Is economic growth slowing down permanently?

-1=e^ipi · August 28, 2015

Cite?

I thought this was common knowledge.

Transistor density is still going, but clock speed has definitely stalled. That's why they started going for multi core computers.

http://cs.stackexchange.com/questions/27875/moores-law-and-clock-speed

Bonam · August 28, 2015

After 2 decades of constant growth clock speed has stalled at 3-4GHz. My understanding is the mechanics of power dissipation make further increases in clock speed implausible. The focus in the last 10 years has been on mobile devices which run at slower clock speeds but draw less power and on increased parallelism with multiple cores. IOW, transistors may continue to shrink but we are not seeing the same jumps in computing performance as we did in the past which leads some to conclude that Moore's Law has reached a saturation point.

Moore's Law says nothing about clock speeds though. It talks about the number of transistors that can fit within a certain area doubling every fixed time scale. That law has continued on through the last decade, despite clock speeds having stayed constant at 3-4 GHz for that decade.

Furthermore, as Kurzweil argues in his writings regarding technological progress, Moore's Law is merely the last and current one of the paradigms of exponential progress in computation, which was preceded by other forms of computation.

I would extend this into the future and predict (as Kurzweil does) that when the physics precludes further exponential progress in silicon integrated circuit density, a 6th paradigm will take over where the 5th left off.

Furthermore, despite the lack of increase in clock speeds, we most certainly have been seeing the same exponential rate of progress in computing performance as we did before. It's just that that progress is in parallel computation rather than serial computation, which is just as valuable but requires software that can take full advantage of it.

Supercomputers have followed the exponential progress line to the present day. You can also find charts showing exponential growth in the computational power of consumer-available graphics cards (which are now tapped by many programs for modeling/computation), as well as exponential growth in the number of computations per unit energy expenditure.

Edited August 28, 2015 by Bonam

Bonam · August 28, 2015

I thought this was common knowledge.

Transistor density is still going, but clock speed has definitely stalled. That's why they started going for multi core computers.

http://cs.stackexchange.com/questions/27875/moores-law-and-clock-speed

See here:

https://www.google.com/webhp?sourceid=chrome-instant&ion=1&espv=2&ie=UTF-8#q=moore%27s%20law

"Moore's law" is the observation that, over the history of computing hardware, the number of transistors in a dense integrated circuit has doubled approximately every two years.

WIP · August 29, 2015

All this Moore's Law crap has missed the point that Moore's Law only applied to miniaturization and data storage - and nothing equivalent to Moore's Law works in areas like energy production and storage. So, fusion power is still 20 to 30 years away from practical application....just like it was 40 years ago, and making batteries even modestly more efficient has come at a high price because of rare metals needed to produce them.

Point being your smart phone isn't going to drive you home tonight! And if there is such a thing as a techno-fix to avert economic and ecological collapse, something close to Moore's Law would have to be happening in the production of cheap, clean energy.

Bonam · August 29, 2015

Point being your smart phone isn't going to drive you home tonight! And if there is such a thing as a techno-fix to avert economic and ecological collapse, something close to Moore's Law would have to be happening in the production of cheap, clean energy.

It is.

Cost of solar panels:

Battery energy density and cost:

Bonam · August 29, 2015

So, fusion power is still 20 to 30 years away from practical application....just like it was 40 years ago, and making batteries even modestly more efficient has come at a high price because of rare metals needed to produce them.

Despite me working in the field of fusion energy research, my personal thought is that fusion will never be relevant for terrestrial grid power production. The energy storage problem is a much easier problem to crack than fusion energy, and once it is cracked, solar can provide all the energy we need, likely much more cheaply than fusion ever could. We've got a giant free fusion reactor in the sky.

Fusion will have its place: naval reactors, space power and propulsion, outposts/bases in the outer solar system. use as a high energy neutron source, production of exotic materials, etc, but I'd say the chance of it ever providing a substantial % of Earth's power needs is pretty low.

TimG · August 29, 2015

Despite me working in the field of fusion energy research, my personal thought is that fusion will never be relevant for terrestrial grid power production.

Interesting. Why exactly?

The energy storage problem is a much easier problem to crack than fusion energy, and once it is cracked, solar can provide all the energy we need, likely much more cheaply than fusion ever could. We've got a giant free fusion reactor in the sky.

Interesting. I look at the problems with grid scale energy storage and I find it hard to believe that evolution of the currently available technology will bring the price down enough to make it economic. We need a yet-to-be-discovered breakthrough in material science to solve the price problem.

-1=e^ipi · August 29, 2015

Despite me working in the field of fusion energy research, my personal thought is that fusion will never be relevant for terrestrial grid power production.

What if we mine He3 from the moon?

Bonam · August 31, 2015

Interesting. Why exactly?

