Hacker News Reader

> It takes more energy to produce more stuff

Energy consumption is also about using that stuff. And in rich countries like the USA, a lot of energy goes to using things or just moving people. So it's possible to build more and use less energy, even if you don't reduce the energy cost of building things. They can simply become more efficient to use and move.

In USA, 30% is just homes and commercial sector. 40% is transportation. 30% is "industry". By a sort-of inverse-amdahl's law, it's possible to get lower energy use with more throughput even if we don't make "industry" more efficient.

>So it's possible to build more and use less energy, even if you don't reduce the energy cost of building things.

In principle I can imagine that being true - but that's not represented in our world, so it is an open question if you can 'build more and use less energy' in the real world.

It's represented everywhere that efficiency counts. I haven't fact-checked it, but one source[0] claims that refinements in the design of soda cans (thinner walls, different shapes) since 1960 saves "at least 90 million kilograms of aluminum annually."

[0] https://www.youtube.com/watch?v=hUhisi2FBuw

Exactly!! There a uncountable millions of examples, and you're right, anywhere efficiency counts, which is pretty much everywhere.

It's also irrelevant as anything other than a limiting case. The issue is with the statement that energy use must go up to produce more GDP or that not increasing energy use means we cannot raise GDP or even that GDP is limited by a flat energy use.

USA GDP is still on basically the same exponential curve, yet energy is flat. That's the counter example. The rest is exposition.

It's really really not, you can see it in the data. I took GDP per capita and kg oil equivalent energy consumption per capita and in 1970 the very year the article highlights as us falling off the HA curve and representing 'lost' GDP the exact opposite happens. We begin producing way more GDP per capita using the same unit of energy.

It's basically a piece wise function if you graph it. 1960 to 1970 the GPD/energy unit ratio is largely stable then after that it begins increasing monotonically going from .69 to 7.94.

https://news.ycombinator.com/item?id=41392222

I will note that the minimum energy needed to move people, as dictated by the laws of physics, is zero. There is no net change in average gravitational potential energy per capita, and any energy that goes into kinetic energy can in principle be recovered with arbitrarily high efficiency.

There are losses in practice, of course, to friction, air resistance, etc. but the laws of physics don't impose any lower bound on these.

> Jevons Paradox, named for the 19th- century English economist William Stanley Jevons, who noticed that as steam engines became ever more efficient, Britain’s appetite for coal increased rather than decreased.

> Yet the amount of electricity we consume for light globally is roughly the same today as it was in 2010. That’s partly because of population and economic growth in the developing world. But another big reason is there on the Las Vegas Strip: Instead of merely replacing our existing bulbs with LED alternatives, we have come up with ever more extravagant uses for these ever-cheaper lights

NYT: The Paradox Holding Back the Clean Energy Revolution

https://archive.is/rb2DX

> To light the factories, more energy used to correlate with more light, until we swap out the incandescent/halogen lighting sources for LEDs

this is nonsense. lighting hasn't been a significant fraction of the energy usage of factories since they switched from being lighted by fireplaces to gaslighting. not in 02024, not in 01974, not in 01924, not in 01874

> Seems that measuring [gdp] growth by energy consumption is like Bill Gates' famous example

it's true that higher efficiency is better, of course, but your comment embeds the false assumption that higher energy efficiency reduces energy use. in fact, higher energy efficiency usually increases energy use, because it increases the scope of things to which marketed energy can be economically applied more than it reduces the use of marketed energy for things it was already being used for. (this is the well-known jevons paradox mentioned in https://news.ycombinator.com/item?id=41392248). so, even today, it turns out that the countries with the lowest gdp and lowest energy use also have the lowest energy efficiency

similarly, using high-level languages reduces the number of lines of code to implement some given functionality; but you would be completely mistaken if you used that fact to predict that the vast majority of programmers spend their time writing assembly language instead of python because python requires one twentieth of the code to do whatever. many things are done with python not just because companies writing python outcompete companies writing assembly, but also because programs that would be unprofitable to write in assembly language become profitable to write when you can write them in high-level languages!

>> lighting hasn't been a significant fraction of the energy usage of factories

Right, but it is an easy example to explain at the outset and why in the literal next sentence I mentioned: "Same for more efficient motors, swapping ovens for inductive heating, more efficient processes, etc."

>> embeds the false assumption that higher energy efficiency reduces energy use

NO. I am describing a COMPLETELY different effect. I am not talking about marketing anything, but the results of improved process efficiencies.

If Company X, or an entire industry, switch from large inefficient ovens to spot-heating, and still produce the same number of widgets for 10% of the energy, the energy used has entirely decoupled from economic output. Same GDP, lower energy consumption. Or, if the new widgets are cheaper and more get sold, then greater GDP and smaller energy consumption number.

The point is, regardless of all the knock-on and sometimes paradoxical effects, the previously strong 1:1 correlation of energy use to GDP starts to break when efficiency-enabling technologies gain widespread adoption.

There are also paradoxical questions such as what happens when Company X switches entirely from oil heat and electricity to on-site solar & batteries, producing the same number of widgets using the same total energy? The GDP seems to go down because oil is no longer being pumped, transported, refined, and transported again to Company X's factories, but the output is the same, so the same net goods flow into the economy. The entire enterprise of getting refined oil to Company X is now shown to be redundant to production. But also, the energy usage as measured at the grid went to zero.

Again, the point is, the not only is the relationship of energy-GDP unstable, the entire measurement of GDP is problematic.

> Bill Gates' famous example of saying that "measuring software progress

Oh the irony.