Skip to content

How Reliable Are Chinese Emissions Statistics?

June 9, 2016

By Paul Homewood 




From The American Interest:


A study conducted by researchers in one of China’s megacities found that lab estimates of car emissions are way, way off. As Reuters reports, by focusing on data culled from taxis and ridesharing services, scientists were able to expose the huge discrepancy:

The findings from the study in the central city were supported by a research arm of China’s top planning body using data from Uber and taxi firms. It found that standard laboratory estimates of carbon emissions from cars were off by about 6,500 metric tonnes per day, or about 59 percent.

“In the (the Chinese government’s most recent) 13th five-year plan, every region and city has the target, but the question is how you meet those targets and how do you verify the target has been met?” said An Feng, head of the Innovation Center for Energy and Transportation, a think-tank that led the study. “You cannot manage if you can’t measure it.”

Measuring emissions from static sources like power plants is relatively easy, but even there China’s self-reported numbers have proven to be unreliable. When you start trying to encapsulate the collective emissions of hundreds of thousands of cars and trucks, the complexity of the task is necessarily going to introduce a larger margin of error. In this sense, it’s not surprising to hear that Chengdu’s car emissions may be more than twice as high as previously estimated.

But it’s vitally important to keep the inaccuracy of Chinese statistics in mind when considering global climate efforts. The country is far and away the world’s biggest emitter of greenhouse gases, but its long history of statistical opacity and the strong incentives its local governments have to fudge the numbers they report to Beijing’s central planners make it difficult to get a handle on precisely how much China is emitting, and in the coming years how much it’s doing to curtail those emissions.

The Paris climate deal lacks any real enforcement mechanisms, and if ratified will merely require countries to submit national plans for reducing emissions. The only way the UN hopes to keep countries to their word is by naming and shaming them if and when they fail to meet targets, but in China’s case we might not be able to know how committed the country is to mitigated climate change.


Concerns about statistical opacity are very real. We have already had problems with China’s measurements of coal consumption, and many economists have expressed doubts about the accuracy of GDP figures for a number of years.


Let us be totally clear. The Communist Party in China only has one objective, and that is its survival. This in turn depends on maintaining and strengthening China’s economic and military power.

The idea that the Chinese government is bothered about the welfare of other countries is patently absurd. Everything they do is designed to further the interests of China itself.

If they have to fudge their CO2 figures to achieve that, they will do just that.

If you are in any doubt about this, ask yourself why China were so adamant that independent auditing of emissions should not be made part of the Paris Treaty. Instead, each country is now responsible for its own stocktaking.

  1. markl permalink
    June 9, 2016 10:21 pm

    Ah…yes, but those asking for a reduction in CO2 emissions know it won’t make any difference so are happy to report that “even China” is going with the AGW flow. Why they even ask is another question since you can extrapolate the real numbers from known usage anyway. And China probably isn’t the only country fudging the numbers to keep the environmentalists at bay.

  2. June 10, 2016 12:24 am

    If you want more accurate measures of emissions, give rewards and gold stars for outstanding outputs of CO2 to feed the plants

  3. June 10, 2016 1:16 pm

    What did you expect? If I was born in east Asia I would be deceiving about statistics.
    There is such strong cultural pressure to give a pleasing answer, rather than a truthful one.
    (If that sounds racist I can still defend it)

  4. It doesn't add up... permalink
    June 10, 2016 1:51 pm

    Here’s some of the history of how BP have reported Chinese emissions:

    Lots of changes. Not that China hold an exclusive on this. Lots of post-facto revision is common – even in countries where you might think statistical standards are a little more robust, such as the USA.

    • AndyG55 permalink
      June 10, 2016 9:19 pm

      “Lots of post-facto revision is common ”

      Sorta like temperatures, hey 😉

  5. June 10, 2016 7:58 pm

    Maybe they could try measuring what goes in rather than trying to measure what comes out

    • It doesn't add up... permalink
      June 10, 2016 8:33 pm

      That is I think the basis on which BP calculate their figures. The problem is that the input side is equally a statistical mess. I posted here before about very substantial revisions to data on Chinese coal consumption – not merely large swings in tonnages, but huge swings in its alleged calorific content.

      • It doesn't add up... permalink
        June 10, 2016 8:34 pm


      • It doesn't add up... permalink
        June 10, 2016 8:49 pm

        click to see larger version

  6. June 11, 2016 11:55 am

    Trying to understand the BP historical graphs, do the mtoe allow for the falling calorific value, BP seem to use a constant conversion factor for toe to tce)? If so the total energy input (i.e. Chinese growth) is not increasing as fast as it seems and the CO2 will be increasing faster than calculated through using low cv coal.

    This is the common problem of assessing the data that economists and politicians use where very large numbers are adjusted to provide fictional growth data. Got to admire BP’s attempts to make sense of the data!

    • It doesn't add up... permalink
      June 11, 2016 8:34 pm

      The mtoe data are in energy. To derive actual tonnages of coal, you have to divide by the calorific value (and adjust for the fact that there are about 10^7 kcal in a tonne of oil equivalent and 1000 kg in a tonne, so coal tonnages are in fact of the order of twice the mtoe figure). What is notable is the oldest data suggest coal quality rising close to 5,600kcal/kg, whereas now the same year shows about 4,900 kcal/kg – a huge change, made even greater in terms on tonnes of coal by the higher mtoe figures.

      Changes in calorific value are almost entirely driven by the proportions of hard coal and lignite consumed. It seems very odd that there is no reasonably consistent data on this: each mine tends to have fairly stable quality, and imports will be measured for quality to determine the price paid.

Comments are closed.

%d bloggers like this: