Water Resources Development,
Vol. 24, No. 4, 593–607, December 2008
Cost – Benefit Analysis of Restoring
Buriganga River, Bangladesh
KHORSHED ALAM
School of Accounting, Economics & Finance, Faculty of Business, University of Southern Queensland,
Toowoomba, Queensland, Australia
ABSTRACT The focus of this paper is to perform a cost – benefit analysis to determine the economic
efficiency of the restoration of a dying river in Bangladesh, namely the Buriganga River. The benefits
of the restoration programme are derived by using market data and employing benefit transfer and
contingent valuation techniques. The values generated by this approach are integrated into the
framework of a cost – benefit analysis, which showed a benefit-cost ratio of 4.35. This demonstrates
that the restoration of dying rivers in developing countries is not only an environmental imperative,
but is also socially and economically justifiable.
Introduction
National and local governments throughout the world are now engaged in restoring
degraded rivers and rehabilitating some of the ecological damage being done to their
environment. This is traditionally perceived as a public responsibility with very limited
contribution from the private sector. The issue is particularly difficult to tackle in
developing countries where environmental restoration activities compete with other
priorities such as poverty alleviation, basic education or health care. The scarcity of
financial resources in these countries, including in the public domain, is often an
unsurmountable barrier to the restoration work aimed at improving the ecological health of
rivers. One particular problem authorities often encounter is to evaluate the desirability of
public funding for such restoration programmes. There is a scarcity of information about
both the desirability of funding and the benefits, particularly non-market benefits that such
activities can generate. Costs of such activities are always considered enormous, while
benefits are often underestimated, because part of these are not transacted through market
mechanism and thus do not bear any price tag. This makes it difficult to assess the viability
of undertaking any river restoration programme. As a consequence, restoration of
ecosystems, preservation of natural resources and development of degraded environmental
resources frequently bear the burden.
There has been significant progress in developing concepts and theories to assess the
value of the natural environment, however, their application is largely limited to
Correspondence Address: Khorshed Alam, School of Accounting, Economics & Finance, Faculty of Business,
University of Southern Queensland, Toowoomba, Queensland 4350, Australia. Email: [email protected]
0790-0627 Print/1360-0648 Online/08/040593-15 q 2008 Taylor & Francis
DOI: 10.1080/07900620802224486
594
K. Alam
developed economies. The differences between developed and developing countries in
terms of socio-economic parameters may imply that there might be corresponding
differences in the relative importance of specific sources of benefits (Russell, 2001).
Another dimension that is lacking in the benefit estimation is to perceive the river
restoration not only at the watershed scale, but also catchment scale. Sources of pollution
include both point and non-point pollutants that spread throughout the catchment.
A number of different terminologies have been used to describe a variety of
interventions to improve the river ecosystems, such as ‘cleanup programme’ (Markandya
& Murty, 2000), ‘river restoration’ (Tunstall et al., 1999) and ‘water quality improvement’
(Carson & Mitchell, 1993; Barton, 1998). In this study, the term ‘river restoration’ has
been adopted to indicate an integrated and holistic approach. Traditionally, restoration of
river ecosystem issues has been perceived as maintaining of river flow, controlling
pollution (particularly industrial effluents) or establishing wastewater treatment plants. In
some studies, restoration is perceived as a return to the original condition which is a
‘complex and difficult task’ (Gore & Shields, 1995). Being a new delta, the river
morphology is continuously changing in Bangladesh and thus it would be hard for some
river ecosystems to return to their previous conditions. Therefore, in this study, the term
restoration is used to refer to a wide range of activities, including the recovery of selected
ecological and biological functions, creation of some physical structures and promotion of
biotic community formation in river ecosystems. The approach taken in this study
addresses both point and non-point sources of pollution.
To demonstrate the economic rational for an intervention in restoring vulnerable rivers,
a case study was chosen for the restoration of the ecological health of the Buriganga River
in Bangladesh. This study’s primary objective was to determine the economic value of
restoring the ecological health of the river and to assess the desirability of investment from
the perspective of the society. Although the background is from a vulnerable river in
Bangladesh, the methodological framework is widely applicable to degraded rivers
elsewhere.
Materials and Methods
The Case Study
Dhaka City, now the capital of Bangladesh, was established as a provincial capital of the
Mughal ruler on the southern bank of the Buriganga River in 1610, although the settlement
is known to have been in existence since the 7th century as a small riverside township.
Since then, the city gradually grew on the south and western banks of the river (known as
‘old Dhaka’).
The Buriganga River is an important part of Dhaka City’s urban landscape, ecology and
economy. Not only early settlements developed, concentrating on the riverbank, it has
been sources of domestic water supply, groundwater recharge and recreation and fishing
sites, and served as major transportation route and flood control and drainage outlet. It has
also been used for agricultural, sanitary and industrial purposes. Always a threat through
periodic floods, the river remained an intrinsic part of the city until the early 1980s.
