Inspection Panel's Report and Findings
on the Qinghai Project
Executive Summary
The Qinghai Project
1. The objective of the China: Western Poverty Reduction Project
is to reduce the incidence of absolute poverty in three provinces
of China: the Inner Mongolia Autonomous Region ("Part
A"), Gansu ("Part B") and Qinghai ("Part
C"), shown on Map 1. This Report is concerned exclusively
with "Part C," the Qinghai component of the Project.
2. When the Western Poverty Reduction Project was placed before
the World Bank in 1997, the basic design of the Qinghai component
(hereinafter referred to more simply as the Project) was already
substantially developed. Located entirely in Qinghai Province,
the project comprised two parts: the so called Move-out area to
the east of Xining city as shown on Map 1, and the Move-in area
some 450 kilometers to the west, as shown on Map 2. The Project
aims to alleviate poverty by voluntarily resettling 57,775 poor
farmers who currently practice high-altitude rain-fed agriculture
in five counties in Hadong Prefecture and one county in Xining
City Prefecture. They will be resettled into a new irrigation
project in the dryland area of the Haixi Tibetan and Mongolian
Autonomous Prefecture in Dulan County. Within the Move-in area,
the Project proposes to renovate an existing 8 m dam, and
construct a new 40 m dam at Keri. It will also build a 29 km
canal from the Xiangride river to the Keri Dam and a further 56
km canal from the Dam to supply water to irrigate some 26,500 ha
in Dulan county - all shown on Map 2.
Panel Asked to Conduct the Investigation
3. The Qinghai component became the subject of public controversy
in late April 1999, following the publication of an article by
the Tibet Information Network. In response to this article and a
growing volume of external criticism, Bank Management conducted
an intensive internal review of environment and social aspects of
the Qinghai Project. As a result, some "refinements" or
"improvements" were proposed. Negotiations with the
Chinese Government were reopened and the agreed modifications
were presented in a revised loan package for Board approval.
4. On June 18, 1999, the International Campaign for Tibet (ICT),
a U.S. based non-governmental organization, acting on behalf of
affected people who live in the Project area, submitted a Request
for Inspection to the Panel. The Requesters claimed that the
Tibetan and Mongolian ethnic peoples would suffer potentially
irreversible harm from the Project and that this harm stemmed
from Management's failure to comply with several Bank policies
and operational procedures.
5. On June 24, 1999, the World Bank Board decided to proceed with
the financing of the Western Poverty Reduction Project. This
decision came with a caveat, however. In an unusual move, the
Board agreed "that no work be done and no funds be disbursed
for the $40 million Qinghai component of the project until the
Board decides on the results of any review by the independent
Inspection Panel."
6. On July 19, 1999, management replied to the allegations in the
Request. In brief, it considered the Project in compliance with
all relevant Bank policies except that of the disclosure of
information. Management pointed out that a number of
"refinements" had been made to the Project since public
concerns had been raised in late April. Internal reviews
confirmed compliance with safeguard policies, but it had
nevertheless identified areas that would benefit from additional
qualitative strengthening during implementation.
7. In its report to the Board on August 18, 1999, the Panel found
the Request eligible (except that the Board itself had to decide
the question of external representation by ICT of protect
affected people) and recommended that he Executive Directors
authorize an investigation. On September 9, 1999, the World Band
Board of Executive Directors authorized the Panel to conduct and
investigation. The Terms of Reference required that the Panel
investigate whether the Bank has violated one or more of its
operational polices and procedures with respect to the design and
appraisal of the Project: BP 17.50 (Disclosure of Information);
OD 4.01 (Environmental Assessment); OD 4.20 (Indigenous Peoples);
OD 4.30 (Involuntary Resettlement); OP 4.09 (Pest Management);
OP/BP 4.37 (Safety of Dams); OP/BP 12.10 (Retroactive Financing);
and OP/BP 10.00 (Investment Lending - Identification to the Board
Presentation).
8. The Panel was assisted by a small team of senior consultants
and a set of interpreters from outside China. Between September,
1999, and January 2000, the Panel undertook a field visit of
Qinghai Province and conducted interviews in Washington, D.C.,
Beijing, Xining and many villages in the Project area. It also
examined relevant documents, including published and unpublished
literature - a process that continued through the third week of
April, 2000.
General Compliance
Interpretation of the Bank's Policies and Procedures
9. During the course of examining some 20 projects over the past
five years, the Panel has encountered certain differences in
views among staff on just how the Bank's operational policies and
procedures should be applied. In this case, however, the Panel's
interviews revealed an unusually and disturbingly wide range of
divergent and, often, opposing views. These large differences
pervade all ranks of the staff, from senior management to
front-line professionals. And they apply to virtually all of the
major decisions required by the policies. The implications of
this for a reasonable application of the Bank's policies and
procedures became a matter of serious concern to the Panel, and
ought to be of concern to the Bank generally since there is no
way that the policies can by applied with reasonable consistency
in the face of such wide divergences of opinion.
10. For example, a number of staff members felt that the Bank's
Operational Directives and other policies were simple idealized
policy statements, and should be seen largely as a set of goals
to be striven after. Others of equal or more senior rank
disagreed with this view. They felt that this interpretation
could render the policies virtually meaningless and certainly
incapable of being employed as benchmarks against which to
measure compliance.
11. In discussions about compliance, staff often pointed out that
the policies allow for flexibility of interpretation. The
decisions made on the specific matters were thus covered and in
compliance. It was simply a matter of "judgement and
Management's sole discretion." The Management Response
itself makes several claims in this respect. Other staff argued,
however, that the policies are clear enough to distinguish areas
that are binding from areas where some reasonable flexibility in
interpretation is called for. Read in their entirely, the Panel
feels that the directives cannot possibly by taken to authorize a
level of "interpretation" and "flexibility"
that would permit those who must follow these directives to
simply override the portions of the directives that are clearly
binding.
12. Most staff at all levels agreed that, in appraising
compliance, Management had an obligation to satisfy itself not
only that the process and procedures mandated by the policies had
been followed, but also that the world under review met
professionally acceptable standards of quality. In other words,
both process and quality were essential components of compliance.
Some staff, however, took the opposite view that, while it might
be appropriate to ensure professionally acceptable standards of
quality, it is not required for compliance. This 'check-list' or
'process' approach to compliance represents a minority view but,
in the Panel's judgement, it is of concern when senior persons in
the project decision-chain hold it.
13. Questions about compliance were often answered with claims of
"precedent." The Management Response itself cites
"…past practice with … a large number of similar
integrated agricultural development projects financed by the Bank
in China over the last 10 years," as a reason, for example,
for assigning an environmental Category "B" rather than
"A." Management's past experience in a country can
obviously provide the basis for a certain level of comfort that
the work required by the policies will be undertaken
successfully. In the Panel's view, it is an entirely different
matter, however, to suggest that experience and precedent can
determine what is required by the policies.
14. Interviews with some staff were punctuated by the refrain
that "in China things are done differently." This is
echoed in the Management Response which states that: "The
level and quality of preparation and analysis for this Project
were very much in line with Bank practice in applying social and
environmental policies to projects in China in the Context of its
political and social systems." The Panel has carefully
examined the policies and has failed to find any grounds for the
view that precedents in a country, or a country's "social
and political systems," can in any way determine what is
required by the policies.
15. Faced with these widely divergent views among the staff, the
Panel was forced to revisit its views on and experience with Bank
policies and compliance. In the end, it returned to the approach
reflected in its earlier reports. There is indeed room for some
flexibility and interpretation but, as provided in the Resolution
that established the Panel, the Operational Directives (and
updated OPs, BPs, GPs, etc.) are the primary source of Bank
policy for purposes of assessing compliance.
The Qinghai Project in Space and Time
16. The Environmental Assessment does not distinguish between
short term impacts and those that will only occur at some time in
the future. This raises serious questions about the time horizons
over which the Project was evaluated.
17. Moreover, in examining the Project documentation, the Panel
found a high level of ambiguity, uncertainty and inconsistency in
the use of the term "project area." This confusion is
compounded by the fact that the documentation is poorly supported
by maps. (The Panel had eventually to prepare its own set of
maps; i.e. maps 1 and 2.) As a result, it appears that
significant numbers of people, including members of minority
nationalities, have been left out of the environmental and social
assessments required by Bank policy.
18. The "project area," as interpreted in the EA and
other documentation, includes neither the commercial and service
centers of Xiangride, nor several of the townships surrounding
it, including some Tibetan and Mongol villages visited by the
Inspection Team. Similarly, the "project area" does not
include several villages between Xiangride and Dulan and between
Xiangride and the new towns to be created in the Irrigation site.
These omissions are difficult to understand. A network of social,
commercial and political interactions clearly exists in Dulan
County and in Xiangride Township. Yet no assessment has been made
of how these linkages and interactions will be affected, for
better or worse, by a Project that will completely change the
economy and demography of the County.
19. Speaking to merchants in Xiangride, the Panel found that
those who were even partially informed about the project were
full of expectations that would increase their trade and commerce
and that it would act as a pole of attraction for further people
and investment. The same is true of the townships adjacent to the
Xiangride prison farm. Their interdependence is clear.
