Friday, February 5, 2010

Energy use and carbon emissions per capita

I stumbled upon these two graphs showing the evolution in time of energy use per capita and CO2 emissions per capita in different countries, and I was baffled. I have to confess that I had a wrong idea about economic development and I thought I would share this with you

First, the energy consumption per capita

It seems that the citizens of the developed world are increasing their consumption more rapidly than the developing world, with some exceptions that can be explained by the particular circumstances of some countries, for instance Quatar. Nevertheless, the growth is much slower than I would have expected. In France or Switzerland it has increased by something like 15% in the last 30 years. A Chinese or Brazilian citizen is consuming now about the same amount of energy as a Swiss citizen in 1960, and their consumption is not increasing that rapidly. The overall picture is, however, that the energy consumption per individual is fairly constant trough time and it mostly depends on the country one is living. I interpret this as a sign that it is the infrastructure of a country and not the morning choice car-or-bike what is relevant. am I mislead?
The second graph shows the CO2 emissions per capita in each country and the total CO2 emissions per country
For European countries we see an amazingly constant behavior through time. The overall picture for all countries is that, with some particular exceptions, there is no 'explosive growth' in CO2 emissions per capita. The explosive growth, for instance in Chinese emissions, stems from population growth. am I correct to interpret this again as a sign that national policies (source energy production, for instance) are the key factor and not individual choices?


Roger Pielke, Jr. said...

Hi Eduardo, yes you are correct. The different levels of national energy consumption per capita are not really a function of individual decisions, but of macro-factors such as you describe.

The same sort of wide distribution can be seen when you look across the 50 United States -- e.g., Wyoming has a high per capita level, New York much lower.

To understand why this is so, consider the energy consumption habits of immigrants. In general, their habits reflect where they live and their local context. A Haitian immigrant to NYC will more likely approximate per capita consumption in NYC than Haiti.

Anna said...

This was indeed interesting graphs!
And very different from the picture painted in media.

I would have expected very large per capita increases at least from the sixties or seventies and on, especially in the industrialised countries, and not this very slow, steady increase.

(And I had no idea that the population of China increased like that, despite their "1 child" laws).

But I guess the most common "concern" in media is: "what happens when all of the Chinese and Indian want the same standard of living as we have here in the West?", (and I don't know whether this is a real problem or not).

Any ideas?

Falk Schützenmeister said...

Eduardo, there is a little circle in your argument. The conclusion that the individual energy consumption depends on the country you live in is true. However you cannot conclude this from the per capita consumption because you assume that there would be no individual variance within a country. Energy consumption varies widely. The inhabitant of a favela does not use a lot of energy, but the rich family in Brazil is probably closer to an US family than to the countries average.

Individual choices matter too. People can make choices if they have them (provided by the infrastructure and also somewhat by culture). The morning decision to use the bicycle or the car is not available within a far away suburb or in Atlanta, GA. I name this city in lieu of others that do not have sidewalks or shoulders and no through streets in neighborhoods (cul-de-sacs).

By the way, the best predictor for energy consumption in the US (and probably throughout the world) is income. For this reason, I think Pielke's calculation about the Haiti immigrant to NYC is not entirely correct. There is even a hypothetical chance that the immigrant's energy consumption would be lower than in Haiti: 1. The immigrant would start from a low income and housing level. 2. The energy supply system might be more efficient. 3. If assumption 1 would be wrong and the immigrant were indeed wealthy, his or her energy consumption back in Haiti would have been much higher than the average.

Of course, once the immigrant could afford the American way of life (which is a choice, he could also stay in his densely populated city neighborhood), Pielke is right. I would see it as a short cut in economic development.

The question is not infrastructure OR individual behavior. Both have to be considered in relation to each other. There are some interesting approaches in city planning (smart growth, new urbanism, etc.) and in the social studies of technology (Urry).

In which extent individual behavior matters depends on development itself. It matters more in a postindustrial consumer society than in a traditional rural setting. In the US about 40% of the energy consumption is household related!

I am happy about the new discussion. Thanks Eduardo.

Anonymous said...

With no evidence, I suspect there are three reasons for most of the high-consumption countries: natural gas pipelines, coal deposits, and nuclear power. All three make energy transport and creation cheap. Cheap energy helps people be paid more at work, and also they can better afford it at home.

Others have pointed out the weather. Cold countries need heating. Quatar can afford cooling.

Marco said...

the one-child policy in China did not start until 1979. Moreover, several exceptions were available, especially to Chinese in rural regions. Finally, life expectancy has risen considerably. Thus, even with fewer births a population can (at least 'temporarily') grow.

eduardo said...

@ 3
Falk, thank you for your explanation.

'the best predictor for energy consumption in the US (and probably throughout the world) is income.'

This is what I would have expected, but not only across countries, or within the US, now, but also through time. That was what surprised me The energy consumption per head seems to vary very little along time, although income certainly does. For some reason, energy efficiency also increases to offset the expected increase of 'gross' energy use.

