the quality of
life by achieving basic capabilities such as better
health status from avoidable morbidity and escapable
mortality, better nutrition and lower poverty levels,
higher longevity, and higher social status. The impact
would be the highest, if hygiene education also forms
part of the provision. Because of the basic nature
of the commodity, it is provided as merit goods and
administered centrally by the government for the poor
in the developing world. However, the decentralized
planning brings an alternative institutional arrangement
for its provision and its maintenance through local
self-governing institutions. The paper provides a
methodology based on the traditional cost-benefits
analysis for evaluating the " centralized and
decentralized " provision of water supply, sanitation
and hygiene education by a case study of two coastal
fishing villages in Kerala.
The paper is mainly concerned with
the evaluation of gross benefits from the provision
of water supply, sanitation and hygiene education
(WATSANGENE). In other words, our objective is to
devise a method for the quantification of likely gains
that accrue to a region (households) if provided with
WATSANGENE. The theoretical framework for achieving
the objective is the Cost-Benefit Analysis (CBA).
CBA is usually employed in two occasions, ex ante
and ex post. In the ex antle analysis, it is meant
for ranking projects for the allocation of scarce
resources so as to maximize social welfare. In the
second situation, it is used for ex post evaluation.
The result of ex post analysis could become input
in the future investment decision in the sector without
making past mistakes. Our study is an exercise in
the ex ante sense, since we are concerned with gross
benefits before investing in the sector.
*The paper forms part of the report,
written jointly with Mr. G. Murugan, on "Social
costs of not providing sanitation in coastal Kerala,
to be submitted to the International Technology Development
Group (ITDG), U.K. We acknowledge the financial help
from ITDG for making this study possible. Of course,
the usual disclaimer follows.
** Associate Fellow, Centre
for Development Studies, Thiruvananthapuram.
CBA technique follows two important
principles in evaluating the costs and benefits. First
it compares with and without project situations in
the estimation of benefits and costs. Second, it makes
a distinction of benefits and costs as incremental
and non-incremental. The method becomes relevant if
the benefits are estimated from the demand functions
of the commodities. If the demand function is not
available/ obtainable from existing data, as is the
case of WATSANGENE in rural areas, then contingent
evaluation technique should be used for obtaining
it. In the case of water supply, it may be estimated
reliably without much difficulty by combining willingness
to pay and the existing tariff for urban supply1.
Similar method, in principle, can be used for sanitation
as well. But the method has serious limitation in
the case of sanitation. It will be extremely difficult
to elucidate the marginal worth of sanitation since
its benefits are not immediate unlike drinking water.
Moreover, the demand for sanitation is derived one
depending on the hygiene awareness and levels of income
of the households. Hence, consumer may judge the worth
of drinking water more accurately than that of sanitation.
Contingent evaluation method may not yield the correct
demand function for sanitation. The method is certainly
not suitable for hygiene education since it only shifts
the demand curve for water supply and sanitation.
In other words, it is an important parameter in the
demand for water supply and sanitation. To sum up,
the demand curve analysis for obtaining the benefits
may be reliable for water supply, less so far sanitation
and certainly not for hygiene education.
The applicability of willingness
to pay method, with all its limitations, is still
possible if the benefits are separately measurable
"with and without" project. This is further
complicated by the economies of scope in the provision
of WATSANGENE. It is well known that impact is much
higher if they are provided jointly rather than separately.
This is attributed to the synergy and complimentarity
among each other in the provision2. Hence,
stable health benefits can be estimated only in the
case of joint provision. Under joint provision, the
aggregate demand function is not properly defined
for arriving at the impacts. Alternative ways should
be devised for the valuation of impacts. An indirect
way of computing the benefits is proposed which involves
the valuation of the factors the beneficiaries would
consider in the willingness to pay for the composite
commodity, WATSANGENE.
The remainder of the chapter is
organized as follows. Section 2 gives an introduction
followed by the analytical framework. Methodology
of the study is provided in section 3 followed by
summary and conclusions in the final section.
2.