20. Tibetan and Mongol village leaders in Xiangride township,
whose villages lie directly on the proposed supply-canal route to
the irrigated area, told the Panel that they had neither been
officially contacted nor asked about the Project. These villages
will be impacted directly by the Project's physical
infrastructure and indirectly by, at the very least, a doubling
of their county's population.
21. Many parts of the Move-in area have been overlooked or
glossed over in the EA and in other documentation of the Project.
Little or no attention was paid to the environmental impacts on
areas other than the proposed new irrigation areas and the Balong
Soak. Compliance with OD 4.01 requires that many other areas
receive detailed consideration.
22. An adequate environmental and social assessment would require
a much greater consideration of the effects of the Qinghai
Project on nearby townships including Xiangride and Dulan, as
well as a much larger number of villages, some visited by the
Inspection Team. It would require consideration of the multiplier
effects of impacts caused by the construction of the new dam and
reservoir, irrigation system, roads, and townships with all their
infrastructure; and also the likely population increases as the
nearly 60,000 resettled farmers will have growing families and
will attract more migrants, merchants, teachers, medical
personnel, etc.
23. In the panel's view, given the letter and intent of ODs 4.01,
4.20 and 4.30, the actual scale of the area to be impacted by the
Qinghai Project, the ethnic composition of the Project's impacted
populations, and the boundaries of the "project area"
were far too narrowly defined by Management. As a result, the
assessments fail to address many of the most significant social
and environmental impacts of the Project on the Potentially
affected populations, including those who are members of minority
nationalities. The Panel finds that this is not in compliance
with these ODs.
The Consultation and Survey Method
24. A comparison of the three populations in the Move-in area
that will be involuntarily resettled by this Project with the
populations that were surveyed, is vital to an understanding of
the sampling procedures. The Plan lists these impacted
populations as follows: (1) 63 herder households (352 persons)
who use the land "under formal lease agreements with the
townships;" (2) 289 herder households (2,411 persons) who
pass twice yearly, with their herds, through the area proposed
for transformation into irrigated farmland and townships for the
resettled population; and (3) 248 households (1,237 persons) who
currently farm on the area served by an existing irrigation
system, which will be integrated into the Project's new
irrigation systems.
25. The extent to which these involuntarily resettled groups were
"consulted" through the survey method is very
different. Take the second group, for example. No survey was
designed for, and no survey carried out among, the estimated 289
nomadic pastoralist households (2,411 persons) in this group,
whose twice-yearly passage through project site with their herds
will be disrupted. That is to say, zero percent of this group was
surveyed. In stark contrast, 100 percent of the 63 herder
households in the first group were surveyed. Of the 248
households (1,237 persons) in the third group, who are currently
farming an irrigated area in Balong township, and whose
irrigation system will be expanded and integrated into the new
system, 80 households or 32 percent were surveyed.
26. Thus, the largest group of persons involuntarily resettled by
this Project (i.e., the pastoralists in the second group whose
passage through the irrigation site with their herds will be
disrupted) received the least consultation, at least as measured
by the survey data. Although this group makes up 60 percent of
the people and 72 percent of the households to be involuntarily
resettled, none (zero percent) of this group was surveyed, using
Management's own population estimates and survey data.
27. Management also failed to consider weighting the surveys by
ethnicity, or to assure adequate representation in the survey
sample of affected minority groups. Most striking is the fact
that in the entire move-in area only three Tibetan households
were included in the survey. This is in spite of the fact that
the Project occurs in a Tibetan and Mongolian Autonomous region,
and the physical infrastructure for the Project (its
supply-canals) passes through Tibetan villages that were not
included within Management's definition of the "project
area."
28. Four points can be made concerning the survey in the Move-out
area. First, the questionnaires are not confidential. (All four
surveys required the respondent to put his or her name on the
survey.) Second, from the internal evidence of the questionnaires
themselves, they must have been filled out by someone other than
the individual respondents. Third, the very limited source of
information about the subject matter of the survey is striking;
93% of respondents indicated that they learned of the
resettlement from "government propaganda." Fourth, an
examination of the questions asked, and the context in which they
were asked, indicates that opinions and information gathered are
probably not reliable because respondents will probably think
that this questionnaire could directly influence whether they get
selected for the resettlement project.
29. The hazards of using the methods of consultation discussed
above are even greater in the Move-in area, because here
respondents are being asked, without guarantee of
confidentiality, whether they would welcome the influx of
settlers. The Bank must be aware that if there is even a
perception of potential adverse effects that could result from a
truthful statement of opposition to this Bank-financed project,
then Bank staff has a responsibility to guarantee confidentiality
of the respondent. This responsibility derives from OD 4.20 and
from the requirements for "full and informed"
consultation in OD 4.20 (esp. par. 8), 4.30, and 4.01, since full
and informed consultation is impossible if those consulted even
perceive that they could be adversely affected for expressing
their opposition to, or honest opinions about, a Bank-financed
project.