Because I think it will be inherently too expensive. Even putting aside regulatory and political issues, conventional nuclear power plants are quite expensive compared to fossil fuel plants. And fusion will be even more so. The tokamak, currently the fusion topology undergoing the most intense level of research, inherently requires gigantic superconducting magnets, immense defect-free vacuum vessels, incredibly complex real time diagnostics and control, and vast quantities of exotic and expensive materials. Even once the science is proven and the designs for the technology are already made, the high costs will remain.

Meanwhile, the cost of solar energy is declining steadily.

Interesting. I look at the problems with grid scale energy storage and I find it hard to believe that evolution of the currently available technology will bring the price down enough to make it economic. We need a yet-to-be-discovered breakthrough in material science to solve the price problem.

There is certainly much additional progress that needs to be made in energy storage. Although, even today it would not be hard to achieve, as creating an artificial lake and a hydro dam with pumped storage, even where terrain is otherwise flat, is merely an engineering and construction undertaking and can be done economically.

Anyway, batteries are a commercial product and enjoy economies of scale and major investment in further R&D by private companies, who know that if they can make a better battery they can generate more profit. Battery technology is advancing continuously and so is materials science. See the graph I posted on the previous page regarding battery cost and energy density. That trend is not stopping.

Edited August 31, 2015 by Bonam

Bonam · August 31, 2015

What if we mine He3 from the moon?

He3 doesn't really help.

First, the easiest fusion reaction to achieve is the deuterium-tritium (DT) reaction. Deuterium is abundant on Earth, while tritium has a short half life and needs to be produced. It can be bred from lithium using neutron bombardment. He3 is not necessary for this process.

But the DT reaction releases most of its energy in a high energy (14 MeV) neutron. The high energy neutron tends to penetrate right through everything and it is hard to use this energy for productive purposes. The high intensity neutron radiation from the DT reaction would also cause the reactor walls and structure to become radioactive over time, which is not ideal.

The holy grail of fusion energy research is achieving the hydrogen(proton)-Boron (pB) reaction, because it is aneutronic (produces no neutrons). This reaction is much harder to achieve because of its much smaller reaction cross-section. The plasma would need to be orders of magnitude hotter (and so more difficult to contain) compared to a DT plasma. Nonetheless, it is the ultimate goal for some research groups in the field. It also does not require He3.

The reaction that uses He3 (D-3He), is intermediate between the two above. It is still much harder to achieve than the DT reaction (though easier than pB) and still produces some neutrons due to side reactions (though much less than DT).

The result is that first generation / demonstration fusion reactors will work on DT, which is the only thing even remotely in reach of current fusion energy research schemes. Meanwhile, as we learn more about how to contain HEDPs (high energy density plasmas), people will continue to try to achieve the pB reaction. Maybe someone will want to try out D-3He as an intermediate generation of fusion energy production, but it's not a critical path nor would it make fusion a reality sooner, nor cheaper.

https://en.wikipedia.org/wiki/Nuclear_fusion#Criteria_and_candidates_for_terrestrial_reactions

eyeball · August 31, 2015

But the US will grow again. And so will the world.

When you say the world will grow you're not including the planet are you? They're clearly two different things.

It would be easier to buy into the idea of growth continuing on and on and on if we were busy mining asteroids, beaming energy down to Earth, building cities on the moon and preparing the first colony ships to the stars but given none of that is happening or looks like its about to...so what about the planet? If the planet were a company or commodity you could buy stock in what would the fundamentals look like?

msj · August 31, 2015

When you say the world will grow you're not including the planet are you? They're clearly two different things.

It would be easier to buy into the idea of growth continuing on and on and on if we were busy mining asteroids, beaming energy down to Earth, building cities on the moon and preparing the first colony ships to the stars but given none of that is happening or looks like its about to...so what about the planet? If the planet were a company or commodity you could buy stock in what would the fundamentals look like?

I'm talking about old fashioned, problematic, GDP.

The world generally does grow, economically.

With the great fusion reactor in the sky I think there are going to be lots of opportunities for growth in energy.

While there are concerns about certain elements becoming rare (helium for example) most commodities are not going to run out.

With population trends declining it may only take 4 generations to reduce population levels down to a reasonable level anyway.

That would be a headwind for economic growth - but I would still expect that GDP per capita would still rise - so there would still be growth.

eyeball · August 31, 2015

Measuring growth in GDP per capita is the same as measuring growth in the economy? I can certainly see how there would be more to go around if there were less people but that's not the reality growth is currently predicated on.

I still think the planet's ability to produce our goods and absorb our wastes is what at's issue and we're playing chicken and betting we'll win the race towards the planet's ability to sustain us and some sort of imagined financial/technological singularity and Cornucopian existence where nothing can hurt us.

