Exponential Economist Meets Finite Physicist (2012)
dothemath.ucsd.edu
dothemath.ucsd.edu
Physicist: "Right, if you plot the U.S. energy consumption in all forms from 1650 until now, you see a phenomenally faithful exponential at about 3% per year over that whole span."
It's not phenomenally faithful. There's an inflection point visible even in his large graph going back all the way to 1650. US primary energy consumption in 1977 was 78 quadrillion BTU (quads). At 2.9% annual growth it would have risen to 245 quads by 2017. Actual US primary energy consumption in 2017 was only 98 quads. The per capita primary energy consumption in the US was actually higher in 1977.
Physicists don't need to imagine future limits to growth. The limits are already visible in the historical record. But, contra the fears of many scientists circa 1960, the limits to growth showed up on the demand side before the supply side. We ended up with a world where widespread obesity is a problem and predicted gigadeaths from starvation did not actually happen.
For any product you can imagine consuming, there are "obesity"-type limits to how much more of that same product you can consume before the marginal utility goes negative. For any service you might use, similar limits apply since there are only 24 hours in a day.
The physicist character gives the more correct answer but his supporting evidence is flawed. The economist has the much worse answers, but only (I suspect) because he's a strawman constructed for didactic purposes. Economists generally don't make 1400 year forecasts of any sort.
And I think some people were... "disappointed", for lack of a better word, that this happened. That their fantasies of global collapse Soylent Green-style didn't pan out, that the lackadaisical masses who had the temerity to not give up all modern advances weren't punished by the Dread God Malthus. Call it Puritanical Luddism, call it Apocalyptic Deep Green, call it a dogmatic insistence that any system not under totalitarian control must collapse.
And here we are, not even 20 years later, and America could potentially become a net exporter of oil. economics is hard because supply and demand are incredibly dynamic, and react far quicker than anyone seems to think.
That said, a non-Malthusian ending does not mean a happy ending, or even a better world than previously existed. Huge (long overdue, IMHO) investments in the commodity markets in 2006-2009 are going to keep us going for another few decades until another cycle of underinvestment bites us again. However, each time the invisible hand flings money from one pot into another there is uncounted costs in disrupted lives, environmental impacts, and national upheaval. Capitalism is far from perfect, but--at least in terms of resource allocation in dynamic information-starved systems--its the least imperfect solution we've got.
Not to mention capitalism is an umbrella term with many variants making it a works claim. As if you said "the best we have is cars".
That's part of the strength of Capitalism: You can tweak it, modify the regulatory regime, and it's still Capitalism. You don't need a revolution to modify some deep theory in order to improve how you regulate a Capitalist economy, or to adopt some ideas from the Socialist platform, like you would to modify a specific Socialist ideology.
King Hubbert was laughed at when he predicted in 1956 that USA oil production would peak in a decade or so. It did.
http://www.hubbertpeak.com/hubbert/1956/1956.pdf
In 1998 Campbell & Laherrère predicted "The End of Cheap Oil" before 2010, due to the peak of conventional oil.
http://www.oilcrisis.com/campbell/endofcheapoil.pdf
The International Energy Agency called them "pessimists" :
http://petrole.blog.lemonde.fr/files/2011/04/weo1998.1302253...
In 1998-1999 the price of an oil barrel fell from $30 to $20.
In 2008 it reached $160.
These days it fluctuates between $40 and $70.
https://www.macrotrends.net/1369/crude-oil-price-history-cha...
In 2010 the same IEA said that conventional oil production had peaked in 2006.
https://www.iea.org/publications/freepublications/publicatio...
And while the USA is fairly close to be a [net exporter of liquid hydrocarbons](https://www.forbes.com/sites/rrapier/2018/12/09/no-the-u-s-i...) (not crude oil), the future of tight oil doesn't look good - Laherrere seems to forecast peak tight oil for 2018 :
https://aspofrance.files.wordpress.com/2018/03/lahall19march...
(it takes a while to collect the data...)
Sure, maybe we'll figure out how to economically extract even less conventional hydrocarbons (like shale oil = kerogen, which you have to "cook" before it can be used), but these will be even more expensive and polluting !
Naturally, the most rigorous analyses are going to be heavily weighted toward historical data, but will carefully ignore predicting future breakthroughs and paradigm shifts. Any analysis that tries to highlight humanity's adaptive ability would be laughably ignored, or at least minimized as being overly speculative.
So here we sit, with our broken pessimistic models, because they're the only ones the data supports.
He imagined that the world might come to demand 20 kilowatts per capita (more than twice the present per capita use in the US) and have 20 billion people living in it. Neither population growth nor energy consumption per capita growth have kept up with his estimates. Accordingly, nuclear fuel reprocessing and the breeder reactor alike have remained niche technologies without broad industrial impact. There was never enough uranium consumption growth to economically justify complicated reactors or fuel cycles.
It is difficult to make predictions, especially about the future.
https://xkcd.com/793/ (scroll over text is on point here)
Especially when they get older, they try to apply the 'make a cow a sphere of water, one meter in radius' trick all sorts of other areas. The issue is that they can't deal with 'history' in real life. Most of life is chaotic and the initial conditions and stochastic nature of the world make it almost impossible to predict anything. Not all things can be made into neat theories, in fact, it's a miracle that anything does fit into neat little theories. Physicists don't like that. Typically, they start throwing around all sorts of math-y sounding words when they realize that their 'model' won't work at all (Oh, just take the Hamiltonian, that'll solve it). If that still doesn't work, they make an excuse to go to a bar as soon as possible.
Not if products are finanicalized (like equity, debt or land ownership). There's no obesity-type limit to net worth of an individual and that money allows the person to have unprecedented power to exploit actual physical resources and energy. Some, like Bezos for his Blue Origin project, decide to do so.
> We ended up with a world where widespread obesity is a problem and predicted gigadeaths from starvation did not actually happen.
