How can digital services be used to create open, fair and democratic governments? Richard Pope, a new Digital State affiliated researcher at the Bennett Institute, shares his insights.
It is a characteristic of a democratic system that people have clear opportunities to both understand and influence how that system works – democracy, it has been said, is government by explanation.
When it comes to our built environment, the planning process mandates the publication of notices that explain potential changes to people’s neighbourhoods, and public meetings let people have their say, understand and influence decisions. People have clear opportunities for recourse.
Erskine May, the book of parliamentary procedure (finally made available for free in January), along with Hansard and the UK’s corpus of legislation, expose the workings of parliament and the laws it passes (even if those workings are sometimes esoteric). Public registers, like those maintained by the Land Registry and Companies House put important information and events on the public record. Those facts can be checked and verified by the public. Finally, there are organisations like The National Audit Office, along with regulators and parliamentary committees, that report on the health of public services and hold them to account.
These principles – notifying people of important changes that might affect them, opening up the rules, giving clear routes to recourse, publishing important facts and auditing – are ones that we can see repeated across different aspects of public life. Together, they help make government legible.
What’s different about digital services?
More and more government services are being delivered digitally, with the Institute for Government putting the number at 800 at the last count. At the same time, there are concerns being raised by academics and campaigners about the fairness and transparency of some of them, particularly the use of automated decisions in areas of welfare and immigration.
As more law is transcribed to software code in this way, or expressed through the choices of designers in websites and apps, how should we apply the principles listed above to digital services? How can digital services be designed in such a way that they invite understanding?
Digital services have some characteristics that make this challenging: updates to the code can change how the service works over time; automated decisions can be inherently opaque; small changes to the design of interactions and words used can have big impacts on how people use a service; and data collected for one purpose can be used for another.
Open rules, open code
One way of doing this is by publishing the code for a service in the open so that researchers and regulators can understand how it works.
Since 2013, the UK government’s Service Standard has mandated that all new code is made open under an appropriate licence. In practice, it is not clear how often this happens so, while you can find the code for parliament’s petitions website on GitHub, big services like Universal Credit are often not open. The government identity platform MOSIP, which is being developed by international funders and aimed at helping governments around the world implement more digital services, is being developed in the open, in part for this reason. The explicit aim is to “ensure transparency” of the system.
Publishing code in the open should become the default for all government services.
A register of digital services
A complementary approach would be to publish a list of the current digital services operated by government. This would make it a matter of public fact which services are in use and when they change, and could include a link to the source code.
Such lists are beginning to be produced, albeit for slightly different reasons. The Home Office maintain a list of their digital and non-digital services as the public experience them, and many local government digital teams publish information about new services they are developing on the Pipeline website.
Central government should maintain a central public register of digital services. It should include information about the current versions of software deployed, along with an explanation of any changes made over time.
Explaining how data has been used
Putting users in control of the data held about them and telling them how that data has been used is an approach that several countries have taken to increase trust and transparency. Users of India’s digital identity Aadhaar can view the history of how their credentials have been used online. While in Estonia, citizens can see when public officials have accessed data about them.
UK government should follow this approach and provide ways for people to understand how data about them is being used. Alongside this, there should be clear routes to report misuse of data.
Expecting individuals, alone, to audit how data about them is obviously not sufficient. Regulators and auditors should gain new powers to review both code and data use.
Transparency at the point of use
Finally, there is a huge opportunity to design openness and transparency directly into the services people use. Digital services could easily contain links to information about how the service is performing, who the minister responsible is, and upcoming changes and the underlying law. Given the UK now has a single central website, this could be done systematically across all the services linked to from GOV.UK.
Government by explanation
It is important to note that none of this is a substitute for the design of policy and services in an empathetic and fair manner. However, government services will never work for all the people all of the time. And in those (hopefully rare) situations being able to understand why is critical.
Digital services are now a core part of how government tries to meet the needs of the public. As such, they should no longer just be seen through the lens of making things cheaper or easier. Creating opportunities for public understanding and accountability should become a core design principle of the digital government movement.
The Digital State project, led by Dr Tanya Filer, sets out both to lead policy research and provide a forum for broad-ranging discussion with academics and policymakers on the opportunities and challenges that digital technologies pose to policymaking, governance and democracy.
The views and opinions expressed in this post are those of the author(s) and not necessarily those of the Bennett Institute for Public Policy.
