GLOBAL ENVIRONMENT FACILITY
CONCEPT DOCUMENT FOR GEF PIPELINE
June 2003
Country Brazil
GEF Focal Area:
International Waters, OP9 Integrated Land and Water
Multiple Focal Area Program
Project Title:
Igarape 40 Cleanup, Manaus
Requesting Agency:
IFC
Total Project Cost (tentative):
US$ 28 million
Financing Plan (tentative):
US$ 9 million, GEF
US$ 19 million (Aguas do Amazonas and other sources,
including public ones)
Project Duration (tentative) 5
years
Preparation Costs (tentative):
US$ 745 000
PDF Block B Funds Requested:
US$ 350 000
PDF Co-financing:
US$ 395 000
Information on Project Proposers:
- Aguas do Amazonas (AdA) is a fully owned affiliate of Suez Environnement. It is expected to
play a key role in project preparation and implementation, both as directly in charge of
implementing project activities (investments) and as advisor to the State of Amazonas and
Municipality of Manaus.
- IFC the private sector affiliate of the World Bank Group - will be the executing agency, acting
through the World Bank as GEF Implementing Agency..
A. BACKGROUND
General background:
Manaus has a population of about 1.5 million, of which 60% is low income. The city is cut by
numerous streams called igarapés, flowing into the Rio Negro. The proposed project area is the basin
of one of the longest of these streams, Igarapé do Quarenta (I40), close to the city center.
Land use in the I40 basin is both residential (270 000 people, high density) and industrial (Manaus'
Industrial District -187 factories, small to large). Historically, low-income migrants to the city have
arrived by boat and settled irregularly near the banks of I40, or directly above the water in houses
called palafitas built on wooden piles. The closer to the igarapé, the further from the main water and
sanitation networks and the more precarious the living conditions. Close to the bank, raw sewage and
solid waste are thrown directly into the igarapé, causing a significant , documented, health hazard
especially affecting children (mostly diarrhea, some hepatitis A, typhoid fever and leptospirose). Most
other houses located inland have septic tanks or pits for toilets, but discharge their gray water in the
drainage network or in open ditches, which in turn flow into the igarapé.
Discharge of industrial and domestic wastewater has drastically altered the physical/chemical
conditions of the water and of the sediments of I40, including in particular a high concentration of
heavy metals. In addition to the igarapé itself, at least 48 of the 70 natural springs in its basin are
polluted.
The population is all the more sensitive to this long lasting pollution problem as I40 is an emblematic
stream in Manaus, like Rio Tietê in São Paulo or Baía de Guanabara in Rio (people used to swim and
fish in the igarapé, which is now impossible).
Background on Aguas do Amazonas' concession contract:
Water & sanitation services in Manaus have been under the responsibility of a private concessionaire,
Aguas do Amazonas (AdA), since July 2000. The main features of this concession contract, its
evolution since 2000, and the parallel evolution of water pollution legislation in Brazil must be taken
into account in order to understand the rationale and objectives of the proposed project. The main
elements of this context are the following:
- competition for the concession was based on an economic model built by the conceding authority
and its advisors, using key economic assumptions. As is typical in the water sector, some of these
key assumptions were made with high confidence, others depended on data of unknown quality
that could only be verified after several months of operation.
- contract design took this uncertainty of baseline data into account and provided for adjustment
mechanisms, overseen by the regulator, aiming at maintaining the initial economical equilibrium
of the concession (balancing revenues, operating costs and capital investments), in accordance
with the legislation in force in Brazil. In short, when disrupted, the economic equilibrium can be
restored by adjusting either the level of tariffs (the only source of revenues of the concessionaire)
or the service targets (in timing and/or sizing).
- the main contractual service targets (those requesting capital expenditures) were set as follows,
with the objective to maximize private financing of the infrastructure needed to reach them, while
maintaining tariffs at an affordable level:
INDICATOR
UNIT
2000
2006
2011
2016
2021
2026
2029
POPULATION
SERVED WITH
%
91 95 98 98 98 98 98
WATER
POPULATION
SERVED WITH
%
11 31 51 71 90 90 90
SEWERAGE
COLLECTION
TREATED PORTION
OF COLLECTED
%
- 35 50 70 80 90 90
SEWERAGE
- customers being initially unwilling to pay for sewerage services, and the poorest part of the
population being unable to pay the connection cost to the sewerage network, the economic
equilibrium of the contract was therefore based on the following overall priority order, reflected in
the targets: first improving and expanding water supply, then wastewater collection, and finally
wastewater treatment.