However, since the 1980s, intensive human intervention, unplanned urbanization and
population pressure have greatly influenced its flow and ecological function. Once a
determining factor for trade and urbanization and source of industrial growth
Cost – Benefit Analysis of Restoring Buriganga River
595
(predominantly small and medium scale enterprises), it was later degraded by industrial
(particularly tannery) and sprawling residential developments along its banks and in its
catchment, and due to the use of waterways as sewers for carrying urban solid and liquid
wastes. The situation deteriorated further because of the lack of appropriate waste
management infrastructure and sewerage disposal system in the city, particularly in the
vicinity of the river. Not only were many landfill sites established close to the river,
households along the river also directly dispose of their waste (both solid and liquid) into the
river. Furthermore, since the early 1980s, unscrupulous people started to seize the off-shore
land, building illegal encroachments without waste disposal and sanitation facilities. As a
result of these human actions on the one hand, and failure by the authority to enforce rules
and regulations to save the river on the other, the Buriganga River is dying biologically and
hydrologically. Studies show a steady decline of major parameters of water quality (Alam,
2002). In recent years, no fish and other aquatic species are found in the river during the dry
season and in many sections, the river looks like a canal due to shrinking of its width.
A hypothetical restoration programme (Buriganga River Restoration Programme or
BRRP) was designed to improve the ecological health of the Buriganga River. Unlike other
rivers, to be effective a restoration programme for the Buriganga River requires addressing
both point and non-point sources of pollution. Both benefits and costs are derived from this
hypothetical programme and then collated within the framework of an economic analysis
to assess the economic viability of public funding which is elaborated below.
Cost – Benefit Analysis
Cost – benefit analysis (CBA) is the method commonly used to analyze and evaluate public
interventions. CBA is an analytical tool used to quantify and value the cost and benefit of
an intervention (be it a project, programme or policy) occurring at different points in time
within the framework of a common yardstick (e.g. net present value). The key objective is
to facilitate the decision making with the efficient allocation of scarce resources from the
perspective of the society (Boardman et al., 1996). Therefore, it does not dictate choices,
rather it provides information to policy makers to decide whether allocation of resources
for a particular project is economically efficient.
The CBA is a straightforward appraisal technique when all the costs and benefits of the
intervention are identified and valued. It becomes complicated when either components of
cost or benefit are not easily quantified or valued. The costs of implementing
environmental programmes can be a private good in the sense that the information is
available in the market. Therefore, the observed prices can be used in the CBA. On the
other hand, benefits that have market value can be monetized using market data, whilst
there is no information either from conventional markets or related markets for some
benefit items, termed as non-market benefits. These benefits need to be measured using
non-market valuation technique, which is elaborated later.
Once all benefits and costs are expressed in a common measurement unit (i.e. money),
these can be incorporated into a decision-making rule that will reflect the perspective of
the society as a whole. This means both costs and benefits can be used together to take an
integrated decision about whether benefits outweigh costs and such an undertaking from
the point of view of society as a whole is economically efficient. There are three criteria
commonly employed within the framework of the CBA. These are net present value
(NPV), internal rate of return (IRR) and benefit-cost ratio (BCR).
596
K. Alam
Estimating Benefits
The Buriganga River restoration programme has been put in a broader perspective in this
study, beyond mere water quality improvement. That means the policy scope is much more
holistic. The restoration programme is considered as a mix of public and private good; part of
which will be passed onto the residents through market transactions and the rest will be
available free of charge. Benefits of the restoration are expected to accrue to the residents both
directly (mainly use values) and indirectly (mainly non-use values), such as:
. directly through increased provision of goods and services (e.g. an increase of
navigability), improved quality of the existing facilities (e.g. improvement of water
quality), and newly created facilities (e.g. use of river water for domestic needs, and
tourism and recreation); and
. indirectly through the existence of a healthier environment both for current and
future generations.
Benefits of the restoration programme are identified within the framework of total economic
value (TEV) which is categorized into use and non-use benefits. Both generic and specific
categories of these benefits and possible methods of deriving values of these benefits are
illustrated in Table 1. Although theoretically all categories of use value can be estimated
using market data as markets exist for them and this is true for almost all Western
economies, however, some components of use value do not have any established market in
developing countries such as Bangladesh (e.g. improved water quality). Therefore, these
benefits along with conventional non-use benefits are measured using contingent valuation
method. Furthermore, in some cases reliable data for quantification of use benefits is absent
(e.g. existing or projected figure of fish catch in the river). In that case, benefit transfer (BT)
approach is used. Details of the CVM and BT are elaborated below.
Contingent Valuation Method
Contingent valuation method (CVM) is used to estimate the non-market benefits of the
BRRP. This survey-based technique has been increasingly used in the economic valuation
of water resources (Carson & Mitchell, 1993; Garrod & Willis, 1999). An advantage of the
CVM is that it can measure both the use and non-use values of an environmental change.
As explained earlier, due to the lack of established market and information about monetary
values associated with many of the uses, only some direct use benefits are estimated using
market data; all other components of TEV, shown in Table 1, are measured using the CVM.
