This blog first appeared on Policy Exchange’s PolicyBytes.
Of the thirty-three recommendations in Policy Exchange’s recent Technology Manifesto, two have elicited the greatest reaction. One was our call for the UK to have the world’s highest rate of basic digital skills by 2020. The other was for the establishment of a Local GDS, an idea first outlined in detail by Richard Copley. Much of the debate on the latter has focused on the merits or shortcomings of having one website for all of local government. That’s an interesting and important question, but it is also just the start of a much more pressing and far reaching debate, namely: what should digital government look like in the local government sector?
Since it has prompted recent comment, let’s start with the idea of having one website – a LOCALGOV.UK – to replace the hundreds currently maintained by local authorities (LAs). Socitm has called the idea ‘ill-conceived’ and argued that it ‘should not be attempted’. They reason that:
Local authorities are independent, democratically accountable bodies. Citizens and businesses pay taxes and other charges to consume services delivered by their local authority. A direct, digitally enabled relationship with the accountable body and its elected representatives is therefore essential.
Whilst I completely accept that local councils have their own democratic link and local priorities, I do not see how this entails having a unique web presence. Of all the concerns that citizens have about the democratic bond between themselves and their local elected representatives, I would imagine that their council’s choice of Content Management System is very near the bottom of the list. To be clear, we need to distinguish between the platform and the content: the platform can be common to all local authorities; the content will always – and rightly – be local and specific.
But why bother? For two reasons. Firstly, if our primary concern is for the vitality of local representative democracy, then a key priority should be having clear communication channels between governors and the governed. Today, many people’s preferred means of finding information and communicating is online. I would argue that it makes local government less transparent, accountable and democratic when every council has a different front face, making it harder to find and compare information, services and contacts. This is especially the case given that people do not conveniently live out their lives within one local authority area. They find it baffling that relocating to – or working in – a different area entails having a completely different means of communication with local government.
A second benefit would be cost savings. Citizens and businesses do indeed ‘pay taxes and other charges to consume services delivered by their local authority’. It is for precisely that reason that they are justified in demanding that those services are provided as cost-effectively as possible. Reducing the vast duplication in hosting packages, web-designers – and yes, even some in-house IT staff – would be a step in the right direction.
Of course, the easy part is having one website platform for publishing written content. The harder task is using that website to provide common digital transactions (e.g. for applying for parking permits, paying council tax, reporting pot holes, etc.). Socitm have correctly pointed out that there would be costs associated with integrating those transactions with councils’ different back-end IT systems. As someone who spent a previous career working on large-scale IT integration projects, I can entirely sympathise with that argument. But let us not forget that there is also a significant cost associated with each council independently commissioning or designing its own bespoke transactions to work with their particular IT setup. It is at this point that conversations about having one website lead into the wider – and far more important – debate about the need for open standards behind the scenes of local government IT.
A well-rehearsed criticism of government’s past approach to technology has been that it has spent years buying bespoke IT to perform its vital functions. Where it hasn’t, there is at least significant variability in the IT systems procured by different councils for tracking benefits, plotting maps, handling customer relationship management, accounting, estates management and so on. That will not change overnight. But we should ask ourselves this: as we move forward, does it make sense to keep everything different, or would it be better to converge on some common platforms that would see harmonisation across local government? Local councils are essentially hundreds of organisations performing very similar tasks across the country, so why have completely different IT?
To my mind, the answer is not to insist that councils all use the same IT, but that whatever systems they do use be based on common, open standards to ensure interoperability. (I have been criticised for not calling for open source. I am genuinely agnostic in this regard, but I share Mark Thompson’s concern that a bespoke software package based on open source is just as bad as bespoke proprietary software.) The benefits of adopting open standards would be fourfold:
1) It would enable councils to more easily use common software and digital transactions created by a Local GDS and / or external suppliers, making possible the advantages of having a single local government website with standard transactions, as described above.
2) It would reduce the cost of IT for three reasons. Firstly, because it would enable councils to make increasing use of common IT solutions as described above. Secondly, it would make it cheaper for developers to make new products for local government. Without open standards, the fragmented nature of local government means that developers cannot build once and sell many times. Instead they frequently have to adapt their product for each council’s unique IT needs or provide the middleware to make it compatible – all of which carries a cost. Thirdly, it would prevent lock-in to any single package or vendor, as the data would be portable between different systems. This would incentivise companies to compete on prices to retain and win new customers.
3) It would enable better data sharing within councils, leading to improved internal communication and joined-up working between departments.
4) It would enable data better sharing between local authorities.
As we shall see, it is the last two points that should be the ultimate goals of local government digitisation.
Viewing the digital maps on which local authorities plot their data reminds me of playing the computer game Command and Conquer: everything beyond an army’s territory lies in complete darkness. It’s the digital equivalent of medieval atlases which labelled unknown regions with: ‘Here be Dragons!’. Put simply: councils have detailed records which enable them to plot the location of parks, buildings and parking spaces; they can show the addresses of constituents with particular needs, from education to welfare, and much more besides. But – for the most part – they have little or no data on those same things beyond their boundaries. That’s a problem for a number of reasons.
