Comments on: Economics meet Thermodynamics

Economics meet Thermodynamics

jeffburdges — Thu, 12 Apr 2012 12:27:44 -0800

"We do not share the view of many of our economics colleagues that growth will solve the economic problem, that narrow self-interest is the only dependable human motive, that technology will always find a substitute for any depleted resource, that the market can efficiently allocate all types of goods, that free markets always lead to an equilibrium balancing supply and demand, or that the laws of thermodynamics are irrelevant to economics."

By: gauche

gauche — Thu, 12 Apr 2012 12:40:57 -0800

I am neither a physicist nor an economist. I thought that was very interesting. Thank you.

By: gauche

gauche — Thu, 12 Apr 2012 12:47:39 -0800

And now, a thought: The science of economics qua science post-dates the Industrial Revolution, does it not? Are the baseline measurements from which economics draws its normative assumptions are drawn, not on a baseline at all, but on a curve? Is it possible that those measurements were drawn in a time that is an exception rather than a rule? Can economics perhaps be in a condition similar to that described in the first chapter of After Virtue?

By: Edgewise

Edgewise — Thu, 12 Apr 2012 12:47:47 -0800

Very interesting ideas, and I liked the telling, as well. Reminds me of how many of Plato's works were framed as dialogues, but I think Dr. Murphy is much more fair to his (actual) opposition.

By: oneswellfoop

oneswellfoop — Thu, 12 Apr 2012 12:56:46 -0800

Economics is not a science. It is a religion with the local currency as its God. This applies to Marxist/Socialist/Keynsian Economics as well as Randian/Capitalist/Friedmanist Economics. If we all spent less time concerned with money, we would all be better off, regardless of our economic status. But way too often the High Priests of Economics won't let us (It's like mandatory church attendance 5-6 days a week). Bitcoins and Greek TEMs are not the answer. If I had the answer, I'd write a book. But the publishers would never buy it.

By: grobstein

grobstein — Thu, 12 Apr 2012 12:58:25 -0800

This is a pretty good exchange; I like it better than previous times this dude has been posted to the blue. But the economist's point is important and deserves to be drawn out further: economic growth doesn't correspond to any easily identified physical quantity. Economic growth does not require energy growth because some sources of growth are energy negative. Tele-conferencing vs. flying is a good example -- if we hold everything constant and replace flying with tele-conferencing, we are plausibly richer while consuming less energy and less of most any resource you care to name. Wealth has, in a sense, come out of nowhere. This doesn't violate physical laws because wealth is not a quantity to which those laws apply. Neither is economic growth in a rigid correspondence with GDP growth. GDP growth is a widely used estimate of economic growth and nothing more. We can in principle become richer while GDP declines or stagnates. I think Murphy (the physicist) is right, on the ultimate question, insofar as it even makes sense. But the unnamed economist concedes the ultimate question at the beginning of the dialog -- the earth will turn into ash at some point, and economic growth will be over. The debate is really about qualitatively how much growth is left, and how much growth is threatened by resource scarcity of the kind we can now identify. And here I think Murphy is a little bit crazy. The problem is neatly illustrated by the exchange over fancy desserts:

Economist: Well, for instance, look at this dessert, with its decorative syrup swirls on the plate. It is marvelous to behold. Physicist: I'm smiling because this reminds me of a related story. I was observing at Palomar Observatory with an amazing instrumentation guru named Keith who taught me much. Keith's night lunch—prepared in the evening by the observatory kitchen and placed in a brown bag—was a tuna-fish sandwich in two parts: bread slices in a plastic baggie, and the tuna salad in a small plastic container (so the tuna would not make the bread soggy after hours in the bag). Keith plopped the tuna onto the bread in an inverted container-shaped lump, then put the other piece of bread on top without first spreading the tuna. It looked like a snake had just eaten a rat. Perplexed, I asked if he intended to spread the tuna before eating it. He looked at me quizzically (like Morpheus in the Matrix: "You think that's air you're breathing? Hmm."), and said—memorably, "It all goes in the same place." My point is that the stunning presentation of desserts will not have universal value to society. It all goes in the same place, after all. [I'll share a little-known secret. It's hard to beat a Hostess Ding Dong for dessert. At 5% the cost of fancy desserts, it's not clear how much value the fancy things add.]

The physicist basically thinks that the dessert is a lump of raw materials; arrangement doesn't matter; a Hostess cake is basically the same thing but way cheaper; you're a sucker if you'd actually buy a fancy dessert. Well of course if you think this way the possibility of growth without increased resource exploitation seems impossible. But it's insane to view the world this way.

By: triggerfinger

triggerfinger — Thu, 12 Apr 2012 12:58:52 -0800

This is interesting. I would love to see an actual economist weigh on on it here.