30. Though the Bank seems to have accepted these methods of
consultation, the Inspection Team found that other international
organizations working in the same Qinghai province did not accept
them.
31. A believable guarantee of respondent confidentiality might
have yielded very different results, especially in the Move-in
area. The Inspection Team's short field visit, though it recorded
many positive comments about the proposed Project, also yielded
some disturbing and dramatic examples of what can only be
described as a climate of fear, through which some individuals
nevertheless managed, at great perceived risk, to express their
opposition to this Project.
32. The Project Appraisal Document admits, as does the Management
Response, that the consultative process "was not always
smooth nor did it fulfill all best practice guidelines. At times
during the social assessment process respondent confidentiality
was not fully observed, and the data gathering process was less
than ideal." Nonetheless, Management still concluded that
"the will of the move-in host population was adequately
expressed."
33. The Panel obviously did not conduct a scientific sample of
opinion during its brief stay in Qinghai. Nonetheless, in the
Panel's view, the expressions of opinion it heard and the
incidents it witnessed indicate the need for far greater efforts
to obtain public consultation under adequate conditions, before
Management can be said to have met the requirements for public
consultation in the Operational Directives. The mere fact that
opinions expressed were so strikingly different, and especially
the fact that there was a strong perception of risk from those
expressing opposition to the Project during the Inspection Team's
visit, indicates that methods of public consultation used for
this Project have so far been inadequate.
The Consideration of Project Alternatives
34. If there is no alternative there can be no choice. The Bank's
policies and procedures leave no room for doubt as to the need
for a careful and systematic consideration of a number of
different types of alternatives, including investment
alternatives, alternative sites, alternative project designs,
alternative implementation plans. The purpose of considering
these alternatives is to ensure that option is supported with
will achieve the project's objectives most cost effectively,
while meeting the Bank's safeguard policies.
35. One of the most noticeable and significant weaknesses of the
assessments is that investment and project alternatives are
neither identified nor systematically compared. For all practical
purposes, the Environmental Assessment avoids consideration of
alternatives, both for poverty reduction in the Move-out areas
and for sites in the Move-in area. From the documentation, it is
not possible to deduce whether the Qinghai Project as proposed is
the best way for the Bank to meet the Project's objectives or to
ensure that the Bank's safeguard policies are being respected.
36. Management failed to ensure that those responsible for the EA
understood their brief to include an examination of alternatives
to resettlement in both the Move-out and Move-in areas. Instead,
the Panel found that they understood the main purpose of their
studies to be to assist the optimal resettlement of around 60,000
people from the Move-out area into the Balong-Xiangride
irrigation area. The same is true of the Social Assessment. There
is no systematic study of in situ alternatives to resettlement,
or of alternative resettlement sites, or of alternative
development plans for the national minorities affected within the
Move-in area.
37. Why the Bank accepted Assessments conducted in such
circumscribed and limiting manner is unclear. Whatever the
reasons, the Panel finds that the Assessments do not make any
meaningful analysis of realistic project alternatives as required
by Bank policy.
Environmental Compliance
Environmental Screening of the Qinghai Project
38. One of the most important decisions (perhaps the most
crucial) that Management must make concerning the environmental
assessment of any project is the category of the assessment that
will be undertaken. Under OD 4.01, this critical judgement is
made in the first instance by the Task Manager (now Task Team
Leader), with the concurrence of the regional environment unit.
The project is assigned to Category "A," in which case
a full Environment Assessment (EA) is required, or a Category
"B," in which case a full EA is not required, but an
environmental analysis is, or a Category "C," in which
case no environmental analysis is required. An Annex E to the OD
provides illustrative lists of the types of projects best
classified in Categories "A," "B" and
"C." Twelve types of projects are listed for
"A," of which eight are found in the entire Western
Poverty Reduction Project and four in the Qinghai component of
the Project. The four are: dams and reservoirs, irrigation, land
clearance and leveling, resettlement and all projects with
potentially major impacts on people. The OD further provides that
a full EA is required if a project is likely to have significant
adverse impacts that may be sensitive, irreversible, and diverse.
A footnote explains that impacts qualify as "sensitive"
if they affect vulnerable ethnic minorities or involve
involuntary resettlement.
39. This critical decision of the category of the EA is made at a
very early stage in the project cycle, but the OD permits a later
revision of the category as new information becomes available.
Neither the OD nor the Annex provides for exceptions on the basis
of past experience or precedent in the country in question, or
the area of the project relative to the area of the geographic or
political jurisdiction involved.