This is looking at a very specific metric of human well-being. The limits of nature assert themselves in various ways. The loss of biodiversity and climate change is directly attributable to economic growth. https://www.theguardian.com/commentisfree/2019/apr/25/capita...
I think its legitimate to draw a trend line ignoring the kink in the last 10% of the graph. I mean yes we could now be on a different trajectory, but that's just 'this time it's different' thinking, that's not to say it isn't a legitimate discussion, just that its a separate discussion.
No energy growth was anyway covered in a later 'course'.
Per capita energy use seems to be declining in advanced economies[1].
> But if energy became arbitrarily cheap, someone could buy all of it, and suddenly the activities that comprise the economy would grind to a halt...There will be a floor to how low energy prices can go as a fraction of GDP.
That seems to be a bizarre understanding of how the economy works. Advances in efficiency or energy production aren't going to be stopped at some point because people are worried about someone buying up all the energy. You're not going to get an inventor to say, "Hey, this new product would be too efficient, I'm not going to introduce it to the market because then energy is going to get so cheap that someone will buy it all up."
And someone could already try committing energy blackmail now if they really wanted. Pepco powers Washington, D.C. and parts of Maryland, and Exelon bought them for $6.8 billion. Some one, or a group of people could pay that and try to blackmail the government. Of course, they wouldn't get anything for it but a quick trip to prison.
[1] https://data.worldbank.org/indicator/EG.USE.PCAP.KG.OE?locat...
If I buy my bread from a baker, I can say I don't need any energy to run an oven anymore, but that doesn't imply that when the baker gets as wealthy as I am the world won't need ovens anymore.
CO2 emissions are a decent proxy for primary energy in our present fossil-dominated era, and when accounting for CO2 embodied in imports, the US does look worse. But it doesn't make a big difference:
https://www.carbonbrief.org/mapped-worlds-largest-co2-import...
"Despite the large total of CO2 imports and exports, US emissions are only 6% higher and Chinese emissions are 13% lower when CO2 transfers are taken into account."
If the energy/CO2 really should be increased 6% to account for imports, that would put American energy consumption at 6804 * 1.06 = 7212 kgoe per capita in 2015, still lower than in the 1970s.
Anyway, it's great when you compare yourself to second worst.
That wasn’t really his point. His point was that even if energy itself is cheap there’s a floor to how it is valued relative to the total economy for such a vital commodity. If you don’t like the buying it up argument you can also think along the line of delivery cost, capital cost etc. Bottled waters are not free, e.g.
I don't think that's true though. There are vital commodities that have been effectively free in some societies (water and salt, for example). Even energy - wind energy is free and some societies used it extensively, and biomass for burning was abundant in some places.
We can argue whether or not it's likely that energy would get to be so cheap in the future, but there doesn't seem to be much basis for his claim that it's economically impossible.
I think the easier arguments are that transformational technologies like 1000x improvements can dramatically extend growth, even if they don't stop it, and that transformational energy technologies that don't rely on the sun would frankly imply starflight. So the timelines might be off by factors of ten, and in such timelines these things don't feel so necessarily absurd.
If you're seriously at the technological stage where you can put computers with 1000x more power than ours running AI and using 1000x as energy dense of technology, expanding beyond the solar system seems pretty approximately the right place to be going.
Just unlikely. And finite. Even a very long time is finite. Perhaps this is where the economist and physicist differ most.
Price is not universal indicator of value, neither being cheap implies that you can't run out of it. Many Indian cities are quickly running out of fresh ground water. The worst thing a city could do is let market determine the price so rich could use it more. Rationing at a less price is the way municipalities are going, which is more ethical way of doing things.
The same could happen for energy or other resources, while they still being cheap.
Yeah we moved a bunch of heavy industry to China and switched to low carbon finance.
Right. Things only become arbitrary cheap because there is a practically infinite supply of them.
Even for oil the supply is too diverse nowadays, and we're rapidly becoming much less dependent on oil than we were. Even if you could buy a plurality of the gulf oil, what about Russia, shale oil, etc? How do you also dominate coal, solar, wind, nuclear, etc? They are too varied, too geographically and politically diversified.
For example, he first insists that the conversation about energy consumption be confined to Earth and ignore space exploration, which is reasonably sensible because of the large energy cost of space travel. He moves from there to a thermodynamic bound on energy dissipation from Earth into space via black-body radiation. This argument is self-contradictory: if we were producing so much energy that it were able to heat the atmosphere (>1000x today), we would certainly have enough energy to fly into space. (The linked article "Why not space?" contains no math and few real arguments: https://dothemath.ucsd.edu/2011/10/why-not-space/ )
Also, this paragraph is relatively good evidence he didn't talk to a real economist:
>But if energy became arbitrarily cheap, someone could buy all of it, and suddenly the activities that comprise the economy would grind to a halt. Food would stop arriving at the plate without energy for purchase, so people would pay attention to this. Someone would be willing to pay more for it. Everyone would. There will be a floor to how low energy prices can go as a fraction of GDP.
None of this hemming and hawing about infinity answers the real question: is it possible for the global economy to grow until nobody has to live in mud-huts without air conditioning? The answer is almost certainly yes.
That said, I think you misunderstood the point of dismissing space colonization. That was done to examine the consequences of that assumption, not as an assertion that space colonization can't be economical.
Many people might think exponential growth can continue even without expanding into space, and trying to dispute this opinion without first drawing a line would just result in confusion as the discussion retreats into "but we'll just expand into space then!".
None of this hemming and hawing about infinity answers the real question...
In what world was that the real question? The one where limits to growth are a broadly accepted fact acknowledged by politicians, economists, and business leaders?
Besides, I think that pessimism (what the physicist probably thinks is realism) rests on one key point, that the price of energy will have a non-zero floor. If you remove that, everything is possible, even assuming steady that energy use, and requiring continued economic growth.
