[yojo]

I’m skeptical of the closing claim that exponential growth keeps happening forever. Yes, you can grow GDP 2% for 200 years, that results in an economy 50x the start size. Expand it to 1000 years and you’re talking about an economy 400 million times as large. After 2100 years you’re up to an economy a QUINTILLION times as large.

At some point the exponential curve has to go S-shaped. Maybe we’re still in the happy exponential looking part of the curve. There are also signs we might be transitioning. Population growth has slowed, productivity growth has slowed, and the marginal return on capital seems to be somewhere around zero given modern interest rates.

[shoo]

> I claim that economic growth cannot continue indefinitely. [...] the Earth has only one mechanism for releasing heat to space, and that’s via (infrared) radiation. We understand the phenomenon perfectly well, and can predict the surface temperature of the planet as a function of how much energy the human race produces. The upshot is that at a 2.3% growth rate (conveniently chosen to represent a 10× increase every century), we would reach boiling temperature in about 400 years. [Pained expression from economist.] And this statement is independent of technology. Even if we don’t have a name for the energy source yet, as long as it obeys thermodynamics, we cook ourselves with perpetual energy increase.

https://dothemath.ucsd.edu/2012/04/economist-meets-physicist...

[lern_too_spel]

The physicist says that per capita energy use has surged, but that is only true in the developing world. California's per capita consumption of electrical energy has not changed since 1975. https://www.google.com/url?q=https://www.nrdc.org/sites/defa...

[crisdux]

Consider that are dealing with a bias metric. California's weather makes it the least energy intensive state. In 1975 about 50% of the homes in my state had air-conditioning. Now its about 95%. Similar story with heating modern systems. Even in the 70s, older generations dealt with a much colder home in the winter than most people would normally allow today. The days that require climate control in my state are something like 2-3x the average number of days in California and at a much higher intensity.

Additionally, California has exported much of it's energy use to other states and countries.

[lern_too_spel]

We're talking about the growth rate of energy usage. California's weather has not changed since 1975 in a way that would reduce per capita energy usage.

[crisdux]

I'm not talking about the weather changing. I'm talking about the growth of climate control using electricity in states that aren't California has led to continued energy demands.

Honestly your reply is frustrating because you sound like you don't understand my point. Please re read my comment from before.

[lern_too_spel]

Even in your scenario, we wouldn't expect per capita energy usage to increase faster than population growth (as the physicist claimed) because now everybody who wants climate control has it. The whole point is that per capita energy usage isn't growing exponentially in the developed world.

[dredmorbius]

The decoupling argument breaks down when imports are considered.

"The material footprint of nations", Thomas O. Wiedmann, Heinz Schandl, Manfred Lenzen, Daniel Moran, Sangwon Suh, James West, and Keiichiro Kanemoto

PNAS first published September 3, 2013; https://doi.org/10.1073/pnas.1220362110

https://www.pnas.org/content/early/2013/08/28/1220362110

https://web.archive.org/web/20130906063246/https://newsroom....

[lern_too_spel]

That study shows zero growth in resource consumption instead of the negative growth observed without their methodology. This does not match the exponential growth in per capita consumption that the physicist assumes.

[fouc]

> First, Tom Murphy is confused by what economic growth is. You don't need increased energy consumption for economic growth.

> Economic growth measures the number of ways resources (material and non-material) are used hence resources are like the universe; "finite but unbounded." Tom Murphy is a typical Newtonian in his mindset and this imperial reasoning is so pre-Planck.

> Even in a so-called steady state any substitution from one material to another would seen as a kind of economic growth simply because markets the place more value on the newer item.

> Second. He assume that exponential economic growth is tied to some physical exponential. It is not. What is the basis for the recent exponential growth of physical consumption. It is growing population. It has been shown that energy growth of consumption doesn't increase much beyond 150k dollars. So once population stabilizes the physical growth model switches from exponential to linear. Tom Murphy never makes a linear growth model and thus his short-term peakerism.

http://mikenormaneconomics.blogspot.com/2012/04/tom-murphy-e...

[an_opabinia]

The stock market has grown significantly while demand for energy has widely declined in the last five months. Every energy long bet has been a disaster.

Economic growth may not be able to continue indefinitely, it’s inconclusive, your computers can create more economic value with declining watts even if you can’t. However accounting value, which is what a stock market is, can definitely grow indefinitely.

[roenxi]

> The stock market has grown significantly while demand for energy has widely declined in the last five months. Every energy long bet has been a disaster.

The stock market isn't backward looking, or even short term future looking. Note that someone only watching the S&P 500 would know that the coronavirus broke out in early March but probably think the situation had been completely resolved by August. We can't be sure what the stock market thinks it is seeing (or if it is right for that matter).

