By Julian Wong Aug.11.2009
In: energy efficiency

Deconstructing China's Energy Intensity--A Lesson in Fuzzy Math

Guest blogger John Romankiewicz  a/k/a Sustainable John (pictured right), a carbon markets analyst at New Energy Finance and director of the China’s Green Beat video blog, questions the consistency of NRDC’s announced progress on energy intensity reductions with his own calculations using NBS data.

China’s energy intensity target is perennially referred to by Chinese negotiators in international talks, but also regularly mistaken by the foreign press as some carbon emissions reduction target. Here’s the low down on what it is, progress to date, including the most recent announcement, and exploration of some curious inconsistencies.

What is China’s energy intensity target?

From 1980 to 2000, China led itself down a path of increasing energy efficiency and concern for the bottom line by introducing market incentives and competition such that by 2000, GDP output required two-thirds less energy than it did in 1978. From 2001-2005 however, China – with a realization of its new found wealth – went to a tremendous building boom. Industry could not keep its eye on such rising demand and energy efficiency improvements at the same time, so energy efficiency took the back seat. The government, realizing this in 2005, set a plan in the 11th Five Year Plan to once again improve on energy efficiency. Specifically, the headline target was to reduce energy intensity (the amount of primary energy consumed per unit of GDP produced) by 20% from 2006 to end of 2010. Using energy intensity as the cursor of progress leaves room for GDP to grow as it will even as energy efficiency is being improved.

Two of the most noteworthy programs have been the shutting down of old, inefficient coal-fired capacity and the Top 1000 Energy-Consuming Enterprises program. Additionally, many industrial enterprises have been installing waste heat and gas recovery equipment at their facilities. There are plenty of other smaller initiatives like putting in CFL’s, motor system improvements in manufacturing, and efficiency standards for appliances. It is no doubt that these initiatives are all contributing to China’s improvement in energy efficiency. I only question how much China’s energy efficiency has actually improved.

Announced versus calculated progress to date

According to the National Development & Reform Commission (NDRC), energy intensity has decreased by roughly 10% in the three year period 2006-2008. They have just announced another 3.35% year-on-year reduction for H1 2009 in comparison to H1 2008.

I have analyzed National Bureau of Statistics (NBS) data from 2004-2008 which is summarized in the charts below. The two sources of my GDP (in nominal terms), GDP growth rate (in real terms), and primary energy consumption data are the 2007 and 2008 official economic and social development statistics reports off the NBS website.

You’ll note that I have no idea about H1 2009 energy consumption because the figure has not been reported to my knowledge. Only the percentage drop in energy intensity has been reported. We’ll come back to this in the next section.

I decided to perform an analysis of my own, using these data to calculate energy intensity over the years. For each year, energy consumption is divided by GDP. However, I did not simply take the nominal GDP numbers for each year, but instead, converted the GDP for each year into real terms (i.e. 2005 renminbi) by applying to each nominal GDP figure the applicable GDP growth rate, which was already expressed in real terms. The result is the following:

Given these figures, which are calculated using NBS data, the energy intensity drop for the three year period 2006-2008 is only 7.7%, not the 10% commonly reported. It looks like they still have a long way to go to reach their goal. One of the main discrepancies I have found is with the release of 2008 data. The NDRC announced a 4% increase in energy consumption in 2008 (2.85btce), and coordinated with a 9% increase in GDP, this led to a 4.2% drop in energy intensity. However, if you use 2007 and 2008’s numbers for energy consumption off the NBS website, the increase in energy consumption is actually 7.4% which leads to only a 1.5% drop in energy intensity. Maybe the NDRC and NBS should talk more. Or maybe it’s just a politically sensitive subject.

Latest announcement of progress

As mentioned above, the NDRC has just announced a 3.35% year-on-year reduction for H1 2009 in comparison to H1 2008. According to the NDRC’s News Center, energy intensities for specific sectors have fallen, notably: 1.5% for power sector, 1.7% for steel sector, and 7.7% for cement sector. Meanwhile, GDP continues to grow at 7.1% year-on-year. Data for total primary energy consumption cannot be found (please notify me if you find it), but power consumption is down 1.7%. NBS has released a special report (news coverage in English, Chinese report) explaining just how on earth power consumption could shrink while GDP still continue to grow, defying nearly all energy economics knowledge to date. Here are some summary points for what they came up with.

First, they begin by explaining how primary (think agriculture), secondary (think mining, manufacturing, and power), and tertiary (think banking, consulting, and commerce) industries are using different amounts of electricity per GDP output: respectively, 400kWh/10000RMB, 2460kWh/10000RMB, and 370kWh/10000RMB. Value added index (a proxy for GDP, but why they refer to “Value added index” instead of GDP escapes me) and electricity consumption went up in tandem (one-for-one) in both primary and tertiary industries. Secondary industry is the “culprit” then, as value added index went up 5.2% while electricity consumption fell 5.9%. They explain a lot of statistics magic about why things are the way they are, which honestly doesn’t leave any impression on the reader. The most concrete reasons I could pick out were:

They explain a lot of statistics magic about why things are the way they are, which honestly doesn’t leave any impression on the reader. The most concrete reasons I could pick out were

  • Industry is consuming 29TWh of electricity they made themselves from waste heat recovery power facilities, which are not tracked by electricity companies. That’s 2.5% of all electricity they consumed.
  • Small scale industry reduced their electricity consumption by a whopping 48.9%.
  • Of 44 energy intensive products, 30 of them reduced their electricity consumption/product value which had a net effect of reducing electricity consumption by 1.5%.

