Government as a Platform, the hard problems: part 2 — the design of services & public policy

Government as a Platform is the approach of reorganizing the work of government around a network of shared APIs and components, open-standards and canonical data registers. The hope is that this will allow public servants, businesses, and others to deliver radically better services to the public, and to do so safely, efficiently, democratically, and in a more accountable way. This article is part of a series looking at some of the difficult design, policy and technology questions posed by the Government as a Platform concept.

The affordances of Government as a Platform could change both the nature of the services built on top of it, and how those services are created. To take full advantage of platforms, governments will have to understand the new types of services likely to become possible, how it could break some old certainties about who delivers services, and how the process can remain accountable.

The efficiency trap — could we miss the opportunity for better services?

Because shared platforms solve problems once, rather than addressing them multiple times in different parts of government, it will become simpler and quicker to design public-facing services. With this infrastructure in place, designing new services largely becomes a job of joining together existing systems. With this reduced effort and duplication comes the potential to save time and money.

However, increased efficiency is not the most significant opportunity arising from Government as a Platform. The biggest opportunity is the potential for radically better public services. That’s because it does not just become simpler and quicker to design services, it also becomes simpler to design new types of service.

For example, someone wanting to start a company today may need to register the company with one government agency, register as an employer with another agency, for tax with yet another, then apply for different licences from central and local government. They may also need to get the proof they need to open a bank account or buy insurance. The same applies in other areas of people’s lives too — having a child, moving house or applying for benefits. In short, people’s lives don’t reflect the organizational structure of government.

Platforms could change this. That’s because they make it simpler to design public facing services that are orientated around the needs of citizens rather than the organizational structure of government. So, it becomes realistic to design a single service for starting a company that handles many of the above steps in one go.¹

This service-design approach is different from that of the traditional policy maker and may represent a challenge to the profession. A service design approach aims to design end-to-end services, use multiple ‘channels’ (online, phone, face-to-face, etc), and consider the design of both public-facing and back-office functions (and any processes in between).²

A risk is that this opportunity for better services gains less mindshare among public servants than the more familiar one of prioritizing projects based on cost saving and efficiency. It may also be reinforced by the effect of challenging existing power structures between government agencies as new end-to-end services likely require new institutions that sit outside existing government institutions.³

The end of ‘public vs private’ and ‘central vs local’?

Historically, policymakers have had a couple of clear choices to make when it comes to the delivery of services to the public:

1. Should this service be delivered centrally or locally?⁴
2. Should this be done by government or the private sector?

Platforms could start to turn these into false choices as it becomes possible to build multiple services (public and private, central and local) on top of shared foundations. For example, it could become possible to get a parking permit directly from a local government website or bundled as part of a commercial property rental service so it’s there when you move in. That’s because the ‘heavy lifting’ is done by APIs that encapsulate the business rules relating to issuing a parking permit, and information is exchanged to agreed open-standards.

Rather than having to choose between privatizing or outsourcing a given service, governments can follow both a wholesale and a retail approach to providing services.⁵ ⁶

Similarly, municipal, regional and national governments relying on the same shared platforms changes the nature of the local-vs-national or state-vs-federal question.

Platforms make it easier to interact with multiple layers of government at once. A given policy can be implemented by a “composite service” made up of steps from different layers of government. What would it mean if governments operated more like a network and less like a hierarchy? Or if open standards started to make the geographical organization of government less relevant?

Platforms as new zones of political contest

This comes with some clear opportunities. New classes of service made up of parts from public and private, central and local become possible. The long-tail of needs can be met by services tailored to particular users. Services can be provided by any layer of government, and by commercial or third sector organizations and it’s ok when they overlap, complement and duplicate.

However, it also comes with new risks and new questions.

How will a citizen interacting with the government as a part of a non-government service know that it can be trusted? Will they know which agencies are ultimately responsible for the different parts that make up the service? (And does it even matter?). How will governments mitigate the risk of the private sector becoming too dominant and prioritizing a subset of users at the expense of others?

Central and local government may have conflicting opinions about the design of a service. Centralized platforms may offer new opportunities to restrict the activities of local government.

Private companies may attempt to disintermediate the state entirely, and debates such as that surrounding the US tax filing industry start to appear in more domains.⁷ Privatization and nationalization could become harder to spot.

Rather than being inert bits of technology, they will create new zones where political contests play out.⁸ If the old certainties of central vs local, public vs private start to fade, working out what the rules should be will take effort and may itself be open to power plays and political contests too.

When services change in days but legislation changes in years

Modern ways of working have sped up the delivery and iteration of software.⁹ Continuous delivery and agile project management have become defining characteristics of the various ‘digital service units’ around the world.¹⁰

Platforms have the potential to shorten the development of entire services still further. With the heavy lifting done elsewhere, organizations tasked with providing public facing services can focus on solving the problems that are unique to their domain.