40. The initial decision to assign "B," taken on
January 8, 1998, was made before the Task Team leader, or any
other Bank official associated with the decision, had an
opportunity to visit the Move-in area. It was also made without
detailed maps and other basic information being available and, it
would appear, without regard to the illustrative examples of
"A" projects mentioned above. Furthermore, the
assignment was made on the stated assumption that there would be
no involuntary resettlement.
41. The first Bank official involved in the decision to visit the
Project site was the consultant who was engaged to be responsible
for the environmental impact assessment aspects of the Project.
Following his visit, on March 13, 1998, he raised a number of
issues with senior staff and recommended the re-classification of
the Qinghai component as an "A." Senior staff responded
at length, citing reasons why it should remain a "B."
Management was aware of the A/B debate, but did not intervene.
The Project concept Document meeting of April 14, 1998, in effect
confirmed a "B."
42. The EAs for the three components of the Western Poverty
Reduction Project were received prior to the December, 1998
Decision Meeting. The same Consultant who had prepared their
Terms of Reference, negotiated them with the local institutes,
and overseen their implementation, also reviewed the resulting
assessments and advised the Task Team Leader than he approved
them - in effect, recommending them for clearance. Subsequently,
the documentation for the entire Western Poverty Reduction
Project was presented to the East Asia Country Director and Legal
Department for clearance and authorization to issue an invitation
to negotiate.
43. Following completion of the negotiations during the week of
April 12, external criticism of the Qinghai component of the
Project erupted, starting with publication of the Tibet
Information Network (TIN) article on April 27, 1999. This
triggered an intensive internal review of, inter alia, the
environmental classification of the Project as "B"
rather than "A."
44. Management decided to ask several senior experts from the
central departments to advise on classification and compliance.
One of these experts informed the Panel that from the time that
he first saw any documentation related to the Qinghai project, he
had been convinced that he Project should have been classified an
"A" on at least eight grounds. He recommended that the
project be immediately reclassified as "A" as
preferable to further jeopardizing the Bank's reputation by
defending the category "B." Another expert agreed that
the Project should have been classified as Category
"A," and told the Panel that he had advised Management
that it should simply acknowledge that a
"well-intentioned" mistake was made. Another expert
took a slightly different view. Agreeing that if the Project were
to be categorized now, under OP 4.01, it would be an
"A," he understood that the size of the Project was not
out of line with others in the China portfolio and that its
categorization as a "B" was consistent with accepted
practice in the Region.
45. A senior official, also with strong credentials in the field,
informed the Panel that in his view the Project was quite clearly
an "A," whether under old or new rules. The Region, he
felt, was in non-compliance, not with whether or not the Project
was categorized (obviously, it was), but with the A/B judgement
and, in late May, he so advised Senior Management. Expressing
understanding with the position of colleagues who must contract
their services to, and are paid in part out of, project budgets,
he went on to say that "Frankly, they don't want to bite the
hand that feeds them (for cross support) by taking a hardline
view."
46. After reviewing the screening process, the Panel finds that
management's decision to classify the project, as a "B"
was not in compliance with OD 4.01. Several components of the
Project fall within the illustrative list of "A"
projects in Annex E, i.e., dams and reservoirs, irrigation, and
resettlement. And the impacts qualify as "sensitive"
since vulnerable ethnic minorities are affected and involuntary
resettlement is involved.
Environmental Assessment of the Qinghai Project
47. Management adopted a very limited definition of
"environment" in the Project with the result that the
Assessment fails to analyze the full range of Project effects.
The assessment also fails completely to place the Project in
proper time frames. As noted above, the spatial boundaries of the
Move-out and Move-in areas are defined narrowly, or not at all
with the result that whole communities and populations, whose
lives will be impacted by the project have been left out of the
Environmental and Social assessments.
48. Concerning the Move-out areas the EA and other Project
documentation are moot with respect to both the plans and levels
of financing for improving the living conditions of those
remaining behind. There is also a lack of documentation on the
social and economic implications for a village of "moving
out" a significant proportion of their most productive and
economically active population. This voluntary resettlement could
have major, long-lasting, and irreversible social impacts.
Considerably more detailed analysis of the social and
environmental problems of the move-out villages and of how these
all will be addressed by the Project are required in order to
comply with OD 4.01.
