Decoupling what!?

Adding some perspectives to the decoupling debate

I suppose you are aware of the discussion on decoupling, which is if it is possible to decouple economic output from the ecological footprint. There is absolute decoupling which means that the ecological footprint is shrinking while the economy is growing and there is relative decoupling which means that the ecological footprint per economic output unit is shrinking. Often, the decoupling discourse is limited to ”carbon dioxide emissions”, but it can also be about material use in the economy, impact on biodiverstity, or the composite footprint index.

I want to add some nuances and perspectives on that topic. Contrary to what I mostly do, I will not use a lot of references and links to my statements. That doesn’t mean I can’t find any supporting facts, just that I am busy and it does take a lot of extra time. In addition, sometimes the wealth of references and figures can make the reasoning disappear in a haze of data, both for me and the reader.

There are many issues to consider here both regarding the numerator (the economic output) and the denominator (the ecological footprint) as well as the scale, the time and the boundaries applied.

Undoubtedly, there is quite some reliable statistics showing that carbon dioxide emissions in a number of countries have gone down, while the Gross Domestic Product, the most common measure of economic output, has gone down in the last decades. Sweden, where I live is one such country.

Let me first state clearly that I can find no theoretical basis for why carbon dioxide emissions should have to increase with increasing economic output. For sure, one could believe so if just looking to the data from a global perspective and from most of the epoch of industrial capitalism. But it is possible to “decarbonize” economies to a large extent (how far is another question), with hydropower, nuclear, solar, wind and biomass. New energy forms often add to the old ones rather than replacing them, but it is not an iron rule. For example in Sweden, the lower carbon dioxide emissions are to a large extent attributable to a massive switch from oil to biomass for heating, starting already in the seventies. It is even possible to “decarbonise” by substituting one fossil fuel for another. A substantial share of the reduction of carbon dioxide emissions in many economies are due to that coal fired power stations have been replaced with gas fired ditto.

For sure, there are issues around to what extent a total shift to renewables would at all allow for the same kind of economy, and growth, as the prevailing one. I also have my doubts about that. And it is also true that almost all wind and solar equipment and construction is based on fossil-fuelled production and will be so for decades to come. But this is another discussion I will not enter today.

I can find no theoretical basis for why carbon dioxide emissions should have to increase with increasing economic output. I believe the link between energy and economic output is much stronger.

I believe the link between energy and economic output is much stronger. But even with energy and not carbon dioxide as a measure, it is apparent that there are some countries that have managed to grow the economy substantially without increasing the energy consumption at al. Again, Sweden is such a country. But the use of materials in the Swedish economy has increased substantially at the same time.

ADDED 11 jan: In the context of energy, it is worth noting that in a period after WWII oils became so cheap that there was an enormous waste. For the first time in human history, energy was really superabundant. This changed with the oil crises of the seventies. After that, energy efficiency became much more prominent both in production and consumption stages. In the Soviet Empire, oil and gas remained superabundant until the collapse and anybody who visited those countries before 1990 can testify how much energy was wasted. This development is also visible with the use of nitrogen fertilizer a very energy intensive input in agriculture. In the advanced economies the use peaked in the seventies and in the Soviet Union at the time of the collapse. This also means that there is no linear relationship with energy input and economic output. But it also means that some of the decoupling that can be observed can be attributed to the reduction of a ridiculous waste. The proponents of the market economy will of course see this as an indicator of that “the market works”. END

Sometimes decoupling is discussed in various sectors of the economy and then decoupling often boils down to gains in efficiency. Undoubtedly, for most production, the competition and the profit motive makes increased efficiency a no-brainer. Historically, it mostly meant less labour per produced unit, accomplished through a higher degree of mechanisation, i.e. more use of energy. In a more mature situation, increased efficiency is often about saving materials and energy, in the production. Almost all industries can show impressive such efficiency gains. But efficiency gains in a production process or a whole sector doesn’t mean that the total economy will use less. There are two reasons for this. One is the rebound effect (often referred to as Jevon’s paradox) which means that increased efficiency in one process will reduce the cost (that is the whole point of it from the producer’s side) and thus the price. This will increase the consumption, so the total resource consumption might end up the same or even higher. Perhaps even more important are the effects on a system level: If the consumption doesn’t increase, there will still be an increased production and consumption space in the economy as a whole, as resources that are freed up will be deployed elsewhere.

Population is of course also very relevant: It makes a difference if we compare footprint per capita or total for a country and if we compare footprint per GDP for a country or for its inhabitants.

Another aspect of decoupling is the question of what system is studied and its boundary. Most statistics are national and sometimes these are added up to get global figures. But economic output from the production processes as measured by the GDP is in many cases not national but comes from global value chains. In global value chains there is a mismatch between where the value added is generated and where the resources are used. Mostly, there is a reversed relationship so that most value is added where the resource use is lowest and little value is added where most of the resources are used.