By: grobstein

grobstein — Thu, 12 Apr 2012 13:05:20 -0800

Having now read the epilogue, I see that the physicist concedes my point. He had not previously realized that growth was not a physical quantity of the sort he is used to, so he thought it was subject to hard limits that it's not. That explains why his previous posts on the subject have been so obtuse. All that said I think it's fairly likely that 2% annual growth will fade into the past. The historical average growth rate over human history is a small fraction of 1%.

if we accumulate knowledge, improve the quality of life, and thus create an unambiguously more desirable world within which to live, doesn't this constitute a form of economic growth?" I had to concede that yes—it does. This often falls under the title of "development" rather than "growth." I ran into the economist the next day and we continued the conversation, wrapping up loose ends that were cut short by the keynote speech. I related to him my still-forming position that yes, we can continue tweaking quality of life under a steady regime. I don't think I ever would have explicitly thought otherwise, but I did not consider this to be a form of economic growth. One way to frame it is by asking if future people living in a steady-state economy—yet separated by 400 years—would always make the same, obvious trades? Would the future life be objectively better, even for the same energy, same GDP, same income, etc.? If the answer is yes, then the far-future person gets more for their money: more for their energy outlay. Can this continue indefinitely (thousands of years)? Perhaps. Will it be at the 2% per year level (factor of ten better every 100 years)? I doubt that. So I can twist my head into thinking of quality of life development in an otherwise steady-state as being a form of indefinite growth. But it's not your father's growth. It's not growing GDP, growing energy use, interest on bank accounts, loans, fractional reserve money, investment. It's a whole different ballgame, folks. Of that, I am convinced. Big changes await us. An unrecognizable economy. The main lesson for me is that growth is not a "good quantum number," as physicists will say: it's not an invariant of our world. Cling to it at your own peril.

By: grobstein

grobstein — Thu, 12 Apr 2012 13:10:11 -0800

A more economically sophisticated version of this argument has been slowly raging between Bryan Caplan and Robin Hanson, with Caplan favoring "indefinite" growth and Hanson favoring physically constrained growth. I think Hanson has the right of it but the disagreements between them are not likely to have any consequences in the near future. Sample post from 2009; I believe the debate is ongoing.

By: grobstein

grobstein — Thu, 12 Apr 2012 13:13:12 -0800

And it's worth acknowledging that in the sci-fi upper reaches of this argument, we start running into metaphysical problems, versions of Parfit's "repugnant conclusion" argument.

By: valkyryn

valkyryn — Thu, 12 Apr 2012 13:25:49 -0800

Umm. People have been trying to apply physics to economics like, forever. Previously. Interesting, this seems less like an attempt to do economics like physics than to simply examine the physics implications of various economic models. This is a different project altogether, and, if anything, it underscores the fact that economics is not like physics. When an economist comes up with a model that suggests unbounded growth he says, "I've solved the economy!" A physicist does the same thing and says, "Well that's not right. I must have screwed up somewhere."

By: -harlequin-

-harlequin- — Thu, 12 Apr 2012 13:30:07 -0800

This is important. People who profess to experts in economics need to stop living in their collective delusion and start working on the hard questions and coming up with useful advice, such as possible ways to steer a resource-consumption-growth economy into a steady-state resource-consumption economy without killing growth in quality of living. QFT: If humans are successful in the long term, it is clear that a steady-state economic theory will far outlive the transient growth-based economic frameworks of today. Forget Smith, Keynes, Friedman, and that lot. The economists who devise a functioning steady-state economic system stand to be remembered for a longer eternity than the growth dudes. [Economist stares into the distance as he contemplates this alluring thought.] As the rest of us should be thinking about the things we're experts in - how those things can be changed to allow increasing quality of life without creating incentive for increasing resource consumption.

By: jeffburdges

jeffburdges — Thu, 12 Apr 2012 13:34:36 -0800

If you accept "knowledge" as your measure of economic growth, then yes we all hope that can grow effectively indefinitely, but.. First, "expanding knowledge" simply isn't what the economists are selling to our politicians. Second, these arguments demonstrate that steady state economics, or more likely decline economics, must soonishly become vastly more important than exponential or knowledge economics, perhaps that's Hanson point. Third, there are already several subjects that study the limits of knowledge, such as complexity theory, mathematical logic, etc. and epistemology, but none resemble economics in the slightest. And amongst the the most successful of these is information theory, which resemble thermodynamics through and through.

By: grobstein

grobstein — Thu, 12 Apr 2012 13:39:24 -0800

So what? You don't think the economy is near any kind of information-theoretic limit, do you? Stated most generally, the pro-growth point is: we're far away from optimums along many dimensions; moving towards those optimums would constitute economic growth.

By: euphorb

euphorb — Thu, 12 Apr 2012 13:46:35 -0800

Its not really controversial to say that exponential growth doesn't go on forever. And it's somewhat ridiculous to ask a scientist or economist to make claims about events 500 years from now. Extrapolation is always dangerous but if we imagine that energy use on earth is bounded by the amount of solar insolation (because any more would heat the planet), then we arrive at that magic number of 10^15 watts at around 2200. This is from the chart in the link. But that limit only applies to earth. By that time we could have factories on the far side of the moon producing everything we need and dumping any waste heat produced into space or into the moon, assuming there is any waste heat. And waste heat may not be an issue because we could be using isothermal processes instead such as DNA computing. It's useless now but it shows that computation can be vastly more efficient than can be accomplished by pushing electrons around. Switching to pushing photons around or to quantum computing would be a huge improvement. Essentially the theoretical limits that we bump into with exponential growth or so far into the future that those theoretical limits will be obsolete by then. Also, here's the obligatory SBMC comic.

