Last week the Prime Minister is reported to have said that economist Richard Blandy had estimated Labor’s plan to abolish Work Choices would destroy between 200,000 and 400,000 jobs. This claim has provoked a response from Australia’s (and one of the world’s) leading econometrician, Adrian Pagan. My one line summary: Blandy has derived this estimate by comparing levels, but has ignored the pre-existing upward trend in employment.
The full piece is over the fold – if there are any opinion editors among my readers, this would make a great contribution for the final week of the campaign.
Economic tricks: A comment by Adrian Pagan
Many years ago a visiting econometrician from the U.S. gave a public lecture in which he mentioned that a secretary who was typing his material thought that â€œeconometricsâ€? should be typed as â€œeconomic tricksâ€?. Last week the Prime Minister is reported to have said that economist Richard Blandy had estimated Labor’s plan to abolish Work Choices would destroy between 200,000 and 400,000 jobs. This was apparently based on an econometric model that was reported on in the AFR of 29/8/07. It needs to be appreciated that the conclusion is indeed a type of â€œtrickâ€?.
Blandy claimed that Work Choices has added something like 290,000 jobs. When teaching econometrics to students one always argues that they should plot the data to get some idea of what they are trying to explain. Later, after they have finished the modeling, it is a good idea to see what the model did explain.
Following this advice one would get the figure below.
The quite volatile dashed line is the variable Blandy was explaining – deviations from employment after trend – and the solid line called â€œfitâ€? are the predictions of his model. Work Choices was introduced in May 2006 where the jump in the solid line takes place.
Now the first observation is that there was a high growth in employment before Work Choices. Moreover, the model looks singularly unimpressive as a model of employment over that period. Things seem to change after the advent of Work Choices. Such marked differences inevitably makes one suspicious.
There is a simple explanation of the â€œimproved fitâ€?. It is a consequence of the way in which Blandy allows for Work Choices in his regression – through a so-called â€œdummy variableâ€?. This variable goes from zero before the introduction of Work Choices to the value of one when Work Choices comes in during March 2006. No matter what the effect of Work Choices is one will find that the amount attributed to it by Blandyâ€™s regression will be very close to the average of the employment change after its introduction. This is simply an artifact of the way in which the analysis is done and has no causal significance whatever.
Obviously there was some other factor (or factors) â€“ X – that caused the big rise in employment before Work Choices. If X is still operating after March 2006 then its effects will be attributed in Blandyâ€™s regression to Work Choices. Unless an attempt is made to control for X, there can be no confidence that Work Choices has had any effect.
To emphasize the foolishness of all this I replaced the Work Choices dummy with a Rudd dummy. It is constructed in the same way as the Work Choices dummy – zero until Rudd took over as leader of the ALP in December 2006 and then unity. Interpreted in the same way as Blandy interprets the Work Choices results it would imply that Rudd has added many jobs- in fact 10% more jobs per month than Work Choices did.
This type of work is akin to the famous quote from Wallace Irwin that â€œStatistics show that of those who contract the habit of eating, very few surviveâ€?
Professor of Economics, UNSW
Processor of Economics, QUT
Chair, National Centre for Econometric Research