Report No. PID9725
Project Name             Mongolia-Sustainable Livelihoods Project
Region                   East Asia and Pacific Region
Sector                   Livestock;Decentralization; Other Social
Protection
Project                  MNPE67770
Borrower(s)              MONGOLIA
Implementing
Agency(ies)              HOUSEHOLD LIVELIHOODS SUPPORT PROGRAM OFFICE
Address: Khuvsgalchdyn Avenue,
Ulaanbaatar-38, Mongolia
Contact Person: Mr Ch. Khurelbaatar, Director
Tel: 976-11-323500, 322465
Fax: 976-11-328107
Email: mppl@magicnet.mn
Environment Category     B (Partial Assessment)
Date PID Prepared        May 21, 2002
Auth Appr/Negs Date      February 14, 2002
Bank Approval Date       June 11, 2002
1. Country and Sector Background
Increasing poverty and inequality
Following the exceptionally difficult years of the early 1990s, and despite
four years of reasonable growth and low inflation in the mid- to late-1990s,
poverty and inequality continue to be serious problems in both rural and
urban areas. The Living Standards Measurement Surveys conducted in 1995 and
1998 found that the overall headcount of poor people remained stable over
1995-98 at around 36% of the population; the depth and severity of poverty,
however, worsened over this period.
The findings of the Bank-supported Participatory Living Standards Assessment
2000 (PLSA) are consistent with these trends and confirm that polarization
between rich and poor intensified between 1995 and 2000. While some
households managed to improve their living standards and climb out of poverty
over the period 1995-2000, it appears that many more fell behind, and those
already below the poverty line fell even further behind. Vulnerability to
risk was regarded as the most important dimension of poverty to address.
Analysis of the various shocks and stresses that triggered a downward trend
of impoverishment among 180 low-income middle, poor and very poor households
suggests that loss of employment, natural hazards (dzud), and the cost of
health and education were the most common risk factors. Fuel-price increases
leading to declining producer terms of trade, theft of assets (especially
livestock), and other lifecycle events (e.g. death, illness) were other
commonly experienced factors that could trigger a process of impoverishment.
A clear, common perception emerged from over 200 focus group discussions
conducted throughout the country that by 2000, in urban areas, poor and very
poor households accounted for the majority of residents; while in rural
areas, the poor and very poor made up around half of all households. These



findings suggest that government efforts should be stepped up to ensure that
the gains from economic growth are more widely shared.
Between 1994 and 2000, the principal means by which poverty reduction was
pursued in Mongolia was through the National Poverty Alleviation Program
(NPAP), a US$19 million inter-sectoral program supported by a range of
multilateral and bilateral agencies and international NGOs. World Bank was
the major donor to NPAP, contributing US$10 million through the Poverty
Alleviation for Vulnerable Groups Project (PAVGP) (Cr. 2760-MN) from 1996 to
2000. The intensive-learning ICR for this project and a joint Government-
donor evaluation of NPAP as a whole concluded that NPAP had provided valuable
support to local governments for the rehabilitation of social and economic
infrastructure, but the direct income-generation support to poor households
was much smaller than ought to have been realized, and benefited perhaps not
more than 20t of poor households. Achievements in reducing rural poverty were
particularly limited. NPAP was unable to reduce poverty on a significant
scale in the absence of a supportive economic policy environment. Since July
2000, however, the new Government has made poverty reduction one of the
highest priorities in its Action Plan and has begun to address marked rural-
urban disparities with a stronger focus on rural development. The Interim
Poverty Reduction Strategy Paper (I-PRSP) has been completed and the full
PRSP is in preparation. Government anticipates that the proposed Sustainable
Livelihoods Program will make a major operational contribution to the
achievement of the PRSP's goals.
Ambiguous institutional structure
During the early years of economic transition in the early 1990s, poverty was
considered mainly a 'social policy' issue affecting 'vulnerable groups' such
as orphans, the disabled, the elderly, the unemployed, low-income households
with many children, and female-headed households. Accordingly, a single line
ministry (currently named the Ministry of Social Welfare and Labor) was
designated under the Law on Government to take the lead on poverty issues.
The Poverty Alleviation Program Office (PAPO) - the executing agency for NPAP
- was constituted as "an autonomous entity under the supervision of the
National Poverty Alleviation Commitee (NPAC), reporting directly to the
Minister of MoSWL in his capacity as Deputy Chairman of NPAC". This proved to
be ambiguous in practice. MoSWL considered PAPO to be under its own
authority, while the donors considered PAPO to be an autonomous entity
reporting to NPAC as an inter-ministerial committee responsible for
coordinating anti-poverty efforts. Throughout the course of NPAP, MoSWL's
repeated efforts to block decisions taken by PAPO proved highly disruptive to
implementation. In an effort to resolve the issue in early 1996, GoM issued a
statement that PAPO would thereafter report directly to the Prime Minister,
in his capacity as Chairman of NPAC, thus making PAPO operationally
independent of the MoSWL. This failed to resolve the issue, however, and
inter-agency struggles for control over NPAP seriously threatened PAPO's
status at several points, at significant cost to effective program
management.
Although the original intention was that NPAP should embrace all government
activities pertaining to poverty reduction, with NPAC acting as a Cabinet
committee on poverty, in practice, NPAP served as a unified 'window' for
attracting international donor support for poverty reduction. NPAC dealt
solely with oversight of NPAP rather than with government actions and
policies that influence poverty as a whole. As a result, NPAP came to be seen
-2 -



as the totality of those internationally financed projects for poverty
reduction that were brought under the unified management of PAPO, which acted
as Secretariat to NPAC. Opportunities were missed to address two of the other
original objectives of NPAP: (a) to coordinate the poverty-related actions of
all government ministries and agencies; and (b) to establish closer links
between policy-making at central level and the on-the-ground realities at
local level. The Government's current efforts to prepare a Poverty Reduction
Strategy Paper (PRSP), with the support of World Bank and IMF, should be seen
in this light, since the PRSP aims to achieve precisely what NPAP failed to
do: namely, to 'mainstream' poverty reduction in the Government's overall
program.
Inadequate capacity building
Solid progress was made under NPAP in building local capacity to implement
development projects, through inter-departmental Poverty Alleviation Councils
(PACs) at aimag and sum levels and their equivalents (duureg and khoroo
respectively) in the capital city, Ulaanbaatar. However, since GoM was
reluctant to use loan funds for capacity building, PAVGP had to rely on
parallel financing from UNDP and bilateral donors for this purpose. Donor
support focused primarily on capacity building at national and aimag levels,
while from 1998 the sum and khoroo level became the main locus of decisions
concerning income-generation projects. The situation was exacerbated by high
turnover of sum and khoroo PAC members as the incoming political party
replaced all local-level line ministry department heads in the wake of
national elections.
Weakness in pastoral risk management
Events of the 1999/2000 and 2000/2001 winter/spring periods heightened
awareness of the importance of pastoral risk management among the Government,
donors, and civil society. The last two winter/spring dzud emergencies were
perhaps the worst in living memory, in both cases following severe summer
drought. In 1999/2000, livestock mortality was close to three million head,
or around 9t of the national herd, concentrated in six south-central and
western aimags (see Map 2). In winter/spring 2000/2001, drought was again
followed by an even worse dzud. This resulted in the loss of over four
million additional animals, amounting to almost 14t of the national herd. The
devastating impact of these consective years of dzud was exacerbated by
historically high stocking rates. Having remained at a relatively stable
level of around 25 million head for several preceding decades, largely owing
to high autumn off-take encouraged by state-guaranteed livestock- and meat-
marketing under agricultural collectivization, total livestock numbers
increased by 33t over the six-year period 1993-99, reaching a peak of over 33
million head.
It would be wrong to view dzud as simply 'natural disasters', however. The
impacts of dzud have become more severe as a greater proportion of the
national population has come to depend on livestock for a livelihood, and as
herding livelihoods have themselves become more vulnerable to risk. In the
absence of alternative livelihood opportunities, many former state-sector
employees turned to herding to support their families. The number of herding
families more than doubled from 75,000 in 1990 (accounting for 351,000
people, or 17t of the total population) to 192,000 in 2000 (representing
824,000 people, or 35t of the national population). By comparison with
existing herders, the new herders tended to be less skilled and experienced
in herding. Their herds were, on average, smaller and less likely to increase
- 3 -