By: valkyryn

valkyryn — Thu, 12 Apr 2012 13:52:19 -0800

Of course, it was also an economist who said that "If something cannot go on forever, it will stop."

By: saulgoodman

saulgoodman — Thu, 12 Apr 2012 14:20:42 -0800

The physicist basically thinks that the dessert is a lump of raw materials; arrangement doesn't matter; a Hostess cake is basically the same thing but way cheaper; you're a sucker if you'd actually buy a fancy dessert. I think you're wrong. The point from a physicist's point of view is you can't rearrange those raw materials for the Nth time to create an even fancier dessert without expending some amount of physical energy, even if it's less than you might expend building a new factory. Even thinking about how to arrange things actually has a literal, describable physical energy cost--there's no physical-cost-free activity, period, despite what economists in their scientifically challenged view of reality might like to theorize. There is no ghost in the machine in the world of economic reality anymore than in the human body; everything at some level is physically describable, even "services."

By: polymodus

polymodus — Thu, 12 Apr 2012 14:42:40 -0800

I think the article gets it mostly right. But the author is an outsider, which explains why he doesn't see that economics is actually fundamentally about entropic and computational constraints on human well-being. It's just that economics in practice is in comparison very limited in scope, concerned with modeling man-made economic systems and pragmatic implications of that. The contribution of this article is that probably economists have left the deeper philosophical problems on their backburners a little too long, and that it would be wise to return to first principles and try to grapple with some of these futuristic problems.

By: George_Spiggott

George_Spiggott — Thu, 12 Apr 2012 14:44:45 -0800

Economics needs to have its Allan Sokal moment.

By: gauche

gauche — Thu, 12 Apr 2012 14:45:45 -0800

Why? Greenspan has already made Stanley Fish look like a mathematician.

By: polymodus

polymodus — Thu, 12 Apr 2012 14:47:33 -0800

The dessert example is great but could have been presented clearer. It is basically an observation about the incomparability of art. Art is the only human endeavor that subverts a lot conceptions about progress or growth.

By: -harlequin-

-harlequin- — Thu, 12 Apr 2012 15:22:25 -0800

One of the challenges is that wealth made firmly from resources tend to enrich my life for a long time (eg a pair of shoes, or an ipad, a house), while wealth made primarily of services tend to enrich my life only momemtarily (eg a massage, a great cocktail, a concert). Yet it's only a matter of degree - the concert ends before the shoes do, but shoes still end up losing the quality of enriching my life (ie end up in landfill) There are exceptions to this trend, but I think we'll need to find ways to turn the trend entirely on its head if we are to have any hope of a world where desire for improved quality of life doesn't absolutely ensure resource destruction.

By: valkyryn

valkyryn — Thu, 12 Apr 2012 15:36:03 -0800

The point from a physicist's point of view is you can't rearrange those raw materials for the Nth time to create an even fancier dessert without expending some amount of physical energy, even if it's less than you might expend building a new factory. Not the point. Rather, the amount of energy that goes into making a Ding Dong is actually not that different from the energy that goes into making a fancier dessert. Same amount of raw materials, and a trivial difference--if any--in human labor. The fancier dessert is potentially less energy costly, as it doesn't include the amortization costs of a huge factory of massive transportation costs from said factory to the bodega. So we're looking at an instance where less energy expenditure is actually more economic value. True, we're still looking at a finite expenditure of energy, and that can't go on increasing forever. But the fact that energy expenditure and economic productivity are not directly related in any predictable way does pose some problems for the physicist here. I think the way to answer that is to say "Okay, maybe that's true. But the 3% annual growth we've seen for the past x years has been related to increased energy expenditures, in particular, population growth. Massively so. That's the low hanging fruit. The suggestion that any significant portion of the economy can be devoted to the sort of thing which is both high-"value" and low-energy doesn't hold water. So even if it's true, it's trivial. The underlying fact that most growth comes from increased energy expenditure remains true on the macro level."

By: TypographicalError

TypographicalError — Thu, 12 Apr 2012 15:36:54 -0800

I think you're wrong. The point from a physicist's point of view is you can't rearrange those raw materials for the Nth time to create an even fancier dessert without expending some amount of physical energy, even if it's less than you might expend building a new factory. Even thinking about how to arrange things actually has a literal, describable physical energy cost--there's no physical-cost-free activity, period, despite what economists in their scientifically challenged view of reality might like to theorize. There is no ghost in the machine in the world of economic reality anymore than in the human body; everything at some level is physically describable, even "services." You don't rearrange the raw materials N times, you rearrange them once. Consider that an economic agent finds things in state A and transmutes them to state B, in the process consuming x energy and adding y wealth to the economy. Then a transformative figure comes along and discovers the process to take things in state A and transmute them to state B', in the process consuming x' < x energy and adding y' > y wealth to the economy. Since state B' is a substitute for B in most cases, the relevant economic agents change their behavior to the more lucrative opportunity, and we have economic growth and less energy consumption. Note that even if not all consumers of B become consumers of B', anyone switching from B to B' is a net energy savings for the world.

By: -harlequin-

-harlequin- — Thu, 12 Apr 2012 16:09:47 -0800

I think the point of the dessert (lets make it "a" point) was that gains top out after a while. With greater knowledge and skill, a desert that is valued more highly to some for the same or less resource cost, but that even these gains are finite and limited and cannot open the door to unlimited growth. Dessert is an illustration that energy is not the same as wealth, but the limits on how much value the chef can add to the resources when not everyone even views that value as positive, means that energy not necessarily correlating to wealth still can't handwave away the underlying problem.

