Assessing infrastructure readiness for scaling digital cash transfers toward ending poverty

Summary

Noting the significant major recent disruptions in global development policies and practice, unconditional digital cash transfers (DCTs) have received growing attention over recent years as a new technology and decisive intervention in reducing extreme poverty. In 2023, under the auspices of the 17 Rooms initiative, a working group drafted an initial plan for a potential global fund to scale up DCTs for this purpose. To help inform deliberations linked to the proposed fund, this piece takes advantage of a new data source to present an initial benchmarking of digital infrastructure readiness for DCT scale-up in 72 countries still grappling with substantial extreme poverty. The analysis offers grounds for cautious optimism that the country-level infrastructure and policy preconditions are increasingly viable for rapidly scaling-up DCTs toward a goal of ending extreme poverty.¹

Introduction

Amid a period of substantial disruption in global development policy and practice, it is important to consider the evolving evidence base around new technologies that could make a decisive difference in future development cooperation efforts. In this regard, a substantial body of literature has demonstrated the efficacy of unconditional cash transfers, including digital cash transfers (DCTs), as a new technological option in dramatically reducing extreme poverty (e.g., Chowdhury et al. 2022; Banerjee et al., 2023). For instance, a recent meta-analysis of 72 randomized evaluations of unconditional cash transfer programs in 34 low- and middle-income countries suggests a positive and persistent impact in alleviating multiple dimensions of poverty (Crosta et al., 2024, 2025). This rigorous empirical work complements practical delivery innovations in low-income settings, such as the pioneering success of Togo’s Novissi program, which rapidly delivered contactless mobile money during the COVID-19 pandemic.

This evidence is highly relevant in the context of prominent international pledges to end extreme poverty by 2030. For example, in 2013 the World Bank adopted the objective as a headline organizational goal for 2030. Two years later, all 193 UN member states affirmed the same ambition as Target 1.1 of the Sustainable Development Goals. Nonetheless, in the 2020s, the world’s progress toward this ambition has stalled, with around 600 million people still in extreme poverty.

To help reinvigorate global progress on extreme poverty, a proposal has recently been put forward to create a new purpose-driven international fund to support the dramatic expansion of DCTs (Kharas and McArthur 2023, Lawson and Stewart 2024). Here we dub this proposal the Digital Fund for Ending Extreme Poverty (hereafter D-FEEP, or Fund). A first draft design of the Fund suggested that operations at a scale of $6-7 billion per year for ten years—much less than 5% of global aid budgets, even after recent reductions—could, if matched by domestic resources, potentially be enough to eradicate extreme poverty (i.e., below a critical threshold) in countries with a poverty gap of over 1% of GDP (Lawson, Stewart et al., 2023).

The role of digital public infrastructure for digital cash transfers

In considering the D-FEEP’s proposed design and implementation systems, digital infrastructure issues are of first-order importance. Successful scale-up of DCTs requires efficient mechanisms both for targeting beneficiaries and ensuring accurate financial transfers at low cost. At the country level, this implies at least four preconditions:

• A digital identification (ID) system
• A digital payment system
• A mechanism for identifying beneficiaries
• Strategies for enrolling beneficiaries in relevant programs²

Some of these ingredients can be more challenging than others to implement, depending on the local situation. In Nigeria, for example, there is no unified legal or operational mandate compelling states to participate in or update the National Social Registry, leading to patchy coverage and inconsistent data quality for DCT scale up across the country (World Bank 2019).

In this context, “digital public infrastructure” (DPI) has emerged as a specific concept and approach to digital infrastructure that prioritizes a publicly managed or publicly governed country-level backbone for digital identification (ID), payments, and data exchange, upon which both public and private actors can deliver services (Eaves, Mazzucato, and Vasconcellos, 2024; Eaves and Sandman, 2024; Ingram, McArthur, and Vora 2022). Countries like India and Estonia are often held up as leading exemplars of DPI, although many more countries are advancing key dimensions of DPI (Kant and Mishra, 2023; Leosk, 2022; Kotschwar and Colter 2024).

The common backbone nature of DPI offers a contrast to countries where, for example, private entities offer their own fully independent competing infrastructures, or where each government ministry builds or uses its own independent digital infrastructure systems for delivering services. In this spirit, digital infrastructure for public services will often not meet the common backbone standard of DPI.

