In many low- and middle-income countries (LMICs), strengthening state capacity and expanding the tax base are fundamental to improving public sector functions. A state’s ability to govern effectively often hinges on making society “legible”—systematically recording and utilising information about its population and activities. Digital public infrastructure (DPI), with digital ID systems at its core – together with digital payment and data exchange -, offers a significant opportunity to improve State capacity and boost tax administration. Digital IDs can facilitate the identification of taxpayers, streamline registration, and enhance compliance. Yet, while the potential of digital IDs is clear, evidence on their impact remains limited and mixed.
Case studies: Ghana, Uganda, and Ethiopia
Several LMICs are integrating digital ID systems into their tax administrations with varying degrees of success. In Ghana, the Revenue Authority (GRA) partnered with the National Identification Authority to replace its old taxpayer identification number (TIN) system with the national ID PIN, known as the Ghana Card. This change, introduced in April 2021, gave GRA access to the ID database and enabled automatic enrolment of individuals turning 18 with income-generating potential. The same strategy was adopted by India, where, as of July 2017, PAN cards for tax purposes got linked to the ID number from the Adhaar system.
In Uganda, in 2022, the tax administration has instead developed an Instant TIN interface that connects to the National Identification and Registration Authority (NIRA). When an individual applies for a TIN online, the system retrieves data from NIRA to pre-fill and authenticate the registration. This reduces errors and speeds up onboarding. Unlike Ghana, however, Uganda’s approach remains demand-driven, initiated by the individual rather than mass auto-enrolment.
In Ethiopia, the Ministry of Revenue integrated tax registration with the national digital ID system, Fayda, in December 2023. The system links each TIN to a corresponding Fayda number. Previously, biometric data was used for TIN registration, but the high costs of maintaining this system prompted a shift to the more scalable Fayda system, partially financed by external donors.
What did we learn so far?
- Massive Expansion of the Tax Base: The integration of digital ID systems has significantly increased the number of taxpayers. In Uganda, 35% of all new registrations in 2022 were ID-based, while in Ghana, the number of new registrations tripled from April 2021 to March 2022. Digital IDs are an effective tool for identifying taxpayers who would otherwise remain outside the formal tax system. The success of these reforms lies in the capacity of the ID system to seamlessly link individuals to their tax obligations, simplifying the registration process for new entrants.
- Inclusivity and Accessibility: Digital ID systems have been shown to bring in different categories of taxpayers compared to traditional, manual registration processes. In both Ghana and Uganda, those registering through digital IDs were more likely to be younger, female, and previously working in the informal sector. This suggests that streamlined digital registration systems reduce barriers to entry, particularly for groups that face difficulties navigating more traditional tax processes.
- Data Quality – A Mixed Picture: Digital ID systems improve the accuracy of taxpayer data by eliminating duplicates by design and enhancing contact information. However, sector-level information, crucial for targeted tax enforcement, is often not available in national ID databases. This means that while ID-based systems improve general tax data quality, they fall short of providing the detailed sectoral data needed for tax administration. Also, in both countries, significant staff resources have been used to clean the new ID data, match it with existing records, and onboard the new entrants.
- Tax Revenue Impact – Still Unclear: Research on tax impact is currently limited to Ghana, analysis in Uganda is still ongoing. In Ghana, while taxpayers migrating from TINs to PINs showed improved compliance, new entrants exhibited mixed or worsened filing and payment behaviour, at least in the short term. This suggests that while more people are being brought into the system, effective tax administration still requires monitoring, enforcement, and support. The rapid expansion of the tax base can overwhelm tax authorities’ capacity to manage new registrations, leading to a situation where many new taxpayers remain inactive or underreport their income. Similar research from South Africa shows that while new business registrations increased, revenue impact was limited due to persistent non-compliance among new entrants.
Next steps for successful ID Integration
While Ghana and Uganda provide valuable insights into the potential of digital ID systems for tax administration, much more evidence is needed to fully assess their impact, especially in Ethiopia, where no research on ID-based tax registration has been conducted so far.
It is also interesting to notice that impacts are quite similar regardless of whether TINs are still in place but coupled with an ID number, or whether they have been fully replaced by IDs. Similar patterns across different contexts, like Uganda and Ghana, suggest that the key to success lies not much in the technology itself, but in other crucial factors needed to unlock the potential of that technology – such as the capacity to manage a growing taxpayer base. Tax administrations must ensure they have the resources and infrastructure to monitor, enforce compliance, and engage with taxpayers effectively. Without this capacity, expanding the tax base can result in a bloated and inefficient system with limited revenue impact. Likewise, technology should not be used to reach short-term policy goals of mass registrations disproportionately capturing small businesses, but rather to focus on hard-to-reach high income taxpayers already in the tax net. Beyond tax administration, the success of ID integration ultimately relies on the universal adoption of IDs, with LMICs still struggling to achieve full coverage.
As the DPI approach evolves, our research on IDs for tax contributes to the international debate by providing empirical lessons that can be applied to other government institutions as well.
The authors would like to acknowledge Celeste Scarpini for her insightful contributions to the writing of this blog.