By: -harlequin-

-harlequin- — Thu, 12 Apr 2012 16:19:15 -0800

I also think it's a gross misunderstanding to take the physicist's and economist's mutual agreement that there are forms of growth that need not be limited, and somehow leap to thinking that the elephant towering over our entire societies will take care of itself. Unless "take care of itself" is in the same sense as "flattens". We don't want to be crushed and destroyed. We want a transition (to a currently-unknown but probably frighteningly different basis of economics) that is pleasant and acceptable. And the path we are on right now does not deliver anything of the sort. This will be one of the great problems of our age, and yet few seem to be grappling with it.

By: bukvich

bukvich — Thu, 12 Apr 2012 16:21:01 -0800

I like the analogy of expensive pastry compared to a Hostess Ding Dong. That is almost the perfect illustration of decreasing marginal utility (at least for those who do not have strong cravings for sugary food).

By: dhartung

dhartung — Thu, 12 Apr 2012 16:38:40 -0800

In the long run we are all dead.

By: -harlequin-

-harlequin- — Thu, 12 Apr 2012 16:45:24 -0800

Some of us would like to give a positive legacy to future people to enjoy during those moments when they're not dead, as we have enjoyed positive things left to us by those before us. :)

By: nixerman

nixerman — Thu, 12 Apr 2012 16:47:03 -0800

The focus on growth in modern economics is unfortunate as "growth" is, as people should now realize after the "Bush Boom" of 2003-2008, not nearly as real and enduring as economists would have you believe. There is another way to model what's being created within an economy and that is to focus on risk. Unfortunately risk is an even more difficult to understand concept -- people have some intuitive notion of a society "growing" -- but ask them to visualize a society maximizing its risk optima, and, well, they'll look at you like they're insane. What's interesting is that there is indeed a very real connection between risk and entropy and I think it's slowly starting to dawn on people that neoclassical economics truly fails when it comes to actually understanding risk and will simply have to go. The most fundamental assumption, the belief that "wealth" is profits and thus profits must be maximized, is broken. Real wealth is the ability to absorb loss/risk which is why, even after the crash wiped out trillions of dollars of "wealth," the best positioned were actually able to become significantly more wealthy as the market plummeted. One of the challenges is that wealth made firmly from resources tend to enrich my life for a long time (eg a pair of shoes, or an ipad, a house), while wealth made primarily of services tend to enrich my life only momemtarily (eg a massage, a great cocktail, a concert). Of course this is completely untrue and, unless you're a 3rd world peasant farmer, the exact opposite is true. The vast majority of your real wealth is tied up in your education, your social connections, your bodily health and what the ancient Greeks would call your 'character'. This is what lets you feed yourself and these are the real reserves you will draw upon in a time of crisis. But again this really shows just how broken neoclassical economics, with its inability to even successfully identify the real wealth that most people enjoy, really is and why something else is desperately needed.

By: anigbrowl

anigbrowl — Thu, 12 Apr 2012 17:02:16 -0800

First, I'll just mention that energy growth has far outstripped population growth, so that per-capita energy use has surged dramatically over time—our energy lives today are far richer than those of our great-great-grandparents a century ago [economist nods]. So even if population stabilizes, we are accustomed to per-capita energy growth: total energy would have to continue growing to maintain such a trend [another nod]. Well, wrong, because technology can also shrink your power consumption: computation per watt continues to rise, advances in design and operation allow the consumption of less power in motor vehicles and planes for the same number of persons traveling the same distance, and so forth. While the overall trend of human power consumption has been upward, that is by no means inevitable and many thermodynamically-neutral technologies (solar and ind power, for example) have only been deployed on any large scale very very recently in historical terms, so have not yet had much opportunity to reshape the trend. I like the Do the Math blog but a lot of his economic reasoning is extremely facile.

By: aubilenon

aubilenon — Thu, 12 Apr 2012 17:45:25 -0800

Yeah, I'm not convinced either. I see that fancy dessert example as a weak illustration of the value of design. A laptop computer today is much more capable than a desktop computer from 10 years ago, while using less energy, less matter, and less manufacturing effort. And that exact same laptop can potentially become more valuable to me, if software is written to enable me to do things I can not do now. There may be some informational-theoretical limit to how much raw computing power one can do with a Watt-hour of power, but I don't know of any reason to think there's a limit on how useful that computation can be.

By: Diablevert

Diablevert — Thu, 12 Apr 2012 17:47:58 -0800

Well, wrong, because technology can also shrink your power consumption: computation per watt continues to rise, advances in design and operation allow the consumption of less power in motor vehicles and planes for the same number of persons traveling the same distance, and so forth. Such increases in efficiency have been happening for decades; the increase in per-capita energy usage continues apace. Jevon's Paradox in action: As the cost of energy drops, new activities become economically viable. Switching from transistor to microchip enables magnitudes more available computational power per joule; it also transformed computers from something only an elite university fat on wartime project funding could possibly afford to something I can have six of. Me and my 3 billion friends. Have you no rebuttal to the physicist's comments on the prospect for exponentially improving the efficiency of energy-generating technologies? If current solar is currently 25% efficient, then at max it can quintuple.

