This Can’t Go On

OP takes a historical view about the actual economy, and where it in real life can go from here. Your argument – a very common one and very influential in today's society – however is purely theoretical. As John O'Neil points out:

"It is logically possible to have increasing GDP and a decreasing physical and energy throughput in an economy. However, it is a fallacy to move from claims about what is logically possible to claims about what is physically possible and another from what is physically possible to what is empirically actual."

If you actually look at the coupling of environmental pressure with historical empirical economic growth measured by GDP, we can't go on like this for even a few decades more, if we want to avoid complete ecological collapse. Trusting humankind's future on theoretical armchair economics magically becoming a reality, despite decades of evidence to the contrary, is complete madness.

The rebuttal to this is presented in "Exponential Economist vs Finite Physicist": https://dothemath.ucsd.edu/2012/04/economist-meets-physicist...

The argument boils down to that all economic activity, whether material or just information, requires energy as an input, and at least on one planet, the ability to heat-sink that energy (let alone generate it) must set some finite growth limit.

Right, I don't get what atoms have to do with it at all. Today, I can buy an iPhone for $1000 that's a phone, camera, encyclopedia, game console and more and more. Just 30 years ago this would have required a lot more atoms (including energy input, I suspect) and a lot more money.

This is a tired old argument.

Economic growth due to technological advance is completely swamped by economic growth due to increased throughput.

It is the throughput, the increase in entropy, that is at issue. Almost all economic growth comes from increasing the flows from sources to sinks. E.g. Turning coal into CO2

I agree. The book "Factfulness" by Hans Rosling is an excellent read about the topic.

Yes, growth in economy and population can go on and will likely result in an S-bend curve in the distant future.

Demographic leveling or collapse is in a category I've heard called "demand limits to growth."

Everyone always thinks about supply limits, with some making dire Malthusian prophecies of what happens if we exhaust our resources with runaway growth. I've seen surprisingly little discussion of demand limits.

Other possible sources of demand limits include: satiation (people just feeling satisfied and not wanting much more), diminishing marginal utility of wealth, diminishing impact of new technologies (compare the next CPU density node to jet aircraft or antibiotics), consolidation of functionality into fewer more versatile products (e.g. computers and smart phones eating TVs, tape decks, radios, book shelves, CD collections, etc.), cultural shifts toward simplicity, replacement of complex labor-intensive technology with simple low-maintenance technology (e.g. EVs have 1/10th the moving parts of typical ICE cars), etc.

The eternal growth assumption assumes both infinite supply and infinite demand.

> The number of computer programs is incredibly large. Even assuming a dense packed instruction set, even 512 bits is already enough to contain more programs than can be assigned to individual atoms in a universe! That's right. We could label every single atom with a number that would fit in the AVX instruction registers. That scale is mind-blowing, but it's true.

You're going the wrong direction. The issue isn't fitting atoms into data. It's fitting data into atoms. You can give every atom a GUID, but you can't give every GUID an atom!

You don't have to put it in terms of economy. Our rate of growth for energy and data are unsustainable. Current growth would hit a wall informed by our understanding of physics somewhere very roughly in the thousands-of-years-range. So either we go beyond our needs for space and energy or growth slows down. Both of those are a big change from the status quo.

You can only run so much code with so many electrons. At a certain point, you run out of electrons.

Imagine a world in which there is just one person and the economy is just how big of a number this person can make on a computer.

If there is more than one person or more than one computer - you have less electrons at your disposal - because they're making up other people & computers respectively.

This is obviously very far into the future - but so is 8000 years!

Between 1960 and 1970 - global GDP increased by about ~$1T. That was about the same increase between 2018 and 2019 - even after adjusting for inflation.

That was a low-growth year for the last decade.

In the 1960s we had the entire space race and a lot of the modern computer (including the Internet).

If the trend continues, in another 60 years, we could accomplish about as much in a month. In another 120 years, about as much in a day. In another 120 years, about as much in an hour. Another 180 years, about as much in a minute, and in another 180 years, about as much in a second.

It's hard to argue you haven't hit the singularity at that point - and that's only 660 years away.

8000 years at this current growth rate would be truly unimaginably alien - and I'm not sure why everyone is convinced it could keep growing or would even need to.