over time as new herders were more likely to slaughter animals to meet
immediate livelihood needs. They also tended to move less frequently than
more experienced herders, and to remain closer to urban settlements, roads,
and other points of market access.
Under the collectivized planning of agriculture that prevailed until 1991,
the former State Emergency Fodder Fund (SEFF) and other forms of State
assistance to affected areas during dzud conditions did a great deal to
minimize livestock mortality. This system was heavily subsidized, and
collapsed once Soviet support was withdrawn in the early 1990s, which shifted
the burden of risk management from the agricultural collectives to individual
herding households. Customary land management practices involving the
seasonal separation and rotation of pastures, which were maintained under
collectivization, also began to disintegrate from the early-mid 1990s as the
number of herding families and total livestock numbers rose, and as pastoral
mobility and administrative support for livestock production declined.
Little investment was made in the 1990s in developing an alternative strategy
for mitigating and responding to risk in the livestock sector. Until
1999/2000, the need for such a system was not seriously put to the test, as
there had not been a serious dzud in Mongolia since 1993, when the remnants
of the SEFF still functioned to some extent. The Democratic Coalition
government between 1996 and 2000 was opposed to State intervention in
pastoral risk management, whereas the new Mongolian People's Revolutionary
Party (MPRP) Government (since June 2000) has clearly outlined a role for the
State to complement herders' individual efforts to prepare for and better
manage the effects of dzud. In May 2001, GoM Resolution No. 47 enacted a
policy for the protection of livestock against drought and dzud. GoM requires
external financial support to enable it to fully operationalize and implement
this new policy. Key elements of the program include: pasture land tenure and
management; hay and fodder production, conservation and marketing; livestock
marketing; early-warning and emergency response systems; and livestock
productivity-enhancing investments such as livestock breeding, veterinary and
other livestock services.
Considerable experience has been gained in developing household herd-
restocking programs as the main approach to reducing rural poverty in
Mongolia, in combination with efforts to diversify the livelihoods of those
who are unlikely to achieve sustainable livelihoods based on livestock
production. These programs have been piloted since 1995 with initial
financial and technical assistance from FAO, IFAD, and Save the Children Fund
(UK). World Bank (under the PAVGP) supported the extension and scaling-up of
the approach in five of the most severely affected aimags in response to the
1999/2000 dzud. Common elements of the approach developed in Mongolia are
that it is organized as an in-kind credit scheme administered by local
governments, with repayments in the form of animals or cash; procurement of
animals is done locally, often within the same communities (i.e. richer
herders sell animals to externally financed projects for on-lending to
poorer, eligible herders), thereby ensuring that animals are already adapted
to local ecological conditions, and that there is a neutral impact on local
animal stocking densities; and a transparent, community-driven process of
selecting eligible beneficiaries is followed.
While restocking provided an important short-term safety-net in rural
Mongolia in the late 1990s, GoM and donors have now come to appreciate that
- 4 -



it also raises serious concerns about long-run sustainability. The main
concerns relate to the disincentive effect of state-led restocking for risk
management in livestock production (a 'moral hazard' issue), and for the
wider development of the micro-finance industry. The gains from restocking
could easily be wiped out by future dzud events. The areas in which
restocking had initially been piloted escaped the most serious impacts of the
1999/2000 dzud, and it still remains to be seen how restocked households
would cope with future risk episodes. In principle, livestock insurance is
the primary means by which risk is offset under Mongolian restocking
programs, but in practice voluntary coverage has been near-zero. The
commercial viability of the conventional approach to livestock insurance,
which attempts to cover losses at the individual household level, is
seriously questioned. There is an urgent need for further work to develop and
implement an alternative approach to livestock insurance in Mongolia. The
most promising approach would base indemnities on some form of index, perhaps
based on weather data, which is independently verifiable and which would
promote rather than undermine incentives for good livestock management.
Inadequate community participation, sustainability and decentralization
As a social fund-type mechanism, the Local Development Fund (LDF) under NPAP
suffered from two serious problems. The first was that genuine community
participation in decision-making remained weak, with the notable exception of
the community-based restocking initiatives undertaken in 2000. The Poverty
Alleviation Councils reach down only as far as district-level administration
which, as the PLSA revealed, community members regard as remote and
unresponsive to their needs. Under NPAP, resources tended to be allocated
equally to all areas for a pre-defined, and rather limited, menu of sub-
projects. Another weakness was the lack of sustainability. Community
participation was intended as a key objective and mechanism under the LDF,
but was not always achieved beyond contributions of short-term, wage-labor
inputs.
Inadequate community participation is also correlated with the low level of
fiscal decentralization, which prevents local communities from actively
making local development decisions because of limited self-generated revenues
or inter-governmental transfers over which they have discretionary control.
While each of the three levels of government nominally have independent
budgets, the fiscal structure of Mongolia remains highly centralized, as
revealed by a Decentralization Assessment conducted in support of project
preparation. In 1997, for example, revenues accruing to central government
accounted for 77% of total government revenue, and central government
accounted for 71% of total public expenditure (excluding transfers to sub-
national governments). Although this left only 299 of total public spending
in the hands of sub-national governments, half of this amount still came from
central government transfers. In other words, only half of the already small
fraction of total public spending that sub-national governments control is
generated from local revenue sources, and essentially all of this goes to
meet recurrent expenditures. In fact, there is circular relationship between
community participation and fiscal decentralization. The currently low level
of fiscal decentralization inhibits the voice and influence of local
communities over public expenditure, while inadequate community participation
makes fiscal decentralization less feasible because local capacity remains
under-developed.
Decentralization in policy and practice
- 5-