How might the proposed D-FEEP approach issues of digital infrastructure in the context of a growing global movement to promote DPI (United Nations, 2025)? There are a range of embedded issues. For example, should the Fund be restricted to supporting DCTs only in countries with DPI already in place? Should it also help countries build DPI to deliver DCTs? Would the prospect of external support for DCTs incentivize countries to build DPI faster? Does it matter if a country has adequate digital infrastructure that doesn’t qualify as DPI?

Assessing infrastructure readiness for digital cash transfers

Until now, a lack of systematic cross-country information on digital infrastructure readiness has made it difficult to assess practical options for global scale-up of DCTs to reach people in extreme poverty. To this end, we here present a first estimate of digital infrastructure status for 72 countries still grappling with substantial extreme poverty, generally defined as at least 3% of the population living on less than $2.15 per day using 2017 purchasing power parity prices.³

We assess digital infrastructure by taking advantage of the March 2025 edition of the global DPI Map, a new and evolving data set published by David Eaves, Krisstina Rao, and colleagues at University College London.⁴ The core of our analysis draws from DPI Map’s assessments of country-level digital ID and digital payment systems and cross-references these with poverty estimates drawn mainly from the World Bank’s online World Development Indicators (WDI).⁵

Due to the nature of the data embedded in the DPI Map, we note that our stock-taking exercise is skewed toward diagnosing DPI forms of digital infrastructure, rather than digital infrastructure more broadly. More nuanced assessments regarding the nature, quality, and extent of each country’s digital infrastructure would be valuable but are beyond the scope of this analysis. Nonetheless, we assume DPI to be the highest standard of relevant digital infrastructure. So at least in theory, if a country has national scale, DPI-caliber digital ID and digital payments infrastructure then the core digital conditions should be in place to deliver DCTs at national scale.

In this context, we segment our sample of countries into three overarching categories of digital infrastructure readiness, as also shown in Figure 1:

• DPI in motion. These countries are deemed to have both a DPI-caliber ID system and real-time payment system, either operating at national scale or in the process of national scale up.
• DPI in pilot or planning. These countries have at least partially developed DPI-caliber digital ID or digital payment systems, including in pilot or planning phases.
• Digital infrastructure to be built. These are countries with no (or unknown) evidence of digital ID or digital payment systems.

Figure 1. Three categories describing status of digital infrastructure implementation

We further segment countries by identifying which ones are eligible for concessional financing through the World Bank’s International Development Association (IDA). Of the 72 countries in the sample, 59 were IDA eligible as of June 2025. These 59 countries are also home to a large majority of the world’s extremely poor people (roughly 550 million people) and represent an indicative group that could be initially eligible for potential D-FEEP support.⁶ The other 13 non-IDA countries have a total of roughly 70 million people in extreme poverty and could potentially be eligible for technical support from the Fund, but not likely for direct funding. Scale-up of DCTs in these non-IDA countries could likely be financed domestically.

As a final dimension of the analysis, we include information on which countries have recent experience in implementing some form of unconditional cash transfer (UCT) programs. Unfortunately, data are not available regarding which of these were delivered in digital form. We assess the presence of recent UCT policies by drawing from the World Bank ASPIRE database and Ugo Gentilini’s review of cash transfer expansion associated with the COVID-19 pandemic (2022: 85-93). Within the ASPIRE data, we designate countries with evidence of unconditional transfers reaching at least 0.5% of the population as having relevant experience. Within the Gentilini data, we make a similar designation for countries that had either universal coverage policies during the pandemic or where we were able to identify substantial population coverage of unconditional transfers. As a caveat, this designation of recent UCT experience is indicative, since it might exclude some countries with relevant policy experience.