If the population keeps growing at current rates - we'd have enough productivity for everyone to have a higher quality of life than Jeff Bezos does today in <700 years - at least, inflation adjusted, the average person on the planet would have >$15Bn in annual income.

This doesn’t really apply to housing or congestion of any sort. Space can be recycled but only when you’re not using it. Everyone and everything takes up space. Things like AirBnb only go so far to make cities more efficient. people don’t like getting packed in, so they will pay for more space.

> The fallacy here is the equivalence of economic goods to material goods. Many economic goods are not material.

But what part of the economy does this represents? Most of the Internet run through ads, whose objective is to sell stuff. The real fallacy is the knowledge economy here. The value of Internet is that it sells physical stuff, not that it makes people smarter or happier.

Yes, but renewables would only give us a chance to see growth plateau without our civilization disintegrating. The point the article makes is the year over year multiple-percent economic growth is impossible, and the limiting factor becomes the number of atoms in the galaxy on a timescale shorter than human civilization's total history.

The other part is increasing knowledge, which allows us to use energy more efficiently. This could theoretically allow continued growth with the same amount of energy.

> At the end of the day "this" (economic growth) is powered by using energy to multiply the impact of human labor.

That's just a restatement of the Victorian/Marxist model of industrial labor.

The actual reality is that energy intensity of the OECD economies has been dropping for decades. As well attested in several sources such as, say, EIA data.

I'm a big fan of energy consumption BTW, and think it should increase significantly on a per capita basis, especially in poorer regions (though not from fossil sources). I just have been following this particular statistic for years and see that growth is not proportional to energy consumption.

Ultimately an economy has to meet the fundamental physical needs of human beings (at least until we're all Singularitied into Hive Mind Cryostatsis Upload Tanks or equivalent). That's food, clothing, shelter, and the rest of it --- the base of Maslow's Pyramid.

More money can solve some distributional probems, but ultimately not the problem of insufficient production or supply.

The classic "toy economy" example of this is "The Economic Organisation of a P.O.W. Camp", which looks at the use of cigarette-based currency within a World War II prisoner of war camp.

https://www.jstor.org/stable/2550133

It's not a perfect analogue of a true economy (goods were largely supplied to the camp through the German authorities and Red Cross shipments, there was little actual production or labour). But what was illustrated were both price fluctuations as the currency (cigarettes) increased and decreased in prevalence, subject to an innate destructive demand (habitual smokers). And when the actual supply of goods (food and other items) dried up late in the war, no amount of currency could in fact make the camp economy function.

The point isn't that physical limits will only be a problem in 8200 years, it is to counter the defective idea that such growth can be sustained indefinitely. We don't have access to all the atoms in the galaxy, and can't even make use of most of the ones we have here on earth. We will hit a wall much, much sooner.

For some perspective, 8000 years ago we were still in the Stone Age.

I think stagnation, or at least a change through successively smaller exponential regimes, should be the base case. I'm bullish on our future tech prospects and I know we have the entire universe from which to draw resources. But for all the real world cases of exponential growth, I think they are all limited.

And I only exclude collapse as a base case because I'm an optimist.

This! In a world where oil discovery, extraction and distribution was ever increasing, betting on a bigger energy pie in the future made sense. All of our macro-economic theories were developed under the naive assumption of an increasing fossil-fueled bonanza.

But in a world of no new oil fields and increasing costs for extraction, we no longer have the ever-growing pie to fund our future financial obligations. Credit looks risky and economic decline seems inevitable based on the energy availability alone.

That's not even mentioning the externalities of fossil fuel consumption: climate impacts, ocean acidification, mass extinction, etc. make these economic issues even harder. Or is it the other way around? Either way, "this can't go on" and fossil fuels are a central part of the equation.

Is it not interesting because there's nothing to argue? If the author focused on something from Limits to Growth, it would be possible to argue about whether the world was actually that way. As a logical argument based only on historical trends and physical impossibility, it's a much stronger argument than it would be were it possible to pick holes in the detail.

https://dash.harvard.edu/handle/1/37364868 my digital person from the CT camp will intersect with this BAU2 version or ... not

The world3 model is not really something you can use to make predictions- it’s got variables like “resources” with no units. And the conclusion you get from it was programmed into it.

I’d put my money on “steady state” or maybe “steady state but after a contraction due to climate change”, but not due to the world3 model.