After coming to power in June 2000, the current Government set out in its
Action Program an ambitious agenda to promote decentralization. The
mechanisms envisaged included the promotion of regional and rural
development, in part to address the over-concentration of economic
opportunity in the capital city and central, more market-accessible regions
of the country; increasing accountability in the public sector; increasing
the capacity of local governments to generate revenues over which they have
discretionary control; and offering citizens greater opportunity to directly
monitor the performance of their elected citizen's assemblies (khurals) at
various levels.
The proposed Public Sector Management and Finance Law seeks to increase the
accountability of public sector institutions at all levels, including civil
service reform linked to the budgeting process and overhaul of the existing
system of inter-governmental transfers. Government's 'Good Governance for
Human Security Program' also refers to 'decentralizing and empowering local
self-governance and local administration', and aims to support 'stakeholder
consultation, voice, and participation' in the broader policy process,
including 'monitoring citizen satisfaction and recommendations with regard
to services, rights, and government responsiveness to citizen demands'. The
I-PRSP echoes the broad theme of raising public accountability, notably
through greater public voice in and scrutiny of the budget process.
To date, however, decentralization in Mongolia remains both imbalanced and
incomplete in practice. The Decentralization Assessment conducted in
association with project preparation shows that, measured against a set of
standardized criteria developed through international comparative
benchmarking, Mongolia is currently characterized by substantial though
incomplete political decentralization, but little administrative
decentralization and virtually no fiscal decentralization. Administrative
decentralization follows a deconcentrated pattern in which governors and
their staffs report upwards to the next highest level of government. There is
little opportunity for elected assemblies to exert any influence over
technical department staff and quality of service delivery at their
respective level of government, and therefore little incentive for local
populations to express their views to their elected representatives. Evidence
from the PLSA reveals that community members are widely dissatisfied with the
extent to which local governors take account of their interests.
Current efforts of Government, with the support of IMF and ADB, are focusing
on fiscal 're-centralization' in the interests of fiscal control and to
improve the overall tax collection rate. In parallel, however, efforts are
being made to develop a transparent, formula-driven approach to determining
inter-governmental transfers, which account for the vast majority of most
local authority revenues, particularly in rural areas. The formula is still
under discussion, but is likely to include an element of poverty-targeting.
Once a transparent basis for inter-govermental transfers is established, it
is then imperative to ensure that expenditure allocation at the local level -
between sectors and between alternative investments within individual sectors
- takes local needs and aspirations into account.
Under-development of financial systems, particularly in rural areas
Mongolia's financial system, at this point in its transition, is small,
fragile, and highly dysfunctional, with recurring insolvencies in the banking
system and a high level of vulnerability to economic shocks and crises. These
- 6 -



weaknesses, together with the narrow range of available financial services,
seriously constrain private sector development and growth. In a recent World
Bank-funded survey of Mongolian entrepreneurs, 969 reported the lack of long
term financing and 65t reported inadequate working capital as the main
constraints to their business operations. An IFC-supported study found that
although the informal sector is estimated to generate economic activity
equivalent to as much as 30- to 38t of official GDP, informal sector
operators obtain their finance mainly from personal and family savings (82%),
followed by individuals (11%), and only 0.5% from bank loans. Mongolia's
financial system is characterized by the following themes, as summarized by a
recent IFC report: (a) lack of confidence in banks; (b) predominant state
ownership and concentration of assets; (c) collapse of financial services in
rural areas; (d) poor corporate governance; (e) lack of management experience
and skill; and (f) serious rent-seeking behavior. Furthermore, it was
estimated around 1999 that some 95% of cash circulating in Mongolia's economy
was concentrated in the capital, Ulaanbaatar; the vast proportion of trade in
rural areas remains on a barter basis. Until very recently, those in rural
areas, and the poor virtually everywhere, had practically no access to
financial services of any kind, with the exception of micro-loans being
provided by NPAP through the PACs. The environment for micro-finance service
delivery has improved dramatically over the last two years, however, with the
restructuring of the Agriculture Bank (now opening new rural branches) and
the remarkable success of the (still largely urban-based) XAS-Bank which has
rapidly achieved financial and institutional sustainbility in micro-credit.
Government is committed to increasing outreach of financial services to poor
and near-poor households on a sustainable basis, both in rural and in peri-
urban areas. A new micro-finance law is under preparation.
2. Objectives
The development objective of the project - referring to the first four-year
phase of the overall program - is: "an effective approach to promoting
improved, secure, and sustainable livelihood strategies developed,
demonstrated, and validated in selected areas, and institutional capacity
created so that these strategies can be replicated and scaled-up in Phase II
of the Program".
The project would be national in scope, with intensive pilot-testing of
certain institutional innovations in eight selected aimags (see Map 1). In
parallel with the pilot-testing of new approaches, selected NPAP activities
would continue to be implemented nation-wide, with the aim of facilitating
convergence between the new approaches and existing anti-poverty
interventions over the medium term. The outputs that in combination will
contibute to achieving the project development objective are:
a)      Pastoral Risk Management: an integrated strategy developed, piloted,
and adopted in eight selected aimags for managing covariant risk in pastoral
livestock production, with a primary emphasis on risk preparedness. In
combination with other project components, this component would aim to assist
in addressing the underlying causes of rising vulnerability to drought and
dzud ('winter disasters');
b)      Micro-Finance Outreach: the outreach of financially and institutionally
sustainable micro-finance services (including savings, credit, and insurance)
to targeted poor and vulnerable non-poor households and individuals in rural
areas of selected aimags achieved;
- 7-



c)      Local Initiatives Fund: efficient, socially inclusive, and transparent
community-driven mechanisms identified and established to facilitate
community prioritization, selection, co-financing, and execution of
investments in basic infrastructure and social services provision in rural
areas, combining local resource mobilization with matching government grants
based on clear and transparent criteria for inter-governmental transfers; and
d)      Project Management: a decentralized project management framework
realized through the progressive devolution of project management
responsibilities to aimag and sum (rural district) levels within local
governments, while retaining overall coordination, monitoring and evaluation,
and fiduciary oversight at national level.
3. Rationale for Bank's Involvement
The value added through Bank support for this project is fourfold:
First, the Bank brings broad international experience and technical knowledge
to bear in several areas critical to project success: (a) community-driven
development, including Africa Region's Community Action Programs, demand-
driven rural investment programs particularly in EAP and LAC Regions, and
social funds more broadly; (b) social risk management (particularly through
the Social Protection Family); (c) micro-finance services and strategies,
particularly through the Bank's close association with CGAP; and (d) pastoral
development, particularly through the Animal Resources and NRM Institutions
thematic groups.
Second, the Bank has made a long-term commitment to supporting poverty-
reduction efforts in Mongolia, and acts as the lead development agency in
this field. The proposed project builds directly on lessons learned under
NPAP, for which the Bank was the major financier through PAVGP; and it is
provided for in the CAS. Continuity of support for poverty reduction from the
Bank is regarded as a priority by GoM, and is reinforced through the PRSP
process. The Bank is in a strategic position to help government coordinate
donor efforts with respect to livelihood-based poverty reduction, designing
and financing a local initiatives fund, strengthening the decentralization
process, and securing sustainable financing mechanisms for maintenance.
Third, this would be the first major investment program in international
experience designed explicitly according to Sustainable Livelihoods concepts
and principles. It should yield important lessons on how to translate such a
concept into a concrete set of investments, while respecting the need for
flexible design in order to accommodate community initiatives, process-based
iterative learning, and to explicitly plan for external shocks the timing and
intensity of which are difficult to predict.
Fourth, the Bank would be the first international financial institution to
support the establishment of an integrated pastoral risk management system on
national scale, building on lessons learned under the Kenya Arid Lands
Resource Management Project. Should the pilot initiatives under Phase I of
SLP prove successful, it would open up possibilities for replication not only
in Mongolia but among pastoral peoples of Northwestern China and Central Asia
as well as in drought-prone areas such as the Horn of Africa.
4. Description
Pastoral Risk Management: an integrated strategy developed, piloted, and
- 8 -



