Chinese electricity consumption smacked?

China Securities Journal yesterday reported that China’s electricity output may have dropped 7.5% in January, citing sources close to the China Electricity Council.

It is well known that the electricity output/consumption of China is a good proxy for economic activity.  Particularly for people who are skeptical about the Chinese statistics numbers, power output is an alternative gauge.

As always, one should be cautious about reading too much into January and February macro data because the Chinese New Year distorts the picture (and the Chinese New Year came early this year).  However, a drop of 7.5% in power output/consumption is huge and cannot be explained by Chinese New Year alone, as the chart below illustrates.  The last instance of a drop of comparable magnitude was during the post-Lehman crisis:

Latest posts by __ADAM__ (see all)


  1. …electricity output may have dropped.

    I may have won the lotto.

    Is a Lehmans type plummet credible or may we be better off waiting for actual data before speculating about Armageddon?

    • Shit man, do you want to know there’s a leak telling us this or not? The CSJ is credible. And no, you didn’t win the lotto, unless someone from lotteries leaked it to you, that’s the difference. But in honour of your point, I’ve out a question mark on the title. Alright?

      • and I note on the chart two pre-Lehman dips into negative territory while GDP just sailed along.

        CSJ credible? Even esteemed outlets like Fairfax can get it wrong when they published rumours. Let’s wait for the actual data.

        BTW has this guy ever gone to press on rumours that are positive for China?

          • yes I got the seasonality bit and the fact that the rumour exceed seasonality in spades.

            Is the monitoring of electricity usage automated or does joe sixpack head of the local power station have to report in? In which case is Joe enjoying his new year holiday and forgetting to report?

            Interesting how the speed at which China produces a GDP number is met with skepticism yet they can supposedly bang out a rumoured electricity number much faster (by weeks) than e.g. the EIA, and it is to be believed (if it is bearish). The author presents us with no analysis just gossip.

            Sorry I’ll wait for hard data. I don’t believe rumours …except when they involve celebrities.

          • Sorry I’ll wait for hard data.

            There’s hard data out of China you can trust?

            The only hard data that matters to Australia is iron ore prices and volumes. Everything else is speculation, including the “official” Chinese GDP numbers.

            BTW, for every Zarathustra there’s a dozen China boosters, prattling on about a billion people urbanising etc. If you don’t like what s/he has to say then go read The Cupboard or the Smage after Gina assumes full editorial control.

            FWIW, Zarathustra was posting anecdotes about real estate volumes falling off months before it was being reported in the mainstream media, so s/he has some credibility.

        • dumb_non_economist

          So far I’m with Towelie on this one.

          The following are the dates for a CNY start, the rest are Feb dates. I would have thought a 7.5 % drop was mild. China comes to a halt during CNY and it’s closer to 2 wks than the official 7 days. So as D the Cynic stated checking the av of J + F would be a better indicator.
          2012 Jan 23
          2009 Jan 26
          2006 Jan 29
          2004 Jan 22
          2001 Jan 24
          1998 Jan 28

  2. Diogenes the CynicMEMBER

    Chinese New year means shutdown of factories and this year Chinese New Year was early (Jan 23rd). In 2011 it was February 3rd so there maybe some wrinkles in any month to month or year to year comparison. That said if you sum January and February and compare YoY or average it for a month to month versus December you can smooth this out which is what we used to do when I was at the coal face in Hong Kong.

    However it is a canary in the coal mine and (more) evidence that something is wrong. Electricity is a very good proxy for GDP especially in a place like China where manufacturing is so important.

  3. Other outlets are using the same CSJ source and claiming it it defintivley.

    The use of ‘may’ is perfectly correct grammatically as it is the subjunctive mood in English and that makes the question mark unecssary, except to those that don’t know how the subjunctive is used.

  4. Houses and Holes, I’ll take this information any day.

    My research led me to believe that Electricity Consumption was “the” best metric for production devoid of all the usual measurement problems that expressing same in units of a currency involves.


    1. Can you explain in more detail what the charts are communicating [YOY?, Why chart goes negative when production can’t be negative etc]

    2. You got any other indicators of real GDP you like and where are they available?

  5. Baltic Dry is now down well below where it crashed to after the 2007/8 crash – the idea that valemax has impacted on that has been dismissed, the Brazilian ore exports are down 46%

    China’s construction and infrastructure boom is over, look to direct stimulus into consumption.

  6. Mark Lundeen has monitored US electricity production for years as a guide to what is happening in the real economy:

    “First, Electrical Power (EP) consumption for the US electrical grid is contracting, as we see in the red circle below. In truth EP’s current contraction could be from seasonal factors; last summer was cool, this winter has been warm. But then EP’s last all time high (Terminal BEV Zero) occurred in August 2008, 3.5 years, or 41 months or 180 weeks ago.”

    Query what happens with increasing private solar PV production (but it is de minimis in month on month terms).

    • Most measures of electricity consumption are net of small embedded generation (like roof top solar). So adoption of these technologies on a large scale, if not accounted for properly, can lead to underestimates of consumption.

      Electricity usage growth in NSW was negative for the first time since WWII in 2010/11.