I think the avoidance / supposed maintenance of the "energy price floor" point was too weak. The assumption was that it will still be a purchasable resource ("someone could buy all of it" and hold everyone else hostage), but what if energy becomes unlike that? What if energy in the future is unlimited and no one can buy all of it because everyone can access it? And the only constraint Earth imposes is that no one uses too much so we don't boil off the atmosphere or whatever.
I think that's likely. Zero point energy, free energy from the vacuum. Some whistle-blowers say our secret space program already drives its ships with this kind of power.
What's interesting is that, based on these projections about hitting critical energy output, we won't need this ZPE for another 250 - 600 years. So maybe that's why, if it does exist, it is not going to be released until then. Kind of sad that 40 years ago or so, some bureaucrat planner may have robbed the human race of free energy for the next few centuries because of the projections of an economist and a physicist, not these two, but some other pair tasked with considering the economic and physical implications of then soon-to-be back-engineered alien technology.
Sounds like a pricing mechanism had better appear. Fish are free for anyone to catch, as long as nobody catches the last two...
I feel like the rest of your post was trolling, though, so I'm not sure you were serious about this part.
I wasn't trolling on the rest, just brave. I don't expect you to be able to respond without name calling tho, as I think you'll see free energy as topic you can't seriously touch lest it taint your reputation.
I think generally people agree with you that the answer is almost certainly yes. Given that lack of controversy, why would you choose this as "the real question"?
Perhaps more contentious would be something like "how many years into the future will the so-called 4% safe withdrawal rate remain safe?"
For example to get delta V to go to Mars he adds earth's 11 km/s escape velocity to the 3.6 km/s Mars injection velocity. A freshman aerospace student could tell the speed of a Mars bound ship in a hyperbolic earth orbit is sqrt(11^2 + 3.6^2).
I've done a number of posts calling out Murphy's bad arguments:
https://hopsblog-hop.blogspot.com/2014/02/the-most-common-de...
https://hopsblog-hop.blogspot.com/2014/03/murphys-reply.html
and https://hopsblog-hop.blogspot.com/2018/06/space-meow-boys_30...
I generally agree with him that we live on a fragile, limited planet and need to learn to live within our means.
However I don't see preserving our own planet and opening a new frontier as mutually exclusive. Musk is one of the most passionate space advocates and he's working for a sustainable future (solar panels, electric cars). Some technologies benefit both space development and sustainable comfort here on earth. For example CELSS (Closed Ecological Life Support Systems) technologies needed for space might be used here on earth.
Murphy may be correct that opening a new space frontier is implausible. But he hasn't conclusively demonstrated that's the case.
We already have experience with an economic input which is required for practically all production becoming less important to the economy as efficiency increases, it's called labor. And as efficiency of labor usage increases, the price of labor goes up, not down, see Baumol's Cost Disease. So no, the fact that the price of energy will go up with gdp does not mean the growth is "illusory", and the fact that an "extremely prestigious" econ professor did not point this out is a bit fishy.
The author is saying that physics puts a fairly solid cap on the amount of power we can produce on earth. That's the basic point.
What does a finite energy-consuming, exponentially growing economy look like?
If we accept that all economic transactions consume some energy, and the economy is growing (so the number or the value of economic transactions is increasing), then once we've hit our physical constraint on power generation, how does growth continue? Only if the energy required per unit of transaction-value continually decreases. So the price of energy must continue to drop, measured by the relative to the value it provides.
The author then points out that this is a contradiction because the price of energy would have to rise, and therefore that the problem is in the assumption of continuing exponential growth.
Are we supposed to accept this? Because I do not. I think the way that we allocate resources towards energy collection/generation/consumption is a bigger constraint than what physics can currently allow.
This is a raw physical limitation. How we use and how we generate the energy are irrelevant details. The only thing that matters is Earth’s ambient temperature, the additional wattage being used planetwide, and the Earth’s radius. Past a certain point, Earth’s equilibrium temperature will rise well past current levels, and rapidly.
Just as irrelevant as assuming the basis behind current growth rates where it is impossible to mitigate agaisnt "literally saturating the Earth’s capacity to radiate excess energy as heat into space".
All of this is based on assumptions. You can't just accept one set of assumptions that underlie a model and throw out others that could refute it (well one could, but it isn't intellectually honest).
For instance you can't disrupt the 2nd law, so at some point it's actually true that it is impossible to mitigate overheating and you won't be able to improve Stefan-Boltzmann law (except for increasing our emissivity, at night I'd guess). You can try to play with mass transfer but more thermal escape is not a good idea and a heat exchanger with extraterrestrial masses won't work.
I know it's hard to take and that physicists sound arrogant when they do this, but that's how it is.
We're talking about macroscopic systems here. The laws of thermodynamics depend on the law of large numbers (if you take the stat mech route), they can be considered laws about statistical averages. It doesn't matter that somebody in nanowhatever claims a deviation from the macroscopic behaviour, that's their job. You still have a macroscopic Earth. An Earth brimming with grey goo would still obey the laws of thermodynamics.
I referenced wikipedia and cited an example of such from my experience in physics lab in college nearly a decade ago… I don't get why people try to use some "law" built upon a particular set assumptions that work on a particular set of circumstances expect their leaky abstractions on top of it to govern everything that could ever be possible…
>The laws of thermodynamics depend on the law of large numbers (if you take the stat mech route), they can be considered laws about statistical averages.
On systems operating under certain assumptions… just like navier-stokes predicting singularities from a specific set of IC's in theory that don't happen in real life from said IC's, which have to be handled with a bunch of corrections and adjustments that are more accurate… we're still working out the math as to why that is…
Conservation of momentum is a physical law, that is, we're not making that up, we always see it. The Navier-Stokes equation is a continuum mechanics model of fluids with interesting mathematics, but it's not fundamental, it relies on conservation of momentum and additional reasonable assumptions about the fluid, in the same way that for instance the van der Waals equation is reasonable but it's not a law, you find deviations and there are better models.
