Tuesday, October 26, 2010

CPI surprise

Today’s Australian CPI data, according to the headlines, was ‘lower than expected”.  This was the first part of a forecast I published here back in early September, when I said “Inflation and GDP will surprise on the low side in the September quarter”.  GDP figures come out with the National Accounts on the 1st December so we had a little while to wait before assessing my prediction (1st November is the ABS capital city price index which may also show some surprises).

But the CPI print really shouldn’t have been a surprise.  Maybe most economists have loyal wives and girlfriends (or husbands and boyfriends, although it is a male dominated profession) to do their shopping, so they wouldn’t have noticed the price declines in food, health, communications and transportation in the previous quarter.

It is evidently odd that the US can experience no price growth with a collapsing dollar, while Australia’s currency has gained strength yet our favourite media hungry economists forecast high inflation and multiple interest rate rises. The high dollar was always going to dampen any inflationary pressures.

On a far more interesting note, Google has been experimenting with a real-time price index compiled, I assume, by experimental software that searches for listed prices of items on the web.  Their index has showed a “very clear deflationary trend” for the US, and has the additional benefit of compiling the same (or at least comparable) indexes across countries.  By the same measure the UK has shown a slight inflationary trend, attributable to the weak sterling.

The automatic nature of the index also provides the possibility of releasing multiple indexes with different scope and purpose, to provide a much richer picture of prices changes across the economy.  For example, hedonic price adjustments can be in one index and not in another, and the basket of goods can be quickly changed to suit different social groups.

There has been a strong push for the ABS to publish multiple prices indexes to address these very issues, particular with regard to quality adjustments.  I have demonstrated the Lower Bound Problem of Hedonic Price Indexes before, although Rob Bray makes the argument more concisely:

Revise the approach to quality adjustment to take account of the actual utility consumers achieve from changes in product ‘quality’; and also consider an approach which reflects the extent to which products actually exist in the market place for consumers to purchase

Twice the quality is not the same as half the price.

The benefits of real-time data available to Google are yet to be fully understood by economists, but there is no doubt the Hal Varian, Google’s chief economist, will change that soon enough.

Mr Varian also discussed some of his other work on using Google’s search data for economic forecasting. He said that he is working on “predicting the present” by using real-time search data to forecast official data that are only released with time lags.

For example, searches for “unemployment insurance” may be a good tool to predict actual claims for unemployment insurance, or the unemployment rate.

This is something I have tested before with the US housing bubble, clearly demonstrating that search volumes can be amazing predictive tools.  It won’t be long before these real-time measures become commonplace in mainstream economic publications.


  1. The problem with just about all the statistics produced by the ABS is that it is the only provider of such statistics, and when the budget cuts bite the quality assurance areas are always the first to suffer.

  2. These data makes massive difference to the market, so it is crucial that we trade only after we are fully sure. I usually prefer to trade on lower levels with AUD, it’s very risky yet things are slightly comfortable with having highly rated company like OctaFX broker at my side with their rebate program where I am able to earn 15 dollars profits per lot size trade, so that really helps me work well and far more easier than anything.