49. In May/June 1999, following the outbreak of public concern,
Management tried to compensate for this by requiring a study to
evaluate the environmental and social impacts of the Voluntary
Settlement Implementation Plan. This study (which is to recommend
measures to enhance the environmental sustainability and the
living conditions of the people in the Move-out counties) is to
be undertaken not later than three years after the implementation
of the Plan has commenced. Undertaking an environmental and
social assessment three years after the commencement of
resettlement is a bit late, not only in terms of the policies,
but also in terms of any elemental understanding of the purposes
of such assessments. Within the social arena, this is comparable
to requiring that the safety of a proposed dam should be studied
within three years after it has been built!
50. As to the Move-in area, the proposed in-migration to Dulan
county will more than double its population. The proposed new
towns will each have populations five times as large as
Xiangride, the nearest established town to the main irrigation
site. This will lead to further induced development, on which OD
4.01 lays great stress. The dam, irrigation and resettlement
parts of the Qinghai Project are treated as though they were to
take place in a regional vacuum. The potential impact of this
development on the network of social, commercial and political
interactions that exists in Dulan county and Xiangride Township
has not been considered. There is no indication of how these
communities and their populations will be affected, for better or
worse, by the Project. Without this assessment, the Bank's policy
goal of enhancing Project benefits has no substance or meaning.
51. The EA and other Project documents fail to consider the
appropriate mess of implanting large-scale irrigated agriculture
in this Region. It does not examine its suitability or viability
in comparison with the traditional forms of land use, including
agropastoralism, sedentary pastoralism, semi-sedentary
pastoralism (semi-nomadism), and migratory pastoralism involving
the herding of sheep and other animals. There appears simply to
be an assumption that irrigated agriculture is "a good
thing" without consideration of alternatives and relative
costs.
52. The EA is uninformative about the layout of the new towns and
villages, their infrastructure, and the facilities such as water,
heat and light that will be provided for the settlers. It is
silent on what methods the new towns and villages will use to
manage and dispose of their wastes. Similarly, the EA is
virtually silent on the impacts of the emissions and wastes on
the areas surrounding and downstream from the two new towns (with
initial populations nearing 30,000), and villages. Contrary to OD
4.01, the Bank's Terms of Reference for the EA did not call for
any such assessment.
53. The standard of maps, charts and references is inadequate.
They do not provide a satisfactory record of the settlements,
infrastructure and land use in the areas that will be affected by
the Project. The Qinghai Project involves a total transformation
of the vegetation of some of the project areas, a substantial
impact on others during the construction phases, and a
substantial impact on areas that are adjacent to or downstream of
the construction areas. Management should therefore have required
a proper vegetation survey and map. The assessment provides no
description whatsoever of the vegetation in the Move-out area,
and an inadequate description of the vegetation in the Move-in
area. Attention had been drawn by an FAO consultant to the urgent
need for a detailed soil survey and land classification in
connection with the irrigation scheme, but the recommended map
(at a scale of 1:10,000) had not been complete at the time of the
Panel's visit.
54. The information on the biodiversity of the area is very
sketchy and inadequate. It does not incorporate an assessment of
the diversity encountered with regard to distribution,
frequency/rarity and conservation status. While an exhaustive
inventory of the biodiversity would not be expected, it would
have been appropriate to search the literature and to reference
studies (including lists) as these are not lacking China.
55. Extensive exploitation of oil, natural gas and minerals is
carried out in parts of the Qaidam basin and test drillings have
been undertaken near the Project area. No mention is made in the
EA of the general economic importance of oil and minerals in the
Province, or its possible effects on the Project areas, or of any
drilling activities in the Region.
56. The Panel finds that the Environmental Assessment of the
Qinghai Project is not in compliance with Bank policies as set
out in OD 4.01.
Dam Safety, Pest Management and Natural Habitats
57. While the design and location of the proposed Keri Dam
appears to be in compliance with OP/BP 4.37 and Annex B of OD
4.00, there is no reference to the seismicity of the Project area
and the fact that two earthquakes of magnitude 7+ (equivalent to
the 1999 Turkish (Ismit) earthquakes), occurred in Dulan County
in 1937 and 1963. An assessment of the risk to the nearby Keri
irrigated area should the dam fail (as a result of an earthquake)
should be undertaken in the near future.
58. The use of pesticides form part of the implementation phase.
No Bank funds will be used for the procurement of pesticides, as
these will be provided by the borrower as counterpart funds for
the Project and, as such, acquisition and use of pesticides is an
integral part of the Project. No detailed comments can be made on
the subject, as the exact nature of pest management proposals is
not known. In the Panel's view, the lack of pest management plans
is a violation of OP 4.09. However, assurances were given to the
Inspection Team that an integrated pest management programme
would be undertaken for the Qinghai Project.
59. The increasing pace and scale of habitat conversion in the
Qinghai-Tibetan Plateau requires a broader view to be taken of
the fate of the various ecosystems in the region, not just those
in or adjacent to the Project area.
