ABS revisions

A couple of days ago I was chatting to a journalist from the Financial Times who had picked up on the Christian Kerr v The Kouk debate on tobacco consumption. I thought it a good idea to outline some of the points I made:

  • Christian Kerr is talking about tobacco consumption, The Kouk is talking about tobacco expenditure. Not quite the same thing. Expenditure is price times quantity. If we suspect that consumers are substituting for cheaper brands then quantity consumed could rise even if overall expenditure is down. We know, however, that overall expenditure rose over the first 12 months of the plain packaging regime.
  • The FT journo also asked me about the March 2014 quarter data point. Is it true that this data point is the lowest since 1959. Short answer – yes. Long answer – we can’t be sure.

There are a few things going on. First there was a massive increase in tobacco excise (more on that in a later post). Then we know that the ABS routinely revise their data. We also know that given the seasonal adjustments that the ABS make that the end-points in the data are rubbery.

To highlight the the potential impact of the ABS data revisions I plotted the December 2013 data, and the March 2014 data in a graph.

Tobacco - decline

There are non-trivial differences. For example in the December 2013 vintage we’d have thought that the increase in tobacco expenditure over the year Dec-12 – Dec-13 was 3.98%. Just three months later ABS revisions had reduced that increase to a mere 2.53%. A 1.45% reduction in smoking was entirely due to the ABS revision.

Then look at the differences in the March quarter of last year – revisions account for a nearly 1% increase in smoking, while in the December quarter revisions account for a 1% decline in smoking.

The takeaway is this: The ABS expenditure data doesn’t really address the issue of actual tobacco consumption – because we suspect that effective prices are falling (as consumers substitute to cheaper brands in 2013) and then we have the massive excise increase in the first quarter of 2014. In any event, we also know that the ABS data are rubbery and will only “settle down” later in the year.

As I said the FT journalist – there is more heat than light in the debate right now.

This entry was posted in Take Nanny down, Uncategorized. Bookmark the permalink.

One Response to ABS revisions

  1. PaulL

    Presumably people know the excise changes are coming, and shift a little purchasing forward – this could be enough to make a 5% increase before the excise change, and a 5% decrease in the period immediately following.

Comments are closed.