As far as we can tell the laws of thermodynamics are laws indeed, that's why no one is proposing perpetual motion machines of the second kind any more.
(Look at the CMB spectrum —and refs therein—: https://commons.m.wikimedia.org/wiki/File:Cmbr.svg)
"Oh but we haven't observed it in the cosmos from our cherry picked backtested 'sufficiently black' ones so it doesn't matter"
Rah rah rah, if you want to continue more just, email me cinque [underscore] mcfarlane - blake [ - a - t] brown [- d - o -t] edu
Even if energy becomes incredibly cheap, there's a certain cost to dealing with waste that for now we haven't baked into economic models (because we could simply throw it away). If energy usage, no matter how cheap, keeps on growing, waste will also keep on growing.
Especially since it repeats a line of reasoning, whose premises clearly apply to labor, but whose conclusions are clearly incorrect in the case of labor.
"So the price of energy must continue to drop" simply does not logically entail from the preceding premises.
Bremermann's limit, (or some extension of it?), says that the minimum amount of time needed for a system with average energy E to change to an orthogonal state, is inversely proportional to E (specifically, pi * h_bar, all divided by 2E ).
So, assuming that “I have the money” and “you have the money” are orthogonal (collections of) states, then there is a maximum rate at which the system can switch between those states.
You might say “well, what if we just count the number of transactions that ‘should’ have happened, instead of counting the number of actual events where the money changes hands”, but that seems to obviously not reflect actual benefit?
If me and another person turn away from you, and then turn back, and claim that we exchanged ownership of a dollar bill I’m holding 2^64 times while we were facing away from you, the obvious response would be “no, you didn’t.”.
"This piece of paper represents a chunk of gold!"
"No, it doesn't."
Though, whether one can exchange a particular paper for a particular amount of gold is not a matter of pure imagination.
It's by no means perfect, but by any objective measure it's providing me with more value at a lower energy cost. Energy cost to operate, energy cost to build, energy cost to maintain. And that trend is most certainly not going to reverse any time soon.
Very true, in order to get more progress, one would have to develop systems that are sufficiently different and more efficient at some level which may be odds with how we currently allocate resources.
At some point though friction against road and minimum weight for safety and comfort come into play, and once you optimize these you end up with a train. :)
It's the same problem electronics face right now, when they really butt into quantum physics and consequently have huge yield issues - and you can scale horizontally (more transistors) only for subset of problems.
Breaking both in any major way would require or beget actual teleportation. (Which unlike electricity or moving in atmosphere has not been seen in nature.)
There does exist a series of absolute limits on Earth, such as capacity for growing food limited on sunlight, water and availability of specific minerals. Inserting more planets adds to that limit with a transportation losses... Unknown exactly but bottom floor is easy to figure out for spaceflight and mining (unless teleportation).
You can even limit maximum size of "life speed" cone around Earth which would be the maximum distance we can get things from during lifetime of any person. (Unless immortality or teleportation or high multiplier FTL.)
>Breaking both in any major way would require or beget actual teleportation. (Which unlike electricity or moving in atmosphere has not been seen in nature.)
Well maybe we haven't seen them because we didn't know how to look… recently declassified FLIR from US navy reporting archives suggest things that we haven't typically seen in nature in the past, can be seen.
>There does exist a series of absolute limits on Earth, such as capacity for growing food limited on sunlight, water and availability of specific minerals.
True, but what is the purpose of any of these things, and do we currently pursue the most efficient ways of doing such for those ends in a way that we cannot think about it anymore?
Once we figured out information-theoretic entropy we got tools to set bottom limits on everything. I'm not entirely sure when, but Kolmogorov was quite some time ago. The math is in good company of Shannon and Fischer.
Remember that reducing costs by not doing something has risks. Existential risks in fact. (E.g. reliance on a single source of energy without fallback.)
>Existential risks in fact. (E.g. reliance on a single source of energy without fallback.)
Existential risks for oil companies investors or CSP investors?
If it’s production or goods, there’s obviously an upper limit. Even if in distant future, we begin assembling atoms into making novel rare elements, all that process consumes energy and possibly would produce waste. Production is not free. In the exponential limit, we will hit some constraint.
If we are talking about services, we should realise that even they cost energy. Even if we tomorrow think that long distance travel will be replaced by VR, production and delivery of such experiences cost energy.
Fundamental argument is then about energy. Can we keep harnessing it infinitely? No. Even if we make energy efficient processes, Jevon’s paradox suggests we will simply start using more energy.
The only way physical limits can be respected is by limiting the growth of economy and population. (Otherwise if we keep printing money without increase in production, it isn’t really economic growth)
The thing about infinity is that while it is a limit, it can always be approached based on ones understanding. And I would not state that our knowledge and application of energy collection/generation (and even energy itself) is fairly limited.
>The only way physical limits can be respected is by limiting the growth of economy and population. (Otherwise if we keep printing money without increase in production, it isn’t really economic growth)
I disagree (and debt monetization != printing money), naturally markets and systems will correct. They always have, even when people try to distort it and keep them from correcting.
Markets and systems obviously correct, but the key question is at what cost and to whom. Imagine for a moment that we find ourselves in a situation where energy is limited. Markets will correct themselves by making it expensive, but who'll suffer the most is the poor (while markets might have been reacting to overconsumption by the rich).
Saying markets will correct themselves is like saying things sort out themselves. They do, but usually who takes a hit isn't fair and ethical (as was evident by 2008 crisis).
Is there? I have a difficult time imagining any meaningful inherent upper limit to the value of creative works. Perhaps you intend a narrower point?
Just like we're facing the prospect of peak oil, is it really hard to imagine that we could face the prospect of physical limitations imposing themselves upon creation and distribution of creative works?