adopted in eight selected aimags for managing covariant risk in pastoral
livestock production, with a primary emphasis on risk preparedness. In
combination with other project components (particularly the index-based
livestock insurance scheme under the MFO component, promotion of 'exit
strategies' for new herders through access to micro-finance, and demand-
driven support for public goods in agriculture through the LIF), this
component aims to reduce the vulnerability of herders and enhance their
resilience to drought, dzud and other shocks through four sub-components: (i)
risk forecasting and contingency planning (improving coverage of weather
forecasting data, broadening the range of data used in semi-annual early
warning system bulletins and enhancing their dissemination to local
authorities and herders, and development of sum-level contingency plans);
(ii) grazing and pasture management (support for community-based pasture
management, including mapping and re-instituting seasonal pasture rotation
with credible sanctions for non-compliance, provisions for negotiating access
with other herding communities during times of drought or dzud, and
mechanisms for conflict resolution; rehabilitation of emergency grazing
reserves at inter-aimag and inter-sum levels; and group-based well
rehabilitation in areas of high quality yet currently under-utilized
pasture); (iii) herder self-help initiatives (organizational strengthening
with time-bound, matching loans to assist groups in establishing revolving
funds for livestock productivity-enhancing investments such as breeding and
veterinary services, fodder purchase and/or production, livestock product
marketing); and (iv) hay and fodder enterprise development (testing business
models for hay-making and feed concentrate production under alternative
technological and market-demand conditions). Implementation would begin in 16
pilot sums in the first year of the project, expanding to all 143 sums in the
eight core aimags by the end of project, with the relative emphasis among
sub-components and activities varying according to ecological and market-
access conditions.
Micro-Finance Outreach: the outreach of financially and institutionally
sustainable micro-finance services to targeted poor and vulnerable non-poor
households and individuals achieved in remote rural areas of the eight core
aimags. This objective would be met through three sub-components: (i)
creation of a Micro-Finance Development Fund (MDF, the Fund) as a wholesale
lending facility to accredited Micro-Finance Institutions (MFIs) for on-
lending to the target population (sub-loans), with the aim of expanding and
diversifying livelihood sources and rural incomes. The day-to-day management
and operation of the MDF would be carried out by the Micro-Finance Management
Office (MMO, the Management Office), to be established within the Bank of
Mongolia (BoM, the Borrower's Central Bank), in accordance with a Micro-
Finance Policy Manual (the Policy Manual) acceptable to the Bank. This MMO
would be supervised by a Micro-Finance Development Board (MDB, the Board),
which would review the operation of the Fund, the credit activities (loans
and sub-loans), and the strategic decisions on policy matters. The Board
would comprise representatives of the GOM (MoFE, BoM, HLSPO), the private
sector (Banking Association, commercial banks and micro-finance
institutions), and academia. Proceeds accrued from the repayment of the
subsidiary loans by the MFIs would be maintained in the MDF for re-lending
for the same purposes and under the same terms and conditions as those under
the original IDA credit. The sub-component would also include substantial
technical-assistance resources to promote the strengthening of the MFIs and
to support the development of the microfinance industry in Mongolia,
including: (i) to help selected MFIs prepare and implement adequate business
-9-



plans; (ii) to identify (through market research) and develop new micro-
finance products tailored to local demand and market conditions; and (iii) to
develop the legal and regulatory framework required for the efficient
operation of non-banking financial institutions (NBFIs) as micro-finance
intermediaries. As conditions for the disbursement of the IDA funds allocated
to this sub-component for the micro-finance loans (Category 4, for a total
equivalent to US$3 million), the Borrower would: (a) adopt a Policy Manual
acceptable to IDA for the implementation of the sub-component; (b) establish
and maintain during project implementation the Micro-Finance Development Fund
(including the Board and the Management Office) in terms and conditions
satisfactory to IDA; and (c) accredit MFIs (at least one) and sign Subsidiary
Loan Agreements (SLAs) for the disbursement of subsidiary loans, in
compliance with the accreditation criteria and the terms and conditions set
forth in the Policy Manual.
(ii) Strengthening of Revolving Loan Funds (RLFs) so as to improve the
operational performance and financial health of the existing, local
government-owned and -operated micro-credit RLFs that were created at the sum
level under NPAP. The process of strengthening these RLFs would include: (i)
review of their governance and management arrangements; (ii) collection and
analysis of information on current operations and on-lending arrangements,
quality of loan portfolio, and financial statements; (iii) selection of best-
performing RLFs, which have the best chances of achieving financial
sustainability; (iv) design of business plans for the selected RLFs to help
them improve their efficiency; (v) provision of loans from the Borrower to
help the selected RLFs implement their business plans and improve their
financial situation; and (vi) eventual privatization of the best-performing
RLFs. As a condition for the disbursement of the IDA funds allocated to the
provision of loans to the selected RLFs (Category 5, for a total equivalent
to US$ 500,000), the Borrower would prepare a detailed Action Plan for the
use of these funds, acceptable to the Bank. This Plan would include the
institutional arrangements and responsibilities, the on-lending terms and
conditions, the criteria for the selection of RLFs eligible to receive new
loans, the purposes and uses of the funds, as well as the steps to be taken
towards the eventual privatization of these RLFs.
(iii) Development of an index-based livestock insurance scheme. A risk index
would be developed during the first year of the project, on the basis of
which participating private insurance companies would offer livestock
insurance to individual herders. The index, based on weather and/or livestock
mortality data, would differentiate relative risk at an appropriate level
(most likely using sum-level data). Indemnities under the scheme would be
based on the sum-level index rather than on losses at the household level.
All holders of insurance would therefore receive payments whenever given
thresholds in the index were exceeded in their locality, irrespective of
their individual losses, thereby preserving incentives for good livestock
management. The project would also finance training workshops for
participating insurance companies and the Government officials involved, and
a nation-wide information campaign to publicize the scheme and attract
clients. The scheme would be launched during the second year of the project,
and would be expected to become profitable by the end of the project. During
the start-up phase (project years 2 to 4), Government would offer a stop-loss
provision to participating insurance companies, financed from the proceeds of
the Credit, to limit their losses in the event of a major dzud. This
provision would take the place of international reinsurance, which would be
- 10 -



