i Recommendations for Reorganizing the World Bank Safeguard Function: Priorities for Consultation on a Safeguard Implementation Plan Vince McElhinny Bank Information Center [email protected]October 6, 2014 As the World Bank moves forward on plans to define a Global Practice safeguards model, any options that the Bank is considering for reorganizing the safeguard function should be discussed openly and assessed on the basis of agreed upon criteria before reaching final decisions on the new organizational structure. Based on internal Bank evaluations and management proposals, this paper identifies some of those criteria. Effective implementation of the World Bank safeguards requires a robust, independent critical mass of environmental and social safeguard expertise with adequate budget, skill mix, clear and appropriate reporting line, proper incentives and adequate support for strengthening borrower implementation capacity. Three core reforms are proposed: a 100% increase in safeguard staffing to correct a long-standing capacity deficit, a commensurate increase in overall annual safeguard budget to $80 million, and the reinforcement of the independent review responsibility of the Regional Safeguard Advisors. The paper identifies and makes reform recommendations for seven areas that should be reform priorities in any Bank proposal to reorganize the safeguard function and broader enabling environment.
45
Embed
Reorganizing the World Bank Safeguard · PDF filei Recommendations for Reorganizing the World Bank Safeguard Function: Priorities for Consultation on a Safeguard Implementation Plan
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
4) Clear Senior Management structure and reporting line; 5) Incentives; 6) Increased resources for
safeguard capacity strengthening and 7) Accountability for reporting safeguard results.
Understanding the options for change hinge on understanding how safeguards have been organized at the
World Bank before the current reorganization. The short summary of the Bank's safeguard function that
follows is limited by the absence of any clear description of this topic outside the Bank.4 The Bank's
safeguard function, including staff incentives and accountability that existed prior to the Global Practices
and new Corporate Structure were officially launched on July 1, 2014 has evolved over several recent
organizational reforms. Several complementary and interacting units carry out the Bank's safeguard
duties.
Safeguard Support and Quality Assurance: The World Bank has between 300 and 350 staff with
environmental and social safeguard expertise that devote some percentage of their time to support of
2 A. Dani, A. Freeman, and V. Thomas, Evaluative Directions for the World Bank Group's Safeguards and Sustainability
Policies, Evaluation Brief 15, 2011, The World Bank Independent Evaluation Group. 3 In addition to the lack of management response to prior drafts of the current proposal, see FPP, BIC, Urgewald (April, 2013)
submission on Safeguard Implementation to Safeguard Review, Phase I 4 This section is based on interviews with over 50 Bank staff since 2009 as well as a review of numerous project and policy
documents.
3
project preparation or supervision support.5 Safeguard work is substantially understaffed given the 2400
active investment finance projects and a lending portfolio of $170 Billion.6 These safeguard quality
assurance functions were mapped until recently to one of the six regions, while a small number of staff
were mapped to the Environment or Social Development Network Anchors, and reported to Sector
Managers and Directors. Most of these environmental and social specialists combined two types of work:
the provision of safeguard cross-support to projects and the direct management of climate change, carbon
financing, technical analysis, and other lending projects in their sector, both independently and as part of
teams in other sectors. The Country Management Units and specific Task Team Leaders within each
Regional Vice-Presidency have exercised control over a large share of the Bank's annual administrative
budget to contract the services of safeguard specialist services through work program agreements. The
demand for safeguard support and supervision is determined by the project manager.
The creation of the Quality Assurance and Compliance Unit (QACU) and the Environmental and
International Law Unit (LEGEN) of the Legal Department, ensured more centralized guidance for
safeguard application during project preparation. Since 2004, QACU (which included six Regional
Safeguard Advisors, RSAs) and LEGEN have provided guidance for the most complex safeguard
operational challenges, and retained safeguard monitoring responsibility for the Bank's highest risk
projects. Despite the title Advisor, the RSAs exercise project clearance authority at concept and appraisal
stages of project preparation for the highest risk projects. The responsibility is delegated to Sector
Manager for some Category B and C projects at appraisal.
A significant restructuring of safeguard functions was introduced in 2006. Before then, both the
safeguard support and clearance functions were located in the same network (Environmentally and
Socially Sustainable Development (ESSD). In response to concern over a perceived conflict of interest
resulting from both support functions (that have vested interests in the approval of the project) and
compliance functions reporting to the same Director and VP, the compliance function led by the Regional
Safeguard Advisors was relocated from ESSD to Operational Services group within OPCS. As IEG
observes, "the World Bank consolidated the Environmentally and Socially Sustainable Development
Network (ESSD) and the Infrastructure Network—into the Sustainable Development Network (SDN)
under one vice president, bringing the environmental and social staff and their internal clients from the
infrastructure and agricultural sectors under one umbrella." 7 Under this arrangement, QACU continued
to rely largely on the technical staff in the regional environmental and social units to conduct due
diligence and appraisal.
While some viewed safeguards and other operations activities as too close under ESSD, the reform has
also fueled a perception that the division of labor between OPCS and SDN may have moved safeguards
into an untenable position that is often too far removed from fundamental project cycle decisions.
As IEG explains, understanding how the administrative budget has been allocated helps illustrate the
concerns about the effectiveness of the current safeguard model, particularly for project supervision. The
budget for safeguard staff to support project preparation and supervision is controlled by the project task
team leader. The intensity and range of safeguard appraisal and supervision is determined not by a
Safeguard Manager or Director, but rather by the TTL judgment in what some view as an excessively
subjective and unpredictable resource allocation model. The RSAs receive a small off-the-top budget
from OPCS for review and clearance work and there are some measures for adjusting project budgets to
5 A recent internal audit identified 172 Environmental specialists and 200 Social Development specialists at the Bank, who
presumably had some level of safeguard expertise. See IAD, Advisory Review of the Bank's Safeguard Risk Management, (June
16, 2014) undisclosed summary draft. 6 Based on Bank reported data on Sep. 8, 2014, does not include 115 DPOs, 20 PforR and over 100 unclassified projects. IAD
(2014) reports 2,355 IPF projects active as of Feb. 2014. 7 IEG 2011, pg. 15
4
add Safeguard capacity. Once a project enters the Bank pipeline (within the total budget allocation to a
VPU) and is allocated by each managing sector, the budget for Safeguard work for higher risk or lower
performing projects can be topped up by the RVPs.
Due in part to safeguard budget control residing within the six Regional Vice-Presidencies, safeguard
staff lack a predictable work plan and a single, clear line management. Technical support staff report to
regionally specific sector managers. Regional safeguard advisors report to a Quality Director within
Operational Services for a specific region.8 QACU (recently renamed as OPSOR) reports to another
Director within OPCS. The specific relationships between CMUs/TTLs, QACU, and Safeguard cross-
support varies for each of the Bank's six regions - there is no single Safeguard system.9 This fragmented
and overlapping management structure leads to confusion and perverse strategies for safeguard
application and oversight. that few outside the Bank understand. IEG argues that reliance on this
"market" approach to contracting safeguard services results in "considerable inefficiencies in resource
allocation for safeguards oversight."10
II. Management proposal for reorganizing the World Bank Safeguards
Efforts to retool and reorganize the Bank safeguard organizational structure are intended to ensure
compatibility with the newly reorganized World Bank Group. On July 1, the Bank formally launched the
Global Practices and taken strides to integrate some World Bank Group functions, such as Human
Resources and External Communications. Other organizational reforms are under discussion and are far
less clear or visible.
Some reorganization decisions not taken are equally noteworthy - the decision not to create a single
Safeguards Global Practice or Cross-Cutting Solution Area, for example. A proposal to merge
Environment and Social Practices was resisted, reportedly by Social Directors, and Social was then
merged with a Global Practice on Urban and Rural Development. Splitting Environment and Social
technical staff into separate GPs may have the benefit of balancing the attention given to social risk
assessment given a perception of a bias toward environmental issues. However, lumping social into a
multi-sector GP and beyond arms length coordination with environmental staff also creates the first
coordination transaction cost and may dilute the authority of both GPs.11
a. Single, Centralized Safeguard Unit - the path not taken.
In considering all options for reforming the World Bank Safeguards structure, a proposal to pool the
safeguard specialists into a single, self-standing and independently managed group was also rejected, as
was the complementary option of merging the World Bank Safeguard groups with the IFC's Environment
and Social Development Department (CES).
