In this trip through some basic data on energy and economic output, I’ve looked at the strong relationship between those two things, an argument that energy use causes wealth (rather than wealth causing energy use), and evidence for increasing energy efficiency over time.
The evidence on energy efficiency was in the form of cross sections, as in Figure 1. Each color represents a particular year, and each point on a chart represents an individual country in one of those years. The further to the right, the richer the country, and the further down, the more energy-efficient. (For data sources, see here.)
|Figure 1. Data from BP Statistical Review and Penn World Tables|
(Note also the black dashed line at 50 toe per million dollars of GDP. In 1970 and 1990 there are a couple of very poor countries with better efficiency than that. By 2014, several rich countries have approached that level, but no countries (rich or poor) have surpassed it.)
While Figure 1 suggests overall movement toward improved efficiency, we don’t see the dots from a particular country: we can’t trace the US, or Czechia, or any other individual country as it changes from 1970 to 1990, or 1990 to 2014.
In contrast, Figure 2 shows exactly this kind of path for the US (blue) and the Czech Republic (black). The US data run from 1965 to 2014, whereas the Czech data only go from 1990 to 2014 (the earlier Czech data were bound up with Slovakia’s data as part of Czechoslovakia).
|Figure 2. Data from BP Statistical Review and Penn World Tables|
There are two main things to note here.
First, both countries trend down and to the right. In other words, each of them, over time, has tended to get richer and more efficient. For the U.S., this trend kicks in after 1970. For Czechia, it starts from 1991, as the country recovers from a downturn associated with the transition from communist rule. (The country may well have seen improving efficiency earlier as well, as part of Czechoslovakia.)
Second, although the shapes are similar, it’s not a situation where the two countries are following the same path just at different times. Rather, at any given level of wealth, Czechia’s path lies below the U.S. path—in other words, for a given level of wealth, Czechia’s economy has been more energy efficient than the U.S.’s.
So while each country is getting more efficient as it gets wealthier, the poorer country (Czechia) is significantly more efficient at any given level of wealth.
Looking at 2014, despite the fact that Czechia has only about 60% of the GDP of the U.S. per capita, is slightly more efficient.
One explanation for greater efficiency over time that I mentioned in the previous post was that, as a country develops and gets richer, its goods-producing sectors (agriculture, forestry, manufacturing) tend to shrink as a share of the economy, while the service sector grows. Since running a bank takes less energy than running a steel mill, this shift to services should increase an economy’s energy efficiency.
But that factor doesn’t help in this particular comparison of Czechia and the U.S., because in 2014 the Czech economy was 59% services while the U.S. economy’s service share was 78%. (Service-sector data are from the World Bank, indicator code NV.SRV.TETC.ZS.) So put this greater efficiency of the Czech economy in the “unexplained” bucket for now.
Despite their relatively large differences in wealth, the World Bank classifies both the U.S. and Czechia as “high-income economies.”
Figure 3 shows Mexico and Turkey, both of which are classified as “upper-middle-income economies,” which is the next step below “high-income.” Both countries show a period of rising wealth with worsening efficiency (their path moving up and to the right), but then moving to a trend of rising wealth with improving efficiency (moving down and to the right).
|Figure 3. Data from BP Statistical Review and Penn World Tables|
Figure 4 combines one more “upper-middle-income-economy” (China) with three “lower-middle-income economies” (Bangladesh, India, and Indonesia). Though they are now in different income groups due to China’s growth, they all started off with GDP per capita under $1,500 (measured in 2016 dollars). (Note that Bangladesh’s data start in 1970 rather than 1965, because the country was part of Pakistan before that.)
|Figure 4. Data from BP Statistical Review and Penn World Tables|
Each one peaks in a different year in terms of energy use per million dollars of GDP (Bangladesh’s peak is in 2005, but it was too hard to fit that label on the graph), and yet there is a striking similarity among these four countries.
Bangladesh, China, and India all have periods of essentially zero growth in GDP per capita, but increasing energy use per capita, which means their energy efficiency got worse while GDP stagnated, and this shows up as an ascending vertical piece of each country’s path.
But in each case, as a country moved out of stagnation into GDP growth, it also moved to a trend of increasing energy efficiency, with its path moving down to the right.
Indonesia didn’t have the long upward vertical path of the other three with zero growth, but it did have an extended period of slow growth from 1965 to 1989 with generally worsening efficiency, then slightly faster growth from 1989 to 1996 with slowly improving efficiency, then the disastrous impact of the Southeast Asian economic crisis of 1997, which for the next 6 years reduced its economic output and drove up its energy use per GDP.
But after 2003, even Indonesia enters a long period of high economic growth (moving rapidly to the right) combined with improving efficiency (moving downward).
For all four countries, slow growth or stagnation goes with worsening energy efficiency, while fast growth (or in the case of Bangladesh, any growth at all) goes with improving energy efficiency.
Figure 5 is a mess, with 22 countries crowded onto it. It shows (almost) every country in the dataset* with GDP per capita in 2014 that was over $30,000—in other words, the rich countries.
|Figure 5. Data from BP Statistical Review and Penn World Tables|
The general trend, visible despite the mess, is movement down and to the right. In other words, in this whole group of countries there’s a very strong trend toward getting richer and more efficient.
To see individual countries in the context of that tangle, we can fade out most and highlight a few at a time, as in Figure 6, which shows Ireland and Germany with strongly improving energy efficiency.
|Figure 6. Data from BP Statistical Review and Penn World Tables|
In Figure 7, Canada, the Netherlands, and Spain all show a pattern reminiscent of Turkey and Mexico in Figure 2, where there’s a period of economic growth with worsening efficiency (path rising to the right), then economic growth with improving efficiency (path falling to the right).
|Figure 7. Data from BP Statistical Review and Penn World Tables|
Figure 8 shows Switzerland, getting all the way down to 58 toe per million dollars of GDP, and being very wealthy as well (exceeded on this graph only by Norway).
|Figure 8. Data from BP Statistical Review and Penn World Tables|
Hong Kong is the star in Figure 9, and it is in some sense a “best case” for energy efficiency. The “country” is nothing more than a city and its suburbs. There are no great distances to cover. There is no agriculture and almost no industry, which means that the major energy-using activities of a normal economy (actually making things, from food to cars) happens elsewhere and doesn’t show up on Hong Kong’s account.
And yet the territory has made no meaningful progress since 1980 and is actually less efficient than it was in 2000.
|Figure 9. Data from BP Statistical Review and Penn World Tables|
Remember the black dashed line in Figure 1, the hypothesized “envelope” of 50 toe per million $ GDP? Here we have Switzerland approaching it, but possibly slowing down as it gets closer, and Hong Kong moving along for decades near that level, but not meaningfully improving.
Is 50 toe per million $ GDP some sort of barrier that a modern, wealthy economy can’t get past?
The next post will look at a different set of paths, following energy use per capita rather than energy use per unit of GDP. These paths will indicate that this probably isn’t an impenetrable barrier, but also that this news may not be as promising as it sounds.
If you’re not already too dispirited (or tired of trying to figure out charts),