In this approach, the survey participants are asked to state their preferences, either in the
form of willingness to pay (WTP) or willingness to accept (WTA), for a specified level of
change in the provision of a good. As Alam (2006) raises the issue, the way valuation
questions (e.g. dichotomous choice and payment card) are raised before respondents does
not take into account the local context in developing countries such as Bangladesh where
many of the economic activities are non-monetized, and also many of the transactions are
conducted in non-monetary ways. For the low-income respondents, it does not make sense
to express willingness to pay from their ‘disposable’ income. Inability to offer direct cash
payments probably does not indicate that the poor people care less for an environmental
improvement. These important aspects of preference for an environmental change are
ignored in framing conventional CV questions in the context of developing country.
Cost – Benefit Analysis of Restoring Buriganga River
597
Table 1. Taxonomy of the total economic value of the Buriganga River restoration programme
Components of TEV
Direct use values
Consumptive uses
† Water transport (navigation)
† Increased fish production
† Better quality water for domestic and
industrial uses
Technique to measure value
Market data
Market data/BT
Market data
Non-consumptive uses
† Increased housing and land values
† Jogging and walking along the river
† Tourism and recreation
† Educational, scientific and cultural purposes
† Improved health benefit
† Bathing, washing and boating
Market data/BT
CVM
Market data
CVM
Market data/BT
CVM
Indirect use values
† Riverbank erosion prevention
† Flood control
† Biodiversity
† Ecological function (e.g. watershed protection)
† Pollution assimilative capacity
CVM
CVM
CVM
CVM
CVM
Non-use values
Option values
† Biodiversity
† Future use (e.g. recreation option)
† Potential gene pool
CVM
CVM
CVM
Existence values
† Satisfaction from knowledge of existence of
clean water
† Aesthetic benefits
† Biodiversity
† Spiritual, cultural heritage and religious
Bequest values
† Value arising from the knowledge that
the river remains healthy and viable
and will persist for future generation
† Biodiversity
CVM
CVM
CVM
CVM
CVM
CVM
In order to capture this aspect of the respondents’ preference, the conventional approach
of framing valuation question is extended in this study. Instead of asking monetary
contribution for an environmental improvement, this approach asks respondents’ a twopart question; in the first part it asks respondents about their willingness to contribute in
terms of money, and then in the second part, irrespective of their willingness to contribute
money, it asks whether or not they would be willing to contribute in terms of time. These
two parts question—willingness to contribute money (WTCM) and willingness to
contribute time (WTCT)—together represent a respondent’s total willingness to contribute
(TWTC) to a proposed environmental change. Respondents’ preference for an
environmental improvement deriving by both asking monetary and time contribution is
still consistent with the utility theory (Alam, 2006).
598
K. Alam
On the basis of this theoretical construct, a survey instrument was designed to derive the
non-market benefits of the BRRP from the residents of Dhaka City.
Survey Design
A stratified random sampling technique was employed to select a sample of 400
households, dividing the study area into two constituents: ‘Buriganga River Area (BRA)’
(i.e. people living near the river) and ‘Outside Buriganga River Area (OBA)’ (i.e. those
living at a greater distance). This sample size was considered to be a reasonable number
for the approximation of the population, following other CV studies. This stratification
approximately corresponds to the demarcation of ‘old Dhaka’ and ‘new Dhaka’ in the
study area, although some adjustments were required. Stratification not only ensures
representative sampling, but also helps to examine any variation with regard to various
attitude-behaviour attributes (Czaja & Blair, 1996; Sarantakos, 1998). Whittington (2002)
also recommended split sampling for CV studies in developing countries.
Dhaka City, within the defined boundary of Dhaka City Corporation area, is divided into
ten thanas (police stations or administrative units). The area within each thana is divided
into wards, which are further divided into mahallas. A ward is the lowest administrative
unit and run by elected local government representatives. Each mahalla contains one or a
few streets and a varying number of households. According to the 1991 census, DCC
consists of 14 thanas, 90 wards and 659 mahallas.
Three thanas from a total of five within the BRA and four thanas from a total of nine
within the OBA were selected randomly at the first stage of stratification for the survey.
The sample of 400 households was then drawn from these seven thanas in proportion to
each thana’s population, resulting in 42– 75 households being assigned to each thana. This
process resulted in selecting 197 households from the BRA and 203 households from the
OBA, spread over 14 wards and 29 mahallas within seven thanas of the study area. All
these selections were based on a random procedure. After selecting a household, survey
respondents were chosen within a household using a ‘random number table’.
Two focus group discussions (FGDs), each group consisting of 16 adult resident members,
were conducted in the two constituents of the study area. Participants were recruited in such a
way that the group was representative of the residents by including a balanced number of men
and women, a range of ages, educational attainments and different income groups. In each
FGD, elements of the interview schedule, such as the payment vehicle, the length of payment,
the description of the scenario and environmental improvement and use of photographs were
described. The FGDs were also used to determine what additional information respondents
desired in order to make an informed decision. Upon completion of the FGDs, a draft survey
instrument incorporating the valuation scenario and bid values was developed.