With that said, is it perhaps possible that creation and distribution might not quite be the same as value in every scenario? A given mass of aluminum can be formed - and recycled - into a functionally infinite series of unique artistic works. I can see no reason why the hard limits of the physical world you so wisely point to would inherently force an upper limit on the value of the artwork.
Have I misinterpreted you? Or perhaps I fail to understand your points correctly. Can you help me? I seem to have suffered a failure of imagination.
Increasingly though, modern economies are shifting away from this sort of work. Instead of the energy-constrained primary and secondary sectors, more things are being done on the knowledge and attention-constrained tertiary sectors.
Like, economic growth through giving people better advice is much less energy constrained than you're suggesting. There's still some limits, but overall it's not something that needs anything as drastic as economic growth limits for their own sake.
The real issue though is not about absolute amount of energy. The author is talking about exponential growth - the one where every year economy must grow as a percentage of existing base. At any base, even with a majority service based economy, exponential growth will hit a bottle neck.
For example, take this (from https://www.energuide.be/en/questions-answers/do-i-emit-co2-...)
>In 2018, Information and Communication Technologies (ICT) account for 6 to 10% of global electricity consumption, or 4% of our greenhouse gas emissions. And this figure increases by 5 to 7% each year!
This year-on-year growth is unsustainable, no matter how small the base. At 7% growth, the greenhouse emissions will double every 10 years. We're really bad at wrapping our heads around exponential growth.
The argument that we will find more energy efficient ways to do a certain activity misses the point because for a growing economy, even that efficiency gains will get outstripped by scale. And there's Jevon's paradox too whereby efficiency actually increases resource consumption. So as Internet bandwidth expands, it's not like per-capita Internet usage remained the same. We simply increased our hunger to consume the available bandwidth.
You started off right (i.e. it's not about material limits) but then I think you missed that material value can be created out of thin air.
As long as humans are innovating and improving their conditions, then 'growth' as we understand it continues.
The iPhone is not 'great' because of the amount of physical material we put into it.
The numbers we put growth it are a function of other, relative things, but the fact remains that value is being created.
One can imagine a similar conversation between a geographer and an economist (if one existed) in 1700 in which the geographer made the same argument about land use; at the time economic growth was very much about bringing land under cultivation or other use, and hence (the geographer would argue) economic growth would be limited by the finite amount of land. Same fallacy.
What humans value is built upon energy and resources (even clicking on buttons), and if that base is limited, where does future value creation come from?
It's possible that by pressing the same number of buttons per year (i.e. the same number of programmer hours worked), the current generation of software in the world could be upgraded to a new, more valuable generation of software.
There's a question of whether new software (presumably with extra features) necessarily uses more energy (per user, per year), but it seems possible that one year's more-valuable-software could be more energy-efficient than the previous year's less-valuable-software.
Unless we have an Asimov-esque robot driven “colonization”, but does that fully count?
The fires in California this past fall? That represented a lot going wrong, but you could still go outside in the Bay Area and breathe.
On Mars or in space? Everything goes perfectly except for one little thing, and you are dead instantaneously.
But those ships are not like planes, you can't reuse them multiple times a day, the trip to Mars takes something like 6 months, and that's a launch window that opens only once every 2 years, the rest of the time it takes longer and more propellant, so your ships would make the round trip in 1 or 2 years, so you'd need to build 820k to 1.6 million ships only to be able to shuttle the population to Mars and stay constant on Earth.
That would be a serious industrial investment. A quick search says that there are around 24000 planes in service in the world, and over the course of history, 150000 planes have been built, including all military and commercial aircraft. So we'd need to build a production with many many times more capacity that all the airplane builder that exist today.
Then there is the energy cost of those ships, Earth gravity well is actually pretty deep and chemical propulsion only just is able to escape it. That many launches is going to need a lot of energy, a rocket can hold up to 500 tons of fuel, multiplied by several thousands of launches per day, and you reach a non-negligible percentage of the current daily worldwide energy expenditure.
The idea of being able to mine asteroids or outsource production in a meaningful way on another planet is similarly unrealistic, escaping gravity is not easy. Of course the technology will improve, but not by the several orders of magnitude that would be needed. When we colonize space, it is likely that it will be more like seeds, with only relatively minimal exchanges when the scale reaches a whole planet.
IMO, that's Space X and Blue Origin's goal. The first one to become Weyland Yutuni will also anoint the first trillionaires even without the colonization efforts.
You can see Space X leading the game here with their work on Star Link. That gives them experience managing a constellation orders of magnitude larger than any before, and shows the overall strategy of leveraging extra capacity in a way that people on Earth will continuously pay for, in order to bootstrap serious micro g infrastructure.
In all these visions of techno-utopia, poor suffer the most while the rich exploit the resources, and then expand that exploitation to more areas in hopes of making a profit.
Why do you feel we will not do the same to Mars what has happened to the Earth? It's possible that once we go to Mars we will exploit _both_ Earth and Mars.
The same argument of destruction of solar system, milky way and our galactic super-cluster can be made for different time scales. Even universe is expected to fizzle out.
Though I agree there are no easy answers. Risk will always be there, but we should realize that hedging one type of risk exposes us to other types of risks.
And taking the VERY long view, and galactic civilizations, assuming current physics, you end up running into similar issues (except instead Earth it's ALL matter in reach converted into Nuclear Reactors + Dyson spheres), and further expansion running into speed of light limits.
But isn't it obvious that we will expand into space stations and then eventually onto other planets?
> But if energy became arbitrarily cheap, someone could buy all of it, and suddenly the activities that comprise the economy would grind to a halt.
This segment contained a number of logical leaps that I can't agree with. It is possible for energy cost to fall arbitrarily low without reaching zero. And you can imagine an arbitrarily low price that would skyrocket quickly if someone actually tried to buy all the energy.
> So once our fixed annual energy costs 1% of GDP, the 99% remaining will find itself stuck. If it tries to grow, energy prices must grow in proportion and we have monetary inflation, but no real growth.
