Could IDA21’s New Results Framework Make IDA22 Harder to Champion?

Last week was the start of the 21^st replenishment cycle of IDA, the arm of the World Bank that supports the lowest-income countries. Over the next three years, IDA will provide close to $100 billion to the world’s 78 poorest countries and fragile states. IDA21 offers significant policy continuity from the previous cycle, IDA20, but one respect in which it will differ is results measurement. A key goal of the replenishment negotiations was to re-tool and streamline the results management system, which World Bank management argued had become overly burdensome and incentivized a culture focused on volumes instead of impact. This goal was achieved: the IDA21 results framework has significantly fewer volume targets than IDA20 and a more focused set of results.

In making the case to donors, management emphasized that the new framework would strengthen alignment within IDA and across the World Bank Group (WBG), fostering greater accountability and transparency. Indeed, the IDA results indicators are now fully aligned with the latest World Bank Group scorecard, adopted in 2024. But there is a downside: IDA’s results indicators are strictly limited to the 22 issues covered in the scorecard. As a result, some issues that were prioritized in IDA20 and before are no longer included, creating important accountability and data gaps. For example, IDA now has only one results indicator under health: millions of people receiving quality health, nutrition, and population services. Previously, these were broken down by category, giving stakeholders a better understanding of IDA’s impact.

Nowhere is this gap more apparent than for gender, including sexual and reproductive health (SRH), which no longer has any results indicators. IDA21 has only one gender commitment—to implement the new World Bank gender strategy in all IDA countries—that will be measured against one broad indicator: “millions of people benefiting from actions to advance gender equality, of which (%) from actions that expand and enable economic opportunities.” Across the board implementation of the gender strategy is an ambitious goal, but as discussed below, there is less to it than meets the eye. And the indicator itself will not offer any insights into gender-related outcomes at a programmatic level.

Ultimately, allowing for a few IDA-specific indicators would have enabled a more robust framework for assessing impact in IDA countries. In an environment of shrinking official development assistance (ODA), stakeholders need all the help they can get to make their case to potential funders. Data on outcomes and impact are key to success.

How IDA manages for results

Each IDA cycle includes a set of policy commitments (country-level programming targets) designed to support implementation of key sector and/or thematic priorities set during the replenishment. The term policy commitment is misleading: these are financing targets (e.g., inputs often used to encourage uptake in new areas (e.g., crisis preparedness or childcare) or in areas prioritized by donors but not necessarily by borrowers (e.g., support for refugees). They have proliferated over time: IDA20 included upwards of 1,000 financing targets related to human capital, climate, fragile and conflict situations, gender, jobs, crisis preparedness, and governance. IDA management pointed to this proliferation as evidence of an overly burdensome and volume-oriented system.

IDA also uses three types of indicators to measure performance:

• Country context: country-level data on poverty and development (e.g., percent of a population with access to water, energy, or the internet).
• Results: outcomes of IDA programming; (e.g., millions of additional people with access to water, energy, or the internet because of IDA financing).
• Organizational effectiveness and efficiency dashboard: process assessments (e.g., timing from project approval to disbursement, percentage of projects rated satisfactory).

In theory, the policy (financing) commitments, country context, and results indicators should be linked. For example, if IDA commits to finance reproductive health care programs in 20 countries, there should be a corresponding country context indicator showing the population with access to reproductive care and a results indicator showing increased access to reproductive services due to IDA programming. But in practice, this has often not been the case, something donors sought to correct in IDA21.

IDA21 results framework: Collapsed and condensed

For IDA21, donors supported management’s proposal to increase alignment within IDA and to the WBG scorecard, which aggregates 22 results indicators from all four lending arms of the World Bank Group: IDA, IBRD, IFC, and MIGA. As a result, the IDA21 scorecard will have 22 country-context benchmark indicators that align with 22 results indicators, versus 39 and 33 respectively in IDA20. These indicators will also be aligned with the funding targets. As an example: IDA21 has a funding target to help 50 countries increase access to broadband connectivity, a country context indicator on the percentage of the population using the internet, and a results indicator reporting on the millions of additional users as the result of IDA financing.

The improved alignment is welcome because many IDA20 financing targets lacked corresponding country context or results indicators. (See Table 1 for examples.) However, the rigid adherence to the World Bank scorecard has created accountability gaps for issues especially salient in IDA countries, like access to nutrition and clean cooking, which were embedded in the IDA20 results framework but are not part of the scorecard. The elimination of these indicators also means that stakeholders will no longer have easy access to time-series data which enabled them to compare IDA’s performance over different replenishment cycles.

Gender results management in IDA20: Well-intentioned but flawed

IDA20 included 13 funding targets, 10 country-context benchmark indicators, and 6 outcome indicators related to gender. But while coverage of gender was expansive, the alignment was poor, making it difficult to form a coherent story or message around IDA’s gender engagement. (See Table 1.)

Table 1. IDA20 gender-relevant funding targets and indicators

*Several policy commitments have been condensed for the table; ** Included as a policy commitment but not framed as one.

What stands out is that only five of the thirteen gender-related funding commitments were aligned with country context and results indicators (i.e., SRH, financial inclusion, education, nutrition, and jobs). This could have been easily addressed in IDA21, but shareholders opted for a very different approach instead.