A pilot survey was undertaken to scrutinize the survey design and procedure with 20
households in different parts of the study area. In addition to providing reaction to the bid
levels, it was used for two other functions: (1) to amend questions in order to ensure that they
would be better understood by the respondents (in some cases, sentence structure and
wording were changed); and (2) to amend the interviewing technique so that the methods
could be improved and standardized. Particular focus was given on how participants
responded to the bid values and hypothetical scenario presented to describe potential
environmental improvements. The pilot survey findings resulted in a number of adjustments
to the interview schedule, mostly changes in the wording of some questions, changing the
Cost – Benefit Analysis of Restoring Buriganga River
599
timing of the survey, identifying unclear or ambiguous questions and assessing the adequacy
of response choices. The overall response rate for the main survey was 85.10%.
A statistically significant relationship was found between the sample and the population
values for literacy rate, occupation and household income. The median monthly income of the
sample was Tk7238 in 2001, while the average household income for residents in Dhaka City
in 1999 was Tk7592/month (PC, 2000) and after consumer price index adjustment, Tk7739 in
2001. A comparison for the level of education between the survey and population was not
possible because only respondents aged18 years and above were selected for the survey.
Benefit Transfer
A full-scale benefit estimation of an environmental improvement requires detailed studies,
often involving several methods other than the CVM described above. However, due to
constraints of time and financial resources to conduct a primary study (Desvousges et al.,
1992; Navrud, 1996; Brouwer, 2000), researchers often employ a benefit transfer (BT)
approach to assess values in a case study of interest. Benefit transfer involves the application
of primary non-market valuation estimates to a secondary setting for which the original study
was not expressly designed (Brookshire & Neill, 1992). As such, valuation results may be
‘transferred’ from one location to another location or from one point in time to another or both.
Benefit transfers can take at least three different forms, namely point value transfer,
benefit function transfer and meta-analysis. However, this study employs the point value
transfer technique, which involves calculating the benefits for the policy site based on the
mean willingness to pay (WTP) from the study site. In this case the benefit estimate from
the study site is directly transferred to the policy site. Applying this technique, benefit
estimates are transferred:
. within study area for increased housing and land values and improved health
benefit for the similar environmental good; and
. between similar geographic region in the catchment for the increased fishing
benefits for the similar environmental good.
Results and Discussions
Estimates of Non-market Benefits
To elicit a WTCM from respondents in Dhaka City, the hypothetical scenario was offered
at one of seven different price levels in the form of a payment card. These payments,
ranging from Tk1 to Tk2000 and more, were proposed to be collected each month along
with water bill for the next 10 years (Taka (Tk) is the Bangladesh currency, US$1 ¼ Tk70
as of June 2006).
Respondents expressed a strong willingness to contribute for the restoration of the river.
Out of 400 surveyed respondents, 94.5% supported the restoration programme. However,
25.5% of the respondents said that they would vote in favour of the restoration at the
various prices at which it was offered to them. Extrapolating this value to the resident
households of Dhaka City, this generates approximately Tk176 million per year. On the
other hand, 32.75% of the respondents were found to be willing to contribute time. When
this time contribution is monetized using wage and salary rates of the study area and
extrapolated to the resident households, the annual monetary value stands as Tk270
600
K. Alam
million. The total WTC value for the residents of Dhaka city derived by adding these two
values is about Tk446 million per year.
There is significant unemployment and underemployment in the study area, which
distorts the opportunity cost of labour. Therefore, residents’ contribution in terms of time
for the environmental improvement needs adjustment. This is done using the shadow wage
rate. Residents’ contribution of time (WTCT) is shadow-priced by dividing the categories
of work into skilled and unskilled labour. Respondents’ average value of willingness to
contribute time after adjustment is estimated as Tk48.66 per month which provides per
annum value for the time contribution as Tk211.78 million. Therefore, the total
willingness to contribute per year for the non-market benefits of the Buriganga River
Restoration Programme is Tk387.69 million.
Estimates of Total Benefits
The value of the components of the market benefits is estimated in this section using
market data, secondary sources and benefit transfer. Market data are derived either from
observed market prices or imputed from related markets for the components of the market
benefits. However, in estimating economic value, market prices of goods and services,
where necessary, are corrected for market imperfections and policy failures that distort
them. These are elaborated later. The market benefits as identified are: (1) increased
housing and land values; (2) improved health benefits; (3) cost saving for domestic and
industrial water uses; (4) increased navigation; (5) increased value of recreation and
tourism activities; and (6) increased fish production. The estimated value of these benefits
over 10 years is shown in Table 3.
The benefit flow of the BRRP also includes (1) residents’ willingness to contribute money
(WCTM) (derived in Table 2); (2) residents’ willingness to contribute time (WCTT) (derived
in Table 2); (3) the net sales revenue (total expenditure minus total revenue and user fees)
raised from providing solid waste service to households; and (4) the levy collected from
tannery industries for treating wastewater (an amount of Tk7.40 million revenue per annum).
When using benefit transfer approach, in addition to the consumer price index (CPI)
adjustment, values are adjusted upward by the percentage increase of residents in Dhaka City.
