IFI governance

News

IMF conditionality high, effectiveness low

1 February 2008

A report by the IMF’s evaluation arm faulted the Fund’s overuse of structural conditionality and partially blamed donors for the problem, but civil society critics of conditionality are not satisfied with the scope of the report or the changes accepted by the Fund.

Structural conditions are economic and political reforms that the IMF requires to the structure of a borrower’s economy, such as privatisation of public enterprises, and are different than quantitative conditions such as inflation and deficit targets. The Independent Evaluation Office (IEO) submitted the report to the IMF executive board late in 2007, but the report was not released to the public until early January. The IEO reviewed all of the Fund’s lending operations between 1995 and 2004 and found that IMF programmes, for both middle- and low-income countries, had an average of 17 structural conditions.

Echoing criticism of past IEO reports (see Update 55 ), Soren Ambrose from NGO Bank Information Center’s Africa office lamented the IEO’s limited approach. “Unfortunately, in its evaluation the IEO does not consider the content of the conditions. The question of misguided ideology is, therefore, left aside. … It does not examine the programme goals, nor does it question whether alternative policies could have been used to achieve the same, or better, ends. Likewise, no attempt is made to gauge the intrusiveness of conditions on governments’ latitude in determining their own economic policy.”

all the more reason to believe that it is imposed

“While the focus on structural conditionality was interesting, another IEO study is needed for examining the IMF’s quantitative conditions, which arguably can be even more important,” said Rick Rowden, senior policy analyst at ActionAid USA. “This is of tremendous importance and ought to be examined, especially when mounting evidence suggests that some of the IMF’s quantitative targets are not well founded in the empirical research.”

The report focussed on whether the IMF’s structural conditions were ‘effective’ at furthering structural reform and whether efforts, launched in 2000, to simplify conditionality (see Update 22) were successful. It also highlighted that compliance rates for IMF structural conditionality were quite low at 54 per cent, and less than 33 per cent for conditions that involved reforms of “high structural depth”.

Semantic games

The IEO defined effectiveness as “whether structural conditionality was effective in bringing about follow-up structural reforms”. It found that there was “only a weak link” with 40 per cent of reforms stalling and 5 per cent being reversed. It also noted that “the effectiveness figures are almost identical regardless of whether conditions were met, met partially or after a delay, or not met at all.” The IMF staff response to the report grumbled that this was an inappropriate way to judge IMF conditions and programmes and that they should only be judged on whether they have met programme goals, not on reforms after the programme has ended.

The low effectiveness ratings led Benedicte Bull, a researcher from the University of Oslo, to be sceptical about government ownership of IMF reform programmes: “If conditionality only achieves minor and short-lived reforms there is all the more reason to believe that it is imposed. And indeed I think that such limited and perhaps short-lived reforms may in some cases do more harm than good.” Ambrose felt that the low compliance rate “suggests that the IMF is trying to force countries to take more drastic measures than are required for the realisation of their programme goals. Even accepting the assumption that the programme goals are wisely chosen, this means that, for example, governments may be working to force down inflation rates, often at the cost of economic growth and spending on health and education, to rates that are unnecessarily low.”

While IMF staff, management and some members of the board were wary of playing a “numbers game”, the IEO report also gave an in-depth numerical analysis of conditionality over time and across sectors. It found no statistically significant difference in the number of conditions after the IMF approved its new conditionality guidelines in 2002. It observed that conditionality “shifted out of privatisation of state-owned enterprises and trade reform toward tax policy and administration, public expenditure management, and financial sector reform.”

Aldo Caliari of US-based NGO Center of Concern found the exercise unconvincing: “Most of the trade reform conditionality has disappeared because the WTO has taken up the mantle of forcing reforms in this area. Also, a lot of trade-related conditionality is hidden in other categories. For instance, privatisation conditionality, and the defunding of public services via fiscal deficit targets that tend to precede it, are key to advance the liberalisation of trade in services and, yet, not counted as trade conditionality. The same is true for financial sector reform, considered among the ‘core’ areas of IMF competence.”

Donors in the dock

The IEO laid part of the blame for the lack of progress on reducing conditionality at the door-step of donors. “Fund arrangements were used by donors and others as monitoring and signalling mechanisms for other initiatives. … Some conditions were included which, while important for the monitoring role, were not critical to the explicit programme objectives.” Though claiming that the IMF is good at this signalling function, the IEO questioned whether this was an appropriate role for the Fund’s lending instruments.

Many civil society groups reject this conditionality architecture, arguing against cross-conditionality between multilateral institutions and donors. A recent Eurodad report on aid to Sierra Leone bore out this problem of cross-conditionality. “As far as multi-donor budget support goes, in order to qualify at all, the government of Sierra Leone has to meet three entry level conditions” including “continued good macroeconomic performance, as evidenced by satisfactory progress under an IMF programme.” Thus donors have a stake in what conditions the IMF sets, aside from their own conditions and those imposed by the World Bank. The IEO report particularly fingered Europe for this problem, with IMF programmes in countries hoping to join the European Union serving as monitoring tools for the EU accession process.

The IEO’s main recommendations were for a cap on the number of conditions, the elimination of structural benchmarks and conditions that do not fall in the Fund’s core areas of expertise, and better explanations of the rationale for conditions. The board argued for business as usual, rejecting caps and demanding that the Fund work harder to implement the current conditionality policy and better explain the use of conditions.

European NGO Eurodad concluded its in-depth review of the IEO report with this response: “the Board only reiterates commitments already taken in 2002 when the conditionality guidelines were approved.  However, if the IMF is serious about making progress in the effectiveness of their programmes and about reaching out to wider constituencies beyond its Washington headquarters, it should listen carefully to constructive criticism and open up to change.”