By: Kid Charlemagne

Kid Charlemagne — Thu, 12 Apr 2012 18:59:56 -0800

Rather, the amount of energy that goes into making a Ding Dong is actually not that different from the energy that goes into making a fancier dessert. Remember when the guy tried to make a toaster all on his own from scratch? Economies of scale are economical because once you have a big machine cranking identical things out en masse you get a huge energy advantage. If Hostess churns out 100,000 Ding Dongs a day at the regional Ding Dong factory, that's what, 1000 chefs creating 100 fancy deserts each? Work out the surface to volume ratio of the pass through oven at the Hostess factory vs the 1000 ovens those 1000 chefs use and you'll find a huge difference that is going to translate into a huge waste of energy. And this type of cost difference is going to be there at almost every aspect of the process. For most things, if there is to be decreased power consumption, it's going to happen by running fewer processes at larger scales. Eventually, though, that savings is offset by the cost of getting those products to the consumer.

By: thrako

thrako — Thu, 12 Apr 2012 19:01:40 -0800

Interesting. I think he reaches the right conclusion at the end: Theoretically, human welfare can grow exponentially, using only a fixed amount of resources per unit of time. Ultimately this growth would not come from tangibles (like cars and ipads) but intangibles (art, as someone mentioned, or the glass bead game). Imagine the government fixes the amount of resources that can be used. There can still be a fully functioning market economy that uses those resources, where most people work in the production of intangibles. In short, I think it is clear that human welfare can increase indefinitely without using more resources per unit of time. He appears to be really confused about the definition of GDP, price levels, and how welfare is measured. But it is late.

By: sneebler

sneebler — Thu, 12 Apr 2012 19:06:07 -0800

polymodus: The dessert example is great but could have been presented clearer. It is basically an observation about the incomparability of art. Art is the only human endeavor that subverts a lot conceptions about progress or growth. While I sort of agree with this, the real art seems to be in marketing, which is why per-capita consumption of energy and materials is what gets valued as a measure of sales performance. Maybe if some of that art was used to educate people about how to behave in a world of shrinking resources, we would be better positioned to find out if ^thraco is right.

By: saulgoodman

saulgoodman — Thu, 12 Apr 2012 19:17:35 -0800

In the long run we are all dead. But my children and their children are not, and I for one would like to leave them with the possibility of a life that isn't sheer hell, if possible. There can still be a fully functioning market economy that uses those resources, where most people work in the production of intangibles. In short, I think it is clear that human welfare can increase indefinitely without using more resources per unit of time. There are no actual intangibles, though, until we become a species of psychics who in violation of every known law of physics don't expend any energy in the formation and transmission of our creative ideas, because at a basic level, even thinking consumes real energy. There is nothing intangible that we can touch without getting at it by way of the tangible. We can't see without expending energy to see. We can't make art that doesn't require a physical medium of some kind to host it. Computers aren't going to solve those problems, as they, too, will always need to consume some amount of real energy. Even in the best case scenario imaginable, there are both theoretical and practical limits to energy efficiency; so if the amount of economic activity always continues to increase, we will eventually run out of capacity. As it is, we aren't coming and never have come even close to being able to make up in gains in energy efficiency what we lose through increased consumption over time, and yet, because we're all wearing our economist hats here for purposes of this discussion, rather than thinking like scientists, the baseline assumption of much of the discussion in this thread is not only that it's practically possible to make such hypothetical gains in increased energy efficiency in order to support continuous growth (in the absence of any empirical evidence to support the idea), but what's more, it's somehow inevitable!

By: -harlequin-

-harlequin- — Thu, 12 Apr 2012 19:20:16 -0800

Interesting. I think he reaches the right conclusion at the end: Theoretically, human welfare can grow... I think that "theoretically" is the wrong conclusion. Economics should guide us with advice that is based on perfect spheres in frictionless vacuums, it needs to be reality-based, and whether human welfare can grow theoretically (it can) is missing the much more important issue of whether it can grow actually, and if so, how.

By: -harlequin-

-harlequin- — Thu, 12 Apr 2012 19:21:00 -0800

(Economics should not guide us...)

By: Arandia

Arandia — Thu, 12 Apr 2012 19:21:10 -0800

One thing has been bugging me about the plane travel vs. video conferencing example being thrown around in this article, and in the comments here. Namely, it's not generalizable. Consider for comparison, the situation often found with programming code. Code tends to be reduced to choosing between optimizing for speed, or optimizing for memory use (say). Sometimes you will find code that can be improved in both memory and speed with no downside. For the programming types, consider: bubblesort vs bogosort (bogosort consists of randomly shuffling a list, and then checking to see if it is in order -- not terribly efficient). In this case, Bogosort is taking the flight while bublesort is teleconferencing. Unfortunately, just because we have found examples of genuinely inefficient use does not mean that we can always find a way to further optimize things without tradeoff. Just because teleconferencing is better than flying does not mean that there will be something in the future that is completely 'better' than teleconferencing. You can't use the fact that bubblesort is definitively better on every metric than bogosort to imply that some 'magicsort' will exist that is better than bubblesort. Yes, I know there are algorithms that beat bubblesort. That's not the point.