Population and household growth rates in Dhaka City are taken as 4.06% and 5.73%
respectively according to BBS (2001). Furthermore, it is assumed that a continual
Table 2. Estimate of yearly non-market benefits
Category
Amount in Tk
Households’ average WTCM per month (Proportion of household
WTCM: 25.50%)
Total number of households in
Dhaka City
Annual value of monetary contribution
(WTCM)
Households average WTCT per month (Proportion of household
WTCT: 32.75%)
Annual value of time contribution
(WTCT)
Total estimated annual non-market benefits
51.91
1.11 million
175.91 million
62.04
270.02 million
445.93 million
Cost – Benefit Analysis of Restoring Buriganga River
601
improvement in the water quality and surrounding environment, and development of physical
infrastructure, would lead to a continuous increase in some of the benefit components. An
annual escalation factor between 5–10% is used to capture these incremental benefits. The
components of benefit flow estimated over a 10-year period are shown in Table 3.
In terms of contribution to the level of market benefit, the increase in land and rental
housing values accounts the highest share (77.71%) followed by cost saving for domestic and
industrial water uses (19.20%) and improved health benefit (2.27%). The total economic
benefit of the BRRP is estimated to be Tk15 068 million. The share of the total non-market
benefits is over one-third of the total benefit over the 10-year period. In addition, the adjusted
WTCT accounts for 54.62% of total non-market benefits. This clearly indicates the need to
account for all benefits when estimating the value a river has to a community and that the share
of the residents’ contributions (through payments or time) is too significant to be ignored.
As is the case with many waterfront improvement projects, an investment in
environmental improvement would stimulate re-development of adjacent areas, including
residential, industrial and public spaces. This study focuses on the immediate benefits
within the study area and is hence a conservative estimate of potential advantages from
improving the environmental health of the river. Furthermore, where there are numerous
benefits that may impact on the disposable incomes and standard of living of residents,
subsequent production, income and employment multiplier effects can generate additional
benefits. Although they may be substantial, this study did not consider such effects.
Estimates of Total Cost
The main cost components for the restoration programme include: (1) removal of illegal
structures from the river and construction of an access road and walkways; (2) solid waste
Table 3. Total benefit of the restoration programme over 10 years (million Tk)
Components of benefit
Valuation method (study source)
Market benefit:
Increased housing & land value
Benefit transfer (Haque et al., 1997)
Cost saving for domestic &
Market data
industrial water uses
Improved health benefit
Benefit transfer (Haque et al., 1997)
Increased navigation
Market data
Increased value of recreation &
Market data
tourism
Increased fish production
Benefit transfer (Hill & Hanchett, 1995)
Total market benefit:
Components of non-market benefit:
Willingness to contribute money CVM
Willingness to contribute time
CVM
(adjusted)
Total non-market benefit:
Other revenue:
Net sales revenue from
Market data
solid waste service
Levy collected from tannery
Market data
industries
Total benefit
Estimated value
4366.01
1078.70
127.33
23.60
17.00
6.05
5618.70
2289.62
2756.37
5045.99
4359.28
44.40
15068.36
602
K. Alam
management; (3) wastewater treatment; (4) improved sewerage service; and (5) river-bed
dredging and construction of landing facilities. The estimated cost components are
presented in Table 4 for a period of 10 years. Most prices of the cost components are
collected from available documents (both published and unpublished) prepared by
government departments in Bangladesh. In some cases where information was not
available, appropriate proxies were used after consultation with departmental experts.
Cost estimates are based on the assumption that some items will increase at 5 –15% on a
year-to-year basis. The total cost (financial) is estimated at Tk4 078.80 million in 2001
constant price, which is significantly lower than the estimated benefit (just above a quarter
of the benefit).
Cost – Benefit Analysis of the Buriganga River Restoration
The estimates of benefit and cost are in market (financial) price. However, market prices do
not reflect the opportunity cost of capital because of the existence of market imperfections
and policy distortions of the economy (developing countries are particularly susceptible to
this). This requires appropriate adjustments to convert financial prices into economic values.
The total cost of each component is divided into two broad categories: investment cost,
and operation and maintenance cost. Investment costs are the costs incurred in establishing
the programme, and include costs of equipment, construction, manpower and land.
Operation and maintenance (O&M) costs are those incurred in running and maintaining the
programme, and include raw materials, manpower, utilities, equipment hiring, and repair
and maintenance. All items of the cost components are then divided into tradable and nontradable categories because these are required to be valued in different ways. Tradable goods
include those items that are either imported or exported by the country, and for this particular
programme can consist of imported goods, materials and equipment. Non-tradable items are
neither imported nor exported, but are used locally, such as land, local consultancy, utility
services, local trade and transport. This breakdown allows makes it possible to estimate the
economic costs and different shadow prices that apply to the various categories of costs. For
the conversion of financial prices into economic values, non-tradable goods are valued at
shadow prices while tradable goods are valued at border (world) prices.
In addition, labour (both for investment and O&M) is divided into skilled and unskilled
categories. The opportunity cost of labour is derived by adjusting the prevailing (market)
wage rate by a factor of 0.82 for skilled labour and 0.73 for unskilled labour in line with
estimated levels of unemployment and underemployment in the study area, as per the
guidelines of the Bangladesh Planning Commission (PC, 1997).