- based on this backbone, the Concessionaire must submit every year to the Regulator a (non-
binding) plan for improvement and expansion of the system over the next 5 years.
- 12 months into the contract, it became clear to all parties that some key initial assumptions were
wrong, beyond what bidders could possibly detect during their due diligence. This led to a 14.8%
tariff increase granted by the conceding authority in Feb. 2002, correcting the mistakes that could
be fully documented at that time, but leaving aside other less known emerging issues - in
particular in very low income settlements in which extension work has started more recently.
- the poor quality of initial data on low income population is due in part to the fact that 60% of
Manaus' population is of low income and technically "illegal" (without land title), therefore
making initial consumption estimates difficult. Serving these parts of the population is part of the
obligations of the concessionaire (although it can be argued that very precarious settlements like
palafitas are de facto excluded as unsafe for their inhabitants and better to be relocated before
being served), but their needs and ability to pay for connection costs and for services, could not be
precisely valued at the time of bidding: here again, some assumptions are now proving to have
been wrong (e.g. actual level of consumption per household much lower than assumed), and there
is an on-going learning period both for the municipality and the concessionaire which is likely to
lead to additional adjustments.
- another factor that has evolved since the beginning of the contract is that sanitation and pollution
reduction are today higher in the priority agenda than at the beginning of the concession, due to
increased political sensitivity for health and environment issues. The state of Amazonas and
municipality of Manaus are now willing to clean up I40 as soon as possible.
- the problem is that sanitation investments needed to achieve this cleanup in the coming years
rather than in 10 to 20 years are not compatible with the initial design of the concession and its
underlying economic equilibrium. These investments are indeed under the contractual obligations
of AdA (the exception being the palafitas areas and part of wastewater management within the
industrial district), but as explained above they are mostly scheduled to start only later, especially
in the most difficult areas like highly populated settlements along the banks of the igarapé.
Background on wastewater investment needs in Manaus, and in the I40 basin
A water and sanitation master plan was finalized by AdA in July 2001; options for medium term
investment plan in wastewater management were then reviewed in the 2nd half of 2002. The emerging
conclusions are:
- overall wastewater investment needs over the life of the contract are around $100m ($1=$R3)
- given current tariffs, underlying the economic equilibrium of the concession, only minimum
amounts of sewerage network expansion and upgrading of existing treatment plants can be
envisaged until 2007. Construction of new wastewater treatment plants can only be done much
later;
- wastewater flows generated in the I40 basin can be divided into 3 categories: (1) domestic, and
industrial (2) without or (3) with heavy metals. (heavy metal pollution of I40 was documented in a
study by Dr. Waichman, University of Amazonas, 2000).
- some of the flow from the basin is collected and directed to the Educandos treatment plant
(preliminary treatment), and rejected after preliminary treatment into the Rio Negro; some of the
uncollected flow ends up directly in the Rio Negro; the rest (i.e. most of the total flow) ends up in
the igarapé, either as uncollected sewage or collected but untreated sewage.
- the evolution of environmental standards - a key determinant of treatment levels (and investment
needs) - needs to be discussed in more detail with federal, state and municipal regulatory
authorities. This, in itself, is a complex topic very much interrelated with investment planning.
These discussions are necessary in order to have an effective control on industrial effluents, which
is under the responsibility of the state environmental agency. They must take into account the
specificities of Manaus (e.g. low alkalinity of the water, making nitrification difficult; difference
between dilution capacity of the Rio Negro vs. igarapés) and the fact that national standards
cannot be applied directly without some adjustments.