Similar to IFC, EBRD, and IDB, which have adopted a centralized safeguard model, such an approach
would have mapped all safeguard support and compliance staff to a single unit that most likely would
have reported to the Vice-President of OPCS. At IFC CES, the social and environmental safeguard
8 Regional Quality Directors assumed the responsibilities for the Quality Assurance Group (QAG), which was dissolved in 2010.
These Directors have dual reporting lines to VP OPCS and RVP. Not clear what happens to Quality Directors under the new
structure (in RVPs)? 9 IAD (2014) 10 IEG 2011, pg 16 11 The debate over where the social development focus would fit within the 14 Global Practices was apparently taken quite late in
the design process.
5
specialists are grouped into respective clusters under 1 or 2 safeguard managers, and report primarily to
the Director for E&S Risk, but provide support to Investment Officers. Social and environmental
safeguard managers coordinate cross-support between groups.
As illustrated in Figure 1, IEG underscores two fundamental differences that distinguish the current
World Bank safeguard model from the IFC. At the IFC, all safeguard staff are housed in a central,
independent unit, which controls its own budget and can therefore assert greater autonomy in allocating
staff resources based on risk.
Secondly, the IFC combines compliance, including the clearing most projects at concept and appraisal
and the handling of the highest risk projects, and support within a single reporting line to Senior Director
and a Vice-President (Advisory Services). The IFC provides a firewall of budget independence between
safeguard staff and project managers as well as a much clearer reporting line for safeguard staff. IEG
asserts that IFC has budget authority, which helps resolve the internal conflict of interest presumed to
result from integrating support and compliance functions.12
Figure 1. World Bank Existing Safeguard Organizational Structure (IEG, 2011)
A single CES Director controls the $22 million annual budget for all IFC safeguard function activities.13
The E&S Director prepares an annual project preparation and supervision plan and independently
allocates supervision staff resources based on risk.
The E&S Risk Director is responsible for knowing the risk spread across the actual and projected IFC
portfolio as well as alignment with current CES safeguard staff capacity. The E&S Director and
Managers are responsible for training and certification of CES staff, which includes maintaining and
12 Procurement support and oversight also sit within the newly created Governance Global Practice. Despite sharing many of the
same sensitivities associated with Safeguards, there is much less concern with co-location of both procurements quality assurance
and control under the same line management and budget. 13 The IFC budget figure does not reflect the full cost of safeguards work in that IFC, more so than the Bank, can charge project
preparation costs to the client.
6
reporting annually on the roster or IFC and consultant staff capacity for the safeguard function. IFC's
E&S Director is responsible for reporting on safeguards outcomes for producing knowledge products that
ensure systemic learning from safeguard performance.
Some of the primary differences and possible advantages of a centralized safeguard unit favored by IFC
and others compared to current decentralized safeguard model preferred by the World Bank include:
CES has independent budget control of all safeguard quality assurance/support and quality
control activities, which facilitates annual planning and provides incentives for a more rational
approach to supervision
Compliance and Technical Support are co-located within the same unit, which preserves
independence but enhances consistency and optimizes resource allocation of SG staff
All safeguard staff are mapped full time to same Vice-Presidency with clear reporting line to a
single VP, which allows for a more coherent career path and incentive structure
A robust, critical mass of safeguard staff devote all time to safeguard support or oversight, which
enhances the ability to defend the safeguard value proposition.
The primary risks of the IFC model for the Bank include:
The centralized models hinges on having the right person and adequate budget to effectively lead it.
Safeguard specialist incentives may be effected by perception of a more limited set of career path
options, induced by the lack of operational management options. The result could be a less talented
pool of specialists.
Compliance and support functions located within the same reporting line could present a conflict of
interest.14
Without strong links to the regional operational structures, a central pooled staff may not be well
coordinated with safeguards challenges when they first arise or be close enough to the client to
adequately strengthen and use borrower safeguards systems.
The decision not go farther in integrating the IFC and World Bank safeguard structures may reflect the
limited progress in achieving greater World Bank - IFC integration in general. This has been
accompanied by the migration of some IFC safeguard staff, including a CES senior manager, to the Bank
Environment and Social, Urban, Rural and Resilience Global Practices.
Embedding safeguards within the Global Practices: Instead, Bank management has decided to locate
the majority of environmental and social specialists in the Global Practices. A gradual implementation of
the Global Practices began on July 1, 2014 without full clarity about the pending decisions related to any
reorganization of the safeguard function. A memo from the Directors in the ENRM (GP4) and Social,
Urban, Rural, Resilience (GP13) Global Practices described the interim arrangements for managing
safeguards after July 1.15
Referring to the new Accountability and Decision Making framework, the
memo assures that there will be few immediate changes to the safeguard procedures of the past. The
decision was taken early in 2014 to map all Regional Safeguard Advisors and their team to OPCS, which
would now provide the RSA budget.16
RSAs will continue to review and clear investment concepts
(PCN) and appraisal packages (PAD). RSAs may transfer projects for further review to Sector/Practice
Managers - who have final clearance authority.
14 Worth noting is that procurement and financial management both belong to Governance GP without apparently triggering the
same concerns of a conflict of interest. 15 Maninder Gill and Bilal Rahill, Arrangements for Managing Safeguards Beginning July 1, 2014. 16 The decision was taken in April, 2014. Until FY15, RSA budget was paid by the Regional Vice-Presidencies. Other Regional
Operation Services previously mapped to the Regions with joint reported to OPCS, will be reportedly mapped to the Global
Practices.
7
Assignment of specialists: Interim guidance states that environmental and social (E&S) specialists for
IBRD and IDA regular investment lending projects and P4Rs will be managed by Practice Managers in
GP4 and GP13 responsible for the respective region (see Table 1) . E&S specialists for other projects
(DPLs, GEFs, Carbon Finance, etc) will be assigned by GP4/GP13 PMs for their respective region at the
request of the TTLs.
This distinction for investment and policy lending narrows the intentional allocation of safeguard staff
coverage to as little as 50% - 60% of Bank's overall lending portfolio. Moreover, responsibility to assign
safeguard specialists depends on control over budget.
Table 1. Current World Bank Regional Safeguard Advisors, and E&S Practice Managers
The interim guidance states that "budget fixed / labor costs for social and environmental specialists and
STC fees will be covered by GP4 and GP13." While this appears to indicate that salary and benefit costs
for safeguards will be transferred to the GPs, the is no confirmation that off-the-top budget for safeguards
staff has been approved and is now operational procedure.17
Also, little is understood about the Work
Program Agreement negotiation process that will fund GP budgets.18
Variable costs (travel and short-
term consultant fees) are treated differently. The guidance states, "GP Practice Managers or their
17 Informal reports indicate that safeguards budget will be protected from cuts, and will likely increase. 18 See V. McElhinny, "Fixing World Bank Incentives for Effective Safeguard Implementation," BIC Update, Aug. 11, 2014, pgs.
designees will hire the necessary STCs, as required and agreed through the budget allocation exercise
managed by GP4 and GP13. All travel costs will be covered by the budget of the respective projects to
ensure coordinated preparation of the missions. For DPLs, GEFs, and Carbon Finance, the fixed and
variable costs will continue to be covered by the respective projects." Left unclear is how the budget
allocation exercise to determine variable costs will work.
b. Proposal for locating the safeguard staff within Global Practice Vice-Presidency
Building on the interim guidance and the alternative model for organizing the safeguard function within
the Global Practices, for which the Bank has opted, strengthening the safeguard function in this model
would involve several critical reforms. Under such a model (see Figure 2), all safeguard technical cross-
support specialists would be mapped to the ENRM-GP4 and SURR-GP13 Global Practices and should be
grouped as an autonomous unit within those GPs.
An adequate budget for safeguard support (both preparation and supervision) should be controlled by a
safeguard Unit Manager for each of the environmental and social safeguard units. Allocation of
safeguard staff for support and supervision of projects is decided by safeguard Unit Managers in
coordination with GP Managers and each other. As early as possible in the project cycle, RSAs help
trouble-shoot and mediate potential disagreements between safeguard and GP managers regarding
appropriate level of safeguard staff time on the project and other safeguard policy interpretation issues
(related to RSA project clearance authority). GP Director, in coordination with environmental and social
safeguard Unit Managers, prepares annual project preparation and supervision plans (based on CPF
Safeguard business plans - see Recommendation 7 below) and allocates supervision staff resources based
on risk.