And if energy bets didn't turn out well, notice that that correlates to the US losing its position as the world's largest economy. China invested a bunch in energy and now have a noticeably bigger economy [0].

> Economic growth may not be able to continue indefinitely, it’s inconclusive, your computers can create more economic value with declining watts even if you can’t.

This growth isn't going to involve more people because they need food and isn't going to involve more stuff because that needs energy. It'll be a very abstract form of growth.

[0] https://en.wikipedia.org/wiki/List_of_countries_by_GDP_(PPP)

[tarsinge]

The stock market is not the economy, and even how it is correlated to accounting reality is debatable.

[dredmorbius]

Asset inflation is not economic growth.

(Though the two are very often confused.)

Stock markets occasionally move in directions other than up.

[sudosysgen]

I made exactly that argument on HN a few months ago! I found that the inflection point would be 200-400 years into the future.

[bufferoverflow]

The economy is not limited to Earth.

[dredmorbius]

Practically, it is.

Impractically, that buys you surprisingly little.

"Galactic-Scale Energy"

In 2450 years, we use as much as all hundred-billion stars in the Milky Way galaxy.

https://dothemath.ucsd.edu/2011/07/galactic-scale-energy/

[ericmay]

Well, eventually what will happen is countries will disappear, wars will destroy things, and other events will wipe out economies, savings, or entire companies. The growth model will probably always be exponential, but from time to time things will reset to 0.

[simonsarris]

or 1% more optimistically: https://pbfcomics.com/comics/reset/

[luckman212]

PBF is so genius. I had forgotten about it, thank you for the reminder.

[dstick]

While I agree with your sentiment, how has this worked in the past? Did things reset to 0 during WW1 and 2?

[zamfi]

In much of Europe and east Asia, yes! Germany and Japan in particular were bombed basically to 0...

[est31]

Those two nations still had the right culture and social structure to build blooming new societies. In this context, with "culture" I don't mean political views or such, but work ethic, and things like literacy, education, etc.

The German car industry wouldn't have been thinkable hadn't the Germans invested a lot of research into building machines in a mass produced way, motor research, etc. Sure, the factories themselves were destroyed, but the engineers together with their knowledge remained.

In fact, there are theories that disruptive events are even helpful for economic progress in the long run because they filter out the lazy elite of rich people while giving opportunities for new ideas to be tried out, and hard working people to rise to wealth and influence.

[ericmay]

Sure, but that filtering out means wealth is destroyed and you have to start over. The account resets to $0.00.

Also, these events "filter out" a lot more than just "lazy elite rich people". Plenty of productive rich people and regular every day people are filtered out as well.

[lalos]

Wasn't food rationed and life quality lower in general during the wars? I'm basing this on seeing propaganda of making daily sacrifices for the war but not exactly sure. My general image was that collectively it was an extreme low point in many countries.

[grecy]

The Roman Empire imploded and went back to 0.

As did all the others that can before the current one.

[fennecfoxen]

> At some point the exponential curve has to go S-shaped.

In principle yes, of course.

In practice, you are imagining a world in which we never expand our civilization into space and across the galaxy, because otherwise you would be looking at the S curve and thinking "wow, this thing has only just gotten started" and marveling at what lies ahead.

[paulryanrogers]

Our space-ward expansion has some hard physics limits. And we've not yet eclipsed what was already done decades ago. And even settling Mars will require more sustainable lifestyles than most first world citizens now enjoy.

[xmprt]

We're talking about hundreds and thousands of years though. A few hundred years ago, the mere thought of airplanes and cell phones would be considered witchcraft. I don't think it's a long shot to be reasonably optimistic about where humanity can get in terms of technology.

[paulryanrogers]

Imagination is unlimited. I don't mean to harsh anyone's dreams. But technological development isn't a predictable march forward.

There is less low hanging fruit as things progress. I'd guess we're very close to the peak of attainable technology.

Once sustainability comes into the picture then, IMO, we're beyond the limit.

[xmprt]

It's not supposed to be all easy. A big part of advancement is creating technology that enables us to reach higher with less effort. Sustainability is a big issue now that a lot of people are working towards. Once we unlock more efficient techniques of capturing solar energy, energy becomes almost a non issue until we reach another barrier that's orders of magnitude away. Space travel will reduce resource limitations too.

[BobbyJo]

>I'd guess we're very close to the peak of attainable technology.

Not many people, and pretty much no research scientist, would agree with that statement. I can personally think of several technologies I'm 100% certain will exist in < 40 years that will drastically alter everything about day to day life.

