Tom Holland (in the South China Morning Post – SCMP) discusses the utility of using mainland electricity consumption data as a proxy for electricity consumption.
He makes an interesting and valid point that certain industries are very electricity intensive (smelting – especially aluminum). Holland reports that over the previous months smelting has increased, as has the mainland’s electricity consumption (+4.7% yoy in July). He argues that the growth figure is distorted.
In a previous post I discussed the merit of electricity consumption as a lead indicator of China’s GDP growth, especially considering the nature of other data points. Accounting for Holland’s comments, I would temper my additional enthusiasm. Nonetheless, I still maintain that its a good lead indicator of a direction at least, and as I said last time, it is important to drill down and understand the components of the electricity data point.
If one adds the economic output data for China’s 31 provinces in 1H09, they do not equal the national figures published by Beijing’s National Bureau of Statistics. In fact the sum of the parts is 10% larger than the whole! (Einstein eat your heart out!), with 24 provinces announcing growth rates higher than the national average. This is an improvement on 2008 figures in which 30 provinces claimed greater than average growth.
Chinese macro economic figures have the following difficulties (especially when comparing to other countries’ figures):
1. Retail sales figures are measured at the wholesale level, and often incorporate government and corporate purchases.
2. Fixed investment data (the largest component of China’s GDP) are accounted for when the money is distributed, as opposed to spent.
3. Economic activity is measured in China in terms of how much is produced, as opposed to how much is consumed.
Economist suggest that there are other metrics by which one can judge economic growth, one such metric being electricity consumption. The graph below shows that changes in electricity consumption in China are not highly correlated by with changes in overall output. Granted there are other factors involved in both, but I would expect one to be indicative of the other. From now on I suggest keeping my eyes on this electricity data as a good lead indicator of the Chinese economy, a good way to compare trends with other countries, and breaking it down to which sectors are most effecting the changes in consumption.