The environmental and social (E&S) Risk Director and RSA Cluster Manager (now both part of OPSOR),
in collaboration with LEGEN are accountable for safeguard compliance. As noted in the interim
guidance, all six RSAs will be grouped in a single cluster under a single Director/ Manager with expanded
capacity and budget autonomy. The RSA Manager reports to the E&S Risk Manager/Director. RSA
Manager and in some cases, E&S Risk Manager/Director advise GP VP on resolution of all safeguard
related disagreements between GP Managers or Directors. RSAs clear all projects at concept and
appraisal, exercising authority to transfer some projects to Practice Managers for approval.
9
SG Cluster w/in GP VPMD/ COO
Sri Mulyani
GP VPs
OPCS VP
Kyle Peters
RVPs
E/S Risk Mgr
Mark King
(OPSOR/QACU)
CD/CMU
Program
Leader
ENRM GP
Sr. Dir. & Dir.
U/R/Soc Dev GP
Sr. Dir.& Dir
Governance
GP Sr. Dir.
Mgr: QC,
RSA Cluster
Glenn Morgan
Pr Mgr
Env. SG
Pr Mgr.
Soc. SG
Procure
Mt QC
11 other GP Dir.
Other Sectors
Global
Practice
Managers
Dir. Operation Risk
Stephan Koeberle
LEGEN
Charles DiLeva
VP Legal
Anne Marie
Leroy
Figure 2. Safeguard Organization Option 1 - Structured SG Cluster within Global Practices VP
E&S Risk Manager/Director in collaboration with LEGEN provide cross support, clearance and oversight
to the highest risk, most complex projects (including inspection panel cases). E&S Risk Director controls
the budget for safeguard compliance function activities and allocates RSA safeguard budget based on
risk. In coordination with the RSA Cluster Manager and LEGEN, the E&S Risk Manager/Director is
responsible for hiring, training and certification of safeguard staff, which includes maintaining and
reporting annually on the roster of Bank and consultant staff capacity for the safeguard function.
The Global Practice VP, in collaboration with the E&S Risk Director is responsible for knowing the risk
spread across the actual and projected Bank portfolio as well as alignment with current safeguard staff
capacity.
The GP VP, in collaboration with the E&S Risk Director is responsible for reporting safeguard decisions,
key activities, learning, portfolio E&S risk compared to safeguard staff capacity and activity, outputs and
outcomes in annual World Bank Sustainability Report (see next section for detail). The GP VP, in
collaboration with the E&S Risk Director, is responsible and shall have adequate budget for producing
knowledge products that ensure systemic learning from Safeguard performance.
10
Primary differences and advantages of this proposal compared to current World Bank safeguard model:
Global Practices have independent safeguard budget control for most fixed and variable costs related
to core safeguard function activities
Safeguard compliance and technical support are separated between GP and OPCS to avoid any
possible or perceived conflict of interest.
All safeguard cross-support staff are mapped full time to VP GP with clear reporting line to a single
VP.
All safeguard cross-support staff devote all/most time to SG support or oversight.
RSA Cluster capacity is fully integrated, expanded and coordinated by a single Manager.
Safeguard staff have somewhat more coherent and positive career path and incentive structure (still
some incentive tension in the GPs between Operations and Support technical staff)
Increased reporting requirements on corporate safeguard performance, including knowledge products.
Some of the primary risks that remain with this approach include:
This model is less appealing to safeguard staff that also want to act as project task managers or help
deliver projects. Balancing safeguards with operations management could strike the right balance in
the incentives for safeguard specialists. However, the risk of mixing safeguards and project delivery
actions could be to dilute the safeguard focus into a generalist environmental or social approach. In
other words, a single environmental safeguards expert should not be expected to manage risk
assessment for hydrological flows, deforestation, urban air and water contamination and carbon
taxing policy reforms. Providing high level safeguard advice and support requires both an advanced
level of knowledge of specific complex technical issues that are unique to each sector, as well as
ability to identify inter-sectoral linkages and externalities. This trade-off risk between safeguards and
more career enhancing project delivery duties is evident in the current Bank structure.
Tendency toward fragmentation of safeguard support into many GPs as transport, energy, agriculture
and water directors under pressure to achieve greater efficiency gains recruit trusted safeguard experts
to remain within their GP, diminishing further the critical mass authority needed for effective
safeguards implementation and increasing coordination transaction costs.
Without clear protection or enhanced incentives from a dedicated and independent safeguard
management, embedded safeguard specialists in the GP VP reporting line that has little expertise in
this field may leave the perception that they are less well positioned for career progression or more
susceptible to reprisals for the discharge of their safeguard functions.
As the World Bank moves forward on plans to define a Global Practice safeguards model, any options
that the Bank is considering, including the ones outlined above, should be discussed openly and assessed
on the basis of these criteria before reaching final decisions on the new organizational structure.
Effective implementation of the World Bank safeguards requires a robust, independent critical
mass of environmental and social safeguard expertise with adequate budget, skill mix, clear and
appropriate reporting line, proper incentives and adequate support for strengthening borrower
implementation capacity. In the following section, these safeguard organizational principles are
explained.
11
III. Seven Recommendations for Greater World Bank Safeguard Effectiveness
1. Capacity and skills aligned with development challenges
Recommendation 1: Based on estimates of gaps in current baseline Bank capacity and projected
shifts in portfolio risk and safeguard coverage, increased World Bank safeguard staffing capacity
(by at least 100%), and skill mix are needed to enhance safeguard support and effective policy
compliance during project planning, preparation, supervision, and evaluation.
By many standards and internal accounts, the World Bank lacks the capacity to assure safeguard
management in project preparation and supervision commensurate with risk and environmental impact.19
A gap exists between current capacity and actual or projected risk in the portfolio or the transformational
development challenges that the Bank is attempting. This gap is leaving the Bank vulnerable to an
unacceptable level of risk of non-compliance with current safeguard policies and compromising the
quality and integrity of a number of Bank financed projects. The staffing shortage is also hindering the
ability of Bank Safeguard specialists to adequately undertake other aspects of their mandate, including
addressing emerging issues.
IAD confirms that he Bank lacks an updated, accurate understanding of the spread of safeguard risk
across its portfolio and can not report the total amount or allocation of resources spent on safeguard
implementation. Lack of updated or complete portfolio safeguard risk data make any precise assessment
of this gap difficult to impossible. The Bank lacks an updated baseline of current social and
environmental safeguard staff capacity based on areas of skill and extent of certified expertise. As a
result, IAD reports that "over half of active projects, including some high risk projects do not have
assigned safeguard specialists,... high grade specialists are not necessarily assigned to high risk
projects." 20
The number of staff for IBRD and IDA mapped to ENRM Global Practice is 280 and another 525 to
Rural, Urban and Social Development Global Practice. IAD reports that Environment and Social
Development Sectors identified 172 and 200 specialists, respectively (perhaps only a third of which work
on safeguards full-time). Of the 372 ENV or SD specialists identified by IAD, only 50% were full-time
grade GG or higher.21
When discounting for part-time and staff less specialized in safeguards, a lower
estimate of safeguard specialists at the Bank is approximately 100.22
The Bank has approximately $55 Billion in Category A investment lending in the active portfolio (285
projects, or 12% of all investment projects) and another $92 Billion in Category B investment lending
(1,400 projects, or 58% of all investment projects) out of 2,730 projects valued at approximately $197
Billion on the books. In FY14, the distribution of newly approved IBRD and IDA investment projects is:
19 See IEG (2010) Safeguards and Sustainability Policy in a Changing World, particularly summaries of focus group interviews;
see also, E.Sanchez-Triana, L. Ortolano, G. Ruta, G. Desfuli, and R. Kanakia (March 2011) Implementation of Environmental
Policies, unpublished Environmental Strategy Analytical Background Paper, pgs. 15-30;, more recently, see IAD (2014)
Advisory Review of the Bank's Safeguard Risk Management. 20 IAD (2014: 26-30) 21 IAD (2014: 5) 22 Based on interviews with Bank staff, IAD 2014 Safeguard Review and OPCS Safeguard Diagnostic, "Environment and Social
Development Safeguards in the Global Practice Design," unpublished memo, November, 2013. The Bank refers only to persons
involved in Environmental Safeguards, which may not include social safeguard specialists. However, Sanchez-Triana et al
(2011) argue that of the 229 regular/term staff mapped to environment, only a third of these staff tend to spend a majority of their
time working on safeguards.
12
Category A (41), Category B (269), Category C (144), Category FI (7), and 8 Guarantees.23
Figures 3 &
4 show the distribution of environmental categorization for IBRD and IDA lending over a 25 year history
and for FY14, respectively.
