Recently, I visited the Moroto District in north-east Uganda, home of the (in)famous Karamajong pastoralist. These number half a million people and are isolated geographically, economically and politically, and are widely despised by their compatriots as violent and underdeveloped. There have been efforts to settle the Karamajong in villages, get children to school and make them grow crops. However most of them are dependent on various development programs as well as food aid from the World Food Programme .
I visited a village where the normal kinds of development interventions were taking place. One intervention seemed to be successful; the construction of wood saving stoves. Such stoves are darlings of the development community and can save at least half of the firewood. Clearly a good thing; and they were in use. When I asked a woman if she now didn’t have to collect so much firewood every morning, she said, “oh, I collect the same amount of wood as before, I just sell the wood I don’t need”. This was ironic as the reason for the introduction of wood-saving stoves is to save trees, not generate income.
This is one of many examples of Jevons paradox formulated by English economist William Stanley Jevons in his 1865 book The Coal Question. He observed that England’s consumption of coal soared after James Watt introduced his coal-fired steam engine, which greatly improved the efficiency of Thomas Newcomen’s earlier design. Watt’s innovations made coal a more cost-effective power source, leading to the increased use of the steam engine in a wide range of industries. This in turn increased total coal consumption, even as the amount of coal required for any particular application fell. Jevons argued that improvements in fuel efficiency tend to increase, rather than decrease, fuel use: “It is a confusion of ideas to suppose that the economical use of fuel is equivalent to diminished consumption. The very contrary is the truth….no one must suppose that coal thus saved is spared-it is only saved from one use to be employed in others”. Which seems to describe very accurately the effect of the wood-saving stove in Moroto.
With the advent of petroleum, Jevons paradox fell into oblivion until the first oil price shock and the emerging environmental discourse  in the early 1970s. When you improve efficiency, say improve fuel efficiency in a car, it leads mainly to that people drive more as the cost goes down. Further, it is not only the same drivers that drive more, but more people buy a car and use it instead of going by foot, bicycle or train. On yet another level, the money saved on buying gasoline is used on some other activity which equally is resource demanding, e.g. building a bigger house, take a flight for holiday or just consume more “stuff”. This is referred to as a rebound effect. There are some that see evidence of that total resource consumption increases as a result of improved efficiency .
If we compare efficiency on various systems, e.g. in farming or food processing, it will in most cases show that the bigger and more technological advanced system is more efficient. Larger crop farms perform better financially, on average, than smaller farms. The larger farms don’t have higher revenue or yields per acre, but they simply have lower costs. As expressed by a report (Farm Size and the Organization of U.S. Crop Farming) from USDA: “larger farms appear to be able to realize more production per unit of labor and capital. These ﬁnancial advantages have persisted over time, which suggests that shifts of production to larger crop farms will likely continue in the future.” Their yield per acre is mostly the same as on smaller farms but the research shows that farms with more than 2,000 acres spend 2.7 hours of work per acre of corn and have cost for equipment of $432, while a farmer with 100-249 acres will spend more than four times as much labor and double the amount for equipment per acre. In that sense the larger farms are more “efficient” or “productive”.
The same goes for a farmer who drives his pickup to the farmers market compared to the lorries supplying the supermarkets; she will use more fuel and more machine capital per kilogram (kg) of goods. And embedded in the machine capital are many other resources, metals, more energy and other peoples’ work. But despite all this efficiency our society neither reduces the number of hours worked nor the resources used, not in total and not per capita. This is not even the case for societies that have moved towards more services, as agriculture and manufacturing declines. How come?
There are several ways of tackling this question. In an article  in the Journal of Cleaner Production, Blake Alcott looks critically at the claim that there is less impact from people employed in the service sector than in manufacturing. He says that this claim loses its validity if the full resource use of the workers is taken into account. If we only look at the labor it is quite evident that a hairdresser uses less resources per hour than a car maker. But the barber will use his money earned for buying the same kind of stuff as the car maker, so the resource use embedded in their work hours is more or less the same. Well, the car maker probably earns more, so in that sense she will use more resources. But, on the other hand, if the service job is in real estate or finances the service worker will earn more, and thus, on average use up more resources. With this perspective it is the total resource use for a human being that is of relevance and not how many barrels of oil he or she use in the workplace.
Others see that it is mainly the inherent forces of capitalism, i.e. profit and capital accumulation, which inevitably leads efficiencies being exchanged for expansion. John Bellamy Foster, Brett Clark and Richard York write in the Monthly Review  that: “An economic system devoted to profits, accumulation, and economic expansion without end will tend to use any efficiency gains or cost reductions to expand the overall scale of production … Conservation in the aggregate is impossible for capitalism, however much the output /input ratio may be increased in the engineering of a given product. This is because all savings tend to spur further capital formation….”
As yet another perspective, try this:
If we compare the resource use of a big, highly mechanized farmer with a small-scale farmer, we have ascertained that per kg harvested yield, the labor efficiency of the bigger farm is higher. This is also the case for use of most other resources for area unit. But what happens if we look at resource use per labor-hour? Then it is clear that the big farmer in his 400 hp tractor uses an awful lot more resources than the farmer with a small tractor, or oxen, not to speak about the half a billion farmers still working with their own labor as the main resource. The same goes for the driver of the delivery truck to Walmart, he uses a lot more resources per hour than the farmer loading her pickup to drive to the market.
Now, you could say that nature doesn’t care about this discussion, if we are efficient per hour, per kg or per acre; nature only cares about the absolute use of resources or the total emissions. That is correct. But almost all people have a job of some kind, and in each job the same logic applies, i.e. that the more efficient each person is, he or she uses less resources per produced unit but more resources per hour of labor . The total resource use in society is thus bound to increase despite, or perhaps because of, increased labor efficiency. This is just another way of looking at the same things as Alcott does. He looks into the embedded consumption which follows a person regardless of occupation, while here I look more into embedded resource use per hour of work.
After all, as long as we all continue to work so much, our total resource use is determined by how much resources we use at work and how much we use as consumers combined. There are no free lunches. Or rather the problem is that we have got so many free lunches in the shape of “natural capital” that we have used “for free”, that we believe that we “have the right” to use so many resources, and that the lunch will be free also in the future. But it will not.
 There have been floods, fighting and a number of other reasons for their precarious situation, but the important thing in this article is not to give a complete picture of the fate of the Karamajong. That merits a separate article.
 Such as the Limits to Growth, from the Club of Rome.
 Jevons himself saw that for steam engines. After all the first ones were rather useless and were thus not used much. As their efficiency increased they spread all through the economy.
 Mill’s scissors: structural change and the natural-resource inputs to labor, Journal of Cleaner Production 21 (2012) 83-92
 Capitalism and the curse of energy efficiency: the Return of Jevons Paradox, Monthly Review, 2010/11/01
 It is likely that there are some exceptions to this, but I believe that they are just that, exceptions.