[ipython]

I’m very curious what are the technologies you’re tracking?

[count]

European global expansion also had hard physics limits. The age of sail grew the global GDP tremendously. I wouldn't discount multi-year asteroid mining trips/etc. from having the same impact in 50-100 years.

[ouid]

Really? What do you imagine to be receiving from an asteroid that cannot more easily be obtained from the earth?

[diab0lic]

Large volumes of minerals that are already in space (don't need to expend energy putting them in orbit) to build settlements, ships, etc...

[njarboe]

"settling Mars will require more sustainable lifestyles than most first world citizens now enjoy."

Not if they decide to use nuclear power. Humanity has at least one more dramatic jump in power-use/person that has been ready since the 1960's.

[onion2k]

If I had to bet on which happens first out of the end of growth-driven capitalism or humans leaving the solar system I wouldn't be putting my money on spaceships.

[mjfl]

Imagine a world with 1 trillion people on it. A new Einstein is born every 10 years. Better access to information, the ability to augment their minds with computers makes them 10x smarter than Einstein was. Is it so hard to imagine that the economy would be 50x bigger in that situation?

[randmeerkat]

It’s not just Einstein being born, it’s Einstein being born in a place where the person has access to education, funds, and an environment conducive to a civilization changing breakthrough.

For example Peter Higgs says that what he accomplished would no longer be possible today with how modern academia works: https://www.theguardian.com/science/2013/dec/06/peter-higgs-...

Overall I’m pessimistic that another Einstein is possible in the timeframe that we need them and that they would have the motivations to solve the problems that we need solved.

My cynical view is that realistically another Einstein would probably just get sucked into figuring out how to use big data analytics to get users to click on ads all so they could buy a $2M three bedroom house in the Bay Area.

[yojo]

It is unclear if the Earth’s carrying capacity reaches a trillion, that’s certainly in the high side of forecasts. It does not look like we’re on a trajectory to hit anywhere near that.

I also question the value of genius at driving GDP growth indefinitely. Eventually you run into natural laws that are insurmountable; you can’t genius your way out of entropy.

That said, the GDP wall could well be millennia away and we still have tons of room for growth. Maybe we finally bring cheap fusion online, solve asteroid mining, and terraform anything remotely habitable around us. Or maybe we don’t, investment as a vehicle for income fails, and no one gets to retire in 50 years.

[PopeDotNinja]

I wonder if there’s even remotely enough phosphorus for that many people.

https://en.wikipedia.org/wiki/Peak_phosphorus

[pm90]

It’s been pretty clear for some time now that inability to inhabit anything besides earth is a huge failure mode for the species. It might possibly be the unifying goal for the species to put aside its petty differences and come together on a grand project.

I gotta say though, the chances seem extremely slim and it’s more likely that climate change would trigger a reduction in the production capacity of humanity triggering political changes that may make it impossible to solve these problems.

[antoineMoPa]

This assumes we can really empower brilliant people all over the globe.

[ravi-delia]

Unless more of the world falls out of economic development, the opposite of the trend we've seen for decades, the Einsteins born in areas that can empower them will increase in number proportionally. In fact, our current accounting of Einsteins is limited to the ones that were empowered, so if trends continue we should expect a larger than proportional increase.

[kiba]

Arguably there's less and less to discover, and revolutionary discoveries become less likely.

[dasudasu]

"In the long term we are all dead" so why would it even matter if over 2100 years your perfect 5% compounding hasn't happened as expected? No place on earth experienced nothing short of complete societal overhaul in that span -- most often multiple times at that.

[yojo]

I threw out some numbers to set a baseline of “this can’t go on forever.” I don’t pretend to know when it starts to transition, or the speed of the transition.

The fact that the rate of growth will someday slow means assumptions you make about your 401k may or may not hold if we happen to be at the wrong point on the curve.

[ivalm]

But humans also don’t have infinite demand on production. If we have a sufficiently high amount of production per capita (ie we are post-scarcity) then growth becomes irrelevant (especially if overall population isn’t growing, which with current trends seems reasonable). At that point the only thing is to ensure that output distribution is sufficiently equitable. We can then have millions of years of stable happy humanity with zero growth.

[lovecg]

Would the world of today seem “post-scarcity” to a medieval peasant? I don’t think humanity will ever have “enough” — it’s just not in our nature.

[ivalm]

I don’t agree. I don’t think we currently have enough stuff. As a techy in the US I make more than most, but there are still many material things that I can’t quite afford/have to limit myself. However, if production was 100x what it is now then perhaps I would be able to afford them. Post scarcity is at a production level far greater than current.