The two charts indicate that Category A projects are about one-fifth of the total Bank portfolio by lending
volume. Category B project volume are about two-fifths, increasing at the expense of Category C
projects over time. About 25% - 35% of the Bank's long-term portfolio is not properly categorized for
environmental risk.24
Distribution of Env. Categorization of
IBRD/IDA Lending Volume (1990-2013) $US
DPL
21%
A
17%
C
23%
B
35%
Other
4%
Figure 3. Distribution of IBRD and IDA Lending Volume by Environmental Categorization (FY90-
FY14)
World Bank FY09-FY14 Ave. Lending by Env. Categorization (%)
7%
47%
29%
13%
4%
19%
37%
8%
31%
5%
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
50%
A B C DPL Other
No. of Projects Volume of Lending
Figure 4. Distribution of World Bank FY10-FY14 Lending by Environmental Categorization
23 Based on World Bank data, in addition to 63 DPLs, 11 PforR, 4 Carbon Offset, and 1 uncategorized project. 24 IAD (2014:7) puts the share of projects not categorized for safeguard risk at 50%. For a detailed explanation of the evolution
of Bank lending and environmental risk categorization, see V. McElhinny," Trends in World Bank lending forecast further
decline in safeguard coverage and signs of a return to higher risk and higher reward lending," BIC Info Update. Oct. 7, 2013
This suggests that each of the 100 full-time dedicated and accredited safeguard specialists at the Bank is
covering nearly 14 high to moderate risk projects, which represent nearly $1.5 Billion in lending. My
estimate is higher than that suggested by IAD, which indicated a project workload of about 10 projects on
average per ENV and SD specialist.25
However, my estimate is lower than that from a recent internal
assessment of safeguard implementation capacity, which reports a variation in the workload for safeguard
specialists by region, from about 20 projects per specialists in the EAP and SAR regions to 50 projects
per specialists in Latin America (see Figure 5). 26
IAD reported that 56% of survey participants in their review responded that they have a heavy workload
and are not able to visit their projects to assess safeguard risks and progress in the implementation of
action plans. Inadequate supervision of safeguards was singled out as a problem, where 48% of the
respondents said the TTLs do not always ask ENV/SD specialists to be part of supervision missions. It is
important to emphasize that these estimates indicate a severe shortage of safeguard capacity while not
reflecting the number of projects that were brought to the Bank and for which the Bank undertook
eligibility and due diligence analysis, but ultimately were not approved.
Figure 5. Number of World Bank Projects per Safeguard Specialist by Region (2010)
Based on its own estimates, OPCS concludes that current staffing constitutes a 30-50% shortfall
compared to need. "Bank staff as a group cover less than 50% of corporate safeguard needs. Another
20% is provided by STCs, including many retirees. We need more staff dedicated to environmental
safeguards, and more staff professionally qualified in safeguards" (emphasis in original).27
For the World Bank, a group of 100 safeguard specialists is expected to ensure effective safeguard
management for 1685 Category A or B projects in the active portfolio (totaling $130 Billion).28
On
25 IAD (2014: 15). 26 Data for this graph is explained in Sanchez-Triana et al (2011: 16). 27 OPCS Safeguard Diagnostic, November, 2013 28 This total does not include development policy loans. See Tables 1 & 2 in Annex B for some comparative indicators of
safeguard capacity at different multilateral development banks.
14
average, each Bank specialist has safeguard responsibility for 3.1 new projects with high or moderate
social and environmental risks, and supervision responsibility for nearly 17 active high or moderate risk
projects.
a. Estimating Supply of Safeguard Risk in the World Bank Portfolio
Demand for safeguard expertise is primarily a function of the changing level of environmental and social
risk in the project portfolio as well as the Bank's appetite for tolerating risk with respect to compliance
with its policies and the environment and social integrity of its projects. The most reliable baseline
indication of Bank safeguard capacity would be the actual time allocation of current safeguard specialists.
For several reasons, timesheet data for how safeguard specialists allocate their time is either not reliable
or not available. The Bank timesheet coding system leaves room for considerable inaccuracy due to the
lack of punctuality in completing the form, shifting of cost codes by managers, and the unsupervised use
of contingency funds. It is common as well that some project preparation costs, including safeguards, are
charged to a prior project that is ending - resulting in challenges for the correct project assignment of staff
hours or costs.
In the absence of this information, a second way to estimate demand is to estimate active and projected
portfolio risk, separated into safeguard preparation and supervision activities. Cost coefficients can then
be used to translate the supply of portfolio risk into demand for safeguard support and compliance
oversight during project preparation and supervision. These cost coefficients may vary to reflect the
Bank's risk tolerance level, but for this paper I assume a low risk tolerance.
The World Bank approves about 40 Category A investment loans and about 270 Category B investment
loans each (10 year average) The average number of Category A and B projects approved between
FY10-FY14 is 309. Current FY14 data indicate that the World Bank has 284 Category A projects and an
additional 1,411 Category B projects in the active portfolio, for a total of $130 billion in high to moderate
risk projects. Adding a share of Category C, FI and Guarantee projects brings the total to 1940 high or
moderate risk projects in the active portfolio as of Sep. 1, 2014.29
Given the President's mandate for
IBRD to increase overall lending by $100 Billion over the coming decade and the Bank to take on greater
risk, we would expect the number of Category A and B projects to grow steadily from the current
baseline. Based on these assumptions, Table 2 provides a breakdown of recent and future World Bank
project approvals by risk Categorization.30
Similarly, the level of risk reflected in the change in Category A and B projects to be supervised in the
active portfolio of the Bank is projected for the next two years. Projected Supervision portfolio is
estimated by adding to the previous year (2013) the total number of approved projects for that year. To
adjust for projects reaching completion and exiting the Bank portfolio, I subtract 15% of the that year's
portfolio. For Category C projects, 50% of prior year portfolio is subtracted due to faster completion
rates. The projected increase in Category A & B projects offset somewhat by the overall decline in
overall lending after unprecedented spike in response to the 2008-2011 financial crisis.
b. Estimating Demand for Safeguard Work with Cost Coefficients
29 In addition to Category A and B projects, a modest amount of safeguard preparation is required for an estimated 50% of
Category C projects (accounting for projects in which safeguard policies are not triggered and additional funding projects) and
supervision resources are required for an estimated 25% of active Category C projects, as well as a small number of public
financial intermediary projects and guarantees. Including these additional safeguard related projects brings the active portfolio to
1917 for FY14. 30 Investment project distribution by environmental and social risk is drawn from data provided in the World Bank's project
database.
15
Safeguard staff cross-support for Category A projects can vary, from as little as 4-8 staff weeks per year
during typical preparation but reaching as high as 1-2 FTE working full time on a project the entire year.
I estimate an average of 43 safeguard staff weeks per year for the 5-10 highest risk Category A project
preparation and 18 staff weeks for the Safeguard preparation of all other Category A projects. I estimate
7 staff weeks for the preparation of higher risk Category B projects, and 3.5 weeks for all other Category
Bs. I estimate 2.4 weeks the preparation of the 15% of Category C projects that require more than initial
screening, and 5 weeks for both FI and Guarantee project preparation.31
Using these cost coefficients for
project preparation summarized in Table 3, an estimated demand for safeguard supervision is 56.2 FTE
Safeguard staff per year in FY2014, increasing by 2016 to 59.6 FTE. 32
Table 2. World Bank Safeguard FTE Demand for Operational Support and Oversight
Preparation 2010 2011 2012 2013 2014 2015 2016
Actual***** Projected
Cat. High A (25%) 9.6 10.5 8.25 7.5 10.25 11.25 12
Cat. A (75%) 38.4 31.5 24.75 22.5 30.75 33.75 36
Cat. High B (33%) 139 143 149.5 108.5 134.5 135 135
31 A conservative estimate of 3 staff weeks for the 200+ Category B projects considers the variation for safeguard demands
between "high" and "low" Bs. 32 Although my cost coefficients are higher than those proposed by OPCS, they are conservative for two reasons. I am not
including 20-25 projects per year on average that are brought to the Bank, involve some preparation or due diligence investment
but are then dropped. I am also not accounting for staff weeks needed to offset the accumulated deficit in safeguards activity
that was not properly staffed over that past five year period of above average lending. Finally, I estimate that 48 staff weeks is
equal to 1 FTE person year (MY), which is likely to be higher than the actual rate.
16
FTE req.***
SG Total
FTE req.***
207.2 213.8 217.0 221.9
* Assume only 25% of Category C projects require safeguard preparation or supervision support
** Safeguard Preparation FTE estimated using 10 SW /yr for Cat. A; 3 SW/yr for Cat. B, and 0.25 SW/yr for Cat. A,
and 5 SW /per year for FI and Guarantees. Staff Years (MY) are estimated by dividing SW by 48 weeks/year.