- current estimate of the total domestic sewerage network expansion works needed in the I40 basin
is around US$ 34 million (65 000 new sewerage connections, networks, interceptors, new lifting
stations, and rehabilitation of outfalls)
- if no additional treatment (beyond preliminary) is required, for the part of the flow directed to the
Rio Negro, but if treated flow increases, then grit removal will have to be reconstructed at the
Educandos plant; a possible option that could be tested (e.g. as a pilot project) in informal
settlement would consist in a small decentralized treatment unit serving around 400 households.
- if higher treatment requirement is necessary, a centralized treatment plant for domestic wastewater
from the basin might be the best option,
- in the Industrial District, the best objective could be a specific treatment plant with pretreatment
and maybe a primary treatment with chemicals, combined with some restrictions on the industrial
effluents being collected and specific treatment plants inside the plants for specific pollutants
(heavy metals, dyes,...) - under the responsibility of the industries.
B. RATIONALE FOR THE PROJECT AND PROJECT OBJECTIVES:
As discussed above, cleaning up I40 fast is beyond the scope of the concession; the state and
municipality have also already allocated public resources to various clean up actions (e.g. removal of
solid waste floating in the igarapé) but in insufficient amounts due to budget constraints and with
limited results; there is today a strong case in favor of external support to fill the following gaps, on
which the proposed project would focus:
(1) improving dialogue among stakeholders (state, municipality, Aguas do Amazonas,
communities, industries), including on the appropriate pollution standards and treatment
levels; and improving planning and coordination of their actions at the "igarapé basin" level.
(2) accelerating domestic wastewater collection and treatment investments by financing the
incremental costs due to the acceleration. Such support could take various forms that would
still be compatible with the overall respect of the concession contract (e.g. temporary subsidy
to fill a tariff gap created by the acceleration of the sewerage investment program, financing
contingent to the ability to pay of the beneficiaries during and after a transition period leading
progressively to a full cost recovery tariff level, etc.).
(3) improving environmental compliance in the Industrial District and reinforcing the
environment regulatory capacity of the state (IPAAM), therefore contributing to reduce
industrial pollution discharges in the igarapé.
The rationale of the project is in fact to rebalance the public-private partnership around the objective
of cleaning up I40, and to find the optimal balance between public (municipal / state), private and
"external" (like GEF) sources of financing necessary to achieve the new (accelerated) investment
objectives.
Note that when developing sanitation services in low income areas, public or external financing is also
needed for accompanying social sanitation measures: community training, health and environmental
awareness, maybe some sewerage connection subsidies for the poorest households.
The overall need for financing is currently being discussed between the concessionaire and the public
authority with the view of combining public and private resources, and of seeking together external
support to fill the gaps, as it is largely anticipated that the state and municipal budgets will not be
sufficient. The preparation phase of the proposed project will contribute to this analysis.
Outside support needed to improve collection and treatment of effluents in the Industrial district is
expected to focus on institutional support, mostly to the state / municipality, rather than on
infrastructure investments (which the industries are deemed able to finance themselves): this will
possibly involving a mix of improved regulatory capacity, consensus building and better use of
available resources.
Although the project will focus on wastewater management in the I40 basin, there is a larger range of
issues that need to be addressed, with a direct or indirect bearing on the efficiency and sustainability of
the project. For instance: improved urban planning in the project area, regularization of land titles,
slum upgrading and development of other urban basic services, etc. Although programs to address
these issues will remain outside of the scope of the proposed project, it will be important that the
project's core coordination mechanism (1) also include a -strong interface with them. In particular,
special attention will be given to:
- the complementarity of water and sanitation services (which should be developed jointly)
- the likely need to consider the resettlement of part of the population living in palafitas.
C. PROJECT DESCRIPTION
The breakdown of activities presented below is tentative at this stage, and will be refined and finalized
during project preparation. In terms of financing, the project is expected to have two parts: a public
one (e.g. grant to the state and municipality for capacity building and technical assistance) and a
private one (e.g. contingent loan/grant to Aguas do Amazonas to finance investments and actions
going beyond its contractual arrangements). The final amount, mix (public / private) and form of
proposed GEF financing would be determined during preparation.