By: -harlequin-

-harlequin- — Thu, 12 Apr 2012 19:37:33 -0800

A further diminishing-return issue with the idea of unlimited added-value of the dessert - most of the perceived added value in the artistry in this example is from the dessert standing out among less crafted desserts. When we have attempted to maintain growth through the craftedness of dessert, within a few years almost all desserts must necessarily become so highly crafted that people may be barely aware of anything much further down the scale. At this point, it becomes apparent that appreciation has an equilibrium that is not related to the objective artistry of the food, and with appreciation in equilibrium, growth in the enhancement to your quality of life is static. Artistry has diminishing returns in many forms.

By: Rat Spatula

Rat Spatula — Thu, 12 Apr 2012 19:41:19 -0800

Obligatory Al Bartlett link.

By: saulgoodman

saulgoodman — Thu, 12 Apr 2012 19:51:21 -0800

In short, I think it is clear that human welfare can increase indefinitely without using more resources per unit of time. But since when is the economic argument that human welfare can increase indefinitely...etc? The argument I always hear is that actual economic growth--steady increases in production and consumption year over year--are and in fact somehow must be sustainable indefinitely. Also, I don't think human welfare should necessarily have to increase year after year indefinitely. There's no reason we couldn't achieve some optimum level of human welfare and just maintain it, assuming our population eventually stabilizes. We could still make and trade widgets all our lives--and our children could too--as long as we weren't relentlessly increasing the rates of production and trade in widgets over time. As far as I know, there are no mainstream economic models/theories that would lead to that kind of stable end state for humanity--despite the fact that I think many of us would agree, it seems like the best outcome one could hope for in the circumstances.

By: Arandia

Arandia — Thu, 12 Apr 2012 19:53:16 -0800

I think that's an excellent point, Harlequin. Relative Value (along with it's compatriot, Feather-Preening) explains a lot of the 'intangible value of art' that has been used to boost the 'infinite economic expansion' argument here. Albeit perhaps not all of its value, but certainly most (and 'most' is all that is needed to effectively squash this argument from practical value, in my opinion at least). You may say that your Monet painting is worth millions -- but would it be worth millions if there had been a million painters as good as Monet, with equal reputations? Those millions a painting costs should be attributed to the social value of owning a Monet, and not to the painting itself (note that this social value is a fixed quantity dependent on the number of people in 'society' against whom to measure oneself, and is not something that can be produced or stored).

By: humanfont

humanfont — Thu, 12 Apr 2012 20:46:40 -0800

Typical physicist know it all drivel. Happy to tell you everything wrong with your field of study, because he's so smart. Knows everything about anything you ask, except for the velocity and position of an electron.

By: Kid Charlemagne

Kid Charlemagne — Thu, 12 Apr 2012 23:22:01 -0800

Yeah. If there's one thing we learned in the past 100 years it's that physicists can't be counted on to create, produce or predict anything, while economists are uncanny in their ability to forecast things.

By: jaduncan

jaduncan — Fri, 13 Apr 2012 04:02:29 -0800

On the gripping hand, eventually even societies that have efficiency as an extremely important goal might be expected to hit limits of growth. Malthus gets an awfully bad press for someone who got the timing rather than the gist wrong.

By: thrako

thrako — Fri, 13 Apr 2012 06:12:29 -0800

There are no actual intangibles, though, until we become a species of psychics who in violation of every known law of physics don't expend any energy in the formation and transmission of our creative ideas, because at a basic level, even thinking consumes real energy. I agree. My point, and the conclusion of the physicist, is that each new idea consumes the same amount of energy as the previous idea. If new ideas increases welfare, then welfare can increase over time without using more resources per unit of time. This is world where the "life of the mind" is the most important thing. Although economists have not focused on this situation, it is consistent with standard economic theory (and physical laws, the author concludes). Economics should guide us with advice that is based on perfect spheres in frictionless vacuums, it needs to be reality-based, and whether human welfare can grow theoretically (it can) is missing the much more important issue of whether it can grow actually, and if so, how. I agree. A further diminishing-return issue with the idea of unlimited added-value of the dessert - most of the perceived added value in the artistry in this example is from the dessert standing out among less crafted desserts. When we have attempted to maintain growth through the craftedness of dessert, within a few years almost all desserts must necessarily become so highly crafted that people may be barely aware of anything much further down the scale. I agree that shifting to a no-energy-growth economy would likely reduce the growth rate of welfare. But empirically, new art and music provides people pleasure, even though we have thousands of years of art and music accumulated. I am not sure how to judge the diminishing returns of these things. But since when is the economic argument that human welfare can increase indefinitely...etc? The argument I always hear is that actual economic growth--steady increases in production and consumption year over year--are and in fact somehow must be sustainable indefinitely. This gets at the confusion over GDP, price levels and welfare. I would argue that the correct definition of real (inflation-adjusted) GDP is very close to a measure of human welfare. Real GDP is the inflation-adjusted value of all goods and services sold in the economy. It misses out on non-market transactions, but that is not the important point here. If the inflation adjustment is done correctly, real GDP (actually, real consumption) increases if and only if the average welfare increases. My point is that real GDP, properly defined, is a measure of welfare. However, GDP is usually just used as shorthand for a vague idea of "more stuff", which is fine over the short term. But the rigorous, economic definition of GDP and consumption is much closer to the general idea of "human welfare". (GDP as currently measured does exclude many important aspects of human welfare. But it is extremely difficult to measure even basic industrial production, so I don't really blame the BEA for it.)

