Thursday, July 12, 2012

May 2012 Industrial Production

The latest IPI numbers posted a pleasant surprise (log annual and monthly changes; seasonally adjusted):



In the case of mining, it’s largely due to the base effect from lower output last year, but the other sub-indexes have bounced up as well. Having said that, the increase in the headline flatters to deceive and I don’t think its necessarily sustainable. We should be looking at the IPI coming back to earth for the June numbers.

We’re also not looking at very good news for 2Q2012 GDP either, as the IPI numbers to date suggest an expansion of just 4.2%, lower than the 4.7% in 1Q2012:


One caveat though is that I’ve been playing around with different forecast models and approaches, and the one based on concurrent IPI has the most pessimistic forecast for 2Q GDP. None of the other specifications I’ve tried show anything less than 5%, and the average is 5.2% – the 66% confidence interval is between 4.7%-5.7%. If growth does turn out to be worse than in 1Q2012, it’ll be a downside surprise.

Technical Notes:

May 2012 Industrial Production Index report from the Department of Statistics (Warning: pdf link)


  1. I was trying to discount the base effect, and from my tinkering, I think at most, the IPI would have grown at about 4% without the mining low base.

    Anyway, don't you think the low base from last year a bit problematic? It's giving high growth artificially when demand might be a problem.

    1. That's always been the problem with judging conditions on growth metrics, even after using smoothing techniques. I always like to look at the actual raw data as well, rather than just relying on growth.