The proposed components are the following:
Component 1: stakeholder dialogue and participation
In addition to Aguas do Amazonas, the stakeholders expected to participate in the project are:
- Municipality of Manaus; SEDEMA
- Government of the State of Amazonas
- IPAAM (State Environment Agency)
- ARSAM (Agencia Reguladora dos Serviços Publicos Concedidas do Estado do Amazonas)
- Ministerio de Meio Ambiente, CONAMA 20, Ministerio Publico
- (selected) companies based in the Industrial District
- low income communities in the project area
Information dissemination on the project, consultation of these stakeholders and effective stakeholder
participation will be key elements of the success of the project.
Objectives: specific objectives of this component will be to (1) obtain constructive participation of the
industries of the Industrial District, (2) obtain adhesion of the low income populations living in the
project area, and (3) reach agreement with the regulatory authorities on the appropriate environmental
standards to be applied to the project.
Expected outcomes: several workshops with the stakeholders (one-to-one and/or as a group), training
sessions on topics to be identified during the preparation phase; public awareness campaign(s)
especially on hygiene and sanitation; community building in low income areas along the igarapé;
possible identification of companies in the Industrial District interested in piloting innovative pollution
control activities; creation of a project steering committee.
Detailed planning (number and timing of workshops and training sessions, communication plan, etc.)
will be determined during project preparation. Some of these activities will start during project
preparation (see "project planning" below).
Tentative budget: $ 1 000 000 to $1 500 000
Component 2: support to the public authorities
Objectives: as a complement to component 1, this component will focus on work and resources needed
to improve environmental compliance in the Industrial District and reinforcing the environment
regulatory capacity of the state (IPAAM), and more generally to strengthen the capacity of the public
authorities to deal with wastewater management issues and to interact with the stakeholders in I40.
Combined with component 1, this component will also aim at creating the conditions that will allow to
connect and serve (collection and treatment) more industrial customers in the Industrial District
currently discharging into the igarapé.
Expected outcomes: specific activities needed will be discussed during project preparation, and could
include training sessions for the staff of public agencies involved in monitoring environmental
compliance, some technical assistance to these agencies (e.g. consulting assignments on specific topics
when deemed more efficient than internal capacity building), and if needed the purchase of support
equipment (e.g. pollution monitoring equipment).
Tentative budget: $1 000 000 to $2 000 000
Component 3: design and implementation of the project financing structure
Objectives: to put in place the appropriate financial and legal mechanism to channel funds to the
investment component of the project.
Expected outcomes:
(1) a study / report on the economic feasibility of the project (applicability of polluter-payer
principles, ability of low income population to pay for sewerage collection and treatment, impact
of the project on the tariff structure, assessment of corresponding publics subsidies needed, impact
of the project on the financial equilibrium of the concession, analysis of public budget
constraints);
(2) based on 1, a proposal on the design of appropriate legal and financing mechanisms/vehicles (if
needed, special purpose vehicle to channel funds directed at the project, definition of disbursement
mechanisms, design of the appropriate contingent finance scheme, monitoring); the underlying
principle behind the design of the contingent finance scheme will be the goal of preserving of the
"economic equilibrium" of the concession, which is the global indicator of the sustainability of the
concession. Analysis of possible innovative combination of public (e.g. state / municipality) and
private sources of financing will be taken into account when considering the impact of the
proposed project on this economic equilibrium, adding some flexibility to the concept.
(3) the actual implementation of such mechanisms/vehicles (e.g. incorporation of the a special
purpose vehicle).
Like component 1, activities under this component will start during project preparation (initial
design) and will be finalized and implemented during project implementation.
Tentative budget: $500 000
Component 4: infrastructure investments
Objectives: to identify, design and build the wastewater infrastructure works - collection and treatment
- needed to cleanup I40 as much as possible, given the economic constraints identified under
component 3. As explained earlier, investments actually financed with GEF support will focus on
domestic wastewater; in the area of industrial wastewater management GEF support will not be used
to finance actual investments, but will rather focus on helping create, via component 1 and 2, the
conditions that will allow to collect and treat more effluents from the Industrial District currently
discharging into the igarapé.