The Analytical Frame Work
2.1
Theoretical Background
Possession of commodities, private
/ public, provides the means to achieve certain ends
that a person considers worth living. There are three
ways of judging such achievement: (1) by utility,(2)
by opulence; and (3) by the quality of life2.
The choice depends manly on which perspective of achievement,
actual level or the freedom aspects, is being emphasized
in the evaluation. While utility approach is more
concerned about the actual achievement, opulence and
quality of life, the other two, emphasize more on
the freedom aspect of achievements3. However,
the latter two measures do not address quite the same
dimension of freedom as noted by Sen (1992). While
the measure on opulence provides the means to achieve
freedom, the quality of life captures the extent of
freedom enjoyed by the individuals as reflected in
the living that consists of interrelated functionings.
In other words, Sen argues that functionings are constitutive
of a person's being and assessment of well-being implies
an assessment of the constituent elements. Our study
concentrates on the evaluation of the constituent
elements arising from the commodity, WATSANGENE.
1This method is advocated
by ADB (1999) for the estimation the demand for water
supply.
2See Kumar(1987); Sen
(1992).
The approach can be put in a nutshell
as follows. The commodities provide certain characteristics
that enable the individuals to achieve certain capabilities.
The achieved capabilities, i.e. functionings, of the
persons from the commodity-characteristics depend
on personal as well as social factors. An aggregate
index of the functionings, a measure of well-being,
determines the relative position of the social group
in the social hierarchy. For our commodity, WATSANGENE,
we consider only the basic capabilities. The basic
capabilities would include the following: (1) freedom
from hunger,
(2) freedom from escapable morbidity
and avoidable mortality, (3) freedom to live a full
life, and (4) freedom to get quality education. Let
us study the impacts very closely within the cost-benefit
analysis. This can done in two ways by comparing the
functionings: (1) 'with' and 'without' project on
WATSWAGENE, and (2) 'before' and 'after' the project.
We follow the former method because of the difficulty
in locating the latter for measuring the impact.
3
This section draws heavily from Chapter II of Sen
(1992)
It is well known that travel distance
for fetching water will be much higher without project
than with the project4. In other words,
household could save considerable amount of time with
adequate supply of water and sanitation facilities
with the project. Two aspects of the saved time have
to be valued: (1) the opportunity cost of time, especially
that of females since they usually collect the water
in traditional society; and (2) the reduction in energy
resulting from lower travel time. The latter improves
the nutritional status of the females and reduces
the poverty gap of the beneficiary households. Poverty
gap will be further reduced if the time saved can
be utilized for income generating activities. Hence
both the impacts would reduce poverty levels among
the households. Another significant impact, with the
project, would be the lower incidence of water borne
and sanitation related illness. The household has
the ability to save the amount that otherwise would
have spent on the treatment of water borne diseases.
This would mean, in the long run, households would
have better health status leading to lower mortality
rates and higher longevity. Better health status would
also increase the efficiency of conversion of food
into various nutrients enabling the households to
reduce the poverty gap further. The averted medical
expenses have an income effect on the consumption
of basic items like food, clothing, shelter, education,
etc. If the saved time is that of the girl child,
then it could increase the enrollment rates in school
or improving the performance at school5.
Finally, the quality of environment improves substantially
due to better hygiene awareness on sanitation and
water-handling practices. The improved constituent
elements in the well-being such as lower poverty levels,
higher longevity, lower illiteracy and so on would
take
4 See Pushpangadan et.
al (1998) for explaining lower user rates of public
taps in Kerala in terms of distance traveled.
5 The problem is very
genuine in the northern part of India, unlike in Kerala,
since female literacy rate is very low. However, the
quality aspect may still be relevant for Kerala especially
in the coastal region where considerable amount of
female children's time is spent on fetching drinking
water.
the social group to higher welfare
and social equity. But the level of achievements depends
on the value of the benefits arising from the project.
Among them, three most important benefits are assessed
below.
3. Methodology
The methodology has two parts. First
part deals with the sampling and collection of data.
The second part deals with valuation of the basic
capabilities with and without project.
3.1 Survey and Data Collection
Sample households from two coastal
hamlets - Adimalathura in Kottukal Panchayat and Pulluvila
in Karumkulam Panchayat in Thiruvananthapuram district
of Kerala State - inhabited mainly by fishing community
were selected for the case study1. For selecting sample,
all the households in the hamlets were initially.
From the census of 1892 household, two hundred households
have been selected at random using the circular systematic
sampling technique. The spatial distribution of the
sample is 84 households from Pulluvila, and 116 from
Adimalathura, based on the proportions of the households
in the population. The survey period was from February
1998 to February 1999. On a careful scrutiny, data
from seven households in the sample were found to
be either incomplete or of poor quality. Therefore,
only 193 households were considered for the present
analysis. The issues in the estimation of the benefits
with and without projects are taken up next.
3.2 Valuation of time
Travel time will be saved if the
source of supply become nearer to households which
is the case with the project. The same is also true
of sanitation as well. There are two steps involved
in the valuation process. First step is the estimation
of time saved if the project is implemented and the
second is the valuation of the saved time. Both problems
are dealt with for water supply and sanitation as
shown below.
3.2.1 Water supply
The saved time is estimated as the
difference between time used for fetching water "with"
and "without" project. The total time taken
for fetching water consists of: (1) time taken for
travelling the distance from the source to household,
(2) queuing time at the source: and (3) time for filling
the vessel. The general formula for calculating the
total time saved is given in equation (1)(Abelin,
1997; Eklund and Herman, 1991).
T=(2D/1000 S + q/60 + V/60 Qd)(1000/V)......(1)
Where;
T = Travel time for fetching water
(hours/m3)
D = Distance from home to the source
(meters)
S = walking speed (km/hour)
Q =queuing time (minutes per trip)
V = Volume collected (litres/trip)
Qd = water delivery rate at source
(litres/minute)
m3 =1000 litres.
The formula needs some modification
for our case study. For example, the queuing time
is not valid in our case since the entire water supply
is met from open well with enough space for drawing
the water. This is especially true for Adimalathura.
The discharge time is assumed to be the same with
and without project. Hence, we are concerned only
with time saved due to the reduction in the travel
distance with project. At present, the open well in
Adimalathura is situated about 400 meters away from
the seacoast, which is taken as the travel distance
without project. If the Govt. of India 'norm' for
rural areas is being followed, then water supply household
should get drinking water within a walking distance
of 250 meters in rural areas at the rate of 40 litres
per capita per day (lpcd). Applying this norm, it
can be concluded that the distance saved with project
is 150 meters. For valuation, this has to be expressed
in terms of time saved, the details of which are given
below.
Table 1 Opportunity cost of travel
time for drinking water
The reduction in the travel distance
per trip = 2 (400-250) = 300 meters
Time saved for fetching water per
trip = .3/2 = .15 hrs
Water requirement of an average house-hold
per day per govt. norm 6 = 6.4 *40
Total number of trips per household
per day 7 = 256/20 = 12.5
Travel time saved for fetching water
with project = 12.5* .15 = 1.8 hours
Queuing time with project = 0
Discharge time with project = 0
Total time saved = 1.8 + 0+0 = 1.8
hours
Female time saved per day per household
hours = .63 * 1.8 = 1.08
Opportunity cost of female time per
day per household 8 = 1.08 * 5.5 = Rs.
5.94
Value of female time saved per house
hold per annum with project = 5.94 * 365 = Rs. 2169
The valuation shows that the opportunity
cost of time saved with water nearer to the household
as per Govt. of India norm in rural areas would be
about two thousand one hundred and sixty nine rupees.
Let us examine the case of sanitation.
3.2.2. Sanitation
Baseline survey shows that the traditional
latrine technology is not suitable for coastal belt
especially in water logged regions with shallow ground
water table. In addition, coverage of latrine is very
low because of the high incidence of poverty in the
region. As a result, males go to seashore and females
to nearest open space for defecation and other sanitation
activities. The resource mapping of Adimalathura shows
that the open space is about 100 meters from the drinking
water source, i.e., dug well. Energy expended by males
for sanitation is not assessed because the distance
to seashore is very negligible and we do not have
reliable information on the proportion of males using
the open space for defecation. Hence, we consider
the estimation of energy expenditure by females only.