By: saulgoodman

saulgoodman — Fri, 13 Apr 2012 06:31:18 -0800

Well, in that case, thrako (putting aside problems with defining welfare basically in terms of how much stuff is bought and sold), assuming human populations eventually stabilized, why should the goal always be to increase human welfare, rather than achieve an optimal level of human welfare and maintain it? Why does GDP--as our proxy for welfare--always need to grow uniformly? How does demanding GDP growth everywhere at all times as an economic good in itself jive with the idea that GDP is a proxy for welfare? Suppose the US had near universal employment, a stable population, and a uniformly high standard of living. What benefit would there be in pursuing more GDP growth year after year in such a scenario? What if the problem isn't the sheer size of an economy, but the discriminatory ways its markets allocate resources, disproportionately increasing the welfare (as measured in terms of GDP) of only a small subset of its population, who've captured the benefits of the markets using monopolistic practices and regulatory capture? No amount of raw economic growth could improve the actual welfare of the majority of people in such a scenario because no matter how much nominal growth in welfare/GDP there is, it would only benefit a small subset of the population, so again, why should raw, indiscriminate growth year over year be a desirable end in itself, as mainstream economists always seem to assume?

By: thrako

thrako — Fri, 13 Apr 2012 07:35:13 -0800

How does demanding GDP growth everywhere at all times as an economic good in itself jive with the idea that GDP is a proxy for welfare? I am not arguing that welfare ought to grow at a constant rate, at all times. Obviously, welfare ought to be maximized. If it can reach some global optimum immediately, we should do that rather approach the optimum slowly. I don't think I disagree with you. What if the problem isn't the sheer size of an economy, but the discriminatory ways its markets allocate resources, disproportionately increasing the welfare (as measured in terms of GDP) of only a small subset of its population, who've captured the benefits of the markets using monopolistic practices and regulatory capture? Redistribution is a first order problem, and one that many, many economists work on. It is difficult. I was ignoring it because I think it is not the most interesting or important part of the post. But redistribution is as important as growth. I think growth gets more attention because it is simpler to study, and lets people avoid some difficult value judgements. Economists will be happy to tell you how to run an economy that achieves a particular level of income equality, but people don't agree on what a good amount of equality is. In the absence of a consensus, more attention is paid to the simpler growth models.

By: -harlequin-

-harlequin- — Fri, 13 Apr 2012 11:44:49 -0800

assuming human populations eventually stabilized, why should the goal always be to increase human welfare, rather than achieve an optimal level of human welfare and maintain it? You're right - there's no systemic reason, just human nature - most individuals in a society desire something they don't have, and will attempt to get it, and most of western culture suggests they should be free to try. End result: Whatever society has, it's never enough. A social system that needs to reform human nature before it can function, is at huge risk of failing and/or perpetrating abuse of its members. Does it seem likely that we will need a society that needs to reform/control human nature, for humans to have a chance for humans to prosper long-term? Ouch. That's pretty scary.

By: klangklangston

klangklangston — Sat, 14 Apr 2012 00:11:19 -0800

"Well, wrong, because technology can also shrink your power consumption: computation per watt continues to rise, advances in design and operation allow the consumption of less power in motor vehicles and planes for the same number of persons traveling the same distance, and so forth." But computation per watt increasing has not shrunk our power consumption — it has decreased the relative rate of increase in our power consumption to the work possible from that computation, but it has not shrunk consumption. Likewise, there are real physical limits on how much can be recovered from, say, planes instead of cars. So, not really wrong, and still pretty circular — assumptions of future growth based on the idea that growth will continue and solve the problems necessary to continue growth.

By: grobstein

grobstein — Sun, 15 Apr 2012 11:10:16 -0800

Robin Hanson responded to this post on his blog. Haven't read yet but he's an economist who's actually interested in this stuff (most economists' research has nothing to do with these issues).

By: anigbrowl

anigbrowl — Sun, 15 Apr 2012 23:46:25 -0800

But computation per watt increasing has not shrunk our power consumption — it has decreased the relative rate of increase in our power consumption to the work possible from that computation, but it has not shrunk consumption. Not in the aggregate, because there are loads of poor people who would like to consume more. On the other hand there are increasingly many people who derive utility from lowered consumption of power, resources, etc. DTM's author seems to assume that consumption will always exapnd to make use of all available resources.

By: klangklangston

klangklangston — Mon, 16 Apr 2012 09:52:13 -0800

"DTM's author seems to assume that consumption will always exapnd to make use of all available resources." Which seems, again, like a pretty valid assumption. To have it not be true you would have to have everyone — not just poor people (and there are plenty of rich people who want to consume more) — stop wanting to consume more resources. Because if even one person keeps consuming more resources than they had previously, the overall rate of resource consumption is positive.