Expected outcomes:
(1) preparation of a detailed investment plan, based on the analysis of alternative wastewater
infrastructure investment scenarios, using options already identified in the existing master plan as
well as the outcome of discussions with regulatory authorities on effluent standards and treatment
requirements (component 1 above). This planning work will be conducted during the preparation
phase. The scope and duration of the proposed project will therefore only be precisely known at
the end of the preparation phase.
(2) actual construction of domestic wastewater collection and treatment investments identified in (1).
Network expansion works will have to include: new sewerage connections, new networks,
interceptors, lifting stations, and the rehabilitation of outfalls; Treatment works will have to be
identified during preparation (e.g. see earlier comments on the Educandos plant). A specific
outcome will be to bring sanitation service to a maximum number of households in the I40 basin,
in particular the low income households near the banks of the igarapé. The number of
beneficiaries will be determined during project preparation, out of a total need of 65000 new
sewage connections.
Tentative budget: $24 000 000 to $27 000 000
Project planning
1. preparation phase (around 6 months) to be completed at the latest during the 1st semester of 2004:
project management and creation of a project coordination mechanism, expected to remain in
place during project implementation, consolidation of existing data, surveys and studies,
additional studies (technical, institutional, financial and socio-economic), consultation of
stakeholders and consensus building, analysis of investment scenarios and prioritization of project
activities, completion of project design and implementation framework, and preparation of a
project brief.
These activities would be conducted by Aguas do Amazonas, or contracted out to international or
national individual consultants, NGOs (already working with AdA) and students from the
University of Amazonas. GEF will be asked to support this preparatory work with a US$ 350 000
PDF Block B Grant.
2. implementation phase (coordination with complementary programs, institutional strengthening,
infrastructure works) that will probably be spread over 5 years. A detailed funding request will be
presented to GEF at the end of the preparation period, to support this implementation phase.
Preliminary financing plan
Estimated Preparation Phase Financial Plan:
Source US$ Explanation
GEF
US$ 350 000
PDF Block B Grant
Aguas do Amazonas, Ondeo
US$ 395 000
staff time to support project
preparation, past expenses, previous
studies
State of Amazonas
to be determined.
staff time
Totals
US$ 745 000
Total Projected Project Budget and Sources of Financing:
Component
GEF
Other Sources
Project Total
Project Preparation
US$ 350 000
US$ 395 000
US$ 745 000
Project
To be determined during project preparation
US$ 28 million
Implementation
e.g. GEF between US$9 million
Risks:
The project faces a number of risks throughout its preparation and implementation phases. The three
main ones would be:
1. a failure to build consensus among stakeholders
2. a failure to coordinate properly the project with complementary programs, which could increase
the technical difficulty of building sewerage networks in the areas currently furthest from the
existing trunk networks. This would greatly increase the project costs, up to the possible point of
economic unfeasibility.
3. some reputational risk for GEF and the World Bank group in particular on the issue of
resettlement
4. a failure to reach cost recovery tariff level after the transition period that could be financed using
GEF support, possibly because of continued unwillingness to pay even after the service has been
in place during a few years and environmental and social benefits have been demonstrated.
5. in case of a devaluation, disconnect between project revenues (in local currency) and repayment
obligations, if financing is in the form of a loan in foreign currency.
6. more generally, risk that this project, in the end, will not be compatible with the objective of
preserving the initial economic equilibrium of the concession (risks 4 and 5 above being examples
of this category), and impossibility to reach a new equilibrium - for instance via the combination
of public and private financing.
These risks can be minimized during project preparation and implementation (e.g. preparation of a
resettlement program framework, preparation of a stakeholder involvement plan, etc.) but not totally
eliminated. Risks 1, 2, 3 will be addressed in particular through Components 1 and 2 of the proposed
project. Risks 4, 5 and - more generally 6, will be addressed via the contingent finance approach in
component 3.
D. GEF ELIGIBILITY
Program Conformity:
The proposed project falls under the GEF International Waters Focal Area, and specifically under
Operational Program Number 9: Integrated Land and Water Multiple Focal Area Program.
Although actions of both prevention and remediation of damage to waters could be envisaged in the
Igarapes of Manaus, the project is more oriented towards prevention, as it is usually the case under OP
9. Still, special attention will be given to the existing heavy metal pollution in the Igarape 40.
The coordination among stakeholders, that the project will seek to establish, has direct relevance to the
objective of the OP to integrate the use of sound land and water resources management strategies.
