IEG calls for overhaul of World Bank’s lending criteria

15 February 2010

The withholding of a critical evaluation of the World Bank’s method of allocating lending to low-income countries raises suspicions that the Bank is seeking to avoid public criticism in advance of the upcoming negotiations for replenishment of the Bank’s cheap loan arm.

Last year, the World Bank’s Independent Evaluation Group (IEG) reviewed the Bank’s Country Policy and Institutional Assessment (CPIA). The CPIA is a rating of countries’ policies and governance which largely determines how much money low-income countries can borrow from the International Development Association (IDA), part of the World Bank Group. (see Update 63, 52, 43) A leaked copy of the IEG report reveals a call for a complete overhaul of the CPIA, with a “thorough review” of every single indicator, and the abolition of the index itself in favour of publication of its separate components.

The CPIA is a controversial assessment of the policies and institutions of all the countries the Bank works in, though it only determines loan allocations for low-income countries. It is made up of sixteen indicators, grouped into four clusters: economic management, structural (economic) policies, social inclusion and governance.

The IEG report, due for publication no later than January under Bank guidelines, has not yet been released. The only permitted reason for this delay is if executive directors , representatives of shareholder governments, demand a discussion before publication.  The CPIA has been one area of dispute among donor countries in advance of this year’s IDA funding round which is due to start in March (see Update 69). Ministers from low-income countries at the IDA mid-term review in November 2009 complained about the CPIA, arguing that its implicit conditionality should be reduced. The fact that the publication of the report is only expected in late February suggests that the cause of the delay may be a desire not to air this controversial topic in public before negotiations begin.


While arguing that “the content of the CPIA broadly reflects the determinants of growth and poverty reduction,” the leaked IEG report says that “the literature offers only mixed evidence regarding the relevance of the content of CPIA for aid effectiveness broadly defined – that is, that it represents the policies and institutions important for aid to lead to growth.” This is a damning finding, given that this is exactly the purpose of the CPIA – to guide the allocation of the Bank cheap loans and grants.

The IEG also puzzles over the fact that 68 per cent of the weighting for determining IDA loans rests on the governance part of the CPIA, with only 8 per cent for each of the other three clusters and 8 per cent on existing portfolio performance, concluding that this “is driven much more by fiduciary and possibly other concerns of donors than by the objectives of achieving sustained growth and poverty reduction.”

Trade liberalisation bias

Trade indicators come in for most criticism, where the IEG finds that “the specification of particular tariff rates for different ratings reflects a one-size-fits-all approach to trade liberalisation that is not supported by country experience.”  They argue that “at moderate tariff levels (which practically all countries currently have), complementary factors (macroeconomic stability and trade facilitation) are more important than further tariff reduction to promote integration into the global economy.” Furthermore “the assessment of trade liberalisation needs to take into account the extent of intersectoral labour mobility because the former in the absence of the latter could exacerbate poverty.”

The IEG also recommends a complete reformulation of the financial sector indicator in the light of lessons learned from the financial crisis, giving further ammunition to those who have criticised the Bank’s promotion of financial sector liberalisation and deregulation (see Update 68, 63).

While finding that the Bank has improved the internal systems that produce the CPIA, the IEG notes that “having staff rate the countries on which their work programmes depend could lead to rating biases.”

In September last year, University of London academic Elisa Van Waeyenberge published a rigorous assessment of the CPIA in the European Journal of Development Research.  She found that “the CPIA perpetuates significant (and particular) [World Bank] influence over the policy space of [low- income countries]” and that it steers World Bank lending “to promote the further adoption of the [World Bank’s] traditional (neo-liberal) reform agenda.”   Much of Van Waeyenberge’s critique centres on the economic policies the Bank classifies as ‘good’ through its selection of CPIA indicators, and the ‘guideposts’ that direct staff when preparing CPIA analysis.

She also argues that the governance cluster is also based on a disputed conception of governance, and that the weighting given to it in the IDA allocation formula means that changes in each governance indicator can significantly affect IDA allocations, raising “the spectre of uncertainty and volatility of aid flows with negative effects on investment levels and government expenditures in debtor countries”

She argues that the use of the CPIA to guide IDA allocation is part of a longer term shift in Bank conditionality, away from trying to influence policy reforms towards directing funding to countries which already have policies that the Bank favours.   She concludes that “the policies implied in the CPIA are in general neither sufficient nor necessary for growth and preclude the various types of strategic interventions that were deployed, for example, by the East Asian tiger economies.”

IEG-Bank friction

Further friction between Bank management and its arms-length evaluation unit was highlighted in the Bank management’s October response to the IEG’s 2009 Annual Review of Development Effectiveness (see Update 66). In a move which seems intended to water down the independence of the IEG, management asked them to stop evaluating the Bank’s progress against IEG recommendations, and instead switch to measuring progress only against management commitments.   The IEG has not so far acceded to this request.