Technology, like politics and power, needs to be decentralised

9 Nov 15

Technology and big data offer the seductive promise of better solutions to social problems. But such tools should be devolved to citizens and communities, not kept in the hands of bureaucrats.

In the 1960s, American cities hit upon a brilliant new idea for solving their social problems. Using new insights from cybernetics gurus like Herman Kahn, combined with the thrilling new science of computing they would cobble together accurate models of the urban environment, creating mathematical copies of the city that would help mayors to predict and provide the right policies and services. Needless to say, the shiny new models that expensive consultants fed into vast supercomputers were garbage. The results were generally meaningless at best, and sometimes led to actively damaging decisions such as closing badly needed fire stations in poor areas of New York.

The idea of being able to create the perfect plan is one of the most seductive and dangerous ideas of the 20th century. It promises a world in which beneficent experts sitting in government departments can watch over the population, anticipating and meeting every social need. But to see the reality you have only to look at Soviet five year plans, the bleak council estates of the 1960s and 70s, modernist disasters such as Brasilia, and Tony Blair’s empire of targetry and inspection.

In the 2010s, it would appear that the planners are finally on the run. George Osborne has embraced the idea of devolution, handing more power to cities to address local problems. Melanie Dawes, the permanent secretary of the communities department, has pointed out that many social problems are simply too complex to be addressed by national policy. But just as it looks like the technocratic plan is on the ropes, new technology is offering a fresh lease of life. As the Boston Consulting Group’s Adrian Brown recently pointed out, the rise of big data means that central planners have more information at their fingertips than ever before.

If they can combine better data with behavioural insights, smart cities technology and human centred design, then it might finally be possible to deliver a plan that works. There were echoes of this view in a recent speech by Cabinet Office minister Matt Hancock, who hymned the joys of small, crack teams of experts working from the centre of government to redesign services around the needs of the citizen. It is a worldview that we might term ‘techno-Fabianism’, a belief that technology can revitalised the waning power of the well-meaning national policy elite.

One of the key concepts behind techno-Fabianism is that government should become a platform: no longer a collection of bulky and expensive services, but a set of organisations that enable others to deliver public services. We seldom ask what kind of platform it should be. Google looks like a simple tool, but the way it sets the rules of its search engine makes it a hugely powerful organisation. The same can be said of Apple or Amazon.

These are organisations that give us a sense of power and autonomy, but whose very real power and control is hidden behind algorithms, slick user experience and anonymised exploitation of user data. Apply their metaphors to government and you will end up with the same old central control hidden behind a better website, rather than the decentralised world that Osborne and Dawes say they want.

Friedrich von Hayek famously argued that central planners can never have enough information to create an effective set of policies. Big data might yet provide a way to solve that problem. But even if it could, would we want it to? The more important part of Hayek’s argument was that even a functioning central plan could easily lead to tyranny. Imagine what Stalin would have done with big data.

The way to avoid creating a new panopticon of ubiquitous sensors and clever paternalism is to decentralise the technology as well as the politics and power. We need to use big data to fuel applications that help citizens make better decisions for themselves. This might include apps that help people to match their skills development to the likely supply of future jobs and new platforms that enable people to trade time and help their neighbours.

Better still, we could learn from the emergence of platform cooperatives like Loconomics, a cooperative version of TaskRabbit owned by its users, or New Zealand’s Enspiral support network for social enterprise. Council customer service centres could be turned into community-owned mutuals which use big data to help local people work with their neighbours to solve their own problems. Rather than being watched over by what hippie poet Richard Brautigan once described as ‘machines of loving grace’, we could use decentralised technology to build a new kind of civic commons, and achieve huge amounts with nary a sensor or planner in sight.

Technology is a tool, and like any tool it can put to different uses. The techno-Fabians would have us live in a world in which clever wonks solve society’s problems using ever more complicated mathematical models. But what if the real opportunity is a world in which data and behavioural insights do not empower the technocrats and the ministers, but are put at the disposal of ordinary people?


Simon Parker’s book Taking Power Back: Putting People in Charge of Politics is available here

Did you enjoy this article?