How much electricity is lost in transmission?

Posted on Updated on

If you want to understand energy issues it is always worth having some key numbers memorised. This let’s you do quick, and often powerful, mental sums, and it’s also a vital tool in your bullshit detection kit*.

It’s worth knowing that in typical western economies per-capita CO2 emissions are 7-10 tonnes – of course, North Americans aren’t typical – and that the average person in Britain, Germany or Japan consumes the equivalent of 3 or 4 tonnes of oil each year, roughly half that in America and double that in China.

Here is another one that is probably worth having somewhere in you mind: The percentage of electricity generation lost in transmission.

Nothing is absolutely perfect, and that includes transmission lines and transformers. Run electrical energy through a wire and some of it will end up becoming heat. As a result, not all all of the electricity generated at a giant rural coal power plant makes its way to the cities that consume most of our electricity.

How much energy is lost in transmission? This can be calculated by looking at EIA data for generation and transmission (available here). EIA publishes generation and transmission figures (in billion kWhs, a rather unseemly unit) from 1980 to 2012 for most countries. I could show them all to you, but that would be labouring a point.

Instead, I will show data for 12 representative countries. Below is the percentage of electricity lost in transmission for the most relevant modernised economies, the B, R, I and C of BRICS, and Mexico.


The numbers are relatively clear. In typical modernised economies roughly 6-9% of electricity generated is lost in tranmission. None of the modernised economies lose more than 10%. Japan and South Korea have lower losses than anywhere else, and I guess this is down to high population density.

Things are different in developing economies. In 2000, India lost around 30% of its electricity in transmission. However, this has improved significantly, and it is now down to around 18%. Meanwhile, Russia and Mexico can still improve a lot on their losses, but not as much as India has.

China however appears to have always had losses at the levels of typical modernised economies, and its transmission losses are essentially identical to America’s today . Whether this is real or simply a result of deliberate misreporting of statistics is not clear. The unreliability of China’s official statistics has long been a topic of academic research – just look at pork statistics – but I cannot find any research on whether their transmission loss figures are reliable. Perhaps there is a paper waiting to be written.

So, there you go. At most 10% of electricity is lost in transmission, unless you don’t have the luck of living in a modernised economy.

*Ernest Hemingway, I believe invented the term, despite it often being attached to Carl Sagan. Though Hemingway cut the bull, he simply it a “shit detector”.

Note on calculation

Data was analysed in R and plotted using ggplot2. The original data is taken from the Energy Information Agency, which in turn takes most of their data from national statistical agencies. The plot shown simply compares generation and transmission losses in each year. It could be more accurate by accounting for electricity imports, but the story would only change marginally (and I am putting the final layer of paint on to my thesis, and should not even be writing this blog post.)

7 thoughts on “How much electricity is lost in transmission?

    Mike Hemsley said:
    February 16, 2015 at 11:46 am

    Hey Robert,

    Thanks for this! Out of interest, any thoughts on why Germany is comparatively low? The fact that they are four very separate transmission systems?




      Robert Wilson said:
      February 16, 2015 at 12:14 pm

      Thanks Mike,

      I’ve never seen studies comparing transmission losses in different countries. Maybe one exists, but I haven’t seen it.

      However, there is at least one study which looked at grid reliability, finding that Germany’s grid was more reliable than most, and that it was probably because more transmission cables are underground. That might have a similar impact on transmission losses, but it would really need to be studied properly.

      Liked by 1 person

        Mike Hemsley said:
        February 16, 2015 at 12:37 pm

        ah interesting – ok thanks!
        All the best,



    David MacKay said:
    February 20, 2015 at 10:17 am

    Could you disaggregate transmission losses and distribution losses? My understanding, for the UK, is that the long-distance high-voltage system is responsible for a smaller part of the losses (roughly 1.5% of generated electricity) and the local distribution system is responsible for more (roughly 6% of generated electricity). See for reference.


      Robert Wilson said:
      February 20, 2015 at 11:27 am

      Good question. DECC produces these statistics each year and the statistics aren’t difficult to find in the Annual Digest. Last year around 74% of the losses were local, 3% were due to theft etc. and the rest was long distance.

      But I don’t think this can be disaggregated at the global level without first compiling the statistics. No organisation seems to do it. And some of these national stats might not even exist. Unless I’m missing something from the EIA website they don’t split losses up. I’ve tracked down the source of US figures shown in the post. It’s Table 8.1 on page 221 of their Annual Energy Review. They just seem to calculate generation and direct use of electricity and then estimate the losses from there. This probably makes sense given how many independent grids there are in America.

      Click to access aer.pdf

      My educated guess is that the breakdown is only official reported in countries which have a single national grid, such as Britain or France. I think France actually reports the losses on the high voltage network in real time, but it’s a while since I looked at the RTE website.


    Silvester said:
    February 21, 2015 at 1:59 pm

    Dear Robert,
    The numbers you present are surprisingly high. ENTSEO reports for 2013 a percentage of 1.6% of transmission losses. See page 2, in small letters below the title (

    I don’t know what could be the reason for this difference. I don’t think the difference can be due to bundling transmission with distribution losses, as the latter are rather high, in the order of 10-20%.

    I don’t think that lines in the ground are more reliable than overground ones. The underground lines are generally harder to repair. The fact that Germany has four TSOs is interesting, but should not affect reliability. I generally hear from people working for TSOs or regulators that keeping the German network stable takes much more work presently than 10 years ago, due to the rise of intermittent (renewable) generation.


      Robert Wilson said:
      February 21, 2015 at 3:12 pm


      I think this is just a question of the terminology used. I was transmission as shorthand for transmission and distribution. EIA appear to use distribution losses as shorthand for transmission and distribution losses. But I think the ENTSOE numbers are referring purely to the losses on the transmission network and ignoring the distribution network. However, these losses are typically much greater than on the transmission network. I also can’t see ENTSOE reporting the transmission losses by country anywhere in their data pages. That would be useful because you would then estimate the distribution losses based on EIA’s data.


Comments are closed.