Development Informatics

Editorial commentary on ICT for development research. Not affiliated with the former IDIA conference series.

Universal Service Funds in the DPI Era: Repurposing an Old Financing Tool for a New Connectivity Mandate

Abstract visualization of financial flows connecting a network of infrastructure nodes

Universal service funds are among the oldest financing instruments in telecommunications policy: a levy on operator revenue, pooled by government or an independent regulator, and earmarked to extend service into areas or to populations the market alone will not reach. The mechanism predates mobile broadband by decades, and as digital public infrastructure — digital identity, payment rails, data-exchange platforms — has become the dominant framework for donor and multilateral investment in digital development, a live policy question is whether these existing funds can be redirected toward DPI rather than replaced by new financing mechanisms built for the purpose.

The evidence on how universal service funds have actually performed at their original mandate is the necessary starting point for that question, and it is not encouraging.

Coverage is broad; disbursement is not

By GSMA’s count, 51 of 54 African countries have introduced or are introducing a universal service fund, and Alliance for Affordable Internet (A4AI) research finds 37 of those funds formally established, with 62% classified as active. The mechanism’s institutional reach, in other words, is close to universal across the continent, and comparable adoption patterns hold across much of Latin America and South and Southeast Asia.

Reach has not translated into spending. A4AI’s research on African funds, conducted with the World Wide Web Foundation and UN Women, found that in 2016 African universal service funds disbursed just 54% of the revenue they collected, leaving an estimated $408 million sitting unspent across the continent’s 37 funds — enough, by the report’s own calculation, to have brought an estimated 6 million additional women online. A separate A4AI study of Latin America and the Caribbean found unused universal service and access fund balances of roughly $7 billion at present value, heavily concentrated in Brazil’s long-running fund. GSMA’s own survey work across 40 African countries, alongside parallel ITU analysis, describes disbursement rates as persistently low worldwide, with some funds unable to distribute any of the levies they had collected.

Why the funds sit idle

The research on this pattern converges on a small number of recurring causes rather than a single explanation. Political interference features prominently: government influence over fund staffing, project selection, and disbursement timing is documented across multiple country studies, and in some cases funds have been used to finance projects unrelated to telecommunications entirely, effectively diverting the levy to general government revenue. Weak institutional capacity within the fund administrator is a second recurring factor — several country studies find fund managers lacking the technical or procurement capacity to design and let contracts for the rural-infrastructure projects the fund is meant to finance. Transparency is a third, related problem: only 23 countries were found to publish operational detail on their fund’s activities, which limits both external accountability and researchers’ ability to compare fund performance across countries with any confidence.

These are not problems specific to broadband-era funds; they are the same governance failures documented in universal service fund literature going back to the funds’ original fixed-line and rural-telephony mandates. That continuity matters directly for the DPI question: a financing mechanism that has struggled for decades to disburse against a comparatively simple mandate — subsidize rural network buildout — is being asked to take on a more institutionally demanding one, since DPI projects typically require sustained multi-agency coordination (identity, finance, and communications ministries working jointly) rather than a single sector regulator managing a construction subsidy.

Where repurposing is actually happening

Repurposing is not merely theoretical. India’s Universal Service Obligation Fund has moved well beyond basic rural telephony subsidy toward broader digital-inclusion infrastructure, reporting by mid-2024 over 213,000 gram panchayats connected, nearly 690,000 kilometers of optical fiber laid, more than a million fiber-to-the-home connections commissioned, and over 100,000 public Wi-Fi hotspots installed under its ongoing schemes. Nigeria’s Universal Service Provision Fund operates alongside newer financing, including a $500 million World Bank facility approved in 2025 for broadband expansion in underserved areas, in a policy environment where the country’s communications ministry has explicitly framed DPI — broadband networks together with identity systems and payment rails — as a unified national goal rather than separate funding lines.

These examples show redirection of institutional mandate and, in India’s case, substantial delivery at scale. They do not resolve the disbursement-and-governance problem the broader research base documents; they demonstrate that a well-resourced, well-governed fund can expand its mandate successfully, which is a different finding from evidence that the median universal service fund can do the same.

What this means for evaluating USF-to-DPI proposals

For donor program officers and researchers assessing a proposal that routes DPI financing through an existing universal service fund, the disbursement-rate literature suggests specific due-diligence questions rather than a blanket recommendation for or against the approach. Ask for the fund’s historical disbursement rate over multiple years, not a single favorable year. Ask whether the fund publishes operational and financial detail, since the transparency gap the A4AI research documents is itself a leading indicator of governance quality. Ask whether the fund’s institutional capacity has been assessed against the DPI mandate specifically, given that identity- and payments-layer projects require coordination the fund’s founding rural-telephony mandate did not.

A4AI’s recommendation that governments modernize the fund mandate — including targeting a defined share of spending at underserved groups such as women, and building a governing coalition beyond the telecommunications regulator alone — is a reasonable response to the governance evidence, but it is a reform agenda, not a description of current practice. Treating a universal service fund as a ready-made DPI financing channel without evaluating its disbursement record first repeats the same evaluation error the mobile money and telecentre literatures have already documented: mistaking the existence of an institution for evidence that the institution delivers.