What is digital public infrastructure?

Think about the road outside your local market. Nobody built that road to sell tomatoes. It was built so anyone could reach the market, in a bus, on a bike or on foot. The same road carries the butcher, the tailor and the school run.

Now think about the digital version of that road. A country needs a way to prove who a person is. It needs a way to move money, and it needs a way to pass a record from one office to another. Almost every service needs all three.

Digital public infrastructure is that shared road. It is built once, under rules anyone can read, and then used by many services.

What the term means

Digital public infrastructure means the shared digital systems a country builds so that other services can sit on top, and people shorten it to DPI.

The word infrastructure is doing real work here. This is because infrastructure is a thing you use without owning it. You do not own the road or the power line. However, you use them, under public rules, and so does your rival.

Note the word public as well. It does not always mean owned by the state. It means governed for public ends, with open rules and open standards. It can be run by a state agency, a company under licence, or a non-profit body.

The three layers

Most countries build the same three pieces.

• Identity. A way to prove that you are you, online and in person.
• Payments. A way to move money between any two accounts, fast and cheaply.
• Data sharing. A way to pass one record from office to office, with your consent.

The third piece is the least visible and often the most useful. This is because without it you carry paper from desk to desk. With it, the tax office can ask the land office for one fact, and only that fact, once you agree to the request.

How it works, step by step

Take one everyday job. You want to open an account with a small lender.

1. You prove who you are through the national system, on your phone or with a fingerprint.
2. The lender gets back a yes or a no, and keeps no copy of your documents.
3. You agree to share one record, such as proof of income.
4. The record moves from its source to the lender, signed, so both sides can trust it.
5. The account opens, and money moves in over the shared payment system.

What changed here? The lender built none of it, and that is the whole point. Build once, then use it many times.

What it is good at

The gain is speed and cost.

When identity and payment are shared, a new service can start with a small team. A queue that took a day can take a minute. Money can reach a household directly during a flood or a drought, which is why many governments rushed to build these systems during the pandemic.

Competition improves too, because open rails let a small firm plug in beside a large bank. That is the same argument people make for a shared road, and it is the strongest case for DPI.

Mobile money shows the pattern at street level. Read our explainer on mobile money to see how the payment layer feels from the customer side.

What can go wrong

Three failures come up again and again.

The first is exclusion. If one system becomes the only door to food aid or a school place, then anyone the system cannot read is shut out. Fingerprints fade with manual work and with age, and a person without a birth record cannot enrol. Therefore every design needs a fallback on paper, staffed by a human being.

The second is surveillance, because one number used across every service makes it easy to join up records and watch a life. The technical fix is to share the least data possible and to log every request. However, the real fix is law, plus courts willing to enforce it.

The third is the single point of failure. When the shared system is down, everything above it stops. Therefore a country with little cash and one payment system has a very bad day when a cable is cut.

Who owns it

This is the question that settles all the others.

Ownership can sit with a state agency, with a licensed company, or with a non-profit body set up for the job. Each of those has worked somewhere. What matters is whether the rules are public, whether anyone can join on the same terms, and whether an independent body can say no.

Kenya shows both sides of this. Payments there grew out of a private mobile money service, and the country got the benefit early. On the identity side, courts halted an earlier national digital ID scheme until the government carried out a proper data protection assessment. That delay was the system working, not the system failing.

Where you already meet it

You meet DPI more often than you notice.

India runs a national digital ID and an instant payment system that now handles billions of transfers every month. Brazil built an instant payment system that reached most adults within a few years. Estonia has run a data exchange between its agencies for about two decades, which is why people there can do almost every government task online.

Some countries no longer build the identity software alone, and they take shared open-source software and adapt it. That cuts the cost, and it avoids being tied to one supplier for a decade.

What to check in your own country

Ask four questions about any national digital system.

1. Is signing up truly voluntary, and is there a route for people the system cannot read?
2. Is there a data protection law, and a regulator with the power to act on it?
3. Can a small company join the payment rails on the same terms as a big bank?
4. Is the source code, or at least the standard, open for anyone to inspect?

Then do one small thing this week. Find out which office holds your national identity record, and how you would correct a mistake in it. If that takes you more than ten minutes to discover, you have learned something about the system already.

Just Out Tech explains new research in plain language. This article was drafted with AI assistance and checked by a human against the original source.