Table 4. Total cost of the Buriganga River restoration programme (million Tk)
Components of cost
Removal of illegal structures and construction
of an access road and walkways
Establishment of wastewater treatment plant
Improvement of sewerage facilities
River-bed dredging and construction of landing
facilities
Total cost
Total cost
% of column total
957.69
23.48
2323.35
283.20
514.56
56.96
6.94
12.62
4078.80
100.00
Cost – Benefit Analysis of Restoring Buriganga River
603
Table 5 represents the economic cash flow of the programme and summarizes the
breakdown of the total cost into different components and the conversion to economic
values. The estimates are made by apportioning goods and services into tradable and nontradable components, and distinguishing between skilled and unskilled labour.
Although components of both cost and benefit are estimated on an annual basis, Table 5
represents total cost over 10-year period. Physical contingencies are also included in the
total cost of each component. Physical contingencies are the monetary value of additional
resources that may be required beyond the initial estimate to complete the restoration
programme. Although often calculated and expressed as percentage of the initial (base)
costs (ADB, 1997), contingency cost is estimated for this study as 5% of total investment
and O&M costs as this cost is ‘unforeseen’.
The economic benefits and costs are estimated in border prices in domestic currency,
derived from financial (market) prices using conversion factors and shadow prices. The
economic analysis is conducted using real price rather than nominal price. To estimate the
border prices of tradable goods, all transfer payments, such as taxes/duties, subsidy and
compensation, are excluded from the financial prices. Non-tradable inputs are valued at
their domestic price. A standard conversion factor of 0.82 is used to convert the domestic
prices of non-tradable items to their border prices.
The total economic cost of the BRRP is estimated as Tk2564.50 million over the
10-year duration of the programme, which is equivalent to $36.64 million. The total
investment cost is estimated at Tk1755.25 million and the operating and maintenance
costs are Tk676.13 million. All these estimates are at constant 2001 prices.
Table 5. Economic cash flow of the restoration programme
Items
Investment items:
Skilled labour
Unskilled labour
Tradables
Non-tradables
Total investment cost:
Operation and maintenance items:
Skilled labour
Unskilled labour
Tradables
Non-tradables
Total operation & maintenance cost:
Contingency:
Tradables
Non-tradables
Total contingency cost:
Grand total (cost)
Total benefit
Net benefit
NPV @ 10.11% ¼ Tk 6 100.89 million
IRR @ 10.11 discount rate ¼ 822
BCR ¼ 4.35
Total economic cost (million Tk)
26.24
29.38
849.45
850.18
1755.25
57.04
88.29
179.20
351.60
676.13
28.66
104.46
133.13
2564.50
15 068.36
12 503.86
604
K. Alam
The total cost is 74.17% of total benefit in Year 1, declining to 4.36% of total benefit in
Year 10. The incremental net benefit is Tk121.96 million ($1.74 million) in Year 1, rising
to Tk2 715.21 million ($38.79 million) by the end of the programme (i.e. in Year 10). This
indicates that the restoration programme will generate more benefits in the later years of
the implementation period.
The summary statistics of the economic analysis of the BRRP are the net present value
(NPV), internal rate of return (IRR) and benefit cost ratio (BCR). These are presented in
Table 5. The discount rate (nominal) is 15% as determined by the Bangladesh Planning
Commission and an inflation-adjusted real rate is estimated at 10.11% per annum. Therefore,
the restoration programme will be acceptable if the IRR is above 10.11%, or if the NPV is
positive, using a 10.11% discount rate as a measure of the social opportunity cost of capital.
The NPV at 10.11% discount rate is Tk6 100.89 million ($87.16 million). The IRR is 822%,
which is well above the opportunity cost of capital of 10.11%. The BCR is 4.35 at 10.11%
discount rate. All three criteria of the CBA are satisfied. Therefore, it can be concluded that
the Buriganga River restoration programme is economically viable (Figure 1).
Sensitivity Analysis
A sensitivity analysis is carried out to test the “possible impact of uncertainty by posing
‘what if’ questions. These questions pertain to what would happen to the project’s viability
if some or all of the key parameter values happen to be different from the original values”
(Asafu-Adjaye, 2000, p. 155). It is difficult to examine the impact of all the potential
parameters on the viability of the BRRP. There may be many variables that can affect the
viability of the BRRP. However, on a limited scale, this can be examined with regard to its
sensitivity to some exogenous factors. For a sensitivity test, four possible scenarios are
considered. The outcome of the sensitivity analysis is presented in Table 6.
The programme performance criterion used in the sensitivity analysis is NPV, i.e. the
effect of possible changes on the viability of the BRRP is calculated in the form of NPV,
Figure 1. Discounted cash flow
Cost – Benefit Analysis of Restoring Buriganga River
605
while varying one variable and holding the others constant. Table 6 shows that in all four
scenarios the NPV remains positive. Therefore, the viability of the BRRP is not sensitive
to any of the scenarios considered here. Thus, the BRRP allows for negative changes,
which will reduce its net benefit but will still remain positive.