*** Safeguard Preparation FTE estimated using 10 SW /yr for Cat. A; 3 SW/yr for Cat. B, and 0.25 SW/yr for Cat.
A, and 5 SW /per year for FI and Guarantees. Staff Years (MY) are estimated by dividing SW by 48 weeks/year.
**** Projected Supervision portfolio is estimated by adding to the total number of projects in the active portfolio for
the previous year the total number of approved projects for that year, then subtracting 15% of the prior year's
portfolio. For Category C projects, 50% of prior year portfolio is subtracted due to faster completion rates.
***** Actual portfolio and new projects do not include Development Policy Operations or other projects
uncategorized for risk.
Similarly for safeguard supervision, my estimates of safeguard staff demand are based on the required
supervision for the active portfolio, as opposed to the actual inadequate level of supervision that is now
done. For the 1,875 active projects, a supervision report is produced semi-annually and site visits are
conducted on a periodic basis.33
However, the additional attention that is required to trouble-shoot
problems, consider course corrections, staff more frequent site visits, and analysis for the higher risk and
problem projects implies considerable additional time commitment. For supervision, an estimate of 25
staff weeks of supervision is required for the small share of highest risk Category A project, compared to
9 weeks for other Category As. For the high risk Category B projects, 2 weeks is required, while 1
weeks is required for low risk Cat. B and C. I estimate that 5 weeks for FI and Guarantee projects.
Combined, these coefficients produce a demand for 157.7 FTE staff of safeguard supervision support for
FY2014 with a projected demand of 162.4 FTE by FY16.
Table 3. Cost Coefficients for Low Corporate Risk Tolerance for Safeguard Cross-Support
(average over all projects, includes RSA and OPSOR time)
Category Preparation Supervision
High A 43 weeks (1 FTE) 25 weeks
A 18 weeks 9 weeks
High B 7 weeks 2 weeks
Moderate B 3.5 week 1 week
C 2.5 weeks 1 week
FI 5 weeks 5 weeks
G 5 weeks 5 weeks
Combined, the current and projected social and environmental risk in the World Bank investment lending
portfolio suggests a demand for approximately 214 - 222 FTE for safeguard support.
This estimate is double the existing Bank capacity.34
Even before accounting for other duties assigned
to WB Safeguard staff and the risk management activities associated with 25-35% of non-
investment lending, the severe existing under-capacity is evident.
33 The frequency or intensity of project site visits is difficult to estimate precisely, so I assume 2 visits per year for 2 day mission
on average. 34 The equivalent estimate here of approximately 9,000 staff weeks is three times higher than the 3,150 staff weeks that were
estimated in 2010 to be devoted to environmental safeguard activities.
17
However, operational support duties are only part of Safeguard staff responsibilities. Because of the
overwhelming focus for Bank Safeguard specialists to keep up with the demand for project safeguard
services, safeguard staff have not been able to properly attend to other mandates. An estimated 35-40
additional full-time safeguard staff plus an equivalent support capacity of 60 MY of short-term
consultants would be required to assume responsibilities for additional tasks, including:
policy development and interpretation,
support for corporate initiatives
assess and manage portfolio risk
identifying and helping the Bank respond to emerging issues,
upstream environmental and social risk management in the context of regional and country
programming,
training and accreditation of safeguard staff and consultants;
developing, sharing and disseminating knowledge products;
undertaking the full range of social impact assessment options;
analyzing institutional capacity, strengthening the capacity of borrower risk management and
using country systems;
analyzing the proper inclusion of economic cost-benefit assessments of project's environmental
impact; and
improving the reporting of project safeguard and sustainability outcomes.
In practice, the undersupply of qualified, dedicated safeguard staff, exacerbated by perverse budgeting
incentives, leads to performance breakdowns in implementation. OPCS argues, "Because of the shortage,
most safeguard resources are allocated to project preparation, to get projects on our books, and less on
supervision to improve results."35
IAD adds that the Bank lacks an updated, accurate understanding of the
spread of safeguard risk across its portfolio and can not report the total amount or allocation of resources
spent on safeguard implementation. Over half of active projects, including some high risk projects do not
have assigned safeguard specialists.36
In other words, under the current safeguard system, human and
budget resources for proper safeguard supervision are often not available.
Problems caused by the undersupply of environmental safeguards specialists are compounded by
deficiencies in the mix of skills possessed by safeguards specialists. The Bank will need to factor in the
many areas of expanded safeguard coverage under consideration but not reflected in the current range of
skills available among safeguard staff. Some of the emerging issue areas identified in the safeguard
review would imply acquiring additional staff capacity in a number of areas of social risk (community
health and safety, disability, labor, gender and sexual orientation, child and human rights) as well as
environmental risk (climate, SESA, cumulative impact, biodiversity, ecological flows) and knowledge of
borrower risk management systems.37
A recent internal report highlighted other factors complicating the undersupply of safeguard expertise:
"Safeguard specialists often operate in areas outside the specialties they possess. Often,
one safeguards specialist is expected to provide operational support on projects in a
variety of sectors, ranging from hydropower to roads. Supervising preparation of
35 OPCS (2013) Diagnostic, pg. 1. 36 IAD (2014: 30). 37 In the 2009 revision of ADB Safeguard Policy Statement, an explicit commitment to increase staff resources over a three year
period was spelled out. "The estimates suggest that incremental staff resources ranging from 1253-1749 person-weeks per year
will be required for safeguard review work in the immediate to medium term (2010–2012)." See ADB SPS, 2010, pg. 27-29).
18
environmental documentation for even a single project type, such as a dam for
hydroelectric power, requires a number of technical specialties, such as environmental
flows, limnology, ionic equilibrium at low temperatures, land and aquatic biodiversity,
dam safety, occupational health and safety, geomorphology and sediment transport, and
watershed management. A single safeguards specialist, no matter how well trained,
cannot provide all the necessary support for twenty or thirty projects in different sectors,
much less in all of the subject areas required by the Bank‘s safeguards policies” 38
Not ensuring adequate safeguard staff capacity also constrains efforts to recruit and retain a senior
safeguard team, while continuing to train younger staff. Similarly, without a clear or updated baseline for
how environmental and social risk is spread across the Bank's active and projected portfolio, planning for
a risk based approach to safeguarding projects is difficult. An efficient risk based allocation of safeguard
resources requires precise, regularly updated assessment of portfolio risk to properly align with available
supply of expertise.
Greater reliance on short-term or long-term consultants can only be a temporary solution to the safeguard
capacity problem. Several factors limit the use of consultants:
The prospects of finding qualified consultants available for a temporary period of time are limited
given the nature of specialized expertise required. High demand for the expertise sought often
exceeds supply.
The limited duration of consecutive days that consultants are allowed to work means that after the
maximum is reached the Bank must find new people with the necessary skills and experience.
There will be a high investment cost on Bank safeguard staff in terms of orientation and
supervision.
The transaction costs associated with finding consultants, and processing and managing the
contracts will be significant.
If demand for operational support already exceeds the total Bank Safeguard capacity even before
accounting for other responsibilities that could consume a substantial share of staff resources, the result is
that an alarming part of the Bank's existing portfolio is not receiving adequate or any safeguard support.
Clearly, the Bank reorganization must correct the current undersupply of safeguard staff to ensure
adequate risk management.
The lack of adequate safeguard capacity at the Bank leads indirectly to other inefficiencies and ineffective
uses of existing resources.
IEG argues that “the artificial separation of environmental and social staff between those who work on
safeguards and those who work on social or environmental sustainability is a cause for concern….
forcing an unnecessary division of labor among the social and environmental staff. The increasing
divide between most of the Bank staff who work on safeguards and those who work on environmental and
social sustainability has diminished the Bank’s ability to deliver on its sustainability agenda. To expand
further its focus on issues such as biodiversity, climate change, and benefit-sharing to enhance social
impacts on the poor,” IEG recommends that the Bank move beyond the “do-no-harm” approach—to
encourage greater attention to how safeguards are doing good, in terms of enhancing environmental and
social results. 39
38 Sanchez-Triana, Ernesto, Leonard Ortolano. Giovanni Ruta, Ghazal Dezfuli, Rahul Kanakia, (March 2011) “Implementation of
Environmental Policies,” World Bank 2010 Environment Strategy Background paper (unpublished manuscript); 39 IEG (2011: 22)
19
2. Independent Control of Adequate Safeguard Budget by Safeguard Line Management
Recommendation 2. The World Bank should allocate to the appropriate Safeguard Director or
Manager level independent control over adequate, off-the-top resources approved on an annual
basis to ensure effective implementation of the safeguard policies. Based on World Bank and other
MDB safeguard systems, a minimum safeguard annual budget should be between $60 - $80 million,
depending on several factors.
