By Tom Wein
Three years of reading theories of sociology prepared Mike Kelly not at all for “the brutality of power” he experienced in counselling ministers, at the Health Development Agency and latterly as Director of the Centre of Public Health Excellence at NICE. Yet he is resolute in saying that this was not a bad thing; merely politicians doing their jobs, in roughly the way that they ought to. In a remarkably wide-ranging talk at the UCL Centre for Behaviour Change, he argued that the onus is on scientists to understand the messiness of real change, and to adapt accordingly.
Where politicians – and civil servants, and academics – could improve is in the way they conceive of change. They are inclined, Kelly argues, to view change as a linear process, from intervention to action to outcome, leaving out all the endless room for contingency, error and derailment which beset any grand plan. This is the fault, he feels, of the Whig view of history: the simplifying “abridgement of history from the point of view of the victors”, which obscures all the ambiguities and uncertainties of how change actually happened.
In public health, for instance, he summarises the story roughly as follows: cholera was defeated in London by the heroic technocrats Bazalgette, who built the sewers, and Chadwick, who led the campaign for sanitation. Their far-sighted empiricism saved tens of thousands, and we’ve been making progress ever since.
That many lives were saved is not in doubt, but almost every other part of that explanation obscures more than it reveals. The fight against cholera wasn’t one fight at all; it was a series of interlocking and contradictory decades-long arguments, with political coalitions contesting morality and offering competing explanations. If the right policy won the day, it was not because of an exceptionally robust RCT. Yet still today we are inclined to say that evidence is all that matters, and try to condemn our way to victory when politicians fail to live up to the technocrats’ evidentiary standards.
Change comes slower than we typically think. In public health, Kelly suggested as a rule of thumb sixty years – the time taken from the production of evidence to effective state action on cholera, and also on tobacco. ‘Cultures of evidence’ take a great deal of time to build, and on Whitehall he singled out only Health (since the 1980s) and Transport (since the 1940s) as departments that are serious respecters of evidence.
One of the beneficiaries of that culture is Deidre O’Reilly, the evening’s other speaker. Leading a team of social and behavioural researchers at the Department for Transport, she offered an inside view of the experience of confronting those challenges.
The route to achieving buy-in, for her, was as follows: first she produced a simplified Behavioural Insights Toolkit and wrote think-pieces to raise awareness of the value of behavioural insights. Next her team started running tailored masterclasses, in which they help policy teams to strip back assumptions and decide precisely what the problem is. After that comes audience selection, powered by a large segmentation exercise the department carried out to group the public by their transport habits. Only then comes new research, focused around collective/individual and objective/subjective binaries.
Behavioural insights is nothing new, O’Reilly believes; the department has been using social and behavioural research to inform road safety for decades. What has changed is the injection of energy and charisma, which has allowed her more scope to do her work. Yet that energy only goes so far. Beyond the general need for crisp communication, both speakers emphasized the challenges of organisational change. Mike Kelly offered a typology of hurdles: change is only possible at the margins; resources are limited; habits and “sedimented social practices” are strong; and power shapes every level of the system.
O’Reilly noted that a cross-governmental effort is currently underway, to produce a compliment to the MINDSPACE paper, focusing this time on group and organisational change, rather than individual behaviour; its application to government itself is an essential first step.
This was the first of a three-seminar series on policy and evidence, hosted by UCL’s Centre for Behaviour Change and chaired by Professor Robert West. More information is available here.
Tom Wein