60. Given the scale of absolute habitat conversion in the Project
area, involving 19,000 ha of land that will be irrigated, and
construction developments such as the dam, canals, townships,
villages and roads - all adding up to an estimated 21,444 ha - it
cannot be asserted with confidence that possible critical natural
habitats will not be lost. The necessary baseline information is
not available in the Assessment. The Panel is therefore of the
view that the Project is in contravention of the Bank's policy,
OP 4.04, in regard to the significant conversion of critical
natural habitats.
Social Compliance
The Qinghai Project and Indigenous Peoples
61. OD 4.20 is quite clear about Bank policy concerning the
preparation of an Indigenous Peoples Development Plan. Paragraph
13 states that: "For an investment project that affects
indigenous peoples, the borrower should prepare an indigenous
peoples development plan that is consistent with the Bank's
policy. Any project that affects indigenous peoples is expected
to include components or provisions that incorporate such a
plan."
62. Management acknowledges that an Indigenous Peoples
Development Plan (IPDP) is required for the Qinghai Project. It
asserts, however, that the Project as a whole constitutes the
IPDP because a majority or of the Project's beneficiaries are
minorities. This assertion hinges on an interpretation of the
last sentence of the paragraph 13 quoted above; i.e., "...
When the bulk of the direct project beneficiaries are indigenous
people, the Bank's concerns would be addressed by the project
itself and the provisions of this OD would thus apply to the
project in its entirety."
63. In the Panel's view, Management's interpretation of this one
sentence of paragraph 13 of OD 4.20, cannot be accepted. In the
case of this Project, it is inconsistent with other parts of the
OD and especially inconsistent with the objective of Bank policy
towards indigenous people, which is "...to ensure that the
development process fosters full respect for their dignity, human
rights, and cultural uniqueness..." and to "ensure that
indigenous peoples do not suffer adverse effects during the
development process..., and that they receive culturally
compatible social and economic benefits."
64. The Panel's conversations with villagers inside the narrow
"project area," as defined by Management, uncovered a
positive attitude towards the idea of separate IPDPs for each of
the minorities. When the Panel visited project-affected towns
near Xiangride, and Tibetan villages along the proposed supply
canal for the project - all left out of the "project
area" as defined by Management - it received a similarly
probative response to self-standing IPDPs for each of the ethnic
groups. Although the Panel's visit was short it is satisfied that
some, if not all, of the ethnic minorities in the Move-in area
would welcome the opportunity to participate in the development
of separate IPDPs.
65. The "indigenous peoples," or national minorities,
in the Move-in area, the Hui, Mongol, Tibetan, Tu and Salar, are
very different from each other, in their "cultural
uniqueness" as well as their "local patterns of social
organization, religious beliefs, and resource use."
Retroactively lumping together these very different cultures into
a single one-plan-fits-all IPDP, whether that is the
"project in its entirety," or a single IPDP for the
whole of the Move-in area, effectively denies these very
different ethnic groups, especially the numerically weakened and
most vulnerable, an opportunity to participate in a process that
would ensure that their minority cultural traditions are taken
into account in the overall design of the project. No number of
add-on indigenous-culture-boosting activities (bilingual schools,
Islamic religious sites, Tibetan pharmacological centers,
corridors for nomadic herders) can override the fact that the
initial project design is flawed in the failure to recognize,
from the outset, the cultural uniqueness of the separate ethnic
minorities involved. The Panel finds that a serious attempt to
prepare separate, free-standing IPDPs for each of the national
minorities in the Move-out area is required to bring the Project
in compliance with OD 4.20.
66. OD 4.20 also stipulates the process and procedures to be
followed in developing an IPDP within the context of the Bank's
"project cycle," i.e., identification, preparation,
appraisal, negotiations, and presentation to the Board. After
reviewing the actual processing of this project, the Panel find
that Management's decisions on Indigenous Peoples during the
identification, preparation and appraisal were not in compliance
with the process and procedures mandated by paragraphs 16-18 of
OD 4.20. The Panel finds that the Project as a whole does not
constitute the IPDP required by OD 4.20, and that separate,
free-standing IPDPs are required to bring the Project into
compliance with OD 4.20.
Involuntary Resettlement
67. The Qinghai Project involves two types of resettlement,
voluntary and involuntary, and in assessing compliance, a clear
distinction must be made between the two, since OD 4.30 only
applies to involuntary resettlement.