The prevailing “internal labor market” approach to supplying safeguard specialists for Bank projects is
not working, particularly for project supervision. IEG reported that more than a third of World Bank
projects had inadequate environmental and social supervision, manifested mainly in unrealistic safeguards
rating and poor or absent monitoring and evaluation.40
This failure to supervise many Bank projects is
attributed by IEG to lack of off-the-top budget control by safeguard managers. “The budget for
safeguards supervision in the World Bank is controlled by the task team leaders for each project, who
determine the intensity of supervision and choose the team members or consultants for safeguards
supervision. The concern for technical capacity of the task team leader to make this judgment, or the
conflict of interest, is not deemed relevant during the supervision phase.” “The reliance on an internal
market for safeguards supervision," according to IEG, "has resulted in considerable inefficiencies in
resource allocation for safeguards oversight at the World Bank, compared to IFC.” 41
The Environmental Department background study underscores the same concerns:
Putting safeguard budget control in the hands of country directors, forces environmental
(and social) specialists to sell their services in an internal competitive market that naturally
removes incentives for task managers to hire specialists who, in the minds of some TTLs,
call for lengthy and expensive environmental assessments.” 42
“Staff members interviewed consistently said that safeguards experts were generally
selected on the basis of availability. When asked to clarify this point, interviewees
explained that due to the perpetual shortage of environmental specialists, they drew on
specialists in environmental units primarily for high-risk projects. If no environmental
specialists were available, they would draw supplemental staff from a roster of consultants
who would be mentored or backed up by a Bank staff member while performing safeguards
tasks. Interviewees noted the absence of a structured approach for selecting relevant
safeguard specialists (either Bank staff or consultants) when shortages existed in a region.
As one interviewee put it, staff selection was done on an ad hoc basis.”
These organizational tensions are further compounded by existing budgetary arrangements
that lead to low levels of staffing during project supervision. Funding for safeguard-
related activities is drawn primarily from resources earmarked to ensure safeguards
compliance during project preparation but not during supervision. Budgets for supervision
are not sufficient to adequately meet safeguards support needs, which in turn inhibits the
ability of the Bank to effectively target resources toward projects with relatively high
environmental and social risk.43
The Bank‘s projects’ budget does not include adequate
40 IEG 2010, xii. 41 IEG (2011: 21) 42 Weaver, C. and R. Leiteritz. 2005. “Our Poverty Is a World Full of Dreams: Reforming the World Bank.” Global Governance
(July, 2005). 43 Whitford, P. and K. Mathur. 2008. The Effectiveness of World Bank Support for Community-Based and -Driven Development.
Safeguard Policy Review. IEG Background Paper. Washington DC: IEG.
20
funding for regular supervision of category B projects by environmental safeguards
specialists.44
TTLs may opt not to spend budgetary allowances on safeguards issues and, in
fact, they have incentives to cut costs during supervision because of budget constraints.
As if the proliferation of evidence was not enough, IAD was commissioned to look yet again at the
funding of safeguard activities as part of a 2014 audit. The results, while not surprising or divergent from
long held concerns, also serve to underscore the foot-dragging by management to finally address this
structural impediment to effective safeguard implementation. IAD findings are incomplete due to the
lack of systematized safeguard expenditure data. With the exception of RSA budgets and the LCR
region, no safeguard expenditure data was available for safeguard fixed or variable expenses. On
safeguard budgets, IAD finds:
"In most regions, the project specific safeguard budget is included in the project supervision budget held
by the TTL. ENV and SD specialists claim that they are not always funded adequately for safeguard risk
management due to the prioritizations made by the TTLs. In 5 of the 6 regions, there is no safeguards
budget allocation to ENV and SD sectors, although these sectors allocate ENV and SD specialists to
projects to undertake safeguard work." 45
IAD juxtaposes the reality of this chronic absence of safeguard supervision with the fact that "the TTL has
budget authority for safeguard supervision, [and that] 34% of survey participants responded that the TTL
decides on the level of safeguards support for his/her project."46
For a Bank poised to introduce sweeping safeguard reforms the success of which is premised on a more
accountable provision of implementation support, fixing the existing budgetary roadblocks to effective
safeguard supervision can be seen as nothing less than a litmus test for delivering on this promise.
There are expectations among some in the Bank that the reorganization focus on pivoting the Bank to be
more efficient, agile and reliant on use of borrower systems may actually result in a significant
transformation of the project portfolio and a lowered demand for safeguard resources. Such a scenario
could involve a cost savings in safeguard staffing expenses. In fact, the new strategic budgeting process
places strategic factors above other budget determinants such as portfolio risk. Just as ex ante safeguard
requirements are under pressure to be streamlined and eliminated, ex ante budget estimates may be less
appropriate in the new budget model than, say, an emergency fund to permit rapid response to emergent
safeguard needs during a period of transitional uncertainty.
Without predicting what the long-term changes may be for the Bank's portfolio - in part because they rest
on a number of highly debatable assumptions, the current portfolio will not change much in the short-
term. OPCS reports that the World Bank has a "young portfolio," with an average project having been in
the active portfolio about 3.6 of a typical 6.6 year project disbursement cycle.47
Any change in portfolio
risk or risk management resource allocation will be gradual at best. Unless the new approach is simply to
concede that a significant amount of risk will not be managed going forward (which appears to be the
default approach now), then the Bank would be required to take proactive steps to reduce the backlog of
mismanaged safeguard risk that has accumulated in the active and foreseeable portfolio.
44 World Bank LCR. 2006. Lessons from the Field. A Thematic Review of Safeguard Policy Implementation in Rural and Urban
Water Supply and Sanitation, Community Driven Development, Biodiversity Conservation, Land Administration, Roads, and
Health Sector Projects in Latin America and the Caribbean. Unpublished Report. Washington, DC: World Bank. 45 IAD (2014: 37) 46 Op cit, pg. 38 47 OPCS, "Review of the World Bank FY13 Portfolio and Update on Actions to Assure Quality," April 15, 2014., pgs. 11-12. The
average lag in years for investment projects between approval to exit is 6.6 years.
21
Relevant safeguard unit managers or directors must have independent control over budget for staff salary
and travel expenses, in addition to resources for appropriate training, corporate support, and knowledge
activities. Although CMUs will still control the administrative budget, an off-the-top transfer to finance a
multi-year safeguard budget should be part of a Global Practice Vice-President budget envelope. The
amount of safeguard budget required should reflect evidence-based cost coefficients that are informed by
actual expenditures from prior years, as well evidence-based operational risk related demand in the
projected portfolio, and country level safeguard business plans (see below for more).48
An annual
safeguard preparation and supervision plan should indicate staff needs based on projected frequency and
intensity of site visits, again informed by safeguard Director or Manager assessment of operation and
portfolio environmental and social risk in coordination with but not dependent upon Regional VP or
CMU judgment.
The Bank should report on the lessons learned from the pilot initiative for the AFR and LAC regions
where some limited safeguard supervision staff resources were transferred to the Sector Manager, while
the TTLs retained control of travel budget. Informal sources report that the evidence has been mixed
regarding the strengthening the effectiveness of this pilot in enhancing the risk based allocation of
safeguard resources.
Moreover, the current safeguard budgeting process tends to reinforce an exclusive focus on project
clearance, leaving unplanned and underemphasized many other safeguard functions, including more
dedicated attention to strengthening borrower systems and the "do good" sustainability mainstreaming
aspects of safeguards. Greater off-the-top budget control by safeguard managers would ensure more
strategic allocation of these resources.
In optimizing resource utilization and allocation, options should consider the efficient and effective
apportionment of incremental resource requirements across internal administrative budgets, technical
assistance, and loans, determined by the demarcation between World Bank and borrower/client
responsibilities. In other words, the financing of safeguard activities should tap a wider range of possible
funding sources.
