10 February 2011

This Quote Cannot be Accurate


An article on energy efficiency over at Yale e360 has this quote allegedly from Energy Secretary Steven Chu:
"it’s a myth that the wealth of a country is proportional to its energy use."
I do not believe that this can possibly be an accurate quote, for the simple reason that it is wrong (and I can find no evidence of it online other than in the Yale e360 story).  The figure at the top of this post comes from Gapminder (the exact graph can be found here), and shows the very strong relationship of the wealth of a country to its energy use.  No myth.

UPDATE: A colleague emails to suggest that a linear-linear graph might show something different.  Good question, it does not.

UPDATE 2: Here is the source of the quote, a few minutes into Steven Chu's 2008 speech at the National Clean Energy Summit.

Chu is clearly saying that beyond a certain level energy use, metrics of quality of life (and he shows a graph using the Human Development Index) do not increase proportionately to increases in energy consumption. I have no problem with such a claim. However, it is highly misleading to reduce that to "it’s a myth that the wealth of a country is proportional to its energy use" absent the broader context, as was done at Yale e360. Thus, the quote is accurate but used improperly. The wealth of a country is indeed proportional to its energy use, which is quite different than saying, once you are very wealthy, consuming more energy does not increase proportionately indicators of quality of life. Thanks to the folks at Yale e360 for quickly responding to my query!


  1. We can simplify the Yale 360 article to - this is what you must do with your time and money.

  2. Thanks for the graphs. They're really insightful.

    I particularly liked it when the GDP is linear, but the Energy is logarithmic.

    It shows that, when your economy uses more than 100M units of energy (not sure what the units are?), then the GDP starts to skyrocket.

    Yet, all those economies using between 0 and 100M units of energy have very little difference between them. They all seem to stagnate at 50B or less.

    And yet our wise overlords want us all to use less energy. Go figure

  3. There a couple of other myths being propagated in the article as well.

    Under traditional business models, it made no more economic sense for utilities to reduce the volume of product sold — kilowatt hours, or units of natural gas

    That's a myth as well.
    There is every reason for the water, gas, electric or phone companies to discourage use if they are already at 100% of capacity. Adding additional capacity that will have a low utilization rate is not profitable and never has been.

  4. It seems to me that both energy use and income need to be calculated on a per capita basis. Otherwise you get a graph that plots population vs. population.

    There's the URL for such a graph at the bottom of this post.

    When I quickly calculated such a graph I did not see much of a trend for countries with per capita GDP in the $30k to $50k range.


    short url

    long url

  5. Have a look at http://dx.doi.org/10.1126/science.1091939 ("These relationships between wealth and energy consumption suggest that as a country becomes richer, its people tend to consume substantially more energy (table S1). However, looking at energy use within the high-income group alone, the correlation is weaker." and "Moreover, when one examines energy use per dollar of gross domestic product (GDP), the low-income countries use more energy to create a dollar of GDP than do the high-income countries, because of greater use of more energy-efficient technologies as a country develops (table S1) (16).").

    Also, in the context of the climate change, this might be also relevant/interesting:

    Both papers are highly cited and should point people to a vast literature on that subject.

    Seems to me that (as always) the answer is not a simple one...

  6. Maybe it depends on how one defines the wealth of a country.

    China produce a lot of wealth, uses a lot of energy, but its citizen are mainly poor. Or wealth by citizen is low, but for the country is high.

    Maybe that is what he means.

  7. "There is every reason for the water, gas, electric or phone companies to discourage use if they are already at 100% of capacity. Adding additional capacity that will have a low utilization rate is not profitable and never has been."

    That's only true if you have an absolute monopoly. Operating at 100% capacity continuously is a sign of a poorly managed business. New customers will go elsewhere and old customers will leave when you have a manufacturing problem, lose capacity and can't serve them.

  8. #DeWitt said... 7

    "That's only true if you have an absolute monopoly"

    It really depends on the marginal cost of additional capacity. I.E. If I have to add an additional $250,000 a year in costs to accommodate an additional $10,000 a year in business I'm better off just saying 'bye bye' to the additional business.

    Boeing has 2,000 737 orders on backlog. They've never delivered more the 400 in a year.

    Airbus has 6 fulls years worth of production on backlog.

  9. It's another one of these natural S curves isn't it? Everyone can happily pick the part of the curve that suits and ignore the rest.