They end with a very tired explanation, saying that power elasticity is simply constantly changing from year to year, which they provide data for shown in the chart below. Power elasticity is the percent change in power consumption per percent change in GDP. So if power consumption increases 4% and GDP increases 5%, then the power elasticity is 0.8.

Certainly, China’s power elasticity is all over the place. And power elasticities of less than 1 are very common during times of efficiency build out. But, the NBS failed to address the basic fact that in H1 2009, the power elasticity has gone negative to -0.24 (-1.7%/7.1%). This is something never seen before in China’s history, and as far as I’m aware not seen before in world economic history. The bullet points mentioned above in the NBS report would lead to a lower elasticity, but not a negative one.

So, what’s the deal?

Comments (6)

  1. Tao Aug.11.2009@5:46 pm Reply

    Hi John,

    I know there has always been some suspicion around Chinese government massaging the statistic data, and quite rightly so from historical experiences. I am not going to defend for them, well I am not able to either I am quite poor in the nuts and bolts of statistics. But I guess there maybe something I can lend for some of the points to the question you just raised.

    1. The energy consumption and economic growth is not always closely linked, particularly during disruptive times. In fact even just about ten years ago, this happened in China already. During then Asia economic crisis, China’s economic growth reduced from 10% in 1996 to 7.6% in 1999, while its total energy demand reduced about 77mtce during the same time before recover in 2000. This energy consumption data is according to Chinese NBS, of course may raise suspicion, but the data from IEA also showed China’s primary energy demand up and down in these years, with almost the same level between 1996 and 1999, which is also confirmed by the data from EIA.

    2. This case does not only happen in China. During the collapse of USSR, Russian GDP contracted an estimated 40% between 1991 and 1998, but according to IEA its energy demand reduced about a quarter during the same time, same in the EIA’s data. But when Russia’s economy recovers in the last few years, the economic growth is also much quicker than its energy consumption growth. So economy could shrink, as well as grows much quicker, than energy consumption.

    3. If neither Russia nor China seems a very reliable case, here is the brilliant “Danish example”, which has always been advocated by the Danish officials in all climate conferences. Since 1980, the Danish economy has grown by 78%, while energy consumption has remained more or less constant, and CO2 emissions have been reduced. You could found it at the Danish Ministry of Climate and Energy.

    So I guess the Chinese case does not seem to be as splendid as it appears to you now? Remember the cases above are all statistics over a long period, which would be much more difficult to have dramatic change than in a short term. In the short term a response to a sudden collapse of exporting market would be stop manufacturing and continue with selling the stocks already in the warehouse, a process called inventory control or de-stocking? That is very true for many small exporting industries in China, many of which even shut down during the economic crisis last year. I guess this may explain a bit of the whopping 48% fall in electricity consumption in small scale industries, although still surprisingly high.

    Also because it is short term comparison, it could quite easily end up comparing “March’s Apple” with “October’s Pear”. The economic growth data and energy data may have quite different timeframe, for example if the economic data is generated using sales data while energy data is using from production data, these two won’t necessarily match. At last, the industrial structure change means if energy intensive industries are shrinking while less energy intensive industries could make up the loss with much less energy demand, you could end up an increased industrial GDP with reduced energy demand. This is just a possible scenario, but I am not saying that is the exact case China now has.

    Having said all that, I didn’t mean that your suspicion was wrong; in fact it is very useful and necessary to question these figures and dig out exactly why things are the way they are. And I hope it would be able to show us that it is not as difficult as we imagined to decouple the economic growth with energy demand. I hope so…

  2. sustainablejohn Jan.6.2010@10:37 pm Reply

    in the latest news… china’s 2008 GDP growth rate is now 9.6% and GDP is 31.40 trillion CNY. and china’s energy consumption has also been adjusted UPWARD to 2.91btce. If the 2005 numbers remain the same, this would actually make my 7.7% energy intensity reduction lower to an even more modest 6.3%. Well we all know the 2005-2007 numbers have probably been adjusted too. We’ll see what the charts say when the official 2009 statistical bulletin comes out in Feb 2010, well it’ll be 初步 estimates, but ya know…

Trackbacks for this post

  1. China Law & Policy » The U.S. & China on Climate Change at the U.N. General Assembly Sep.24.2009@11:51 pm Reply

    [...] “energy efficiency targets” domestically since 2005.  As John Romankiewicz explained on the Green Leap Forward, in China’s 11th Five Year Plan passed in 2005, the Chinese government established a 20% [...]

  2. The Green Leap Forward 绿跃进 » China to adopt “binding” goal to reduce CO2 emissions per unit GDP by 40 to 45% of 2005 levels by 2020 Nov.26.2009@2:08 pm Reply

    [...] can sometimes be a messy affiar, as Romankiecz has previously explained in a guest post, “Deconstructing China’s Energy Intensity–A Lesson in Fuzzy Math.”  So it is welcome news that China will be partnering up with the Intenational Energy [...]

Leave a comment

Discuss, be nice!
Discuss, be nice!