Validity and Reliability of Benefit Estimates
There are no previously published similar CVM studies conducted in Bangladesh to
provide comparable results to validate the estimated values obtained in this study. One
way to validate respondents’ willingness to contribute for the BRRP is to compare the
WTCM with their income. This is common in the economic valuation literature. Choe et al.
(1996) found WTP for the improvement in surface water quality of less than 1% of stated
income in Davao, Philippines, while sanitation benefits from sewerage in Africa figures
range between 1 –2% (Whittington et al., 1992). Lauria et al. (1999) found that households
were willing to pay less than 1% of their household income for a connection to a sewer
system and a treatment plant in Calamba, the Philippines.
Respondents’ monthly willingness to contribute money as a percentage of the total
household income was found to be less than 1% for the BRRP. In addition, WTCM as a
percentage of stated median household income is only 0.07%. Respondents’ total WTC as
a percent of per capita GDP is 0.64. Compared with other studies, these estimates seem
very realistic.
Next is the question of how realistic the estimated value of the non-market benefits of the
BRRP can be raised. This can be answered in two ways. First, a comparison of the WTCM
estimate can be made with the average payments of utility bills for other public and publicly
provided goods and services for households in Dhaka City. In comparison to charges for
utility services of residents in Dhaka City, the amount respondents expressed as their
willingness to contribute in the form of direct cash payment for the BRRP (Tk51.91) is not
too high (between 7.69 and 50% of other utility bills for an average household) and therefore
seems to be realistic. These figures will be even less if indirect cost of receiving these
services are included in the rates. Furthermore, a World Bank-assisted study stated that poor
people in slum areas, not connected to utility services such as gas, water or electricity,
actually pay 50–100% more than the official rate in Dhaka City (The Daily Jugantor, 24
May 2002). Therefore, the generated benefit estimates seem to be realistic.
Issues of Transaction Cost
Contingent valuation studies conducted in both developed and developing countries so far
have ignored the issues of transaction cost, i.e. cost of collecting and managing residents’
contribution (be it WTP or WTA) for the proposed environmental change. This is
Table 6. Sensitivity analysis for the restoration programme (million Tk)
Scenarios
NPV @ 10.11
Scenario
Scenario
Scenario
Scenario
5919
4516
4467
7524
1: Total cost increases by 10% annually
2: Total benefit expected to be generated fails by 20% annually
3: A higher annual discount rate (real) (i.e. 15%) is used
4: A lower annual discount rate (real) (i.e. 7%) is used
606
K. Alam
particularly important for this study as 60.55% of the non-market benefits are estimated to
be derived in the form of respondents’ time. Now the key question remains to be answered
is: whether generating residents’ contribution, particularly in the form of time, is
economically viable? As this was not the focus of this study, an examination of this issue
was limited. Furthermore, no such studies were found available neither in Bangladesh nor
any developing countries. If the transaction costs of collecting and managing residents’
contribution are greater than the social benefits generated by the restoration programme,
then the restoration programme is not viable.
All the public utility agencies in Dhaka City are heavily subsidized and the operational
costs are substantially higher than the revenue they generate. Improving the performance
of governance is a general issue in the developing countries. Seen in that line, non-market
benefits in particular and total benefit in general are still significant.
Conclusion
Non-market valuation techniques have often been transferred to developing countries
without taking into account their social, economic, political and cultural settings. For
example, the same type of elicitation method is applied in developing countries, although
many of the respondents are extremely poor and many of their economic activities are
outside the purview of the monetary mechanism. This paper reports research in a
developing country context where the conventional contingent valuation method is
extended to include respondents’ preference in terms of time for the restoration of a
vulnerable river, irrespective of their decision to contribute money.
In the case of the Buriganga River, about two-thirds of the expected benefits of the
restoration programme are derived from market benefits. The large share of market
benefits is primarily due to the high land value in the study area. However, in the case of
other rivers in Bangladesh, the share of market benefit to the total benefits might be very
small. In that case, the restoration effort may invariably appear to be ‘non-economic’. This
study reveals that the whole range of benefits, both market and non-market, is relevant to
the economic analysis. The use of direct benefits alone to determine investment worth is in
fact a purely commercial criterion. Therefore, the benefits of any ecosystem restoration
should be measured as the sum of all the components of TEV.
The application of CVM reveals that the BRRP is able to generate significant benefits.
The estimated total benefit is a very significant figure for the economy of a poor country
such as Bangladesh. Thus, a failure to account for such benefits could lead to gross underestimation of the desirability of providing public funding for the restoration of dying
rivers. The cost – benefit analysis also reveals that investment for the BRRP is worth
undertaking. Failure to account fully for the environmental benefits of such intervention
means that its net economic worth is misrepresented. Continuing such practices may have
long-term negative impacts on the ecology and sustainable development. Therefore,
public sector investment in developing countries such as Bangladesh needs redirection to
respond to emerging environmental problems such as degraded ecosystems.
Acknowledgements
The author wishes to thank Professor Dora Marinova for her invaluable help with the research and survey design
of the study on which this paper is drawn.