unlikely to be attracted until commercial viability of the scheme could be
demonstrated. A condition of disbursement of IDA funds to finance the stop-
loss provision (Category 6, for a total equivalent to US$ 1,000,000) would be
that a methodology, including eligibility criteria for participating
insurance companies, institutional responsibilities, and flow of funds, has
been elaborated in a manner acceptable to the Bank. Under Phase II of the
Program, the index-based insurance scheme would be expected to operate in a
financially and institutionally sustainable manner, without the need for
Government to offer a stop-loss provision. Insurance companies offering
index-based livestock insurance would then be expected to seek international
reinsurance.
Local Initiatives Fund: efficient, socially inclusive, and transparent
community-driven mechanisms identified and established to facilitate
community prioritization, selection, co-financing, and execution of
investments in basic infrastructure and social services provision in rural
and peri-urban areas, combining local resource mobilization with matching
government grants based on clear and transparent criteria for inter-
governmental budget transfers. Three sub-components would deliver this
output: (i) two separate investment funds: the Local Initiatives Fund (LIF),
a new facility that would be piloted in the eight core aimags and one pilot
duureg (peri-urban district of Ulaanbaatar); and, in modified form, the Local
Development Fund (LDF) which was initiated nation-wide under NPAP, in the 13
other aimags and 8 other duuregs in the country; (ii) training and capacity-
building activities, focusing primarily on the sum and bag levels in the
eight core aimags, for orientation, skills development in administration and
management (including community-based procurement and disbursement), study
tours and exchanges, refresher training, and the development of training
materials; and (iii) a fiscal decentralization initiative, comprising a
three-year action-research program to explore operational modalities for a
needs-driven approach to public expenditure allocation at the local level.
Implementation of the LIF would begin in the 16 pilot sums in the first year
of the project, expanding to cover all 143 sums in the eight core aimags by
the end of project.
Project Management and Policy Support: an efficient, decentralized, and
capacity building-oriented project management system operational and policy
framework for promoting sustainable livelihoods improved. The project would
not support new or additional project management arrangements at central,
aimag, or sum/duureg levels. Rather, in order fully to capitalize on the
experience and skills developed under NPAP, the project would make use of and
build on existing program management structure for project implementation.
Strengthening existing institutional capacity would be the main priority.
Under the technical and policy guidance of the three Working Groups for each
of the above three components, and ultimately reporting to the National
Committee on the Household Livelihoods Capacity Support Program (NCHLCSP),
the Household Livelihoods Support Program Office (HLSPO) would have overall
responsibility for project implementation and coordination. Local governments
and community groups would carry the main responsibility for executing the
project through the aimag and sum Household Livelihoods Support Councils
(HLSCs). Under this component, the project would finance: (i) the investment
and recurrent costs of the HLSPO, including incremental staff as long-term
consultants on fixed-term contracts at central, aimag and sum levels; (ii)
policy advocacy and legislative support activities; (iii) information
dissemination and communications; (iv) project orientation workshops and
- 11 -



training; (vi) monitoring and evaluation; and (vii) auditing.
Pastoral Risk Management
Micro-Finance Outreach
Local Initiatives Fund
Management and Policy Support
5. Financing
Source (Total ( US$m))
BORROWER ($3.39)
IDA ($18.73)
Total Project Cost: $22.12
6. Implementation
A number of principles have guided the design of institutional and
implementation arrangements for the project: (a) work within the existing
government structure and institutions; (b) build upon the capacity already
created under NPAP; (c) recognize that creating capacity takes time and is
not uniform, therefore build flexibility into the phasing-in of project
activities; (d) keep initial procedures and processes simple so that the
project can innovate and test new approaches for broader adoption in later
stages; (e) establish the management structure in a step-wise fashion,
allowing time to test and refine procedures and develop the necessary
training material and train trainers before expanding to other areas; (f) use
the sum (district) as the primary focus for most project decision-making and
implementation; and (g) create a transparent set of procedures that
facilitates participation by communities, and the poor and vulnerable
households within those communities, so that prioritization and uptake of
project activities is driven by them. The institutional structure for
program/ project oversight, coordination, capacity building and
implementation is further articulated in Annex 13 and in the form of
organograms in Annex 15.
Institutional Structures: Overall project oversight will be provided through
the National Committee for the Household Livelihood Capacity Support Program
(NCHLCSP). This inter-ministerial body supersedes NPAC, which performed the
same function for the NPAP. It includes senior representation from all the
relevant ministries (including Finance and Economy (MoFE), Food and
Agriculture (MoFA), Social Welfare and Labor (MoSWL), Infrastructure (MoI),
Health (MoH), Education (MoSTEC), and Nature and Environment (MoNE));
representative NGOs/ civil society institutions; and directors of other major
poverty-related operational programs.
Coordination of project implementation at the central/ national level will be
the responsibility of the Household Livelihoods Support Program Office
(HLSPO). This office supersedes PAPO, which served as the program
implementation unit for NPAP, and will implement project activities under the
guidance of the NCHLCSP. Incremental staff (as long-term consultants on
fixed-term contracts), office equipment, and a small number of vehicles would
be financed from the proceeds of the Credit for the four year duration of the
project. Some of the key staff proposed for the HLSPO also held key positions
in PAPO and are already familiar both with the lessons from NPAP experience
and with the relevant World Bank policies and procedures concerning
procurement, disbursement, and reporting. The HLSPO has two divisions: 1) a
program implementation division with a full-time coordinator for each of the
- 12 -



three main project components (pastoral risk management, micro-finance
outreach, and the LIF/LDF) and an officer responsible for ensuring
coordination among these components; and 2) a division responsible for the
fiduciary aspects of project management, including: (a) financial management;
(b) procurement; (c) monitoring and evaluation; (d) information
dissemination, training coordination, and public relations; (e) donor
coordination; and (f) overall project planning and budgeting. HLSPO would be
responsible for procurement of all goods to be centrally procured, all works
subject to NCB (rehabilitation of two irrigation schemes under the PRM
component), and all consultant contracting.
For each of the three main components, technical Working Groups at national-
level will provide technical guidance and oversight, and would ensure
consistency with sectoral policies. The PRM Working Group is led by MoFA and
includes members from MoNE, MoSWL, and civil society. The LIF Working Group
is led by MoFE and includes representation from all the ministries listed
above plus civil society. The MFO 'working group' is the Board of the MDF.
Two HLSPO staff members participate in the Working Groups as ex officio
members; each component coordinator serves as the secretary of their
respective working group.
Execution of program/project activities at the aimag, sum and duureg levels
will primarily be the responsibility of sub-national (aimag) and local (sum,
duureg) governments. The existing Poverty Alleviation Councils (PACs) at each
of these levels, which served as inter-departmental steering committees for
program implementation under NPAP, will continue to perform the same function
under the proposed program/project, under the new name Household Livelihoods
Support Councils (HLSCs). Their membership will also be broadened as
appropriate, notably to ensure greater involvement of the agriculture and
environment-related technical staff whose involvement is key to successful
implementation of at least the PRM component. HLSCs in the sums of the eight
core aimags selected for pilot activities would be considerably strengthened.
At aimag level, technical Working Groups for the PRM and LIF components,
analogous to the national-level Working Groups, will be constituted from the
relevant line departments to advise on their respective detailed work plans
and budgets. At bag level and below, civil society community mobilizers will
work with the sum HLSC to facilitate, inter alia, herder-group formation for
self-help initiatives, community-based pasture management, and well-
rehabilitation under the PRM component; and identification and prioritization
of potential sub-projects to be financed under the LIF.
Implementation Arrangements: The PRM and LIF components would begin
activities in two pre-selected, pilot sums in each of the eight core aimags
during the first year of the project (a total of 16 sums), and thereafter
scale-up to include all 143 sums in the eight core aimags over project years
2-4. Full implementation in all sums will only begin once the processes and
procedures have been adequately developed, modified and refined, and all
staff have been trained (including training trainers) over the initial 12
month period. The particular activities to be implemented in each sum under
the pastoral risk management component will vary according to local
ecological and market-access conditions. The MFO component would work with
sum RLFs in the eight core aimags, progressively expanding the number covered
in line with the capacity to provide training to them. In the non-core aimags
and duuregs, project activities would be very similar to those implemented
under the LDF (of NPAP), and the HLSCs would operate in the same way as did
- 13 -