68. Management contends that nearly 58,000 migrants from the
Move-out area are "voluntary" resettlers and therefore
not covered by OD 4.30. In management's view, OD 4.30 applies
only to the 4,000 individuals in the Move-in area considered to
be "involuntarily" affected. And, in its view, an
adequate Involuntary Resettlement Plan has been prepared to cover
them. The elements that now make up this Plan were originally
included within the Voluntary Settlement Implementation Plan
(VSIP). Later, however, following the outbreak of public concern,
those provisions in the VSIP relation to involuntary resettlement
were revised and brought together as separate annex to the VSIP.
69. During its discussions with people in the Move-out area, the
Panel was able to confirm that they felt they had a choice
whether they could move or not, and most of those interviewed
wanted a move to Haiti Prefecture. Although it is difficult to
say whether the choice was an informed one in many instances, OD
4.30 does not give clear guidance on the quality of full and
informed choice that is needed to consider a resettlement as
"voluntary." The Panel thus accepts management's
contention that OD 4.30 does not apply to the migrants from the
Move-out area.
70. OD 4.30 applies to those people who are displaced or
adversely affected by the Project. As noted repeatedly,
Management's narrow definition of the boundaries of the project
area resulted in many people and communities affected by the
project being left out of the assessments. This appears to be
true of the population of persons who will be displaced by the
Project documents. In the Panel's view, the Project is not in
compliance with OD 4.30.
71. Paragraph 4 of OD 4.30 requires the development of a plan
that will assist involuntarily resettled persons with their move,
and will provide fair compensation for their loss (whether they
lose the land itself, or the use of the land, or other productive
resources.) The method of compensation chosen for the herders who
are to be involuntarily resettled is "land-for-land."
Adequate baseline data on pastoralism, including the data on land
use and inheritance that would allow a proper assessment of the
compensation offered, are unfortunately lacking for this Project.
Although envisaged by the OD, it appears that this work was not
done. Without the results of such work, it is difficult to assess
the adequacy of the compensation offered, not only for the Panel
but also, in the first instance, for Management. Indeed, it is
difficult to understand how the OD's policy objectives can be
achieved without this information.
72. For all of the above reasons, and others discussed in the
Report, the Panel finds that the Project's current Involuntary
Resettlement Plan is not in compliance with the requirements of
Bank policy as set out in OD 4.30.
Disclosure of Information
73. The timely disclosure of information on evolving projects is
designed to provide opportunities for comment on and improvement
of Bank projects. When a project is under consideration, Bank
policy requires that a Project Information Document (PID) be
released on it. This needs to be made available at the Bank's
Infoshop at the Project Concept Document (PCD) stage. The very
short PID for the Western Poverty Reduction Project bears the
same date as the PCD, March 25, 1998. It makes one reference to
"minorities" or "minority nationalities,"
saying in passing that China's current "absolute poor"
comprise "largely minority nationalities." There is no
mention of minorities in the Qinghai component of the Project.
Rather, the beneficiaries are defined only in terms of their
poverty and their precarious future in the in the absence of the
Project.
74. The Bank's policy expects that this initial PID will be
updated and expanded periodically as project preparation
proceeds. In any event, it is required to be revised before
formal project appraisal. This was not done. If changes are then
made, a final version of the PID is prepared. A first revision of
the PID was made available at the Bank's Infoshop on May 4, 1999,
nearly four months after the Appraisal Mission of January 26,
1999.
75. Following the TIN article, the Public Information Document
(PID) was revised a second time and released on June 1, 1999. The
June version contained a long, 11-page annex on "Social
Aspects" of the Project. This revision is the first official
public document on the Project that contains a description of the
Social Assessment preparation, the selection of beneficiaries,
measures for land acquisition and compensation, and measures for
protecting minority rights.
76. In its Response to the Request for inspection, Management
agrees that there were shortcomings in the timeliness with which
the required documents were sent to the Infoshop. The EA and the
Involuntary Resettlement Plan were not sent to the Board of
Executive Directors and the Infoshop until June, 1999. As
Management admits, the Qinghai Project is not in compliance with
BP 17.50. And the Panel so finds.
Investment Lending and Retroactive Financing
77. The September 9, 1999 decision of the Board of the Executive
Directors asked the Panel to look into compliance by Management
with the provisions of OP/BP 10.00 on "Investment Lending:
Identification to Board Presentation" and OP/BP 12.10 on
"Retroactive Financing."
78. The Panel is satisfied that the Project is consistent with
the Articles of Agreement and the CAS and is anchored in country
policy/sector analysis, as provided in OP/BP 10.00. This Report
shows, however, a number of instances where the Panel feels that
operational policies and procedures were not followed, casting
doubt as to whether the Project, as it stands, is the best
alternative to contribute to poverty reduction and sustainable
economic growth, and, thus, economically justified, as required
in the same policy.
79. In the Panel's view, the Project is in compliance with OP/BP
12.10 on Retroactive Financing.