There has been a huge kerfuffle (okay – I think that’s a good pun) over Christian Kerr’s article from last week. Last night Media Watch got into the act. So let’s remind ourselves of the basic argument:
Plain packaging laws, which came into force in December 2012, have instead boosted demand for cheaper cigarettes, with reports of a more than 50 per cent rise in the market for lower cost cigarettes.
The research by industry monitor InfoView, which shows a rise in the market share of cheaper cigarettes from 32 per cent to 37 per cent last year, is backed up by retailers, consumer marketers and the industry, with cigarette maker Philip Morris saying its information showed no drop in demand.
Last night the ABC confirmed this to be the case:
Well perhaps part of the answer is that tobacco industry’s profits have been hit by plain packaging as people switch to cheaper brands.
People are switching to cheaper brands of cigarettes? Just like Christian Kerr argued? And this effective decrease in price will have no impact on quantity demanded or consumed? (Hold that thought).
I just want to confirm for our international readers that the Australian Broadcasting Corporation is not some spear carrier for Big Tobacco or part of the great right-wing conspiracy, but is an Australian government agency.
I should also point out some very interesting cherry picking:
But … let’s ask a couple of questions.
Do the figures mean more people are smoking?
Well, no … the industry admits, the number of smokers fell in 2013 by 1.4%.
OK. So are the people who do smoke smoking more?
No again … the average number of cigarettes smoked per person also fell in 2013 by 1.4%.
Hmmm – Media Watch showed a very small copy of the letter and quickly whipped it off the screen. Let’s look at what the source material actually said:
“Over the five years in the lead-up to the introduction of plain packaging, total tobacco industry volumes were declining at an average rate of -4.1 per cent.
“Subsequently, since plain packs were introduced on 1 December 2012, industry volumes have actually grown for the first time in a long time to +0.3 per cent.
“Further, the number of cigarettes smoked on a daily basis declined at a rate of -1.9 per cent in the five years leading up to plain packaging, while it slowed to -1.4 per cent after green packs hit shelves.
“The long term decline of people giving up smoking at a fairly consistent rate and also smoking less has changed for the worse.
Oh dear; quoted the same number – out of context – twice to make different points.
I wonder why Media Watch didn’t quote this sentence?
“With growth in industry volumes, fewer people quitting and a jump in the amount of cheap illegal cigarettes on the streets, you could draw the conclusion that people are actually smoking more now than before plain packaging came into effect.”
Anyway the ABC then go on to report some fantastic conspiracy story as if it were news.
The other thing that really intrigues me about all of this is the naive reliance on ABS statistics. In particular The Kouk has been tweeting and retweeting the ABS address of the seasonally adjusted volume measure of tobacco consumption – right down to the excel spreadsheet column. To be sure the ABS does a magnificent job, but as the great philosopher Obi Won Kenobi said, “These are not the numbers you are looking for”.
As I have argued these numbers are subject to revision. Then we need to understand that Expenditure is a measure of Price times Quantity. Now with some very sophisticated mathematics I’m going to change the subject of the formula and we also know that Quantity is Expenditure divided by Price.
So now if Expenditure is down and Price is a constant then Quantity is down too. That is The Kouk’s argument and that is what the ABC wanted us to believe last night. Small problem – as even the ABC now admits:
… people switch to cheaper brands.
Price, in effect, is falling. Suddenly for a given level of Expenditure we cannot be certain what the impact on Quantity is – it all depends on the magnitude of the fall in price. But it possible to imagine – just possible – that over 2013 when Expenditure was rising and Price falling that Quantity would rise. The secret to understanding this is that we’re dividing a bigger number by a smaller number.
Another problem with the ABS data is that their technique for calculating chain measures doesn’t entirely eliminate substitution bias. That’s the fancy term for consumers switching from more expensive products to less expensive products – as another government agency, the ABC, now acknowledges is happening. The objective in these calculations is to eliminate price movements and get an understanding of the quantities (or volumes) involved.
… the ABS derives its annual and quarterly chain volume estimates using the Laspeyres formula with annual base years. With the exception of the latest quarters, quarterly chain volume estimates are derived by linking together estimates derived in the average prices of the previous year. However, the latest five to eight quarters are derived in the average prices of the latest base year, which is the year before the previous year.
So how old are the prices being used in the calculation? At the very least, not especially up to date. Now this isn’t a criticism of the ABS per se because for most things it probably doesn’t matter much. But sometimes it does:
For aggregates such as gross value added of mining and agriculture, and maybe exports and imports, where volatility in price and volume relativities are common, the advantages of frequent linking may be doubtful, particularly using the Laspeyres (or Paasche) formula. For reasons of practicality and consistency, the same approach to volume aggregation has to be followed throughout the accounts.
Fair enough, but note the caveat – when “volatility in price and volume relativities are common” the benefit of the ABS approach may be doubtful.
The ABS actually provide an example where declines in price actually distort their data – computers. This is an example we know and understand. But the ABS also tell us that it doesn’t matter that the data are distorted because Australia doesn’t produce computers and so the overall GDP figures are not distorted much. Again that is a fair judgement – but it does tell us that we should not simply accept any and every data point as being definitive or decisive in a debate.
A lot of very careful work needs to be done into the efficacy of the plain packaging policy – the early evidence isn’t very supportive of the policy.
All this is over and above the illegitimacy of state-sponsored persecution of that minority who consume tobacco. That is an argument that I think is powerful, but it has been lost.