We assume that females above the age of ten only go
for defecation to open space. It is also assumed that
one trip per day, they combine water supply and sanitation.
Hence, only distance not included for water supply
is being considered here for avoiding double counting.
This would mean that from the total distance of 500
meters to the open field, 150 meters should be excluded
in the travel distance for sanitation.
6 Government of India norm is 40
litres per capita per day in rural areas.
7 It is assumed that a person carries,
on an average, one bucket and one pot of water per
trip. The volume of a bucket is taken as 12 liters
and that of a pot is 8 liters. Hence, the total volume
of water carried per trip is 20 liters.
8 The shadow wage rate for the valuation
of the female time is the rate of the domestic help,
i.e., Rs. 45 per day.
Table 2 Opportunity cost of travel
time for sanitation
Distance traveled by females per
day for sanitation meters = 2 (250+100) = 700
Average number of females (>ten
years of age) per household = 2.4
Total travel distance per day per
household = .7 * 2.4 = 1.68 km
Travel time per household per day9
= 1.68/2 = .84 hrs
Travel time of adult females10
= .84* .63/.77 = .68
Value of female time per day (Rs.)
= .68 * 5.5 = 3.74
Value of female time per household
per year (Rs.) = 3.74 * 365 = 1365
The value of time lost without sanitation
coverage is estimated to be one thousand and three
hundred and sixty five rupees. Let us examine the
energy expenditure
incurred by the households without
project.
3.3
Valuation of Energy
3.3.1
Water supply
We have already estimated that average
travel time saved per day per household for fetching
water is 1.8 hours with project. In addition, it would
also save the energy expended for walking to and from
the source. The following methodology is used for
valuing the energy saved with the project as shown
in Table 3.
Table 3 Value of energy expenditure
for water supply
Total travel time saved per day per
house hold with project = 1.8 hours.
Female time11 = 1.4 hours
Male time = 0.4 hours
Energy expenditure of female12
= 1.4*1.37*60 = 117.6 cal.
Energy expenditure of Males = 1.63*.4*60
= 39 cal.
Total energy expenditure per day
per
household = 117.6 + 39 = 156.6 cal.
Value of energy expended per house
hold per month = 286.1*4698/2400
Value of energy expended per household
per year (Rs.) = 286.1*4698*12/2400
=Rs. 6722
3.3.2 Sanitation
In the case of sanitation, the energy
expenditure is calculated for female adult and female
children. The details are given in Table 4.
9 It is assumed that average speed
of walking in sandy terrain is 2 kms/hour.
10 The survey shows that drinking
water is fetched by adult females about 63% of the
time and 14% of the time by female children.
11 The survey shows that about 77%
of the time adult females or female children bring
the water.
12 This is estimated by applying
he energy loss per minute for an adult female of weight
45 kg for walking in coastal belt. Note that the speed
of walking is assumed to be 2km/hour, half the normal
speed of walking 4km/hour. It is found that walking
in sand is almost twice as costly as walking in a
hard surface (See Merdle, et.al (1998): 150 for the
details). For males the weight is assumed to be 55
kg.
Table 4 Value of energy expenditure
for sanitation
Walking distance (one way) for open
space for defecation = 500 meters
Total travel time of the household
= 500 *2 average females per household
Number of females per household =
1000*2.7 = 2.7kms
Travel time = 2.7/2 = 1.35 hrs.
Energy expended for sanitation per
day per household = 1.37*1.35*60 = 120 cal
Value of energy expended at poverty
line13 = 286.1*120*30/2400
Value of energy expended per household
per year = 12*286.1*120*30/2400 = Rs. 5149
3.4 Estimation of health
expenditure
Health benefit with project consists
mainly of averted treatment expenditure, both private
and public, and the opportunity cost of days lost
during sickness (ADB, 1999; Albin, 1997). Illness
affects the income of the households in three ways.