Participation of Aguas do Amazonas is also relevant as a step toward optimizing, via private sector
participation, the overall use of the full range of international best practices in the technical, economic,
financial, institutional, regulatory and social areas, which is another objective of the OP.
Project Linkage to National Priorities, Action Plans and Programs:
Brazil has initiated 6 years ago a large scale effort to better manage its water resources, with the river
basin as the basic territorial unit (cf. 1997 Water Law); as part of this effort, the federal National
Water Agency (ANA) has been working since its creation in 2000 on developing a national framework
of river basin management, improving legislation on effluent discharge and environmental compliance
and encouraging river cleanup programs. The proposed project, which aims at improved wastewater
management at the level of an urban basin in Manaus, is fully in line with this national action plan. It
is also in line with national priority objectives, and Millennium Development Goal, of improving
access of low-income populations to sanitation services, and reducing their exposure to sanitation
related health hazards. Finally, via its expected replication effect, this project will also contribute to
the national objective of preserving the ecological health of the Amazon basin, in particular by
reducing heavy metal pollution.
It is important to also mention that implementation at the local level of these national priorities is
under the responsibility of the State of Amazonas and the municipality of Manaus. They both have
already launched in Manaus targeted environmental and social actions that the proposed cleanup
project will complement.
Incremental Reasoning:
GEF support to the project will have the following specific additional impacts:
create the missing synergies / coordination between stakeholders
improve pollution monitoring and control capabilities of the state and municipality, with special
attention to heavy metal pollution - I40 being a heavy metal pollution "hot spot" in the Amazon Basin
accelerate sanitation investments in the I40 basin, beyond initial contractual arrangement of Aguas do
Amazonas;
develop an innovative model and financing scheme applicable to public-private partnerships in
sanitation elsewhere in the world - one of the most complex issues that came out in recent discussions
(Johannesburg, Kyoto) on reaching the Millennium Development Goals in sanitation.
Sustainability
The participation of Aguas do Amazonas, and link between the key project objectives (increased
collection and treatment of domestic and industrial effluents) and the concession contract, will greatly
contribute to the sustainability of the project. The concession contract will run until 2029, with an
option to extend it by an additional 15 years. Any new service connection financed under the project
will be operated and maintained by Aguas do Amazonas, with prior assurance in the design of the
project that an optimal cost recovery mechanism will be put in place. In particular, this guiding
principle will be applied to the design of any subsidy scheme deemed necessary (for instance, one-
time subsidies to help low income households pay for sewage connection costs, along with tariff
scheme adapted to ability/willingness-to-pay of these households, rather than recurring "consumption"
subsidies).
Global Replicability:
The project is expected to develop approaches that will be replicable at 3 levels:
in Manaus, through the extension of the urban basin management project in the other Igarapes of the
city
in other cities of the Amazon basin (in Brazil and neighboring countries)
in other cities in the world facing similar urban issues, and in need of a proven approach to deal with
increasing demographic pressure and pollution rejections in urban micro basins comparable to the
igarapes of Manaus. Examples of such cities, that could be potentially interested: Manila, Bangkok...
Again, through its potential for replication, this project may have an important value in reaching the
Millenium Development Goals.
Document Outline
- Project planning
- Preliminary financing plan
- Explanation
- Program Conformity:
- The proposed project falls under the GEF International Waters Focal Area, and specifically under Operational Program Number 9: Integrated Land and Water Multiple Focal Area Program.
- Although actions of both prevention and remediation of damage to waters could be envisaged in the Igarapes of Manaus, the project is more oriented towards prevention, as it is usually the case under OP 9. Still, special attention will be given to the ex
- The coordination among stakeholders, that the project will seek to establish, has direct relevance to the objective of the OP to integrate the use of sound land and water resources management strategies.
- Participation of Aguas do Amazonas is also relevant as a step toward optimizing, via private sector participation, the overall use of the full range of international best practices in the technical, economic, financial, institutional, regulatory and soci