Monday 18 August 2014

The plain packs meta-lie (part two)

The second part of the current plain packaging meta-lie is that "customs and excise data shows [sic] a fall of 3.4% in tobacco sales by volume in the first year of standardised packaging" (Public Health England).

Public Health England gives this webpage from the Australian Department of Health as the source for the claim that sales fell by 3.4 per cent in the first year (ie. December 2012-November 2013). That page says that the treasury has "advised that tobacco clearances (including excise and customs duty) fell by 3.4% in 2013 relative to 2012", but it provides no source and there is no trace of the treasury making such a statement.

We might be tempted to take this claim on trust were it not for the fact that the government has produced detailed figures on tobacco sales. They are published by the Australian Bureau of Statistics and can be seen here (table 8).

These data show that seasonally adjusted tobacco sales in the last five years were as follows:

2009: $16,270,000,000

2010: $15,869,000,000

2011: $14,747,000,000

2012: $14,250,000,000

2013: $14,126,000,000

Conveniently, for the purposes of assessing the impact of plain packaging, the 'years' shown above refer to December to November (eg. 2013 is actually December 2012 to November 2013). Mapped on a graph, they look like this:


It should be obvious that plain packaging did not lead to an acceleration of the longterm decline in tobacco sales. On the contrary, it coincided with an unusually small decline. The year-on-year changes are as follows:

2009 to 2010: 2.5 per cent

2010 to 2011: 7.1 per cent

2011 to 2012: 3.4 per cent

2012 to 2013: 0.9 per cent

In case you think that the seasonal adjustment has somehow altered the trend, these are the year-on-year changes based on the unadjusted data:


2009 to 2010: 2.6 per cent

2010 to 2011: 7.0 per cent

2011 to 2012: 3.4 per cent

2012 to 2013: 1.3 per cent

Whichever of the two datasets you use, three things are clear. Firstly, tobacco sales did not fall by 3.4 per cent in the first year of plain packaging. Secondly, the decline in tobacco sales was considerably smaller in the first year of plain packaging than in the preceding years. Thirdly, a decline of 3.4 per cent would not have been unusual even if it had occurred in the first year of plain packaging (the average decline in the previous three periods was 4.3 per cent). But it didn't so it doesn't matter.

You may have noticed that both sets of figures show a 3.4 per cent decline in tobacco sales in the year before plain packaging came in, so perhaps that is the source of the error. I won't rule out deliberate deceit, but routine incompetence is a usually the best explanation for government misinformation. I can't say the same about those who spread this, and other, lies about the alleged impact of plain packaging. They are probably conscious, deliberate liars. They've been doing it for years.

The plain packs meta-lie (part one)

1 comment:

Chris Oakley said...

You did well under the circumstances Chris and I think that a lot of people watching would have detected that Lean was being evasive and unpleasant. I am a bit surprised because he is one of the more scientific nutrition nuts, but of course, his whole livelihood depends on people buying into the bullshit that your study undermines. Which puts his funding allegations into a perspective that Krishnan Guru-Murthy might do well to think about.