the PACs under NPAP, in all sums and duuregs over the full four years of the
project.
Micro-Finance Development Fund (MDF). The daily management and operation of
the MDF (as a commercial wholesale banking facility) would be the
responsibility of the Micro-Finance Management Office (MMO), to be
established within the Bank of Mongolia (BoM), and which will be adequately
budgeted, well-equipped, and staffed with qualified individuals to be able to
perform its duties and responsibilities. The MMO would maintain separate
accounts (according to internationally acceptable accounting practices) for
the sub-component (to be audited annually by private auditors), and would
present quarterly progress reports to the HLSPO concerning its operations, as
well as the documentation required to fully support the disbursements from
the credit account. MMO would handle day-to-day operations of the funds in
accordance with a Micro-Finance Policy Manual (the Policy Manual) acceptable
to the Bank. The main sections of the Policy Manual include: (i) governance
and management of the MDF; (ii) eligible sub-projects; (iii) type of
financial institutions to be eligible to participate in the project; (iv)
qualifying (accreditation) criteria for MFIs; (v) uses of funds (lending and
sub-lending); (vi) terms and conditions (currency, lending rates, maturity,
sub-project financing, collateral, default, and prepayment) for loans to MFIs
and sub-loans to final beneficiaries; and (vii) environmental protection,
audit, supervision and reporting. A draft Policy Manual was reviewed at
appraisal and found to be satisfactory. In order to ensure a full
coordination and efficient use of credit resources, MMO's policies and
lending operations would be supervised by a inter-institutional Micro-Finance
Development Board (MDB), to be established with representatives from the
Borrower's MoFE (as chair), HLSPO, and BoM, as well as representatives from
the private sector (i.e. Mongolian Banking Association, commercial banks, and
MFIs), and academia. The Director of the MMO would serve as Secretary of the
Board. As part of the Project's Mid-Term Review, the Borrower and the Bank
agreed to carry out a comprehensive assessment of the institutional and
financial performance of the Micro-Finance Outreach component in general, and
of the MDF in particular. Based on the results of this review, Terms of
Reference would be prepared for a more detailed and in-depth study of the
micro-finance sector in Mongolia, with the special objective of outlining
alternative options for the future of the MDF. One of the main options to be
explored would be the privatization of the MDF, where the resources available
could be used by the GoM as its contribution to the equity of a new
independent and financially autonomous apex micro-finance institution.
Consequently, one of the conditions for the implementation of the Phase II of
this APL could be the agreement of a specific Plan of Action to implement the
recommendations arising from this privatization study.
Uses of Funds. The funds under the MDF would be lent to accredited MFIs,
according to a pre-agreed selection criteria, under signed Subsidiary Loan
Agreements (SLAs) acceptable to IDA. Proceeds accrued from the repayment of
the principal of the subsidiary loans by MFIs would be used to establish and
maintain MDF as a revolving fund for re-lending for the same purposes and
under the same terms and conditions as those under the proposed IDA Credit.
These MFIs would provide mainly short-term sub-loans to project beneficiaries
in the target areas (individual households and micro-enterprises) to finance
working capital for their viable initiatives. If needed, medium-terms loans
(up to three years) to finance small capital investment would also be
granted. Sub-loans under this sub-component would not exceed the equivalent
- 14 -



of US$400 for individual households and US$1,000 for micro-enterprises
(employing at least two persons outside the immediate family members). IDA
financing under the loans to MFIs would not exceed 75t of total sub-project
costs; therefore, at least 25t of the total subproject cost would be provided
by either the MFIs or the final sub-borrowers from their own funds.
On-lending Arrangements. The proceeds of the IDA credit allocated to this
sub-component (Category 4, amounting to around $3 million) would be made
available to MDF directly from the Special Account. The Project's HLSPO
would disburse and transfer to the MDF account the amounts (in local
currency) required in accordance to the terms of SLAs signed with accredited
MFIs. These funds in MDF would the used for providing loans to selected MFIs
(under signed SLAs), under lending terms to be adjusted periodically
according to criteria that fully reflect market interest rates. The lending
variable rate of interest will be adjusted monthly and intended to be equal
to the adjusted cost of medium-term deposits in the local market to avoid
introducing distortions in the local market or discouraging the mobilization
of savings by the financial intermediaries. A market-based indicator,
reflecting the actual cost of loanable funds in the banking system, to be
used as a reference for setting and adjusting these interest rates would be
agreed upon, as indicated in the Policy Manual, in a manner acceptable to the
Bank. The Borrower would assume the full foreign exchange risk and the
interest rates on the loans to MFIs would cover this foreign exchange risk,
as well as all administrative costs incurred in the establishment and
operation of the MMO, and the credit risk associated with the subsidiary
loans to MFIs. The on-lending terms from MFIs to final sub-borrowers (on the
sub-loans) would be determined freely by the MFIs, according to market forces
and their assessment of the associated credit risks and adequate margins.
These rates should be sufficient to cover the cost of funds and their
operating expenses, as well as provisions for possible loan losses and
adequate level of profits, to allow for further lending expansion to the
target population without undermining their prudent equity to risk asset
ratio. The Borrower, through MDF/MMO, would exchange views and discuss
possible adjustments with: (i) the Bank, on the interest rates on subsidiary
loans to participating MFIs; and (ii) the participating or potential
candidate MFIs, on the level of interest rates charged by them on sub-loans
to final sub-borrowers. This would be done at least once a year, around June
30, commencing on June 30, 2002.
Monitoring & Evaluation. The monitoring, evaluation and reporting system for
the project will comprise performance and impact monitoring and evaluation,
building on the participatory techniques developed and used under NPAP. It
will continue to rely on a combination of project staff and contracted
specialists from NGOs/research institutions for its execution. A Planning/M&E
Officer within the HLSPO will be charged with overseeing the M&E system,
consolidating the reports from the three components, and compiling
information on management and policy support activities. S/he will be
responsible for providing project management with timely and regular
information on performance and impact. The majority of the inputs for the
project's M&E system will come from M&E systems of the three components. Each
component will rely on the aimag and sum HLSCs to carry out day-to-day
monitoring of component activities, with the aimag HLSCs carrying prime
responsibility for component evaluation. The backbone of the system will be a
newly developed MIS. The more interactive, qualitative methods, incorporating
bag-level working groups and the use of community scorecards, will be
- 15 -



