Claims by paid consulting firms for a firm producing cancer-causing tobacco products that “plain packaging” have failed are wrong. The evidence they produce is inconclusive in statistical terms. Other evidence suggests there have been negative effects on consumption levels and participation in smoking. The problem with the argument of London Economics is that they do not understand the concept of statistical significance. They, in fact, find evidence of a decline in smoking but assert it is too small to be statistically significant. They therefore assert there has been no effect. This is wrong – the conclusion should be that there is currently not enough data to determine an outcome. I am interested to note that the same stupid statistical reasoning was used by climate denialists (many of whom get funding from tobacco companies and who promote false claims about the harmlessness of secondary tobacco smoke) to deduce there has been no statistically significant warming since 1998. This claim is also false.
It is hardly surprising that the effects of plain packaging are as yet unclear. The policy targets young people initiating smoking not primarily the overall stock of smokers. It might take a few years for the aggregate stock data to show much response. On the other hand the data on a surge in calls to “quit” lines suggests the policy is having an impact.