Digital Public Infrastructure (DPI) Map Launch

By Krisstina Rao and Michael Chapman

The G20 Digital Economy Working Group and New Delhi declaration have drawn significant attention to the concept of Digital Public Infrastructure (DPI) and made it a critical topic of discussion in international fora. Here at University College London’s Institute for Innovation and Public Purpose (IIPP), David Eaves and Jordan Sandman defined DPI as a set of digital capabilities that are essential for participation in modern society. Prominent examples of DPI include the Central Bank of Brazil’s Pix (instant payments), India’s Aadhaar (digital ID), and the Estonia-led X-Road (data exchange layer). Others, such as the UNDP, have adopted similar or even broader definitions, often grounded in Brett Frischmann’s view of infrastructure as a “shared means to many ends.”

Despite the growing number of conversations about DPI, there exists an incomplete picture of the global state of DPI. Our team at IIPP noticed a shortage of information surrounding the number and types of DPI projects worldwide. While Pix, Aadhaar, and X-Road serve as the DPI poster children, the researchers suspected DPI was less of a new phenomenon and more of an underreported story.

To assess this hypothesis and fill a gap in public technology scholarship, our research team — supported by a grant from Co-Develop — set out to create the world’s first comprehensive view of DPI deployments. Over the past year, we ran a series of workshops with technologists, public servants, multilateral agencies and civil society organizations to identify how DPI deployments can be studied — what are the common qualities across deployments? What makes them infrastructure pieces that have public-interest at their core? We identified 6 attributes that could describe how safe, inclusive and effective DPI should be designed. This was supplemented with desk research and interviews to study numerous deployments according to these indicators. In the end, we established 68 indicators that articulate how DPI’s attributes may be realised through the design of its technology, governance and adoption.

Last week, this work culminated in the launch of the first version of the ‘DPI Map’ (accessible on dpimap.org), where we used the conceptual investigation described above to study real-world deployments and provide a preliminary picture on the ‘state of DPI’ across 132 countries focusing on the key geographies of Africa, Asia, the Caribbean and Latin America (a total sample of 132 countries: the current sample includes all of Africa, the Caribbean and Latin America, and the South and South East parts of Asia. Access the methodology of this investigation on dpimap.org/methodology.). Here’s the recording if you missed it.

Some preliminary takeaways include the following:

• There are far more ‘DPI-like’ systems than previously believed: 46 identity systems; 38 payment systems; and 46 data exchanges (within the existing sample specific to Global South; there are likely many more globally).
• Many countries claim to have a digital ID, but there exists little consensus on what the features of a digital ID are, and even less around its governance.
• Payment systems are easily the most ubiquitous DPI-like systems. Relatively few are ‘public,’ but when looking only at instant payment systems, over half are managed by public entities.
• Data exchanges are far more common than previously recognized, but this is the DPI category with the blurriest edges. It is remarkably global, with projects proliferating across every continent.

We will publish two additional pieces shortly. One will dive into more detail on our methodology, which could influence definitional and regulatory conversations around DPI. The second will explore in-detail insights we have gleaned from this data, which could influence conversations around DPI’s narrative and potential funding strategies.

One major takeaway from this project is that there is a wonderful community of people thinking both constructively and critically about DPI, ranging from policymakers to private sector stakeholders and NGOs.

To support that community, our team hopes this freely available resource will prove useful for a wide range of stakeholders, including researchers, public servants, technical advisors, and advocates. This resource can help the DPI community review the types of projects that have emerged in various contexts, find cross-jurisdictional allies or case studies for additional research, spot gaps where aid funding could be targeted, or build dashboards and visualizations around issues of concerns to civil society organizations.

Our team went to great lengths to approach DPI measurement in a quantifiable and objective manner. However, we note certain limitations in our data. For example, there is a degree of subjectivity on how to define DPI and which qualities to consider — like what makes a system ‘public’ or ‘digital’.

For this reason, we attempted to focus on clear and objective indicators so users can adjust to their own judgments and weights in determining what counts as DPI. However, in many cases, there was a lack of clear, reliable data on whether the indicator was fulfilled. We therefore invite everyone to review our methodology and would welcome any feedback. It is critical to us, and the broader community, that this work be rigorous.

As next steps, we plan to run virtual workshops over the next several months to show how the DPI dataset can be used to investigate the state of DPI further. The Map will also expand into a global one– including data on deployments across North America, Europe, Oceania and the rest of Asia — by the end of 2024. You can expect to read more about the methodology of the DPI Map in our upcoming white paper on the topic, where we formalize our process and findings and provide a basis for future work.

Finally, our constant effort will be to make this data as comprehensive and updated as possible. If you have information to contribute about a deployment, or want to provide feedback on our methodology, we have tried to make it as seamless as possible to engage with us via our feedback/contribution page. We are especially interested in projects or initiatives we missed, datasets we should consider incorporating, and ideas to make the data more useful or accessible.

Please contact us if you have questions or feedback, and subscribe to our newsletter to keep up with the latest research on DPI and related topics.

With excitement,
The DPI Mapping Team