GostHacked · August 31, 2015

Growth cannot happen forever.

Bonam · August 31, 2015

Nothing can happen forever, but things can happen for a long time. Is the ~10,000 years of the history of human civilization almost the full story, or is it only the beginning?

Edited August 31, 2015 by Bonam

eyeball · September 1, 2015

Nothing can happen forever, but things can happen for a long time. Is the ~10,000 years of the history of human civilization almost the full story, or is it only the beginning?

It could still go either way but the little conservative whispering in my ear says we're pushing our luck.

And what you've described is not really one big long human history it's a whole bunch of short stories about a whole bunch of civilizations. Things stop happening forever a lot more frequently than you're implying.

bush_cheney2004 · September 1, 2015

Nothing can happen forever, but things can happen for a long time. Is the ~10,000 years of the history of human civilization almost the full story, or is it only the beginning?

Same question was asked by Zager and Evans in 1969:

Moonlight Graham · September 1, 2015

It is.

Cost of solar panels:

Interesting, thanks for the graphs. Thing is, I wonder how much solar energy (panels, batteries etc.) would cost after a decade or 2 of mass production & use. I'm sure the price would plummet. Imagine a solar panel on the majority of roofs, powering buildings and electric vehicles. There would also be some health cost savings with improved air quality, ie: decrease in asthma rates. Not to mention trillions in savings over oil wars in the middle-east, and the savings from avoiding climate change adaptation from fossil fuels.

Big Guy · September 1, 2015

It is official - Canada is in a recession.

http://www.cbc.ca/news/business/gdp-economy-recession-1.3210790

blueblood · September 1, 2015

Interesting, thanks for the graphs. Thing is, I wonder how much solar energy (panels, batteries etc.) would cost after a decade or 2 of mass production & use. I'm sure the price would plummet. Imagine a solar panel on the majority of roofs, powering buildings and electric vehicles. There would also be some health cost savings with improved air quality, ie: decrease in asthma rates. Not to mention trillions in savings over oil wars in the middle-east, and the savings from avoiding climate change adaptation from fossil fuels.

Maybe solar panels used as roofs. Just need to deal with cloudy and night... Thats where fossil fuels has its edge. As long as other countries are coming into their own, fossil fuels will have a place.

msj · September 1, 2015

Interesting, thanks for the graphs. Thing is, I wonder how much solar energy (panels, batteries etc.) would cost after a decade or 2 of mass production & use. I'm sure the price would plummet. Imagine a solar panel on the majority of roofs, powering buildings and electric vehicles. There would also be some health cost savings with improved air quality, ie: decrease in asthma rates. Not to mention trillions in savings over oil wars in the middle-east, and the savings from avoiding climate change adaptation from fossil fuels.

Exactly.

Imagine solar/wind coupled with batteries and backed up with hydro/nuclear/natural gas generators.

We could ignore the ME and Canada/US would be entirely energy independent. Even Europe would be a winner as North America sends natural gas over there.

sharkman · February 11, 2016

It is official - Canada is in a recession.

http://www.cbc.ca/news/business/gdp-economy-recession-1.3210790

I wonder what the US voter is thinking these days. Their economy had been bumbling along at a mediocre pace although Obama said he fixed it and was pumping over 70 billion into it per month for 5 years or so. Now that he's finally had to cut back on that and raise the interest rates ever so slightly, the DOW has been troubled, and since January has been losing ground. Today is another bad day for stocks, and money seems to be heading for gold again.

msj · February 12, 2016

I wonder what the US voter is thinking these days. Their economy had been bumbling along at a mediocre pace although Obama said he fixed it and was pumping over 70 billion into it per month for 5 years or so. Now that he's finally had to cut back on that and raise the interest rates ever so slightly, the DOW has been troubled, and since January has been losing ground. Today is another bad day for stocks, and money seems to be heading for gold again.

1) this stinks of the recency effect: yes the markets are volatile right now, it will pass as it always does.

2) US jobs are still being created at something like 69 months in a row (private sector jobs).

3) for those who are on stocktwits you can probably see this: https://stocktwits.com/robertlesnicki/message/49474998?utm_medium=community-twitter&utm_campaign=outboundteam&utm_source=t.co

Basically, the S&P 500 was up 114% over Reagan's two terms, 52% over GH Bush's, 209% under Clinton's two terms, negative 40% over GW Bush's two terms, and, so far, up around 128% or so under Obama.

4) if you think that it matters if a Republican or a Democrat is in the Whitehouse and that it somehow has an effect on the stock market then you don't understand markets.

5) Obama does not set interest rates. The Fed is independent to reduce the effect of politics on monetary policy. If you think otherwise then you don't understand even the basics of how the US system works.