Based on available information and estimates of demand for Bank safeguard support in previous section,
I estimate the total budget needed for OPCS recommendation of a 50% increase in Bank safeguard
capacity and my recommendation for a 100% increase (associated with a demand for 256 MY). See
Annex A for a detailed explanation of these cost estimates. To ensure that Safeguard Managers and
Directors have appropriate budget control for Safeguard cross-support, particularly for Safeguard
supervision, the Bank should:
Increase Safeguard Budget to $80 million (a nearly 100% increase in Bank costs over
current baseline): In line with resolving safeguard staff undersupply and acknowledgement of
past underinvestment in all safeguard work but supervision in particular, the World Bank should
significantly increase the overall safeguard budget by a minimum of 100% per year. The baseline
safeguard expenditure as well as the cost implications of the BIC proposal are outlined in Annex
A. The baseline safeguard costs are estimated at $43.7 million annually. OPCS calls for a 50%
expansion of capacity, which would cost $63.7 million. To fund the seven areas of safeguard
institutional reform provide the greatest assurance of effective safeguard implementation, the
adequate safeguard budget is $81.6 million.
48 Because the Bank‘s regional environment units know best which skills they need for safeguards work, the regions should play
a key role in the design of staff hiring and training programs, and those designs should be elements of each region‘s
Environmental Business Plan. Furthermore, regional environment units should have augmented budgets to fund improved
training and hiring.
22
The Bank should review the need to increase the budget in accordance with challenges of
implementing the integrated safeguard policy framework. A needs assessment for any future
budget adjustments will be conducted with the next policy review.
Special Program for Strengthening Borrower Safeguard Systems: The current Safeguard
policies place major emphasis on the need to assist borrowers to strengthen their own safeguard
systems and develop capacity to implement these. Targeted resources, through establishment of a
trust fund to mobilize external resources, should be sought in the medium term to (i) support the
strengthening of borrower safeguard policies, legal frameworks, regulations, rules and
procedures; (ii) support capacity development of government agencies, borrowers/clients and
civil societies at national, subnational, sector, agency and project-level; and (iii) to conduct
borrower system equivalence and acceptability assessment, and diagnostics on a demand-driven
basis.
Third Party Monitoring and Verification: The Bank should clarify how to systematize and
finance the routine use of third party or community monitoring and verification, based on prior
assurance that social and environmental risks are integrated into project results frameworks.
Recommendation 3: The Safeguard structure must ensure independence and accountability at all
phases of the project cycle, at a minimum, by strengthening the RSA project clearance authority at
concept and appraisal stages.
The current World Bank safeguard decision making structure places greater accountability for safeguard
quality decision at critical stages in the project cycle with the RVPs and CDs, while transferring
accountability to the Networks (now the Global Practice) VPs during implementation (with significant
exceptions). Based on interviews for IEG's Evaluation of the Matrix System, this perception of divided
accountability prevails in most regions and appears to be even more pronounced in recent Management
proposals to clarify decision making authority (see Figures 4 and 5 below).
The Country Director has the authority to chair all project preparation decision meetings (concept and
appraisal). The advisory role of Practice Managers and Directors during project preparation is much less
decisive, in part because of uncertain control over the administrative budget for preparation.49
The Quality
Enhancement Review (QER), which has been mandatory and is convened by the Sector/Practice
Manager, has now been made voluntary. The practice of dividing the responsibility for convening
Concept and Appraisal Review Meetings between Practice/Sector and Country Directors to balance
Region and GP interests was ended to align better with other regions where both meetings are convened
by the Country Director. Recent budget process changes may have only reinforced this power imbalance.
As indicated in Figure 5, the project clearance authority of the RSAs is the single hard check on Country
Director accountability during the project preparation process. Under pressure to eliminate or reduce the
scope of RSA clearance sign-off, the Bank must ensure that this authority be preserved and strengthened
for all but the lowest or no risk operations. This would not only involve preserving the clearance decision
making authority of the RSA, but to provide the resources and organizational coherence to enhance the
49 See McElhinny, Fixing the Incentives for World Bank Safeguard Implementation, for an analysis of the proposed budget
transfer procedures between RVPs and GPs. To date, the Bank has not clarified how the negotiation of the Work Program
Agreement and resource transfer from the Regions to the GPs would work differently that the “old” system described in this
paper.
23
allocation of resources on the basis of portfolio risk. The RSA cluster recommendation is intended to
provide these advantages.
One the reasons safeguard supervision is so poorly done at the World Bank is because RSAs have no role
during supervision. If the Global Practice Directors and Managers are to have any authentic
accountability during project supervision, in addition to independent control over adequate budget for
good project supervision, the responsibility must extend to approval of ISRs and mid-term reviews.50
Figure 6. IEG Depiction of Formal Accountability During World Bank Project Cycle. Matrix
Evaluation (2011, Annex G, Table G.4)
50 OPCS proposal, "ADM Roles" indicates that this is planned.
24
Figure 7. Proposed Accountability in the Investment Lending Project Cycle, OPCS Jan. 30, 2014
25
4. Clear Senior Management structure and reporting line
Recommendation 4: The World Bank requires a Vice-President level champion for safeguards and
a single, clear line management of authority and guidance to ensure that senior, more experienced
Safeguard staff are assigned to higher risk projects.
IAD makes clear that specific, formal accountability for safeguards resides in no single authority at the
Bank. “Fragmented organizational and budgetary arrangements have led to diffused accountability for
safeguard risk management. Responsibilities for safeguard activities are spread among many units
including SD and ENV sectors, OPSOR, RSAs, and industry sector managers without institutional
leadership. Nobody is accountable for safeguard risk management results.” 51
The lack of explicit line management accountability for safeguards implementation contributes to many
of the problems pointed out in this paper (misallocation of human resources, lack of updated information
about risk and capacity to manage risk, unmet demand for safeguard supervision and emerging areas of
risk management). Increasingly, accountability problems result from power imbalances between
safeguard experts and inexperienced Task Team Leader (TTLs). OPCS reported that a significant share
of the Bank's portfolio - 37% of the projects in the IPF portfolio and 42% of operations in the project
pipeline, are managed by less experienced TTLs - those with 2 years or less of experience as an
"implementation TTL."52
Misallocation of safeguard human resources is a common outcome, as
evidenced earlier in IAD findings regarding mismatch between project risk and level of safeguard
expertise. 53
As a result, Environmental and social development specialists do not have sufficient independence to
challenge the loan team leader on safeguards performance during project implementation. IAD reports
that:
77% of survey participants think that management does not value their safeguards work.
21% of surveyed safeguard specialists reported that they are not able to communicate identified
safeguard issues to the right levels of management in a candid manner. 54
To correct the lack of explicit accountability, regional variability and duplication of effort for the delivery
of safeguards at the Bank, the reorganization should clarify a single, primary reporting line for all
safeguard staff. The Bank has been overly focusing on addressing a perceived conflict of interest
between safeguards quality assurance and quality control, when this may not have been as significant a
risk as other defects in the organizational structure.55
Both the safeguard QA and QC functions require a
dedicated senior manager (separately for social QA and environmental QA staff). An effective, high
quality leader at safeguard practice manager positions is essential for defending safeguard staff in the
discharge of their duties that are based on portfolio risk under pressure by operations staff to speed
approval and challenge compliance. With most safeguards staff are mapped primarily to Global
Practices, clear accountability for effective safeguard delivery should rest squarely with a champion at
this level, rather than an ambiguous mandate for a lower level advisory position. The Vice-Presidential
51 IAD (2014: 40) 52 OPCS, "Review of World Bank FY13 Portfolio and Update on Actions to Assure Quality," (April 15, 2014). An
implementation TTL is defined by the number of years the person has archived an ISR. 53 IAD (2014:30) 54 Op cit, 21-23 55 Both the Procurement QA and QC functions have been relocated to the Governance Global Practice without the same level of
concern about managing any risk of conflict of interest.
26
level accountability for safeguards depends on clear and timely advice from LEGEN, OPSOR, as well as
other members of Operations Committees. However, for all safeguards effectiveness indicators, ranging
from resolution of disagreements related to high risk, complex projects to public reporting of safeguard
outcomes at a corporate and project level, ultimate responsibility lies with the respective Vice-President.
5. Safeguard Incentives
Recommendation 5: Revise rewards and recognition for Safeguards work in World Bank
Performance Evaluations. Performance evaluations should transparently reward quality of
operation outcomes or impacts (including safeguards implementation outcomes) in addition to
contribution to volume of lending approvals, among other factors.