First is the income of sick person if he/she is gainfully
employed. Second is the household time lost due to
hospitalisation and/or nursing the sick. Finally,
the loss of income due to decline in the productivity
of the person until he/she recovers fully. Before
proceeding to evaluate the benefits, let us be consider
some of its limitations.
Information collected from the households
shows the prevalence of both long-term and short-term
illness. The long-term diseases are mostly confined
to the reproductive tract of women mainly caused by,
according to public health professionals, the lack
of hygiene awareness and inadequate provision of water
supply and sanitation facilities. The cost of the
long-term illness is very difficult to calculate unless
it reaches an infectious stage requiring medical treatment.
But this information is simply not available from
the households and excluded from the treatment costs.
As a result, only the health expenditure on short-term
illness is considered, the extent
of underestimation is difficult to
assess. The health expenditure can be divided into
two: public and private. The former refers to the
expenditure by the state and the latter by the households.
Our objective is to value both components.
13 the conversion of energy into
value is based on the poverty line in 1999 prices.
3.4.1 Public expenditure
Public health expenditures on diseases
very according to season. This is especially true
for water borne and sanitation related diseases in
water logged regions during the monsoon. Moreover,
the chances of drinking water getting contaminated
through fecal matters during summer is likely to be
very low because of deep ground water table and high
temperature during the season. Our key informant interviews
with medical personnel, private and public, suggest
that the rate of occurrence during rainy season is
about three to four times that of summer. As regards
public expenditure for the sample region there is
no record available either with the primary health
centre on with the headquarters. Even if it is available,
it is highly aggregate and the illness specific cost
may be very difficult to obtain from it. Therefore,
we have taken the annual per capita state health expenditure
in the rural area as the public expenditure per household14.
The details are given below.
Table 5 Estimation of public expenditure
per household
a) Per capita health expenditure
for the year, 1997-98 = Rs. 159
b) Per household expenditure = 159
* average household size = 159 *6.5 = Rs. 1034
c) Attributable cost to water-sanitation
related illness15 = .67*1034 = Rs. 692
d) Attributable annual cost per household
in 1999 prices = Rs. 796
Note that this is recurring public
expenditure per annum that can be complete by saved
if water and sanitation related illness can be prevented
with WATSANGENE. Let us now move on to the estimation
of the second component, private expenditure per household.
14 The health expenditure is taken
from GOK (1998)
15 See the details of this method
of allocation Brown and Silbey (1986). The proportion
of water borne and water related illness treated in
Pulluvila, according to the medical officer, is about
67% of all the illness.
3.4.2 Private expenditure
Private expenditure comprises mainly
of medical expenses incurred by the household and
the opportunity cost of time, indirect as well as
direct, lost during the illness. The annual expenditure
can be assessed only if the seasonal nature of the
illness is known. For assessing this, sample households
were revisited during the North-East monsoon. In the
re-survey, it was found that about 57 percent of the
households had been again infected by water borne
and sanitation related illness. The repeated infection
has the following implications. During rainy season,
these households have a meagre income from fishing
due to their inability to venture into the sea with
the traditional technology. An increase in health
expenditure, in such condition, would mean a corresponding
reduction in expenditure on necessities such as food.
As a result, the poverty levels of households increase
during the season and most of them may have to resort
to borrowing at a usurious rate of interest ending
in perpetual indebtedness. The indirect burden is
extremely difficult to assess and not considered in
the estimation. With these limitations, per household
private cost is estimated below.
Table 6 Estimation of private expenditure
per household
a) Average treatment cost per household
during summer16 = Rs. 163
b) Percentage of households with
recurrence of water-borne diseases in the post monsoon
(North-East) = 57%
c) Additional treatment cost per
household during North-East Mansoon = 163 *0.57 =
Rs. 93
d) Estimated increase in expenditure
per household during South-West mansoon17
= Rs. 93
e) Annual per capita treatment cost
adjusted for seasonality per household = 163+93+93
= Rs. 349
f) Opportunity cost of man days of
bystanders lost per household in summer = Rs. 75
g) " in North-East monsoon =
57*75 = Rs. 43
h) " in South-West monsoon =
Rs. 43
i) Annual cost of man days per household
adjusted for seasonality = 75+43+43 = Rs. 161
m) Annual Private expenditure per
household (e+f) = 349+161 = Rs. 510
The total annual expenditure per
household, private and public, is Rs. 1306 which is
likely to recur until water borne diseases are eliminated
completely by providing water, sanitation and hygiene
education. The present value of the recurring expenditure
at 12% discount rate is twelve thousand one hundred
and forty six. The value of major benefits accruing
from the implementation of the project is given in
Table 7.
