Britain's flagship public health policies of the last fifteen years have not been driven by evidence at all, but rather by political pressure from small, often taxpayer-funded interest groups, according to a new book published by the Institute of Economic Affairs.

In Inside the Sausage Factory: The Illusion of Evidence-Based Policymaking, Dr Christopher Snowdon, Head of Lifestyle Economics at the IEA, examines four landmark public health interventions introduced in the last ten years:

plain packaging for tobacco
the Soft Drinks Industry Levy
minimum unit pricing for alcohol
the reduction of the maximum stake on fixed odds betting terminals (FOBTs)

All four were presented by government as the product of rigorous, dispassionate analysis. All four, Snowdon shows, were not.

The book documents a string of factual claims made by ministers and cited approvingly in the press that were demonstrably false. For example:

The most-cited "real-world" study in the press and the Commons supporting plain packaging for tobacco never measured whether a single person quit smoking. It surveyed 536 Australian smokers on how they felt about their cigarettes, nothing more.

A 6% fall in sugary drink sales in Mexico following a sugar tax was repeatedly cited in debates on the Soft Drinks Industry Levy; the actual reduction in sales over the relevant period was around 3%, and obesity rates continued to rise.

Politicians told Parliament that minimum pricing for alcohol in British Columbia had cut alcohol-related deaths by 32% but this was never an observed fall. It was a number generated by a model, run in one direction. In reality, alcohol-related deaths in British Columbia went up over the period, running 41% higher in 2011 than in 2002.
There was no research from any other country on what cutting the FOBTs stake would do, and the government's own evaluation found scant evidence the earlier limit had reduced losses. The sports minister conceded there was a "lack of evidence" that FOBTs were a "major problem", then backed the policy anyway.

In no instance did the evidence meet the standard of a randomised controlled trial, despite RCTs being both feasible and widely used in comparable policy areas.

Policy-based evidence

Packages of evidence, for example modelling studies, international precedents and expert endorsements, have been used to legitimise decisions politicians had already made on political grounds. When the evidence pointed to something unsavoury or at odds with an interest group's aims, it was simply ignored. Public Health England's advice was followed on plain packaging and the sugar tax but rejected on minimum pricing. When international evidence showed that sugar taxes have never reduced obesity, the government focused on modelling instead.

The squeaky wheel gets the grease

Small, well-funded paternalistic groups, with ample PR resources, are highly organised and motivated. Consumers who would bear the costs have insufficient incentive, information or resources to effectively mobilise. For the government, their priorities lie not in whether a policy will work, but instead whether the political cost of acting is lower than the political cost of doing nothing. With public opposition limited and the policies either revenue-raising or revenue-neutral, the calculation almost always favours acting.

A template still in use

The book's implications extend well beyond public health. The same architecture, a small network of ideologically aligned researchers, selective citation, sympathetic media coverage, and a political class more attuned to headlines than evidence, is visible across contemporary policy debates on issues from housing to net zero.

Governments have built elaborate machinery to give policymaking the appearance of scientific rigour. In practice, the decisions are often made first and the evidence assembled afterwards. The four case studies in this book follow an almost identical pattern: a small group of activist-researchers defines the problem, shapes the evidence base, lobbies ministers, and then evaluates the outcome. That is not evidence-based policymaking — it is the laundering of political preferences through the language of science.

We are told that politics “must follow the science”; and, as a general precept, that is obviously sensible. What Christopher Snowden shows in this book, though, is that the reverse is happening. Ministers decide what they want, and then commission the evidence to justify their policy. And what they want is almost always to increase the power of the state at the expense of the citizen. Here is a brilliant exposition of how the pull the trick off.