> home

Norwegian study examines Eurodad and World Bank conditionality analysis

11 December 2007

On 28 November 2007, the Centre for Development and the Environment of the University of Oslo published an analysis commissioned by the Norwegian Ministry of Foreign Affairs of the two reports “Conditionality in Development Policy Lending (World Bank, December 2007) and “Untying the knots: How the World Bank is failing to deliver real change on conditionality (Eurodad, November 2007).

Eurodad welcomes the findings of the independent assessment of the reports provided by this analysis and acknowledges room for improvement with regards to Eurodad’s report. Eurodad used the World Bank’s own conditionality database for its report and, with very limited resources, managed to challenge the findings published by the World Bank.

As the analysis points out, both reports acknowledge progress in the implementation of the World Bank’s use of conditionality. And both reports suffer from analytical and conceptual weaknesses. However, it is noteworthy that the analysis endorses some of the Eurodad approaches and definitions which have the potential to best capture the adherence of the World Bank to the Good Practice Principles. Most importantly, the analysis warns that the World Bank report may be painting a rosier picture than it should.   

Figures discrepancy in the two reports

The analysis, which aimed to discuss the main points of contention between the reports and possible reasons behind this, finds that the most striking discrepancy between the two reports refers to the number of operations that include conditionalities in “sensitive policy areas” – as the Eurodad report finds that 71% of the operations include such conditionalities whereas the World Bank report finds that 30% of the operations include such conditions. The analysis rightly points at differences in definitions as the main reason for the discrepancy.

As the analysis points out, it is “difficult to formulate a definition that accurately captures sensitive policies without also including policies that are not considered sensitive” and it is also often “difficult to classify conditionalities merely based on the reading of a policy document, without having in depth, case-specific knowledge of the operations involved.” The analysis also warns that “the narrow definition of the World Bank would not capture situations where a sector is opened for private competition and the state owned enterprise is denied funding to compete and therefore gradually looses out.”

According to the author of the analysis, the World Bank report has understated the share of operations that contain conditionalities in “sensitive policy areas”. The World Bank says that only 30% of the operations contain such conditions, whereas the author of the analysis found that this figure could be increased up to 40% even according to the World Bank’s own narrow definitions. With regards to definitions, the Norwegian analysis also points that Eurodad’s report is “right in highlighting that some public sector reforms may move the management of the economy towards a “market economy model” without fitting with the criteria for “sensitive policy reform” as adopted by the World Bank.” Drawing on an example from Burkina Faso where the Bank classified a condition to “issue tender for the selection of private operator for the management of the electricity sector” as “public sector reform”, the author states “Eurodad is probably quite right in classifying this as privatisation and therefore as “sensitive policy conditionality”.

Beyond definitions

Beyond definitions, as the analysis says, discrepancies in the figures may also be explained by the fact that Eurodad assessed a smaller sample of the operations over a longer time frame than the World Bank did. The decision to assess operations over a longer time frame was driven by the assumption that a two years time frame would be statistically more robust and would provide a more accurate analysis of trends in the World Bank’s use of conditionality. The decision to assess a smaller sample was due to capacity constraints. However, it must be noted that Eurodad assessed 20 out of the 38 IDA operations assessed by the World Bank, which adds up to more than 50% of the IDA operations and more than 50% of all conditions contained within the 38 operations. The confidence interval for this sample is smaller than 2, which makes Eurodad’s estimates highly reliable.

The analysis says that “the Eurodad report …does not actually state that 71% of the loans and grants had conditionalities for the adoption of sensitive policy reforms” whilst the Bank strictly assesses the actual conditions attached to the operations containing sensitive policy reforms.  However, both reports do in fact count actual conditions attached to World Bank grants and loans. The difference lies in the fact that the Bank only counts binding conditions and assesses non-binding conditions separately. Eurodad includes both binding (prior actions, triggers) and non-binding (benchmarks) conditions in its definition of conditionality.

Not so Good Practice Principles

The analysis expresses concerns about the definition and implementation of the World Bank Good Practice Principles on conditionality, which are very much in line with concerns raised by the NGO community over the past two years and with section 5 of the Eurodad’s report. “Reforms that the World Bank or other donors bring from abroad will be viewed as imposed irrespective of whether there are conditionalities attached to them or not” the analysis says. It also highlights that the World Bank should avoid attaching conditions to their development finance which may be used by some government’s to facilitate implementation of reforms which face strong domestic opposition. Instead, “it may be a better strategy for the Bank to avoid making loans conditional on the implementation of such reforms.”

On customisation (the good practice principle which calls for adjusting “the modalities of Bank support to country circumstances), the analysis states that “in the current global context, wherein a neo-liberal development strategy is no longer “the only game in town”, the need for policy space should be increasingly stressed.” And, last but not the least, with regards to criticality (the good practice principle which requires to “choose only actions critical for achieving results as conditions for disbursement”), the author recommends that “in accordance with the Eurodad’s recommendations, Norway could advice against “bundling” of conditions as it often lead to lumping together both critical and non-critical conditions. It also makes programmes less transparent and open to external reviews.”

The Bank painting a rosier picture

Beyond discrepancies and weaknesses identified in the two reports, the analysis concludes that progress has been made regarding the reduction of conditionality in sensitive policy areas. However, the Bank should not understate the share of sensitive policy reforms still attached to their grants and loans and, most importantly, that the Good Practice Principles should definitely be improved.

The usefulness of this independent analysis shows, once again, that civil society recommendations to establish a system of independent monitoring for the World Bank conditionality are ever more needed. 

Operations assessed by Eurodad : Kb