In global value chains there is is a reversed relationship between value added and resource use so that most value is added where the resource use is lowest and little value is added where most of the resources are used.

I guess you have seen a famous graph showing that 58 % of the income of the iphone went to Apple in the US, even if the whole production process takes place in other countries (mainly China). This example is often used for demonstrating how unfair the world is. But it also serves as a good illustration of what I said above, because most of the emissions and other environmental effects are linked to the parts of the process where the contribution to GDP is the lowest. This is certainly nothing special for Apple, it is the normal thing for the most global value chains.

For countries with a very high share of such value chains in the economy, for example most EU countries, ecological footprint, carbon emissions, energy use in relationship to the national GDP is grossly misleading. Notably, it also means that consumption based emission figures are misleading. A value based accounting shows that Germany has 40% higher resource use than a consumption based accounting. It is a bit similar when people compare the carbon footprint of people in cities and rural areas. Such comparisons fail to take into account that many (most) people in rural areas are there to services people from the cities, farms and all associated business, mines, people in the tourism industry etc and there is no meaningful way of separating their environmental footprint.

What is economic output anyway?

Far too seldom the measure of economic output is discussed in relation to decoupling. The standard measure of economic output is the GDP. It adds up all the value added in the economic processes in a country (it can be calculated in three different ways but they all give the same result). It might sound like it is quite straight forward, but it isn’t really. I will not expand of all the complication in this but let me mention a few:

In order to compare GDP growth over time, a GDP deflator is applied. It will adjust the GDP for inflation. But it also adjust the GDP for quality changes. It is difficult to compare the price of a mobile phone or a computer today with one twenty years ago as the performance is very different. Through the use of the deflator for quality differences the GDP grows quicker than just based on prices, corrected for inflation. You can read more about it here. This is of course quite a difficult thing to do and involves a high degree of subjectivity. I mostly use my computer as a word processor. The fact that the computer I am using now is hundred times more powerful than the one I used back in the nineties, doesn’t mean that the utility of it for me is hundred times higher, not even double I would say (and I use a stone age Nokia phone).

The GDP doesn’t include work in household or other kinds of voluntary work. If families take care of their children, elderly and sick and if they cook their food from scratch, there is no contribution to the GDP, but if these activities are moved into commercial operation or government operations the GDP will grow considerably. Clearly that has been such a development in most advanced economies. (Illegal or informal economic activities are also managed differently. Italy’s GDP bumped up a lot when they started to include them).

What do you get for the money? As noted you get a lot more (better) phone for your money today than twenty years ago. But in many cases you actually don’t get much at all from the GDP. It is not only the fact that car accidents and natural disasters mostly add to the GDP, it is also the fact that an ever increasing share of the GDP is generated in services of no apparent use for society at large. Insurance, real estate services and financial services make up a big share of the GDP, more than 20 percent in the US. While they might be useful for the individuals using them, it is hard to see that they add any real value to the economy as a whole.

There are many services contributing to the GDP without any apparent utility.

All those services are just moving money between people, and such transactions are in other cases excluded from GDP accounts (when you sell a house to someone, it is only the commission to the real estate agent and any fees of the bank involved that impacts the GDP, while the price of the house has no impact). The health service in the USA is notoriously expensive (and thus contributes a lot to the GDP) but it is hard to see that it result in more health to the people than the countries with much lower health care cost. The renewed arms-race is also a massive contribution to the GDP. There are also all those “bullshit jobs” that David Graeber talks about. Added to this is of course also the debate about whether “we” will be happier by economic growth, but that is a bit outside the decoupling discussion.

Tim Morgan makes the claim that the real prosperity per capita of the “advanced economies” started to go down already some 25 years ago, despite continued economic growth. That seems to coincide with the feeling of many that the young today has got a rough deal compared to their parents. Linked to this is also the notion that the GDP growth today is driven by debt (be it governmental or private), that in some way the GDP is inflated by debt. I must admit that I have not really got my head around that reasoning. If I borrow money and buy a new tractor, clearly that contributes to the GDP seen in isolation. But, in the aggregate, the GDP is made up of labour and capital (not money) and it is hard to see that the availability of those are impacted by debt. The many failed attempts to grow the GDP by government spending seem to give evidence to the contrary. Perhaps the debt induced growth is only about those services which themselves are about debt, i.e. financial services and alike? Overall, the link between the financial sector and the real economy seems like something to investigate further.

Adding up, when discussing decoupling, it is extremely important to look into the definitions, the boundaries and the evidence presented – regardless what message they have – in a very critical way. In my current book project, I will explore the track of disaggregating the GDP to better understand its components and if there is indeed a direct link between the use of biophysical resources and energy to the economy.