By Greg Peel

We know that Beijing's numbers are spurious ? at least one government official has already admitted that in the past. Amazingly, China is able to deliver GDP numbers about a week after the close of a quarter, while developed economies, which have had a lot more experience at this, take three months to settle on their own figures. But China remains a closed shop to a great extent, so all we can do for the most part is take Beijing's numbers as read.

There will remain, however, the disbelievers. There were quite a lot of disbelievers about when China released a 7.6% GDP growth figure this month for the June quarter. GDP figures are typically a conglomeration of many inputs and surveys, which is why The West takes so long to add them up. But for China, a longstanding rule-of-thumb measure of Chinese economic growth has been Chinese electricity production figures, which are more concrete, and which over time have proven to be closely correlated to Chinese GDP growth.

In short, China's electricity production barely grew on a year-on-year basis in the June quarter while Beijing declared GDP to have grown by 7.6%. On that basis, the easy assumption to make is that Beijing's numbers are way off the mark and China is descending rapidly into an oft feared "hard landing" without letting on. ANZ Bank's economists believes such concerns "are real", but after some number crunching of their own they have found that perhaps the disbelievers may be overlooking some realities.

The argument goes that Chinese industrial production has a very close correlation with electricity production and there a significant relationship between steel production and electricity production to a correlation coefficient of 0.7. Heavy industry consumes 61% of China's electricity and given property and infrastructure investment in China has slowed this year, demand for electricity has slowed. Ergo, with heavy industry so fundamental, the GDP result looks overstated.

ANZ notes, however, that the June quarter 2012 is not the first time Chinese industrial production and electricity production have converged. The following chart shows several periods of divergence, with the stand-out being June 2008 to September 2009 when electricity production growth turned negative but industrial production continued to grow at 3-5%. Electricity production then rocketed back to go the other way after Beijing's massive GFC-inspired stimulus package was rolled out.

The correlation coefficient here is 0.8, ANZ calculates, and thus we can conclude that electricity production is not a reliable indicator of industrial production 20% of the time.

The divergence may well be explained by growth in China's light industry and tertiary (services) sector. In the first half of 2012 the former grew by 11% and the latter by 7.7% (year on year). In June, electricity consumption by heavy industry grew by just 2.6% (yoy), but light industry grew by 6.7% and by the services industry by 10.9%, according to China's National Energy Administration.

ANZ also offers that China's energy efficiency has improved a great deal in recent years. Small and inefficient steel mills have been forcibly shut, as have small and heavy polluting electricity generation plants. New and more efficient plants have been opened in their place. This is important, because electricity production actually consumes quite a bit of electricity. ANZ roughly calculates Chinese electricity consumption per unit of GDP fell by 28.5% from the early noughties to 2011.

Finally, if we can't trust the GDP number, based on an electricity production correlation, can we trust the electricity production number? Curiously, recent comparisons of electricity production and consumption show the Chinese are consuming more electricity than is produced, and by more and more each year. Bear in mind that baseload electricity, unlike say wheat, cannot by stored or imported.

So which figures are actually the rubbery ones?

ANZ notes some local and privately run Chinese electricity plants are not counted in the production numbers, some local governments can only fulfill their efficiency targets by understating production, and some heavy consumption industries such as aluminium smelting have been shot out to far flung provincial locations which aren't connected to the national grid.

So yes there are lies, damned lies, statistics and data provided by Beijing, but data provided by Beijing may well be more of a sloppy representation than a lie.