Home > Uncategorized > Can a service transition save the planet?

Can a service transition save the planet?

from Blair Fix

Let’s talk sustainability. Unless you’re an anti-science crank, you probably agree that we’ve got a problem with carbon emissions. We need to drastically cut emissions to avoid catastrophic climate change. On this we should all agree.

The question that’s open for debate is how to cut emissions. I think we actually know very little about how do to this. But even worse than knowing little is thinking we know a lot when we don’t. As the old saying goes, “It ain’t what you don’t know that gets you into trouble. It’s what you know for sure that just ain’t so.”

This post is about false solutions to climate change. These are ideas that seem like plausible ways to reduce carbon emissions, but that fall apart when we look at the evidence. If we’re serious about sustainability, we need to abandon ideas that won’t work.

This may seem obvious. But sustainability advocates sometimes forget that we need to test our ideas against evidence. Driven by the urgent need for action, some people latch on to a ‘solution’ and become fervent advocates for this course of action. They forget that their ‘solution’ may be wrong. We owe it to ourselves to keep a cool head and avoid this mistake. Future generations will thank us.

Here I’ll focus on one false solution to climate change. I’m going to debunk the idea that a service transition can reduce carbon emissions.

Why a service transition seems like a good way to reduce carbon emissions

A service transition seems like a good way to reduce emissions. Why? Because services use less energy than industrial sectors. So if we grow the service sector (and shrink industrial sectors), we should reduce our use of energy, and thereby reduce carbon emissions.

On the face of it, this reasoning seems sound. There’s overwhelming evidence that the service sector uses less energy than industry. Figure 1 shows data from the United States. Over the last 20 years, the US service sector used about 4 times less energy per worker than industry:

Figure 1: Energy use per worker in US industry vs. the US service sector. Each ‘box’ indicates the middle 50% of data. The horizontal line indicates the median. The vertical ‘whiskers’ indicate the range of the data. (Source)

Given the evidence in Figure 1, it seems obvious that a service transition should reduce emissions. But it turns out that a service transition actually increases carbon emissions.

Transitioning to services: a recipe for increasing carbon emissions

We’re going to look at how fossil fuel use and carbon emissions change as the service sector grows. To get the widest possible evidence, we’ll look at trends between all the countries of the world.

Let’s start by comparing fossil fuel use per person to the employment share of the service sector (Figure 2). It’s obvious here that a service transition leads to more fossil fuel use:

Figure 2: Fossil fuel use per person vs service sector employment share. Each dot represents a country in a given year. (Source)

Unsurprisingly, carbon emissions per person also increase as service employment grows (Figure 3):

Figure 3: Carbon emissions per person vs service sector employment share. Each dot represents a country in a given year. (Source)

In Figures 2 and 3, we used employment share to measure the size of the service sector. We can also measure the size of the service sector using its share of value added. We still find the same trend. Figure 4 shows that as the service sector’s share of value added grows, so does fossil fuel use:

Figure 4: Fossil fuel use per person vs service sector share of value added. Each dot represents a country in a given year. (Source)

Figure 5 compares carbon emissions to the service sector’s share of value added. Again, no sign of emissions relief with a service transition:

Figure 5: Carbon emissions per person vs service sector share of value added. Each dot represents a country in a given year. (Source)

The evidence is unequivocal. A service transition fails to reduce emissions. The question is — why?

The service transition: what goes wrong?

We have a paradox on our hands. The service sector uses less energy than industry, so it seems like a service transition should reduce carbon emissions. And yet we find that a service transition actually increases emissions. So what goes wrong?

The flaw in our reasoning is subtle, but devastating. We took evidence at a point in time (the lower energy use of services) and inferred a trend over time (that a service transition would reduce emissions). But in our haste, we forgot something key. We forgot about social change!

Here’s the crux of the problem. When societies transition to services, their structure doesn’t remain the same. Instead, the whole society transforms. People move to cities, businesses use different technology, consumption patterns change, and so on.

I could write a book-length post about the social changes that come with a transition to services. But here I’ll boil it down to something simple. A service transition is a recipe for economic growth.

Interestingly, this is not a new idea. More than 70 years ago, Colin Clark argued that a service transition is a key part of economic growth:

[T]he most important concomitant of economic progress [is] the movement of working population from agriculture to manufacture, and from manufacture to commerce and services. (Colin Clark, 1940)

Now here’s the thing about economic growth: it takes energy. Tim Garrett uses the analogy of a growing child. Why do children eat so much? Because they’re growing, and growing takes energy. The same is true of an economy. Producing stuff takes energy. And producing more stuff takes more energy.

Given this fact, we need to rethink the service transition. If a service transition is a recipe for economic growth, it’s also a recipe for using more energy. Let’s look at the evidence.