I'm not able to follow the claim here. Can anyone shed some light?
If we have finite energy at a fixed price (of say 1% of GDP in total cost), then either the rest of the GDP (which is not energy) does not grow, and so the demand for that energy (and so its price) remains the same, or it tries to grow (and cannot "is stuck") because growth will increase demand for the energy and therefore drive up prices, but since it is energy and fundamental to everything, all prices will go up, so we will have inflation, but inflation is not growth.
Oh ok... I see why I got lost. In my head I didn't hold the assumption that "all growth necessarily must consume more energy". This is assumed throughout the article until the Epilogue.
But if you drop this assumption, then I don't think the "it's just inflation not growth" holds anymore.
Then I had to think a few times about what you gave in response! It seems this part of the topic is quite involved. I think I get some of it now.
If you drop the assumption that grow takes energy, then no extra demand on energy from growth, then no price increase on energy, then no inflation, and so it's just growth, not inflation. Agreed.
But when you and me are examining it like this...the whole situation described by that sentence starts to seem a little contrived to me. Seems there's a lot of error bars on each of the assumed connections in that causal chain, that could lead that path to branch at each of them into many other possible paths, rather than just the one path that condensed sentence proposes.
I still have a hard time imagining a growth that doesn't use more energy (except the one defined in that article as "development", which seems to be a more sophisticated/involved use of existing resources, rather than extracting/creating more new (primary) resources.)
If you could elaborate some more on your images of growth that don't consume more energy, I'd find it helpful to understand this whole thing better.
Can’t you imagine efficiency gains leading to an increase in output (GDP) for constant inputs (resources)? That’s growth. (Note that the article talks about development as something different from GDP growth.)
I still can't really imagine the other scenario because when I see efficiency gains I think they will lead to growth and more energy usage. Examples like "increasing CPU power should have led to faster software, but instead it lead to slower languages and larger software". Same for memory and bandwidth. Our websites keep expanding to fill the next generation of network capacity.
So the gains from efficiency are quickly consumed by increasing growth and energy use. Maybe you could help me imagine some that are not like that? I'd like that.
It's easier for me to imagine the examples like in the article (more fancy dessert, same ingredients) that are there classed as development, distinct from growth. But growth as defined there, with static energy, I can't imagine right now.
I don't know what the article is trying to say, really. It talks about "improving the quality of life" but I'm not sure if that includes factors that are already considered in GDP calculations (https://www.bea.gov/sites/default/files/papers/P2006-6_0.pdf) or it's about immaterial quality of life improvements (more hapiness or whatever).
> But growth as defined there, with static energy, I can't imagine right now.
Say you have a factory producing 1000 widgets per day and you improve the efficiency of the process and now you produce 1100 widgets per day using the same quantities of labour, raw materials and energy. Don't you agree that this is growth?
I'm not actually sure right now. Somehow I think that if you make those extra widgets without using more energy, then the extra people who buy and use those extra widgets will end up using more energy in doing so, and their extra purchase and use will, however incrementally, grow the economy as a whole a tiny bit, which will use that much extra energy.
So if it's growth, then it can't be static energy.
But somehow also I think that you can have static energy and not call it growth, but development, or something else like someone said before (maybe you).
What's development? I don't know. But maybe it's something like we produce 1000 widgets with more intricate decorations on them. And use no more energy doing so.
I grant that it's pretty confusing. And that's how it seems to me right now.
Thanks
Imagine the system is at equilibrium at a certain output and energy usage and you discover a device than increases efficiency in the use of energy by 10%. The system could produce the same output as before using 10% less energy. But the new equilibrium won't be the same output at 10% less energy consumption, because some resources will be freed in the energy generation sector and the price of energy will go down resulting in some activities that were not economically viable before becoming viable. Maybe desalting sea water was too expensive before, but with lower energy prices and higher efficiency it does make sense, for example. You will have an increase in output but whether the new energy consumption is below or above the original level depends on the details. The latter is more likely, I guess.
My factory now makes the same amount of sneakers but uses less energy, and I have more profit which I can spend on more things, maybe making new markets viable. And in the macro picture, that extra energy, lower prices, makes some things become viable. And what happens from there regarding energy depends on the details. Very interesting and good, semi-detailed discussion.
... Oh, wait, or maybe growth actually can be expressed as a quantifiable concept?
In this case, the author chooses energy as an objective measurement, because all economic transactions must consume energy in order to occur.
I am not denying that the economy has changed in meaningful ways; I am merely denying that we can quantify the overall state of the economy over time such as to enable simple comparisons implicit in concepts like “growth”—and that with the caveat that the problem isn’t too severe over short periods of time.
"The fact that two incommensurable collections of miscellaneous objects cannot in themselves provide the material for a quantitative analysis need not, of course, prevent us from making approximate statistical comparisons, depending on some broad element of judgment rather than of strict calculation, which may possess significance and validity within certain limits. But the proper place for such things as net real output and the general level of prices lies within the field of historical and statistical description, and their purpose should be to satisfy historical or social curiosity, a purpose for which perfect precision — such as our causal analysis requires, whether or not our knowledge of the actual values of the relevant quantities is complete or exact — is neither usual nor necessary. To say that net output to-day is greater, but the price-level lower, than ten years ago or one year ago, is a proposition of a similar character to the statement that Queen Victoria was a better queen but not a happier woman than Queen Elizabeth — a proposition not without meaning and not without interest, but unsuitable as material for the differential calculus. Our precision will be a mock precision if we try to use such partly vague and non-quantitative concepts as the basis of a quantitative analysis."
(J.M. Keynes)
And is thereby begging the question.
Economists value goods using prices.
You could take some representative branches of housing to build what is called an energy basket, using similar method used to calculate PPP.
Such rough methods would include communication, transportation, housing, clothing, food, self-expression and a few more mostly timeless needs that haven't changed for many centuries.