Cost – Benefit Analysis of Restoring Buriganga River
607
References
Alam, K. (2002) Valuing the cleanup of the Buriganga River, in: F. Ahmed (Ed.) Bangladesh Environment 2002,
pp. 978–991 (Dhaka: Bangladesh Poribesh Andolon).
Alam, K. (2006) Valuing the environment in developing countries: problems and potentials, Asia Pacific Journal
on Environment and Development, 13(1 & 2), pp. 27 –44.
Asafu-Adjaye, J. (2000) Environmental Economics for Non-Economists (London: World Science).
ADB (Asian Development Bank) (1997) Guidelines for the Economics Analysis of Projects (Manila: Economics
and Development Resource Center, Asian Development Bank).
BBS (Bangladesh Bureau of Statistics) (2001) Population Census 2001: Preliminary Report (Dhaka:
Government of Bangladesh).
Barton, D. N. (1998) Applying NOAA Panel recommendations to contingent valuation studies in developing
countries—a case study of coastal water quality in Costa Rica. Discussion Paper No. D-24/1998 (Ås,
Norway: Department of Economics and Social Sciences, Agriculture University of Norway).
Boardman, A., Greenberg, D., Vinning, A. & Weimer, D. (1996) Cost–Benefit Analysis: Concepts and Practice
(Upper Sadle River, NJ: Prentice-Hall).
Brookshire, D. S. & Neill, H. R. (1992) Benefit transfer: conceptual and empirical issues, Water Resources
Research, 28(3), pp. 651–655.
Brouwer, R. (2000) Environmental value transfer: state of the art and future prospects, Ecological Economics,
32(1), pp. 137–152.
Carson, R. T. & Mitchell, R. C. (1993) The value of clean water: the public’s willingness-to-pay for boatable,
fishable and swimmable quality water, Water Resource Research, 29(7), pp. 2445–2454.
Choe, K. A., Whittington, D. & Lauria, D. T. (1996) The economic benefits of surface water quality improvements in
developing countries: a case study of Davao, Philippines, Land Economics, 72(4), pp. 519–527.
Czaja, R. & Blair, J. (1996) Designing Surveys: A Guide to Decisions and Procedures (New Delhi: Pine Forge Press).
Desvousges, W. H., Naughton, M. C. & Parsons, G. R. (1992) Benefit transfer: conceptual problems in estimating
water quality benefits using existing studies, Water Resources Research, 28(3), pp. 943 –950.
Garrod, G. & Wills, K. G. (1999) Economic Valuation of the Environment: Methods and Case Studies
(Cheltenham: Edward Elgar).
Gore, J. A. & Shields, F. D. Jr. (1995) Can large rivers be restored?, BioScience, 45(3), pp. 142–152.
Haque, A. K. E., Faisal, I. & Bayes, A. (1997) Welfare costs of environmental pollution from the tanning industry in
Dhaka: an EIA study. Paper presented at the mid-term review workshop, Yogyakarta, Indonesia, 3–8 September.
Hill, M. T. & Hanchett, S. (1995) Fisheries Supervision Mission Report on the Closure of the Northern Intake of
the Dhaleswari River: EMAP and Mitigation, JMB Report (Dhaka).
Lauria, D. T., Whittington, D., Choe, K., Turingan, C. & Abiad, V. (1999) Household demand for improved
sanitation services: a case study for Calamba, the Phillipines, in: I. J. Bateman & K. G. Willis (Eds) Valuing
Environmental Preferences: Theory and Practice of the Contingent Valuation Method in the US, EU, and
Developing Countries (New York: Oxford University Press).
Markandya, A. & Murty, M. N. (2000) Cleaning-up the Ganges: A Cost–Benefit Analysis of the Ganga Action
Plan (New Delhi: Oxford University Press).
Navrud, S. (1996) The Benefit Transfer Approach to Environmental Valuation (Singapore: Economic and
Environment Program for Southeast Asia).
PC (Planning Commission) (1997) Project Proforma (PP) and Manual of Instruction for Project Proforma
(Dhaka: Government of Bangladesh).
PC (Planning Commission) (2000) Mid-term Review of the Fifth Five-Year Plan 1997–2002 (Dhaka:
Government of Bangladesh).
Russell, C. S. (2001) Applying Economics to the Environment (New York: Oxford University Press).
Sarantakos, S. (1998) Social Research, 2nd edn. (South Yarra: Macmillan).
Tunstall, S. M., Tapsell, S. M. & Eden, S. (1999) How stable are public responses to changing local environment?
A ‘before’ and ‘after’ case study of river restoration, Journal of Environmental Planning and Management,
42(4), pp. 527–540.
Whittington, D. (2002) Improving the performance of contingent valuation surveys in developing countries,
Environmental and Resource Economics, 22(1), pp. 323–367.
Whittington, D., Smith, V. K., Okorafor, A., Okore, A., Liu, J. L. & McPhail, A. (1992) Giving respondents time
to think in contingent valuation studies: a developing country application, Journal of Environmental
Economics and Management, 22(3), pp. 205–225.