combined with the compilation of formal information geared to component
performance and achievement of targets as specified in annual work programs
and budgets.
Financial Management:    To the maximum extent possible, financial management
of the project would rest with the sum-level HLSC coordinators/secretaries,
with the aimag level providing financial oversight and supervision. Overall
responsibility would rest with the HLSPO, which would ensure an adequate flow
of funds to each component, supervise the financial management of the central
and sub-national project management units, and be responsible for
consolidating financial reports, audits and reporting to GoM (NCHLCSP) and
the World Bank. The most demanding element of the financial management system
would be the implementation the LIF and the processes involved in reviewing,
screening, verifying, and approving sub-projects. The cornerstone of the
system would be an annual funding ceiling allocated to each sum. Within this
ceiling, the sums (and the pilot duureg) would identify and submit sub-
projects for financing. The initial amount of the funding ceiling would be
based on the population of the sum. In subsequent years, the ceiling would be
adjusted according to a formula reflecting agreed performance criteria.
7. Sustainability
Pastoral Risk Management: The PRM component has been designed to ensure the
social, economic, financial, institutional, and environmental sustainability
of all activities under the component. All activities have been designed with
an exit strategy in mind and with the aim of enabling local people to
implement the activities themselves, even in the absence of project support,
by the end of project. In designing pilot activities, the concern was to gain
four years of experience in a reasonable number of field sites in order to
assess sustainability in both dzud and non-dzud years. The grazing and
pasture management sub-component would promote sustainable livestock-based
livelihoods through the sustainable management of herds and natural
resources. The sustainable management of wells, and the rangeland around the
water points, would be ensured by pilot-testing and adopting a community-
based approach to well rehabilitation. The community would own the wells and
form well-user groups that would charge user fees in order to cover the cost
of operation and maintenance. They would also be responsible for conducting
monitoring water quality and quantity, and monitoring vegetation changes
around the wells in accordance with guidelines to be devised in a
participatory manner.
The herder self-help initiatives sub-component would build socially viable
and sustainable forms of cooperation at community level as a basis for
implementing joint activities such as grazing management, well-
rehabilitation, hay growing, fodder stock-piling and veterinary care.
Veterinary services such control of parasites that threaten animals' winter
survival would be offered to herders for a fee. The purpose of creating
multi-purpose revolving funds at herder-group level would be to test whether
a group-level risk fund could be run and managed by herders themselves on a
self-sustaining basis. This might involve buying in and reselling hay at a
profit to cover transport costs and to keep the fund turning over year after
year.
The main objective of the hay and fodder enterprise development sub-component
is to pilot-test various business models to guage whether commercial hay and
fodder making can be financially self-sustaining, not only in dzud years when
- 16 -



demand for fodder is high, but also in non-dzud years when demand is low.
Likewise, the purpose of the aimag-level emergency fodder reserve pilots is
to discover whether they can be run as revolving funds on a financially
sustainable basis. This would involve holding part of the stocks of feed
concentrates over their 3-4 year shelf life, and selling part of the stock in
non-dzud years to renew the stock and to cover operating expenses.
Micro-Finance Outreach: At the development objective level, the micro-finance
outreach component would contribute to sustainability within the livestock
sector as a whole by providing an 'exit strategy' for the newcomers to
herding, and would enhance livelihood security and sustainability for this
group specifically. It is envisaged that the Micro-Finance Development Fund
would become an autonomous apex institution under Phase II of the program,
which would represent a strong move towards sustainability within the micro-
finance industry. In the meantime, the MDF and supporting technical
assistance services would help to ensure that existing MFIs are stronger, are
able to offer services on a sustainable basis to currently under-served
clients in remote rural areas, have access to training and good-practice
information, have improved and more reliable access to funding, operate
within a conducive policy and regulatory environment, and will be able to
finance a higher percentage of their lending operations from savings
deposits. The index-based livestock insurance scheme would be implemented by
private insurance companies in the interests of long-run sustainability;
Government's role would be to underwrite a stop-loss provision only during
the initial pilot phase. Once the feasibility and profitability of index-
based livestock insurance had been proven, the stop-loss provision would be
phased out and replaced by international reinsurance.
Local Initiatives Fund Component: The key to sustainability of the LIF is to
facilitate its incorporation into the government's regular process of
resource allocation, through the budgetary process, and thereby to gradually
reduce its dependence on external funding. This is a medium term goal and one
that will not be achieved in the first phase of the program. However, it will
be necessary to pilot initiatives and build a solid foundation on the basis
of which the fiscal decentralization process can be mainstreamed in the
second and third phases of the program. The pilot fiscal management
initiative, to be implemented through action-research in the last two years
of the project, will assist in accomplishing this goal. MoFE is the
government body responsible for fiscal affairs, the budget process and
allocation of government budget, and in the interests of sustainability will
take the lead in the Working Group for the LIF component and on the fiscal
management initiative directly.
8. Lessons learned from past operations in the country/sector
Detailed lessons have been learned from experience under NPAP and the IDA-
financed Poverty Alleviation for Vulnerable Groups Project (PAVGP), through a
joint evaluation conducted by Government, UNDP and World Bank in late 1999 in
conjunction with identification of the proposed program, and through the
intensive learning ICR conducted for PAVGP. Overall, NPAP was found to have
been a timely response to the economic and social problems of the vulnerable
poor during Mongolia's transition to a market economy. Its success under
extremely difficult conditions during a period of economic transition can be
attributed to:
a highly committed and dedicated implementing agency at national level
- 17 -



(PAPO);
the use of an existing network of local government institutions at
aimag and sum levels as a delivery mechanism, which allowed the project
to work efficiently in spite of severe constraints such as low
population density, scattered settlement, seasonal population
movements, harsh winter conditions, and poorly-developed transportation
and communication networks;
the introduction of a bottom-up approach to project identification and
implementation;
the trust placed in local government's capacity to identify, contract,
supervise and report on public works schemes and basic education and
rural health projects; and
the emphasis on making certain that funding was in place for O&M of
whatever structures and services were rehabilitated by the project.
NPAP's most successful component was the Local Development Fund (LDF), which
financed small projects in support of basic education, rural health and
public works. The rural health and basic education sub-components achieved
their objectives of reducing maternal mortality and school dropouts. The
creation or rehabilitation of local infrastructure had a quick and direct
impact on income earning opportunities and standards of living. However, the
temporary employment created by public works programs was too brief to have
had a significant impact on the livelihoods of the poor.
The key to the success of the LDF was its bottom-up, decentralized approach
and its reliance on a well-developed network of cross-sectoral Poverty
Alleviation Councils (PACs) at aimag/ duureg and sum/ khoroo levels. Local
administrations were empowered to identify and implement projects, contract-
in services, and account for expenditure. The LDF constituted practically the
only capital budget at the discretion of the decentralized administration.
Operation and maintenance was ensured by requiring that local governments
allocate the recurrent budget prior to and as a condition for project
approval. Works were useful and were greatly appreciated by communities.
Beneficiary communities also demonstrated their commitment by contributing to
the operation and maintenance of the works.
The main areas identified in which LDF could be improved were: (a) greater
beneficiary participation; (b) stronger integration with the existing public
administration and budgetary process; and (c) increased capacity of local
government to continue the Fund's operation beyond the end of NPAP.
The income-generating credit schemes were the weakest aspect of NPAP and the
Bank-financed PAVGP. The main cause was poor design. In 1995, in the absence
of an operational bank with a branch structure, or a specialist micro-finance
company serving rural areas, a decision was made to develop a government-
administered credit scheme that operated through local government PACs at
aimag/duureg and sum/khoroo levels. There were three main credit programs:
(a) Vulnerable Group Organization (VGO) scheme, which operated under LDF; (b)
Income Generation Fund (IGF) and (c) Women's Development Fund (WDF). In
addition, loan repayments from the other three schemes were channeled into a
series of sum and khoroo level Revolving Loan Funds (RLFs) under the direct
control of sum and khoroo PACs. Of the four schemes, the most interesting and
successful experience - and the only one with the potential to become self-
sustaining - has been that of the RLFs.
- 18 -