In the US, the components developed by 18F such as login.gov and the US Web Design System have allowed suppliers to move substantially faster.¹¹ Similarly, in the UK, shared components have saved weeks or months in integration time.¹² In Argentina, a digital replacement for physical driving licences was developed from scratch in 65 days, in part because it made use of existing notification, identity and payment platforms.¹³

This raises some important questions for the legislative process. How should politicians retain democratic oversight of public services while also empowering the public servants developing the service? Can legislation be designed in such a way that is more based on outcomes, and become less prescriptive, to allow for experimentation? How should the wider platform ecosystem be policied?

Given the new areas of political contest that open APIs, in particular, could open up, these questions may become urgent quite quickly.

A new paradigm for public-facing services

Radically new types of service could become possible (or at least much simpler) under the Government as a Platform model: services that meet the end-to-end needs of the public; services that allow people to interact with multiple parts of government; and services from the public, private and 3rd sectors that overlap and complement each other.

However, only by accepting that they are dealing with a new paradigm, not a cost-saving measure, will they take full advantage of it and do so in a way that is safe and fair.

Read part 3 — shared components, APIs and the machinery of government

Thanks to David Eaves and Ben McGuire for feedback on this blog post.

1. Tom Loosemoore, “Making government as a platform real”, Public Digital blog, 25th September 2018, http://public.digital/2018/09/25/making-government-as-a-platform-real/ This article includes screenshots of a prototype of a single service for starting a company in the UK
2. Angelica Quicksey, “Service Design for Public Policy”, Medium, 29th September 2018, https://medium.com/@angelquicksey/service-design-for-policy-b0a9408dced1
3. Bryan Glick, “GDS is ‘sidelined’ and government as a platform ‘is dead’, says Francis Maude”, Computer Weekly, 14th September 2017, https://www.computerweekly.com/news/450426316/GDS-is-being-sidelined-and-government-as-a-platform-is-dead-says-Francis-Maude Maude said attempts to build new central functions to coordinate across departments — such as GDS — were thwarted by “mandarins” and “heavily resisted by the HM Treasury”. “For much of the mandarinate this was an assault on their autonomy and empires, and what we know about empires is that they fight back — and boy, are they fighting back,” he said.
4. There are examples of services that are operated in partnership today, such as the UK public library system, where central and local government share accountability for the service, but the service delivery is still local.
5. Aneesh Chopra and Nick Sinai, “Wholesale Government: Open Data and APIs”, Medium, 9th April 2015, https://medium.com/@ShorensteinCtr/wholesale-government-open-data-and-apis-7d5502f9e2be
6. Lindsey Smith, “Booking campsites on Recreation.gov is a mess. Here’s the solution”, San Francisco Chronicle, 4th march 2019, https://www.sfchronicle.com/travel/article/Recreation-gov-is-a-mess-Here-s-the-solution-13655884.php A recent example is recreation.gov, which has a public facing website for finding and booking campsites and tours on federal land. It also has an API that is used by commercial websites to do the same thing. (I think this article is a bit unfair on the government run website)
7. Justin Elliot, “Congress Is About to Ban the Government From Offering Free Online Tax Filing. Thank TurboTax.”, ProPublica, 5th April 2019, https://www.propublica.org/article/congress-is-about-to-ban-the-government-from-offering-free-online-tax-filing-thank-turbotax
8. Christoph Raetzsch, Gabriel Pereira, Lasse S Vestergaard, Martin Brynskov, “Weaving seams with data: Conceptualizing City APIs as elements of infrastructures”, 10th February 2019, Big Data & Society, https://doi.org/10.1177/2053951719827619
9. “2018 State of DevOps Report”, https://puppet.com/resources/whitepaper/state-of-devops-report/thank-you The State of Devops 2018 report found that even relatively low performing organizations update their online services weekly or monthly, and the better performers deploy changes hour-by-hour
10. David Eaves and Ben McGuire, “Part 2: Proposing A Maturity Model for Digital Services”, Medium, 7th September 2018, https://medium.com/digitalhks/part-2-proposing-a-maturity-model-for-digital-services-9b1d429699e7
11. Interview
12. Anais Reding, Chris Pinder and Kaz Hufton, “How Government as a Platform is making things better for users” Government as a platform blog, 13 March 2018, https://governmentasaplatform.blog.gov.uk/2018/03/13/making-things-better-for-users/
13. Mike Bracken, “Argentina just made driving licences digital”, Public Digital blog, 12th February 2019, https://public.digital/2019/02/12/argentina-just-made-driving-licences-digital/