Figure 6 shows data for the world as a whole. As the service sector grows, so does energy use per person:

Figure 6: World energy use per person vs service sector share of employment. (Source)

Figure 7 shows data for all the countries of the world. Again, as the service sector grows, so does energy use per person:

Figure 7: International energy use per person vs service sector share of employment. Each dot represents a country in a given year. (Source)

Figure 8 shows two centuries of US data. Again, no surprises. As the service sector grows, so does energy use per person:

Figure 8: US energy use per person vs service sector share of employment. (Source)

We even find the same trend within US industry. Figure 9 shows the growth of non-production workers within US industry. These workers do service-type activity, but are employed by industrial firms. As service-type activity grows, so does energy use per person.

Figure 9: Energy use per worker in US industry vs the employment share of non-production workers. (Source)

The evidence is overwhelming. A service transition is a recipe for using more energy. So we should not be surprised that transitioning to services fails to deliver emissions relief.

Let’s have a sense of history

The idea that a service transition will increase sustainability is relative new. It dates at most to the 1970s, when rich countries began to de-industrialize. It seemed, at the time, that something new was happening. But the thing is, it wasn’t the service transition that was new.

Figure 10 shows two centuries of sector change in the United States. What do you notice? I see two constant trends. Agriculture employment decreases with time, while service employment increases. These trends continue uninterrupted for two centuries. In other words, the US has been transitioning to services for over 200 years.

Figure 10: Two centuries of sector change in the United States. (Source)

What changed in the 1970s? It had nothing to do with services. Instead, industrial employment collapsed. And, as many observers have noted, this industrial collapse was caused mostly by the offshoring of industrial jobs. China now does the industrial heavy lifting for much of the world.

Given that the US has been transitioning to services for 200 plus years, we need to have a sense of history. Over the last two centuries, has the United States become more sustainable? No! So let’s not delude ourselves that a service transition will lead to sustainability. It hasn’t and it won’t.

Further Reading

Dematerialization Through Services: Evaluating the Evidence. BioPhysical Economics and Resource Quality. 2019; 4(2):6. Preprint at SocArXiv.

  1. December 4, 2019 at 9:21 pm

    The correlations are interesting but it’s not clear what are the reasons for these trends. Is a shift to the service sector concomitant with an increase in earnings and thus an increase in consumption and hence energy and material cost increases? Perhaps its the increase in AI systems and robotics which then raise the inputs required? (Jevons redux?)

    Andrew McAFEE has published the volume, More with Less, a flawed argument in defense of the techno/capitalist solution to resource depletion with much the same type of analysis but from the material side.

    • December 5, 2019 at 5:43 pm

      I think you need to look at the historical realities rather than the current figures. Seeing it as one who used to cycle through the old farming, cotton industry, coal mining and mercantile shipping centres of Lancashire, I saw hamlets housing local farm workers, local mills powered by first by local rivers and then by local coal, in which, perforce, local people walked to work and produce was trans-shipped in horse-drawn carts and canal boats. By my youth, big factories had been built round the trading centres, and railways constructed to bring the goods, engineers and service workers in. As I have grown older, electric tramways gave way to buses, steam railways were largely replaced by road vehicles and my home city became so big and busy as to make cycling out of it almost impracticable. Post 1980, the industries have closed, so jobs for the small towns are largely service jobs in major-city sky-scrapers, with half the day spent in getting to and from work in cars, while international trade has mushroomed, as have motorways to airports and international/interstate commuting. Meanwhile, in the small towns parents are still trying to keep their houses warm and ferrying children to school because it has become too dangerous to walk.

      Is it any wonder the vastly increased number of people now commuting to centralised service work are creating more emissions than small-town industrial workers used to? Even the electrification of railways and cars will leave mountains of scrap and defile pristine areas by quarrying new types of resources. The long-term solution has to be to relocalise living, with scrap cars turned into light railways and city-centre sky-scrapers perhaps pulled down to provide rubble for building sea walls, to keep out sea waters rising as the glaciers melt.

    December 4, 2019 at 10:30 pm

    A very interesting post that in my opinion makes a lot of sense. A comment I would like to make is that getting hysterical about fossil fuels, overlooks facts that modern technowledgey has greatly reduced carbon emissions from coal fired power stations. The other important part of the conversation that is often ignored is the importance of carbon absorption methods. Is planting and nurturing the right kind of trees in the right place, where they can do the greatest good. Ted

  3. December 4, 2019 at 11:06 pm

    There is a GREAT DEAL we know about how to reduce emissions. Why is this misleading claim being posted AGAIN?

    Don’t look in the economics literature for what to do, go and educate yourself about cheap clean energy, pumped hydro storage, regenerative farming, energy end-use efficiency, (clean) hydrogen fuel, on and on.

    I’m sorry, but it’s typical of too much economics to talk about a ‘service economy’ as though no one has to grow stuff and make stuff. Get real.

  4. John deChadenedes
    December 5, 2019 at 1:11 am

    Let’s imagine a transition to an economy where all physical production is done in fully robotic factories powered by coal-fueled power plants. Except for the few people who run the factories, everybody else serves food, polishes shoes, cleans toilets, or provides other services useful to those who can afford them. We would see an amazingly strong correlation there between increasing (and ultimately disastrous) carbon emissions and transition to a “service economy”. Gee….what could possibly have gone wrong?