The main lessons from NPAP are:
Poverty alleviation programs - no matter how sound in concept and
principles - are unlikely to reduce poverty significantly unless they
are supported by broad-based, people-oriented, and labor-intensive
growth strategies. In the absence of supportive economic policies and
pro-poor expenditure allocation, the gains of targeted poverty
alleviation programs are likely to be offset by growing unemployment
and economic deterioration. Poverty reduction is not sustainable if -
for every person raised above the poverty line - another person falls
into poverty to take his or her place.
Since poverty reduction is an issue that transcends sectoral
boundaries, the proper location for a poverty program management unit
is either under the Prime Minister's Office or in a cross-sectoral
ministry such as Finance, Planning, or Local Government. A social
sector ministry was a poor initial choice.
The establishment of a cross-sectoral implementing agency (such as
PAPO) facilitates poverty project implementation but creates conflicts
and uncertainty. Lack of clarity about PAPO's autonomy from MoSWL was a
source of continual institutional conflict throughout the life of the
project.
Decentralization of project management responsibility can be an
effective way of overcoming institutional blockages at national level.
Although controversy over PAPO's autonomy hindered implementation at
central level, it was not a serious problem at aimag level and below.
Open public meetings at grassroots level facilitate successful
implementation of a bottom-up, demand-driven approach. The restocking
component achieved broad-based and inclusive community participation in
beneficiary selection because it worked through the lowest unit of
local government - the bag khural (assembly).
The creation or rehabilitation of local infrastructure can have a quick
and direct impact on income earning opportunities and standards of
living. However, the temporary employment created by public works
programs is too short to have a significant impact on the livelihoods
of the poor.
Sustainable access to credit for the poor requires the creation of a
rural finance system. Local government bodies can help channel funds to
the target group but constitute an inappropriate institutional base for
micro banking.
Herder restocking is not sustainable in the absence of parallel support
for pastoral risk management.
9. Program of Targeted Intervention (PTI)     Y
10. Environment Aspects (including any public consultation)
Issues     :  Consistent with the World Bank's Operational Policy
OP4.01, Environmental Assessment (EA) for the proposed SLP is regarded as an
ongoing process that began around the time of project identification.
Environmental assessment has been considered as part of several steps in the
process of project identification and design: a national-level public
consultation meeting in December 1999, conducted as part of project
identification; an analysis of land policy and land tenure arrangements in
Mongolia; a Participatory Living Standards Assessment (PLSA), conducted by
National Statistical Office of Mongolia with World Bank support between March
and September 2000; and an Intensive-Learning Implementation Completion
- 19 -



Report (ILI), documenting lessons learned under the World Bank-supported
Poverty Alleviation for Vulnerable Groups Project (PAVGP). In July-August
2001, an environmental assessment scoping study of the pastoral risk
management component was carried out, culminating in a dedicated
environmental and social assessment over August-October 2001.
The dedicated environmental assessment was carried out over three stages by
an independent team of specialists. First the EA carried out scoping of
impacts and desk-based review of literature. This included the analysis of
project preparation reports describing proposed project components and
activities, and review of the report of the pastoral risk management scoping
study. Second, in-depth stakeholder consultation was carried out over three
weeks in Mongolia in September 2001. This included a series of interviews and
workshops with secondary and primary stakeholders and key informants in
Ulaanbaatar and rural Mongolia (in three aimags, Dundgobi, Omnogobi, and
Ovorkhangai). In Ulaanbaatar, the EA team worked in close cooperation with
the project preparation support team from FAO. Consultation on the EA's
initial conclusions was facilitated by the use of an aide memoire distributed
to MoFA, PAPO and MoNE. A draft of the EA report was prepared, and detailed
comments from the World Bank, the project preparation team, and GoM were
incorporated to produce a final report which was submitted in December 2001.
The full EA report covers the policy, legal and administrative framework; the
methodology of the assessment; the baseline situation; the significant
environmental impacts; the analysis of alternatives; and the proposed
environmental management plan; complemented by a number of annexes providing
detailed additional information.
Justification /rationale for category rating. The proposed project is B-
rated, consistent with regional practice under other CDD-type projects and
those with significant involvement in the livestock sector. Most
environmental impacts resulting from the project are expected to be
beneficial. No incremental livestock will be produced under the proposed
project, and improved herd and grassland management practices will be
supported, which will be conducive to environmental sustainability.
Overall environmental impact. According to the EA, the proposed project has
the potential to deliver a significant overall positive environmental impact
through its contribution to reduced land degradation, because community-level
grazing management is proposed as a key part of the project's integrated
approach to pastoral risk management. The EA field observations and
discussions with herders indicated that the empowerment of herders or herder
associations to manage their pasture has good potential for reducing land
degradation, both in times of drought and dzud, and at all other times.
Current pasture management patterns in Mongolia are accelerating land
degradation due to concentration of herds around sum and aimag centers,
partly as a result of the impact of dzud and drought, and partly due to the
breakdown of existing wells, resulting in localized patterns of excessive
livestock and consequent pasture over-use that have adverse effects on the
pasture productivity and sustainability. Successful actions to reduce
herders' vulnerability through sustainable pasture management offer a
significant opportunity to reduce the extent and severity of land
degradation.
11. Contact Point:
Task Manager
- 20 -



Robin Mearns
The World Bank
1818 H Street, NW
Washington D.C. 20433
Telephone: 202 458 4010
Fax: 202 477 2733
12. For information on other project related documents contact:
The InfoShop
The World Bank
1818 H Street, NW
Washington, D.C. 20433
Telephone: (202) 458-5454
Fax:        (202) 522-1500
Web: http:// www.worldbank.org/infoshop
Note: This is information on an evolving project. Certain components may not
be necessarily included in the final project.
This PID was processed by the InfoShop during the week ending
June 28, 2002.
- 21 -