By: anigbrowl

anigbrowl — Mon, 16 Apr 2012 11:43:34 -0800

That's not so, unless everyone else holds their consumption steady. If A increases resource consumption by one unit, B consumes the same amount, and C and D lower their resource consumption, the overall trend would be negative. Of course, I expect the overall trend to remain positive for some time, given the number of people around the world who live in relative poverty and who would like to enjoy a more affluent lifestyle, or who live in absolute poverty and would simply like some economic stability. However, to maintain that endless resource consumption is inevitable is like saying that it's not possible to reduce carbon footprints or ozone usage. Sustainable energy sources are thermodynamically neutral (although biofuels are carbon-neutral whereas wind and solar are carbon-negative), so the problem can be attacked from the supply as well as the demand side. It's not that I think there isn't a genuine issue here worth addressing, but that the assumptions involved beg the question. Over the long term, I also think he's wrong about space exploration, but that's another story.

By: klangklangston

klangklangston — Mon, 16 Apr 2012 12:19:43 -0800

"That's not so, unless everyone else holds their consumption steady. If A increases resource consumption by one unit, B consumes the same amount, and C and D lower their resource consumption, the overall trend would be negative. " Fair point; however, I can't think of a single example where a developing or developed country cut their energy or resource consumption. And I'm having trouble thinking of a hypothetical where such cuts wouldn't have to correspond to a massive economic failure and depression. I mean, I guess England's resource consumption went down radically in the 1300s, but banking on a plague seems ghoulish. Further, it seems pretty likely that per capita resource/energy consumption is tied to standard of living; reducing the former without the latter seems like it hasn't happened — not just individual energy consumption decrease, as that's offset by increases elsewhere. Of course, I expect the overall trend to remain positive for some time, given the number of people around the world who live in relative poverty and who would like to enjoy a more affluent lifestyle, or who live in absolute poverty and would simply like some economic stability. However, to maintain that endless resource consumption is inevitable is like saying that it's not possible to reduce carbon footprints or ozone usage. Sustainable energy sources are thermodynamically neutral (although biofuels are carbon-neutral whereas wind and solar are carbon-negative), so the problem can be attacked from the supply as well as the demand side." Sustainable energy sources aren't thermodynamically neutral — all systems are inefficient. And sustainable energy sources also aren't a panacea, in that the construction and maintenance of solar, wind, what have you, also requires further inputs to meet growth of energy consumption, and there's nothing about the source that really impacts usage — energy is essentially fungible. So while the problem of the increase in carbon pollution can be attacked from the supply side, the problem of increased energy demand in total can't be. (And true space colonization won't happen until we can figure out a way to beat cosmic rays, something that's well beyond us now.)

By: anigbrowl

anigbrowl — Mon, 16 Apr 2012 14:30:52 -0800

I can't think of a single example where a developing or developed country cut their energy or resource consumption US energy use per capita has actually decreased by a significant amount. Some of this is linked to economic growth/recession, but things like fuel economy in cars do matter: your average family car of today sips gas compared to the average family car of the 1970s. Add in some other countries (list at lower left of graph) and you can see similar trends of plateau and/or gradual decline for various large European economies, Japan etc. The US could obviously cut its energy usage almost by half to match their per-capita consumption, though accounting for geographic differences like distance and climate means it will probably always be a bit higher in the US. there's a definite trend towards using less energy for the same or better result, so I don't think all cuts have to result from demand destruction in the form of reduced economic good. My basic disagreement with the blog author (and Malthusians in general) is the implicit assumption that consumption and population are likely to increase exponentially. It's my belief that what looks like exponential growth is often logistic. Sustainable energy sources aren't thermodynamically neutral — all systems are inefficient. Yeah, but that doesn't matter with things like solar because if you don't capture the energy inefficiently it would be heating up the earth anyway, whereas when you burn fossil fuels you're adding additional heat to the biosphere. Certainly there's an overhead in the construction & maintenance of harvesting plants like solar, wind, hydro and so on, but the marginal thermodynamic cost is negligible compared to that of fossil fuel.

By: klangklangston

klangklangston — Mon, 16 Apr 2012 15:45:20 -0800

"Yeah, but that doesn't matter with things like solar because if you don't capture the energy inefficiently it would be heating up the earth anyway, whereas when you burn fossil fuels you're adding additional heat to the biosphere. Certainly there's an overhead in the construction & maintenance of harvesting plants like solar, wind, hydro and so on, but the marginal thermodynamic cost is negligible compared to that of fossil fuel." Part of the article was about hitting waste heat limits in around 600 years. But I take your point. US energy use per capita has actually decreased by a significant amount." Yes, but... the total US energy usage over the same period has increased. My basic disagreement with the blog author (and Malthusians in general) is the implicit assumption that consumption and population are likely to increase exponentially. It's my belief that what looks like exponential growth is often logistic." But to have a logistic curve, you either have to have growth also on that curve or come up with a way to truly decouple growth from resource consumption, and likely have to have some external limit that controls the energy usage growth.

By: anigbrowl

anigbrowl — Mon, 16 Apr 2012 16:07:21 -0800

Yes, but... the total US energy usage over the same period has increased. Of course, the population has gone up. But while we don't know future populations the way we do past energy use, projections suggest that population growth is leveling off as well. Not that I'll be around to see it, but my hunch is that total energy consumption will be flat by 2200 AD or so. Not that I think this is a slam-dunk or anything, just that the trend is not as inevitably one-way as the physicist seems to think it is.