Table 7 Gross benefits from WATSANGENE,
1999
|
(a) Value of time
|
|
Water supply
|
Rs. 2169
|
|
Sanitation
|
Rs. 1004
|
|
(b)
Value of Energy
|
|
Water supply
|
Rs. 6722
|
|
Sanitation
|
Rs. 5149
|
|
(c) Health expenditure
|
|
Public
|
Rs. 796
|
|
Private Rs. 510
|
|
Total
|
Rs. 16711
|
Source: Table 1-6
Table 7 clearly shown that the social
benefits per household annually with the project is
Sixteen thousand seven hundred and eleven. In other
words, this is the maximum value of WATSANGENE per
household in the region.
16 Treatment cost is the
sum of hospital cost and transporation coat. In the
hopsital cost includes cost on medicine, laboratory
charges, consultation fee and room rent for hospitalisation.
17 The adjustment for
seasonality is the same as in the case of treatment
cost.
4. Summary and conclusions
The benefits of water supply, sanitation
and hygiene education have been evaluated within the
framework of "commodities and capabilities".
The benefits are estimated using cost-benefit analysis
by a case study of two coastal fishing hamlets in
Thiruvananthapuram district in Kerala. Three major
benefits were evaluated using the survey method: (1)
Time saved: (2) Energy saved; (3) averted health expenditure.
The opportunity cost of time saved per household per
annum with project is estimated to be Rs. 3534. The
value of the energy expended that would be saved with
project would be Rs. 11,871. The averted cost of treatment
for water and sanitation related illness is Rs. 716
for the state and Rs. 510 for the household annually.
The gross value of benefits with project would be
Rs. 16711 annually. The present value of the recurring
benefits per household would at a discount rate of
12% , the standard rate usually meant for infrastructure
projects, is Rs. 1,51,188. This unusually higher value
of present value would even justify market borrowing
for the provision, the piped water supply.
This evaluative method can be used
for the choice of centralized and decentralised provision
of WATSANGENE.
References
Abelin, C., (1997), The application
of a cost benefit approach on rural water supply projects,
Minor filed study series. No. 81, Department of Economics,
University of Lund, Sweden.
Asian Development Bank, (1999), Handbook
for Economic Analysis of Water Supply Projects, Manila.
(mimeo.)
Brown , Stephen J and D. Sibley,
(1986) The theory of public utility pricing, (Cambridge
University) Press, Cambridge).
Government of Kerala (GOK), (1995),
Handbook of Statistics, Rural Development Deptt;,
Trivandrum.
GOK, 1998, Economic Review, State
Planning Board, Thiruvananthapuram.
Sen, A. (1987), Commodities and Capabilities,
Oxford University Press, New Delhi.
Sen. A. (1992), Inequality reexamined,
Oxford University Press, New Delhi.
Sen, A. (1993), Life Expectancy and
Inequality Some Conceptual Issues in Bardhan P, Chaudhuri
MD. And Krishnan TN(ed) Development and Change, Oxford
University Press, Bombay.
----------, 1996, Health, Inequality
and Welfare Economics, B.G. Kumar Endowment Lecture
1995, Centre for Development Studies, Trivandrum.
Sen, A., (1998), Mortality as an
indicator of success and failure, The Economic Journal,
Vol. 108, (1-25).
Pushpangadan, K. (1998), Proceedings
of the International Workshop on Returns from financing
WATSANGENE, Centre for Development Studies, Trivandrum.
UNICEF, (1995), UNICEF strategies
in Water and Environmental Sanitation, New York.