Firstly, the issues councils are tasked to address are rarely contained within one local authority’s boundaries. Communities, areas of deprivation, crime, littering, and school catchment areas (to give just a few examples) can all cut across borders. It is therefore troubling that councils have such limited visibility of how those issues continue next door. How do they know if a particular problem they are tackling, or service need they are meeting, represents the tip of the iceberg or the mass below sea level? Does that area of urban deprivation on their eastern boundary continue on the other side of the street into the neighbouring borough? What’s the demand for library services in the community which falls at the intersection of three councils’ areas? Which other councils across the country have a similar issue and might be able to share ideas on best practice?
Without having data to see the real shape of problems beyond their jurisdiction, it is hard to find ways to resolve them efficiently. (And efficiency really matters. Councils are expecting cuts in the region of 40% by 2018, so they will increasingly need to find different ways of working.) If one council spends £1million on combating a particular problem, might they be better to hire the services of the team of a nearby council that has a far greater incidence of that same issue? Shared services will always be limited without shared data. Whilst recognising that some councils do have ad hoc agreements on data sharing and service collaboration with other councils, that is no longer enough. What we need is systematic data sharing between councils. Why? To answer that, it’s worth taking a moment to look at New York.
Tucked away in City Hall, a small team of graduates has quietly been unleashing a data revolution in New York City. The innocuously titled Mayor’s Office of Data Analytics (MODA) was established by Mayor Michael Bloomberg and tasked with harnessing data to better inform policy making and to improve public services. The team systematically gathered records from the city’s myriad agencies and departments and began processing and mapping the data to identify trends and spot correlations. The results have been nothing short of remarkable.
Consider the team’s work on illegal conversions. Every year New York receives around 25,000 complaints about dwellings that have been illegally subdivided into smaller units. New York has a mere 200 inspectors to handle all these complaints. Before the creation of MODA, only 13% of their investigations resulted in a vacate order, leaving thousands of genuine problem cases unaccounted for. MODA sought to address this problem with data. They started with a list of the city’s 900,000 properties and combined it with datasets from 19 agencies. They added information such as foreclosure proceedings, anomalies in utilities usage, dates of construction, ambulance visits, crime rates and rodent complaints. They compared this information to fire records and developed a model to determine where inspectors should investigate next. In short, they were looking for predictive indicators of illegal conversions. Using their more aggressively data-driven analysis, 70% of cases investigated now warrant a vacate order: a 500% increase in efficiency. This case is just one amongst many which illustrate how scarce local resources can be better allocated simply through the use of data analytics.
Could we create a New York-style data analytics team for UK local government? One which would collect data from all local authorities, stitch it together, overlay it on maps and provide the results back to local authorities so that they could see how their own data fitted in with the nation as a whole? To remove the areas of darkness and reveal the whole map?
The idea is not without its difficulties. Yes, there would be serious challenges in ensuring data privacy and secure data sharing arrangements between councils. My reply is that data has power through integration; government cannot hold out in data silos forever. Yes, it would be hard to compare different datasets which are recorded in different ways by various councils. But New York agencies together use 12 different ways to geocode information and they manage – that is why it is necessary to have a team with the analytical tools and capability to combine and interpret the data. It would also have implications for local authorities’ use of IT and would create a strong push for open standards – but as I have argued above, that is needed anyway.
It is also true that this idea entails centralising data analytics capability in one place. I respond: what is the alternative? How many councils seriously have the time, money or staff in place to analyse data on this scale? And even if they did have the capability, it would not get around the central point that the main benefit comes from analysing data from beyond their borough. Would this mean less power for councils? No. Local authorities are autonomous. They would remain so. They would have the power to decide how they used the information provided by a central data analytics team. But at least they would have the information to make informed decisions. And yes, I acknowledge that the UK is not New York. Our data, political structures and societal challenges are different. But I still believe there are core elements which are worth adapting for the UK. How that could be done will be the subject of a future report by the Policy Exchange Technology unit.
And so we come back to the need for a Local Government Digital Service. A Local GDS would develop the LOCALGOV.UK that would create a clearer, more consistent front face for local government. It would also work with councils to design and support the roll-out of common transactions and open standards in IT and data. Those activities would be a means to the more important end of breaking down silos of data. There is huge potential for smarter use of data within and between councils to lead to better, more targeted and more cost-effective delivery of public services. That would require a dedicated data analytics team, similar to that in New York, but shaped for the specific context of the UK. I believe it would make sense for that team to sit within a Local GDS, making it a different beast from its central government namesake.
All these ideas are based on the observation that the greatest benefits of technology and data come through integration. If all local authorities independently try to undergo their own digital revolutions, they will surely fail to realise them.
Follow Eddie Copeland on Twitter @EddieACopeland