Core results on DCT readiness

Figure 2 shows a key result of our analysis by allocating the 59 IDA-eligible countries in our sample to each of the three DCT readiness categories. This finds that:

• Nine countries, with a total of roughly 205 million people in extreme poverty, have DPI in motion: Bangladesh, Ethiopia, Ghana, Nigeria, Pakistan, Rwanda, Uganda, Tanzania, and Zambia.
• Thirty-eight countries, with a total of roughly 204 million people in extreme poverty, have DPI in pilot or planning phases. This includes Kenya, Madagascar, Malawi, Mozambique, Niger, Sudan, Togo, and 31 other countries.⁷ Togo’s inclusion in this category is notable, in light of the much-lauded Novissi program’s success in rapidly scaling up DCTs.
• For twelve countries, with a total of roughly 140 million people in extreme poverty, digital infrastructure is yet to be built or not enough is known to comment on their digital infrastructure readiness for DCTs: Central African Republic, Chad, DR Congo, Eritrea, Guinea-Bissau, Lao PDR, Mali, Nicaragua, Senegal, South Sudan, Tuvalu, and Yemen.

Figure 2 further shows that at least 39 of the 59 IDA-eligible have recent experience with unconditional cash transfers (denoted by “UCT” in the chart), including all the countries with DPI in motion and a majority of those with DPI in pilot or planning phases. Although we do not have information on which transfer programs were delivered through digital means, the chart does suggest that both infrastructure and policy preconditions for DCT scale-up are increasingly present for countries that are home to roughly two thirds of the world’s extremely poor people.

Figure 2. Mapping DPI status for digital cash transfers across 59 IDA-eligible countries

Practical implications and considerations

The diagnostics above offer broad empirical contours to help inform policy debates regarding the proposed D-FEEP and which countries could potentially scale up DCTs to make rapid and decisive reductions in extreme poverty. At first glance, one takeaway might be to suggest that international efforts should first prioritize DCT expansion in IDA-eligible countries within the “DPI in motion” category, since these countries already show signs of strong national digital infrastructure. However, there are at least three limitations to this view.

One issue is highlighted by the case of Togo, a prominent existing DCT success story, because it falls under the category of “DPI in pilot or planning.” If Togo’s Novissi program represents a leading role model in deploying innovative technologies to reach beneficiaries quickly, then achieving the full DPI standard of infrastructure might not be a precondition for DCT scale-up success. Many other countries that fall under the pilot-planning category might be ready for similarly rapid DCT results.

A second and perhaps more salient issue is that the cross-country infrastructure assessments presented above do not include information on how many people in extreme poverty within each country still lack access to digital IDs or payment systems. A more complete country-by-country assessment needs to assess each country’s coverage gaps.

In Nigeria, for example, a country with roughly 70 million people living in extreme poverty, the National Social Safety Nets Project—an $800 million DCT program supported by the World Bank Group—has reportedly reached 5 million households and is aiming to reach 10 million households. Even before attempting to reach at least another 60 million Nigerians likely in need, the program has reportedly encountered difficulty reaching the poorest potential recipients due to lack of penetration of national digital ID efforts and data gaps on coverage of banked/unbanked people, which are preconditions for verified receipt of transfers. An implication is that many countries with DPI in motion might still need to prioritize nationwide access to digital infrastructure as part of a critical path to DCT expansion.

A third consideration lies in political economy. Unconditional cash transfers require a minimum amount of public or political support. A society’s implicit social contract needs to be conducive to giving unconditional grants to extremely poor people. Notwithstanding emergency situations like the onset of COVID-19, this type of support might not always be present. A D-FEEP would therefore need to have a foundation-type model whereby only interested countries apply for funding. Nonetheless, the large number of countries with at least some policy experience with unconditional cash transfers suggests widespread opportunities for DCT expansion.

Concluding thoughts

Overall, our analysis provides grounds for cautious optimism that the infrastructure conditions for DCT scale-up are both promising and rapidly improving. Evolving technologies and policy experience suggest it is increasingly feasible to reach all extremely poor people, even if concerted efforts to expand digital access are still needed within many countries. Other authors, for instance, have recently emphasized the opportunity to make digital IDs rapidly available for 850 million people who need them (Eaves and Roncaratti 2025). A new mix of evolving digital tools, frontier evidence, and goal-oriented policy efforts could empower DCTs to unlock decisive gains on the world’s otherwise-stalled progress toward ending extreme poverty.

_{Images are for reference only.Images and contents gathered automatic from google or 3rd party sources.All rights on the images and contents are with their legal original owners.}