If this is true - which I do not know - I find intriguing that this offsetting is so perfect. There must be a connection between growth of income and increase in energy efficiency through time.

@ReinerGrundmann said...

I think this can be explained by the Jevons paradox (1865): "It is wholly a confusion of ideas to suppose that the economical use of fuel is equivalent to a diminished consumption. The very contrary is the truth."
Here is an illustaration, taken from an article by Wapner and Willoughby:

"Many environmentalists call for reducing family size and consuming less because they see these as ways to decrease the total amount of spending within an economy. Reduced spending limits the demand for resources and the production of waste, and thus contributes to environmental wellbeing. As we will demonstrate, however, while less spending within an economy may, in fact, minimize environmental harm, it is not the case that cutting back on individual consumption or having a smaller family will reduce overall spending. In most cases, such actions will simply shift the locus of spending. If a family, for example, continues to receive the same amount of income but decides to buy fewer products or have fewer children (and thus forgo the consumption associated with additional family members), they have more savings at their disposal. In fact, much of the literature advocating less consumption and greater consumer simplicity emphasizes the bonus of extra savings. It is crucial to recognize, however, that savings do not simply sit in banks or other financial portfolios; rather, they are deployed by financial institutions to fund investment projects. Consequently, because purchasing power is fungible, it makes little difference ecologically if one saves or invests money rather than spends it. For, aside from placing money under one’s mattress, it will most likely be used by investors to create more economic wealth, and this will be done primarily by funding practices that use resources and create waste."
Paul Wapner and John Willoughby (2006) ‘The irony of environmentalism: the ecological futility and economic necessity of lifestyle change’, Ethics and International Affairs, 19(3): 77-89.

Martin Heimann said...


there is one other factor that is not taken into account when presenting the emissions in this way: "embedded" or "exported" emissions. A substantial fraction of the increase of emissions e.g. in China arose because of the outsourcing of production from Europe and North America to China (and other non-Annex B countries). If one adjusts the numbers based on statistics of the goods transported back to the first world from these countries, the graphics change substantially. Per capita emissions in Europe are then no longer flat... A paper documenting this in detail will soon come out in Science.

Anonymous said...

Bjorn said:

Eduardo, the main driver of energy consumption is simply development. (I'll be curious on how Martin Heimann counts energy consumption per capita to contradict this statement! Hope it will be discussed here?)

I am now referring to one of my favourite booklets. It was written by the Swedisch SIPRI researcher Johann Norberg, "In defense of global capitalism". I believe you would like it, at least Norberg describes the world in a 'honest broker' style.

The simple cause for increasing energy consumption in poor economies is the emergence of a middle class. When you move from the Slum to a real house, you want furniture, a bicycle, maybe even a car, a dish-washer and a fridge, all of which consume energy to produce and to maintain.

In more developed economies, additional growth goes predominantly to the service sector. When there is money over, you will not buy another house, but take your family to a restaurant, a spa, have a short holiday, visit more often your doctor, buy arts, all of which do not need so much additional energy.

The more interesting ratio is hence the energy efficiency measured in kJ of primary energy for each unit of gross domestic product.

I can send you on Monday an analysis I did myself on electricity consumption in Turkey and Germany, two economies that are 3-4 decades apart in development. If I am not mistaken, The McKinsey Global Institute did some deeper analysis into the topic.

Unknown said...

As Falk said, energy consumption within developing countries is very heterogeneous. So it is important to persuade the richer part not to emulate the wasteful style of living in the industrialized countries, but to "leapfrog" to a wiser style.

Though relatively minor, resource use per GDP is not always uniform. I recently heard from a colleague who make projection of water resource demands. (Excuse me, this is not energy resource.) Water resource use per unit industrial production (in the unit same as GDP) in Japan decreased significantly in 1960s and 1970s, but it has not changed significantly since 1980. Though I do not have good data at hand, the saving of water resources occurred together with saving of energy resources. It was not consuming more of one resource to save another.

The following is just my haphazard speculation, not substantiated. Maybe nowadays we use more energy resources trying to save water and we use more water resources trying to save energy, gaining neither....

EliRabett said...

Rates are tricky things

If you look at energy use across countries, the first thing you see is that places with small populations where there are energy extractive industries (oil, gas, coal) such as Qatar, Saudi Arabia, Trinidad and Tobago, etc, are off the top of the chart.

The difference between Wyoming and New York is not immigrants. Things being as they are, Wyoming probably has more immigrants as a fraction of its small population than New York has of its large one, but the fact that there are huge coal mining and gas operations in Wyoming.

What is instructive is that states which have actively worked to reduce energy intensity such as California, have per capita use trends that are below or similar to the EU. Good policy makes a difference without affecting living standards, and indeed can positively influence them.

The trend graph is more an indicator of who is trying than who is growing and you are drawing a false conclusion from it.