The problem is of accurate representative sampling of each category. That gets easier the closer we are to modernity with big data gathering.
The methodology would survive start from before bronze age even, I think.
But, I still see a lot of people who won't even concede the point in the case of unlimited growth. That, to me, suggests that we are looking at a deeper phenomenon than can be resolved via logical arguments.
Think about the responses that believers have when confronted with the argument that there is no god. That is not an argument they can confront objectively. They respond with anger, disgust, etc. You're not making a truth claim to them; you're attacking their happiness in the afterlife. You're suggesting that their grandparents have vanished into nothingness instead of being in heaven. No matter what the facts are, they can never accept such an argument.
I get the exact same feeling from all of you arguing for infinite exponential growth. You're not reacting as if this is a truth claim; you're reacting like I'm trying to take away your happy future, in which everyone is fabulously wealthy and everything continuously improves, forever.
Think about how it makes you feel to consider that economic and scientific growth may have a limit. Can you be certain that you are really objective on this topic?
But in my defense it's worth mentioning that globally it still seems to hold true. The fact that energy consumption stopped in rich countries is apparently mostly due to the fact that manufacturing, which is one of the most energy-intensive parts of the economy, moved to poorer countries. But since the rich countries consume those manufactured products, the energy use should be counted against them too!
I don't think our finite physicist is arguing that exponential growth can't happen temporarily, just that eventually we hit the boundaries of our petri dish. It's a challenge to economic models that sometimes deny in principle that the petri dish even has boundaries.
i.e. In the same way that a computer runs hotter when consuming more wattage, how much has the Earth gotten hotter just because we are consuming more wattage lately?
Energy usage per point of GDP = f(t)
Then we can obviously construct a hypothetical scenario:
lim t->inf g(t) * f(t) = C
This is obviously possible if one flips it over, making economic growth depend on the rate of which we reduce the energy cost for producing a dollar of GDP.
This obviously means that production must be decoupled from the sheer AMOUNT of physical goods, as those are necessarily limited.
To increase GDP, we do not need to increase the amount we produce but the value of what we produce.
Sure, one way to increase GDP would be to produce two cars for the same resource use as one car would need. And that has limits.
But if I can produce a rocket that can propel something to orbit with the resource use of a car, I’m am not only enabling more rockets being built, but also actively reduce the energy cost of one.
The article has a few places where the physicist attempts to ascribe an intrinsic value to an item (say dessert types) independent of the supply and demand curves for that item. The whole concept of a demand curve is that everyone has their own value that they give to an item. The fact that some people do NOT value some particular lifestyle improvements does not affect econ in the slightest.
Huh? This seems to be a critical premise for establishing a crisis, yet it's incoherent and essentially imaginary.
Ok, some ways we use energy have poor thermal efficiency. Light bulbs are the classic example, but look at the improvement from incandescent bulbs to LED's. Most uses of energy just move heat around within the environment, and if anything that would tend towards thermal equilibrium (a sort of microcosmic 'heat death'?), not "cooking".
My physics is admittedly not great. However I though a 100% efficient lightbulb generates just as much heat, but at the place where the photons land (walls, floor, etc) rather than at the bulb.
I think basically any time we use energy, whether by running current through a lightbulb it by respiration after eating food, it ultimately ends up as heat.
Are you thinking of transportation here? We burn fuel to add kinetic energy to a car or airplane but, by the time the vehicle has stopped moving, all of that energy has been converted into diffuse heat in the wheels, brake pads, atmosphere, etc, never to be harnessed again for the mostpart.
There aren't any devices that operate by moving heat around unless they create additional heat to do so, or they do no work. If you're just shuffling heat around, you're not increasing entropy, and you're not achieving any observable mechanical objective, which is pretty much the whole reason we consume energy.
Shuffling bits around takes energy; if your virtual reality services grow exponentially and energy production doesn't, then the price of energy must also increase exponentially or else one of your VR companies could buy all the energy and shut down the competition.
Energy production (and computation) are limited by the capacity of the earth to vent heat into space at a reasonable surface temperature.
So unless you allow for an exponentially expanding real-world physical economy, you can't have an exponentially growing virtual economy.
But information doesn't need to be rationed in that way: what happens if people join an open source virtual reality instead, where consumption is unlimited and production is voluntary? Nothing in this world would be part of the economy as economists measure it.
"this thing humans do cannot continue" is a weak argument. everything humans do is some expression of the natural state of conditions in their immediate environment, technology is just another convolutional cycle on top of the pre-existing ones, squeezing more efficiency from the raw resources existing there. the planet is not in balance, and ecosystems are not fixed. things are in a state of glacial flux with periods of extreme instability due to various cycle renewals - volcanic, tectonic, asteroid/comet impactor, solar, geomagnetic symmetry breaking, viral.
'man made causes' are a misnomer, they are just accelerated selection events, man happens to be a global selector for almost all living things, and man activity on the planet can be a mass extinction event due to many scaling factors, the most obvious one being the technology cycle mentioned previously, the efficiency it provides can be simplified to 'making more heat'.
it's easy to fall prey to goldilocks thinking - due to limited information and the tyranny of the present. try to think in terms beyond a human lifespan. look at the progress achieved in the past 300 years. if we wanted to we could spend the next 300 years returning earth back to a pre industrial epoch and wipe from the surface of the planet all traces of the modern advanced civilization we created, not by a cataclysm or destructive war but by planned intentional decolonisation. given that its possible and very likely to happen (perhaps a catalyst is required, like a new global religion) over the past 300k-1my this maybe happened multiple times. humans are still here in some form or another and they will continue to be here into the foreseeable future x10 my's, because they have shown capable of surviving (milankovitch scales, 100k, 41k, 23k years).