IDA21 upends gender commitments

For IDA21, donors agreed to condense IDA20’s 13 gender commitments into one overarching goal—to implement the new WBG gender strategy in all IDA countries. They also agreed to reduce the country-context gender indicators from ten to one and the outcome indicators from six to two, with a focus on access to finance. As shown in Table 2, this represents a substantial departure from IDA20.

Table 2. IDA21 results framework for gender

What stands out about the new gender component of the results framework is that the alignment within IDA is still poor:

• There is no policy commitment (e.g., country financing target) to support the financial inclusion country context and results indicators.
• There are no country context or IDA results indicators around SRH, despite the commitment to “address” it in 35 countries.
• There is no indicator on the number of IDA countries that implement the gender strategy.

In addition, the “catch all” IDA21 results indicator measuring the beneficiaries of actions to advance gender equality is too vague to show whether or where IDA interventions are having an impact. At the end of the replenishment cycle, it will not answer the question: how has IDA helped to support better outcomes for women?

The World Bank has countered charges that the gender indicators are insufficiently robust by highlighting a new commitment to disaggregate 15 indicators that track beneficiaries by sex (e.g., financial and digital services, education, safety nets, health, jobs, and water). Tracking beneficiaries by sex represents a new dimension to the results framework and has the potential to offer valuable insights over time. But the underlying methodology is problematic: the Bank intends to use the proportion of the female population in each country to estimate and report on female beneficiaries when sex-disaggregated project-level results data are not available, a method that relies on the flawed assumption that the project benefits the same percentage of women as their share in the total population. The Bank will track where it is using estimates versus enumeration (i.e., counting the affected population directly), but this will not be made public. Note, as well, that there are no associated gender targets; these are only reporting requirements.

But the gender strategy!

The World Bank is highlighting the IDA21 commitment to implement the World Bank gender strategy in all IDA countries as a major win. But is it?

In practice, IDA will consider the strategy implemented if a country program does one of the following: 1) includes gender equality outcomes in a new Country Partnership Framework; 2) supports specific institutional and policy reforms toward gender equality; or 3) finances at least one operation that primarily focuses on gender equality, aims to reach gender equality outcomes at scale, or includes a gender transformative intervention.

The goal is to go beyond gender mainstreaming by getting at root causes of gender inequality (e.g., social norms), pushing for significant institutional changes (e.g., removal of legal barriers), and achieving results at scale through the use of ambitious targets (numbers or percent of population). The flexibility in execution is intended to meet countries where they are, but the fact that only one gender-focused operation constitutes implementation of a strategy is a low bar.

Moreover, pressure to implement the strategy in all IDA countries during IDA21 risks incentivizing staff to lower implementation standards, especially in countries that do not prioritize gender equality. A better outcome would have been to adopt a higher implementation standard over a longer period or to adopt a tiered approach: partial implementation, significant implementation, and full implementation. Frankly, more robust implementation of the gender strategy among IDA countries that prioritize gender equality would be a better outcome than trying to secure meaningful commitments from governments that do not.

Finally, the World Bank has noted that the gender strategy’s results framework includes supplemental data to compliment the IDA scorecard indicators: progress in ending all forms of GBV, improved SRH, and advancing women’s participation in decision making. However, these will be reported as part of the gender strategy updates, not as part of the IDA scorecard. The result is that gender monitoring and reporting will be dispersed within the World Bank.

But wait, there’s more

At the corporate level, management has pointed to three new corporate targets as further evidence of a robust institutional commitment to gender (see Table 3). Unfortunately, the corporate indicators lack IDA-specific targets: IBRD, IFC, and MIGA efforts also count towards these corporate goals, and in one case (social protection), the efforts of other development partners count too. No arm of the World Bank is accountable for a specific contribution.

Table 3. New corporate gender targets

What’s the alternative?

It would take little effort to make the IDA results framework for gender more robust and coherent, as key data continues to be collected by the World Bank. Here’s one option:

Table 4. A better results framework for gender

Conclusion

It is reasonable to argue that the IDA results measurement framework needed reform when donors negotiated IDA21. By IDA20, the framework lacked cohesion and included more than 1,000 programmatic funding targets, only a few of which had corresponding indicators showing any connection between inputs (financing) and outcomes. But the insistence on a strict alignment with the World Bank scorecard has swung IDA21 too far in the other direction, leading to the jettisoning of indicators especially relevant in IDA countries.

Here is what stakeholders should do to improve IDA’s results measurement when they next have that opportunity in 2027:

• Call for strengthened implementation standards for the World Bank’s gender strategy and a longer timeline for execution.
• Include an IDA results indicator(s) to track progress towards the gender strategy’s implementation.
• Ask that the World Bank indicate whether it is using project or proxy data in its estimates of beneficiaries disaggregated by sex.
• Incorporate SRH, GBV, and leadership indicators from the gender strategy into the IDA scorecard.
• Add nutrition and clean cooking indicators disaggregated by sex.
• Request that WBG corporate gender targets include IDA-specific targets.

In managing for results, focus and alignment are worthy goals, but not to the exclusion of all else. In today’s environment of declining ODA resources and intense scrutiny over donor funding, stakeholders need a compelling evidence base to make a strong case for IDA based on results and impact. Unfortunately, the decision to condense and reduce IDA’s results indicators could make it harder for some stakeholders to advocate effectively, especially to those who value IDA’s contribution to human capital development and gender equality.