When was the last time you read a party manifesto? I mean all of it. Page-to-page. Not just glimpsed bits that were in the media, but absorbed the entire document. They are not exactly required political reading because we don't really buy-in to their promises.
The credibility of this genre is also undermined by the iffy evidence supporting their claims. And to see just how much the claims stack up, The Alliance for Useful Evidence-has launched Manifesto Check with the The Conversation UK, using university-based academic experts to analyse these documents.
Manifesto Check will be one of many voices in the run up to May; we will see a plethora of fact-checkers rebutting what politicians and pundits say. The UK's Full Fact is setting up a ‘war room' of staff and volunteers who will work 18 hours a day in the run up to voting day.
This is good news for a healthier democratic debate. Voters need to know if the promises stand up to scrutiny.
But we also need to cut politicians some slack. It's essential we try out new ideas. Let promising innovations flourish – even if they are untested, and thus evidence-free. The danger of shooting down any claims because ‘there is no evidence', is that we stifle anybody who wants to get out of the status quo.
That's why local and national government needs more experimentation. This is not about freewheeling trial and error. Too often new public policies are rolled out nationally with little evaluation. In effect, governments experiment on the whole population at once - but without learning if these policies are doing more harm than good. The National Audit Office says that £66bn worth of government projects have no plans to evaluate their impact.
It is unethical to experiment on us in this arbitrary way. We need the best of social science to understand if – and how – policies are working, from Randomised Controlled Trials, to other research approaches, as set out in a in our new report.
The current climate also offers practical opportunities for experimentation. Decentralisation of public services provides more chances to experiment, according to Professor Gerry Stoker at Southampton University. Devolving power away from Whitehall to schools, local authorities and frontline professionals provides opportunities to try out different approaches.
Indeed, experimentation may best be done at local level. Central government finds it almost impossible to do it properly, according to a 2003 Cabinet Office review of government pilots. Too many national policies are set in stone. Manifesto commitments, trailed in the media, debated in parliament, may have too much political capital invested in them to really allow for change.
Of course, the danger of only experimenting in one local area is that can be hard to generalise to the wider world. How much can we really learn from the the falls in re-offending rates coming from Payment-by-Results pilots in Doncaster and Peterborough? Are any benefits just down to the local context and the human geography those areas? In their book Evidence-Based Policy: A Practical Guide to Doing It Better, philosopher Professor Nancy Cartwright and ex-WH Smiths businessman Jeremie Hardie have argued that this is a fundamental flaw in things like Randomised Controlled Trials. It is very hard in practice to move from knowing that something that was ‘effective there, will work here'.
The best way to respond to this is to replicate any trial in multiple locations (and even the same location to see that not a flash in the pan result). Replication is an important part of experimental science and social science. It should be an important part of experimental government.
There was support for this local-level experimentation at our fringe events at the Lib Dem, Labour and Conservative conferences last autumn. But one ex-Government minister told us that it is still important to coordinate these local experiments. He suggested that a body like the Local Government Association could have an oversight role; facilitating parallel experiments in multiple locations, to compare different local approaches. The four ‘Whole Place' pilot projects in Essex, Greater Manchester, West London and West Cheshire engaged the LGA and Central Government.
We also now have institutional support for experimentation through the What Works Centres, co–funded by the Economic and Social Research Council and the UK, Welsh and Scottish governments, covering everything from education to crime reduction.
The Local Economic Growth What Works centres, a collaboration between LSE, Centre for Cities and Arup, are carrying out their own experimentation ‘demonstration projects'. The Cabinet Office has also set up a trials advisory service to help public servants design good experiments.
These practical measures will all help, but we do need a change of political culture. Ultimately, we want to see more humility. Less false promises in manifestos. If the data and evidence doesn't back policies, we should say so. And if we don't have any evidence, commit to experimentation in the future.
In Finland - which also has a general election coming up - three of their main political parties have formally committed to experiments in their manifestos. Yes, launching an experiment is an admission of ignorance. But it's a more honest and credible stance. Without it, we will never really learn, improve and adapt public services.
Jonathan Breckon is head of the Alliance for Useful Evidence, Nesta