Deficiencies in environmental safeguards specialists’ staffing levels and skills mix are also rooted in the
disincentives to pursuing safeguards work as a career path at the Bank. The IEG report found that “staff
promotions are slanted towards own-managed projects more than toward providing safeguard services.”56
A staff survey conducted by IEG for the Matrix System at Work Evaluation reported, Sector and anchor
staff reported receiving substantial encouragement from sector and country management to meet lending
targets and adapt knowledge to country needs (60%), but much less so to collaborate across sectors
and mobilize bank wide expertise (25%- 40%)” (Appendix D, page 152).57
Since the Wapenhans report, the Bank has acknowledged that lending approval amounts are the driving
incentive within the World Bank. The shortage of safeguards expertise available for safeguards work is
often attributed to this work not being recognized as essential at the Bank or treated as an attractive career
choice. Safeguards experts note that they meet more problems inside the Bank than outside when carrying
out their safeguards duties, often confronting approaches to the application of Safeguards as a box-ticking
exercise.
Because safeguards operational support is often not seen as a viable career path within the Bank, most
operational support in some regions is provided by junior staff members that typically lack appropriate
risk management experience. Moreover, in an effort to find a career path that has more promising
opportunities for promotion, many junior environmental safeguards specialists are refocusing their work
programs away from safeguards toward environmental operations, particularly analytical work and
investment loans. The mixed incentives to combine safeguards support with operation delivery are not
inherently opposed, but must be clearly delineated in terms of their purpose to avoid the perceptions of
trade-offs.
The Bank must provide tangible and consistent incentives to retain and develop environmental safeguards
specialists. Improving incentives for high quality safeguards work begins with clear, unequivocal
messaging of Bank Senior Management that safeguards add value to projects. Recognition of that value
can range from public praise to revisions to the Overall Performance Evaluations (OPE) process. From a
recent background report on the matter:
56 IEG “Safeguards and Sustainability Policies in a Changing World,” 2010, 36. 57 For a more in - depth discussion of recent Bank performance in terms of incentives for higher quality outcomes, see V.
McElhinny, "Persistent Decline in World Bank Lending Quality Leaves Results in Doubt," BIC Info Brief, Oct. 1, 2014.
"Safeguards-related components could be included into the protocols for performance
evaluations of TTLs, sector/practice managers and environmental and social safeguards
specialists. This can be done easily by expanding the list of feedback providers during a staff
member‘s OPE to include Bank colleagues in a position to gauge the quality of safeguards
related work for the staff member being evaluated. In addition, the OPE peer comparisons
made in evaluation of environmental safeguards specialists should be altered so that instead of
being compared with TTLs, safeguard specialists only be compared (for purposes of evaluating
their safeguards-related work) with other safeguard specialists. The inclusion of safeguards-
related criteria and feedback providers in OPEs will provide staff with feedback on their
performance in supporting safeguards implementation, influence personnel decisions such as
salary and promotions, and help individual regions identify their training and staff
development needs."58
6. Increase resources for safeguard capacity strengthening
Recommendation 6: Provide greater safeguard implementation training for Bank and Borrower
specialists, managers and executives.
The budget limitations for safeguard work have also contributed to a backlog of internal capacity
building. What training resources that do exist are largely controlled by OPSOR. A more ambitious
training program should not only address skill shortages, but how safeguard specialists become more
involved in upstream country and policy dialogue (where budget is controlled by RVPs) and safeguard
implementation (which has been notoriously underfunded). 59
Training for safeguards has been criticized
as being to process based (as opposed to specific areas of sector or topical expertise) and not as accessible
or suitable to needs from staff based in the country offices.60
Certification systems were also lagging behind the diversification of the Bank’s portfolio. Although
management has reportedly stepped up the accreditation program, IAD find that the initiative is still at an
early stage of development.61
Training resource should be made available to Practice Managers in the
ENV and SD GPs. A modest budget for training was proposed in Annex 1.
Worth noting is that safeguard training is not mandatory for all non-administrative staff. All Bank staff,
including executives, should be knowledgeable about safeguards through a minimum of basic training.
As the World Bank prepares to shift to a safeguard system that relies to a far greater degree on borrower
systems for managing social and environmental risk, Borrowers have expressed concerns that greater
responsibility is unmatched by greater Bank resources to strengthen local capacity. Some Borrowers have
complained that the Bank’s proposed Environment and Social Standards will discriminate against smaller,
less well-resourced clients. Evaluations of the Bank’s Use of Country Systems pilot also pointed to the
lack of resources for assisting Borrowers to fill gaps identified in the equivalence and acceptability
assessments.
58 Sanchez-Triana, et al. (2011). 59 The absence of clear social and environmental risk analysis focus is a observed concern in the new Country Partnership
Framework and Systemic Country Diagnostic, which is being piloted now in several countries based on new guidance issued in
2014. 60 IAD (2014: 35); Sanchez-Triana, et al (2011:31-38) 61 IAD (2014: 32-36).
28
When updating its own safeguard policies in 2009, the Asian Development Bank recognized the need to
allocate resources for capacity building,
“The SPS places major emphasis on the need to assist DMC’s to strengthen their own safeguard
systems and develop capacity to implement these. Such capacity development will be pursued together
with bilateral and multilateral partners, based on demand from DMCs. Targeted resources of $80-100
million, through establishment of a trust fund to mobilize external resources, will be sought in the medium
term to (i) support DMCs to strengthen their safeguard policies, legal frameworks, regulations, rules
and procedures; (ii) support capacity development of DMC government agencies, borrowers/clients
and civil societies at national, subnational, sector, agency and project-level; and (iii) work with
DMCs to conduct CSS equivalence and acceptability assessment, and diagnostics on a demand-driven
basis.”62
Combining multiple sources of possible finance, both reimbursable and non-reimbursable, effective
safeguard implementation will depend on an estimate of potential need by some Borrowers for capacity
building assistance.
7. Strengthen Accountability for better tracking, reporting and learning about Portfolio E&S risk,
Safeguard costs and development benefits
Recommendation 7: Reporting safeguard outcomes: Strengthen accountability for better planning,
tracking, reporting and learning about portfolio environmental and social risk, Safeguard costs and
development benefits, including expanded use of independent and community monitoring of higher
risk projects, input from independent experts, and the publication of an annual Global Reporting
Initiative (GRI) format Sustainability report that discloses disaggregated portfolio risk and Bank
capacity to manage that risk.
Informal reviews of the current Bank portfolio indicate that for a surprisingly large share of projects,
information about safeguard implementation is incomplete or absent altogether. IAD reports that the
Bank lacks an updated, accurate understanding of the spread of safeguard risk across its portfolio and can
not report the total amount or allocation of resources spent on safeguard implementation.
The Bank does not systematically track or document that safeguard risks are effectively mitigated.
58% of safeguard specialists report that environmental and social risk management is not adequately
documented in the project supervision report (ISR). Most safeguard supervision information under the
Bank's new integrated risk rating tool (ORAF) is either missing or not updated.63
These glaring gaps in safeguard performance information suggest a core failure in the World Bank
safeguard system. The problems extend to the lack of candor or relevance regarding information about
environmental and social risk management that rarely appears or is understated in supervision or
completion reports, or project appraisal documents. Far too often, the language in PADs is designed to
"panel proof" a risky project, rather than promote honest risk analysis. The Bank's project information
62 ADB (2009) Safeguard Policy Statement, pg. 27, ¶ 77. The actual amount of resources that the ADB has committed to
capacity strengthening remains unclear. 63 Op cit, pg. 21
29
system fails to document the decision making process regarding risk management, which prevents any
true and full accounting of safeguard costs and benefits.
To capture and learn from knowledge of safeguard application, the Bank must better define the
accountability for procedural requirements to ensure environmental and social risks are integrated into a
projects results framework and are tracked, reported and used to inform subsequent project decisions.
As such, World Bank should strengthen project monitoring and evaluation frameworks to assess overall
safeguard performance against the objectives of the policy. IEG reported that ” One of the main
constraints in assessing the value added of the safeguards and performance standards is that costs and
benefits are not systematically tracked by the World Bank Group.” 64
IEG’s own estimates illustrate that
environmental benefits can be estimated, including those benefits that extend beyond “do not harm.”
Even conservative estimates of the economic benefits of safeguards outweigh the costs of safeguards
implementation. Moreover, IEG shows that with adequate data on costs, benefits, risk, and externalities
incurred during implementation, it is feasible to quantitatively estimate the impacts of safeguards.
Based on updated safeguard cost benefit analyses provided by the Bank, an options menu for covering the
costs of safeguard implementation should be presented, which include:
revenue sources within the Bank's current business model;
borrower capacity to pay a greater share; options for outsourcing supervision to independent
contractors; the use of random environmental audits; and