  10. irt capacity loading, the rule of networks is that the larger the network, the more it will be utilized.
    There is probably a diminishing marginal return in capacity expansion, but I will also speculate that larger capacity implies more flexibility for dynamic situations.

  11. think Diminishing Marginal Returns.

    Works for a lotta stuff, including the benefits of energy use.

  12. Here's a Human Development Index vs energy consumption graph


  13. Hi Roger,

    In GapMinder, I changed the horizontal axis to Energy Use/Capita and the vertical axis to GDP(PPP)/Capita with a linear/linear scale. The linear relationship still holds.

    It's interesting to slide through the years and watch how the slope progresses for the developed countries. The reductions in energy intensities becomes apparent.


  14. Hi Roger,

    Don't you recall these quotes:

    "I don't think the American public has gripped in its gut what could happen," he said. "We're looking at a scenario where there's no more agriculture in California." And, he added, "I don't actually see how they can keep their cities going" either.



    I knew almost nothing about Stephen Chu when I read those quotes. But after I read them, I would certainly treat anything he said about an environmental matter with a great deal of skepticism.

    P.S. There is a bit of wiggle room for Stephen Chu in this diagram. For any given total GDP (if there are a lot of countries at that level) there is usually at least a factor of 2-5 range in energy use.

    For example, go to a total GDP of $20 billion. The total energy use ranges from about 2 M to about 20 M...nearly a factor of 10. So what Stephen Chu said here is nowhere near as ridiculous as what he said about agriculture and cities in California.

  15. Harrywr2

    Boeing fails to expand capacity and Airbus appears.

  16. Hi,

    Another interesting aspect from that presentation by Stephen Chu is that Dr. Chu (despite a PhD in physics) apparently can't do simple conversions between degrees Celsius and degrees Fahrenheit.

    First, he says "climate experts" (he doesn't say which ones) say there is a 50 percent chance that the earth will warm by 3 degrees Celsius in this century. Then he equates that to 11 degrees Fahrenheit (it's actually 5.4 degrees Fahrenheit).

    Then he switches to 5 degrees Celsius (again, not naming the people who claim there's a 50 percent chance of warming by 5 degrees Celsius in this century). But even the 5 degrees Celsius is only 9 degrees Fahrenheit, not 11.

    That's pretty bad for a PhD in physics, for whom climate change is a passionate interest. If a Republican Secretary of Energy made such mistakes (in the opposite direction) it seems likely that the goofs would make the papers or the blogs (as another part of the Republican "War on Science").

  17. Would it not be better to plot energy use per capita against income per capita?

  18. http://www.gapminder.org/world/#$majorMode=chart$is;shi=t;ly=2003;lb=f;il=t;fs=11;al=30;stl=t;st=t;nsl=t;se=t$wst;tts=C$ts;sp=5.59290322580644;ti=2007$zpv;v=0$inc_x;mmid=XCOORDS;iid=pyj6tScZqmEd1G8qI4GpZQg;by=ind$inc_y;mmid=YCOORDS;iid=phAwcNAVuyj1jiMAkmq1iMg;by=ind$inc_s;uniValue=8.21;iid=phAwcNAVuyj0XOoBL_n5tAQ;by=ind$inc_c;uniValue=255;gid=CATID0;by=grp$map_x;scale=lin;dataMin=0.0801;dataMax=37$map_y;scale=log;dataMin=282;dataMax=119849$map_s;sma=49;smi=2.65$cd;bd=0$inds=

    looks more logarithmic to me

  19. This is what it look like on a log scale http://www.gapminder.org/world/#$majorMode=chart$is;shi=t;ly=2003;lb=f;il=t;fs=11;al=30;stl=t;st=t;nsl=t;se=t$wst;tts=C$ts;sp=5.59290322580644;ti=2007$zpv;v=0$inc_x;mmid=XCOORDS;iid=pyj6tScZqmEd1G8qI4GpZQg;by=ind$inc_y;mmid=YCOORDS;iid=phAwcNAVuyj1jiMAkmq1iMg;by=ind$inc_s;uniValue=8.21;iid=phAwcNAVuyj0XOoBL_n5tAQ;by=ind$inc_c;uniValue=255;gid=CATID0;by=grp$map_x;scale=log;dataMin=0.0801;dataMax=37$map_y;scale=log;dataMin=282;dataMax=119849$map_s;sma=49;smi=2.65$cd;bd=0$inds=