  5. December 5, 2019 at 2:25 pm

    Richard Smith, in the WEA volume Green Capitalism, eloquently shows that dematerialization and shifts in technology can’t reduce the GHG’s due to energy production without considering all other sources of GHG’s. And that holds for other types of a per capita life style, particularly in a global society with developed world aspirations. This makes me curious as to what underlies the “correlations” in this article.

    The idea of a shift to a service economy and the idea of a “basic income” have similar issues. Smith’s argument that most of the thinking, including the comments to the article are trying to effect the needed shift by staying within a traditional capitalist framework. His argument needs addressment. That it isn’t is concerning with the number of persons in WEA who have a proclivity towards heterodox thinking.

    The recent efforts of the car companies to join the US gov’t to over-ride the California emission standards (though they currently meet these for the most part) is a paradigmatic example that the issue is not whether there are or can be the technologies but rather the larger political and social issues of a capitalist doxa. It’s similar to trying to fit the ugly sisters’ feet into Cinderella’s slipper.

    • December 5, 2019 at 4:42 pm

      We need to address “Smith’s argument that most of the thinking [is] trying to effect the needed shift by staying within a traditional capitalist framework”.

      Mine doesn’t. Capitalism is about investing in mass production and international distribution of surplus for profit. Mine is about mass production and transportation only where needed, motivated with honorary rewards for good work, financed with credit repaid by doing what is needed to care for ourselves, our societies, our infrastructure and our environment, keeping production and maintenance work local where practicable so commuting is minimised and people learn and get satisfaction from cooperatively doing interesting jobs. More or less E F Schumacher’s “Small is Beautiful” with the money issues sorted out on lines suggested by John Ruskin, Frederick Soddy and the modern understanding of control theory. The needed shift is to “honest money”, i.e. maintaining our own credit worthiness rather than allowing accumulation of banker’s fraudulent “money made from nothing”. Not at all capitalist!

      • December 5, 2019 at 10:14 pm

        I expect that Smith would not disagree. The key is exactly what would be needed to allow your vision in this and your above response to materialize. Bernie Sanders’ “social democracy” or the alternative of democratic socialism point to possible paths. Much of this might be seen in, for example, the Scandinavian countries or other parts of the world where the “Wall Street” driven, globalization, path moves money and goods internationally but restricts the movement of people to the detriment of the “local” which you so well describe. In other words, the techno/capital driven economies, defended by McAFEE and the Mont Pelerin membership have failed to demonstrate that a “capitalist” frame defines a workable boundary in which to construct a path forward.

        That concern is what I had thought was one of the conclusions that gave rise to the Heterodox school of economics. What I see, to a certain extent, are the problems with the publish/perish world of academia, the spirit and thinking of which has a defaulted to legitimize a techno/futurists path to ecological redemption while trying not to color outside conventional economic “wisdom”.

  6. Ikonoclast
    December 5, 2019 at 8:40 pm

    We are in the sixth mass extinction event. This means we are nearly at the beginning (in evolutionary time-spans) of the next mass selection and life-diversification event IF the biosphere proves capable of self-rehabilitating and generating another explosion of life in the next x million years.

    Doctor Ian Malcolm, in Jurassic Park, uttered the non sequiter that dinosaurs had been selected for extinction. It is not possible to be “selected for extinction” as only a survival event is a selection event. But Doctor Malcolm’s non sequiter is quite amusing and we kind of know what he means. In Malcolm-speak, humanity is about to be selected for extinction.

    Why is this so? It seems that we are in the opposite of a Goldilocks’s zone with respect to our intelligence. To survive in the future, we will need to be considerably stupider than we are now or considerably more intelligent than we are now. At our current level of intelligence, we are just intelligent enough to get ourselves into a lot of stupid trouble.

    I expect humanity has three obvious evolutionary possibilities from this point. Extinction, intelligence decline or intelligence rise. Those possibilities are in given in order of probability. It is most likely that we will go extinct. It is least likely that we will survive and evolve to higher intelligence and further cultural complexity. If we do survive as a species, let us hope we evolve in intelligence and retain the wisdom in our cultural memory that the anti-empirical ideology of conventional neoclassical economics played a huge role in trashing our biosphere and leading to the sixth mass extinction event. A future more intelligent species would execrate the stupidity of conventional neoclassical economics .

  7. ghholtham
    December 6, 2019 at 11:03 pm

    Neo-classical economics is a severely limited research strategy that has been over-billed, over-used and extended into areas where it is not illuminating. Some of its precepts have been pushed for political/ideological reasons and caused real damage.
    The growth of world populations and mankind’s increasing ability to harness hydrocarbon-based energy sources threatens to destabilise the climate and cause ecological devastation.
    Still, asserting “the anti-empirical ideology of conventional neoclassical economics played a huge role in trashing our biosphere” is hysterical . Oil majors do not explore in the Artic and people do not drive cars around because they have read economics text books. The biosphere would be under pressure even if no economics book had ever been written. Keep a sense of proportion.

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.