technology is dominant and makes the smartest humans think in brittle ways. the people shaping rocks for millennia were not stupid, they were just incapable of thinking beyond their stone paradigm. pretend to be a godlike alien silently watching the world from the largrangian point in the earth-moon system. millions and billions of years go by. once life begins on earth it continues into the present, the idea that humans, in the next few thousand years, will somehow pose an unconscious risk to a process that has survived and morphed for billions of years is a misreading of the story. what COULD happen is another mass selection event, where a human bottleneck eliminates 99.9% of all living things. that happened before and it's part of some larger cycle that humans are necessarily a part of (as living things) and what is considered causal could also be simply an expression of a deeper reason for living things to exist - to out compete other living things and monopolize the space available to them.
is the physicist a finitist? seems like he substituted growth which is the economists mantra, for change which is his. predicting the future is a waste of time, as an agent in the world you will either cause the future or be slave to it despite your best efforts either way. if it is knowable it will be unchangeable, given all your free actions will lead to the known outcome. it is clearly unknowable and therefore ununchangeable - since the capacity to measure any deviation or change does not exist. what will occur is as likely to occur or not occur had you participated or not.
my pet theory is that any time you see people attempting a cross-disciplinary leap of faith their parachute often fails to open. good 20th century examples being james watson and william shockley - both making the tempting leap from (bio)electrical systems to socio-communication ones, leading to self-embarrassment. the only interesting problem for the 21st century is how do you transform knowledge from one domain to another and retain it, the only way humans will survive their own extinction without reverting back to barbaric primitivism is to plasticize expertise and mass produce it.
Uh,
Thermodynamics.
You can’t destroy heat, just pump it around.
A heater does not need to expel cold. A cooler needs to expel heat.
Stuff can feed on heat gradients. Things cannot feed on uniform heat.
is the sun uniform heat? how do ectothermic animals and plants using photosynthesis work? I'm not sure why you think you can magically break thermodynamic symmetry...
(Edit: by which I mean, it does not decrease the total amount of heat in the system.)
You need something colder on the other side.
The sun-earth system is not uniform in heat, no. The sun on average is hotter than the earth on average.
Edit: what thermodynamic symmetry are you referring to?
it's powered by a heat differential, you are draining the heat from the system and storing it or dissipating it under motion - reducing the overall flux - over time by cooling the hot side and heating the cool side eventually some equilibrium is reached, depending on how efficient the throughput is and how much heat you have, you then exhaust your temperature gradient and need to move somewhere hotter or colder.
one of us is very confused here. i claim "you can reduce heat locally in the earth system", and give examples: capturing and insulating it (trees, plants), converting it into other types of energy (genetically modified sea sponge/robots), as well as venting it into space through an atmospheric/orbital seebeck ring.
im not breaking any laws of thermodynamics, i am turning the planet into a refrigerator.
edit: to clarify i am claiming it's possible to move heat away from the habitable thin boundary layer by conversion: storage, mechanical use or venting.
Reducing the temperature gradient(s) of the system does not reduce the amount of heat in the system. When the temperature gradient decreases, that just means that there isn’t as much variation in the temperatures in the system. The temperatures become more uniform. When the temperatures are uniform, you cannot use the temperature in order to do useful work.
.. I now notice that I missed that you mentioned putting things into orbit?
I misunderstood and thought you meant putting the seebeck stuff in the ocean, as if that would cool down the oceans.
Yeah, if you pump heat into places that you can remove from earth, or have emit black body radiation more effectively, that would help.
But, Unless you are ejecting stuff away from earth, there is still a max rate you can radiate away heat at a given temperature? Oh, but, hm, if you made your thing in orbit really really hot, hm.
Is that what you were saying?
Sorry for misunderstanding what you meant about the ocean.
talking about the earth as a system here is a bit of a misnomer. the core up to the mantle is very hot, but that heat only escapes to the surface through lava tubes/vents because the tectonic plates are good insulators. im specifically talking about the part of the earth we care about affecting the temperature of - and you can do this by shifting the heat up, down, storing it or converting it. the atmosphere is more permissive than the hydrosphere, hadal zone or crust.
it's easier to imagine the earth as a ball and things leaving the surface as exiting the ball system. but really when we talk about the earth it's more like a layer on a ball, and if you go deep into the ocean, underground or exit the atmosphere it's all the same type of thing.
as far as i know the climate cycles are a surface feature of the planet, deep underground is more inaccessible to our technology than reaching another star system.
the term is called geosequestration.
Not sure what you mean by storing heat.
You could store energy obtained from (the smoothing-out of) a temperature gradient, but that isn’t storing the heat, just the energy extracted as the temperature gradient goes away. It doesn’t decrease the amount of heat?
And I suppose if you have some endothermic reaction (like melting ice) you can decrease the amount of heat around, (and then if you freeze it again, you release the heat), which is sort of like storing heat, but it isn’t like you can store arbitrarily large amounts of heat this way. You would need arbitrarily large amounts of stuff to have react.
I’m still fairly sure you cannot “transform heat into kinetic or chemical energy”, but rather, you can transform a difference in temperatures, and a potential for heat to be exchanged, into kinetic or chemical or whatever energy?
In the example of that mechanism for converting a temperature gradient into electricity, suppose you had two regions with particular and different initial temperatures , which are in contact with opposite sides of the mechanism. The two regions have the same mass and are made of the same material. How do you expect the temperature of all the parts after reaching equilibrium to depend on the initial temperature? I expect that the total heat will remain the same (even though you did extract some energy as the difference in temperatures went away).
Also, pumping heat around costs energy. To move things against the tendency for temperatures to equalize, costs energy, right?
I don’t understand why the fact that earth is not homogeneous would be a reason to not call it a system?
I don’t see how the specifics of geology and such are relevant here. The question is: under what circumstances can we keep the part of a system that we care about to stay at a low temperature while waste heat is being added to the system at an arbitrarily fast rate, and the only way we have for heat to leave the system is blackbody radiation? (Assuming the system is connected/contiguous and stuff)
Now it sounds like you are suggesting that we just make the inside of the earth hotter while keeping our part cool?