Also, higher production doesn’t mean higher natural resource consumption —- dematerialization is real. My phone satisfies my needs far more than a landline could but requires far less natural resources.

[lovecg]

But don’t you think your desires would grow accordingly? To go back to my “medieval peasant” example, a simple shirt is something we take completely for granted and is arguably “post scarce”. But before industrialization, people would think of a shirt sort of like we think of cars today. It took months of labor to make one, and it cost orders of magnitude more what it costs today. So to those people “I would own a couple of really nice shirts” would have been a reasonable goal to aspire for. They simply could not imagine cars or iphones.

Your second point is an interesting one. I do think we’re gradually moving to a world of artificial scarcity (brand names, luxury goods, bitcoin etc.), plus some things will always be scarce by nature (access to important people, people’s time). So perhaps we’ll just discover that the connection between resource scarcity and human desires is accidental.

[xchaotic]

The rich will not give away their stuff. You’d probably have to give away 90% of your wealth to be level with the rest of the world. Also it is communism and it doesn’t work in practice.

[ivalm]

Current world is not post-scarcity. However, there is a point where marginal value of “extra” stuff is essentially zero. When everyone can be at that level is what I mean by post-scarcity.

[spullara]

Over that time, you could populate a good portion of nearby systems. It would only take 1m years for humans to populate every possible planet in the galaxy:

https://www.learnastronomyhq.com/articles/how-long-would-it-...

Plenty of room for a very large economy. The next galaxy would take some doing though since it is 2.5 million ly away.

[Robotbeat]

Sure, but then your economy is still constrained to grow at the speed of light. A shell of some thickness expanding at the speed of light. The volume of that shell increases parabolically, not exponentially.

So it still needs to slow down growth, although growth can still occur indefinitely. Well, billions of years until stars naturally start burning out, in which case you get a slowing of growth and degrowth until you're huddled around red dwarfs for a trillion or so years and then harvesting energy from black holes until like a quadrillion years.

[onion2k]

"only a million years" seems like a slightly flippant statement in light of the fact humans have only had about 5000 years of civilisation so far.

[spullara]

After it starts it would hard to stop because of the separation except by some other group of similar creatures.

[emteycz]

Consider the last 2000 years. The global market went from nothing to outlandish emperor-level luxuries available to multiple billions of humans + all the stuff that we have that even emperors didn't imagine. I think quintillion times growth in "actual value to humans" is not that off. And thinking about the next 2000 years, I don't see any hard limits there too.

[ben_w]

A quintillion is larger than the ratio of a synaptic refractory period and a human lifetime by a factor of about 1.5 million. Even with my optimism about transhumanism, I’d be very surprised if that was accurate or will ever be.

[emteycz]

It's not about transhumanism, it's simply the (Kardashev) scale of society.

[irrational]

With the world’s population projected to severely contract in the coming centuries, I wonder how that will effect the possibility of exponential growth.

[marvin]

General artificial intelligence, once we figure it out, will cause an increase in the annual economic growth more extreme than even the industrial revolution caused. Either that or a disaster, but hopefully the former.

Anyone who thinks we will run out of things to create just lacks imagination. There's so much stuff we'd really, really like to have but haven't been able to economically create yet. Medicine, body modifications, automation of all that's boring, cheaper manufacturing of everything, entertainment, recreation, transport, space habitats and almost anything else you can imagine.

[xchaotic]

But as things get automated and cheaper that’s actually negative gdp growth. So it won’t pay your retirement

[chii]

If things get cheaper, you can live off less income from your investments .

[marvin]

Yep, and in such a world, a reasonable level of taxation would eventually be able to pay for a modest lifestyle for everyone who wanted it.

In a world of such dramatic surplus, it's either that or violent revolution, so I'm fairly sure it would sort itself out.

[quickthrower2]

The problem with this is how you measure it. In dollars? Inflation adjusted dollars? How to measure that inflation?

[patjenk]

Sorry to be dense, but where in the article does it claim that GDP will grow forever?

An as an aside, GDP is only a proxy for how prolific an economy/country is. The country's output can continue to grow without generating more dollars.

[gexla]

Parking my comment here to be part of the dense club.

Why would you even need forever growth? Is it good enough that old companies die and new companies replace the old? In that process, there's always growth to find and invest in?

[refurb]

In that scenario, the economy does become a zero-sum game, no? Your returns are just someone else's loses.

[gexla]

There's always winners and losers. We hope the market provides us with ever improving options, which could be a win for everyone. Some may lose by taking down zombies, but the market overall improves. </overly simplified and probably wrong answer>