Energy, Pro­duc­tion and Entropy

Flattr this!

This is the pre­sen­ta­tion I gave at the UN ESCAP meet­ing on the qual­ity of growth today in Bangkok.

The key theme is the need to make energy the basis of the model of pro­duc­tion, as argued by Bob Ayres and col­leagues.

About Steve Keen

I am Professor of Economics and Head of Economics, History and Politics at Kingston University London, and a long time critic of conventional economic thought. As well as attacking mainstream thought in Debunking Economics, I am also developing an alternative dynamic approach to economic modelling. The key issue I am tackling here is the prospect for a debt-deflation on the back of the enormous private debts accumulated globally, and our very low rate of inflation.
Bookmark the permalink.
  • Hi Ed,

    I think I had bet­ter elab­o­rate in a later post on my pur­pose there. It was not to argue that entropy poses a sig­nif­i­cant lim­i­ta­tion to cur­rent production–far from it, as Drew observes in an ear­lier com­ment (thanks for that Drew, I’ll fol­low the links). It was that to explain how pro­duc­tion can func­tion in the first place, one has to work from free energy as the source–not labor, etc.

  • PS read John Joe McFadden’s Quan­tum Evo­lu­tion on the “con­sis­tency of life and ther­mo­dy­nam­ics” point. He argues that life’s func­tion is to increase the entropic process of the uni­verse.

  • Thanks Drew, as noted above, I was not see­ing entropy as a limit to production–far from it. I was instead say­ing that a the­ory of pro­duc­tion has to explain the pos­si­bil­ity of pro­duc­tion itself on the basis of the exploita­tion of free energy.

    Thanks for those links. I’ll fol­low up on them, and of course I’ll make myself famil­iar with the maths before I develop the argu­ment any fur­ther. The core role of entropy now I would see as link­ing pro­duc­tion and pres­sure on the environment–not entropy and any lim­its to pro­duc­tion.

  • Steve Hum­mel

    He argues that life’s func­tion is to increase the entropic process of the uni­verse.”

    That’s prob­a­bly just the other end of the bar mag­net nature of real­ity. The other and just as real end would prob­a­bly say that the func­tion of life is the inten­tion to be bet­ter aware of itself via a merg­ing with its actu­ally not so solid bliz­zard of.….whatever sci­ence keeps posit­ing is ulti­mate real­ity.

    T.S. Elliot’s quote is again apropo.

    We shall not cease from explo­ration. And the end of all our explor­ing will be to arrive where we started and know the place for the first time.

  • Jeremy Dawes

    I rec­om­mend read­ing Stu­art Kauffman’s At Home in the Uni­verse for a very lucid expla­na­tion of how com­plex­ity (e.g. life) can arise with­out break­ing the laws of ther­mo­dy­nam­ics. While entropy can only increase in a closed sys­tem, locally entropy can decrease (i.e. order can be cre­ated) when energy flows through a sys­tem.

    We have been able to sus­tain a mas­sive increase the com­plex­ity of our sys­tems (eco­nomic growth) through tap­ping in to a mas­sive new flow of energy (fos­sil fuels). The phys­i­cal lim­its that need to be mod­elled are there­fore not entropy as such, but the rate of flow of energy through the sys­tem.

    I think this may also change the nature of how waste is mod­elled. To my mind waste should be mod­elled in two parts. 

    Firstly it is an energy drain, in that the more we throw away the greater the rate of energy flow that is required to main­tain our cur­rent level of com­plex­ity.

    Sec­ondly, when waste causes pol­lu­tion then we need to divert energy to com­bat­ing the polu­tion , e.g. CO2 waste causes global warm­ing, which increases the water in the atmos­phere, which increases the amount of water in rain­fall events, which dam­ages infra­struc­ture, which needs rebuild­ing, which reduces the energy/money avail­able for main­tain­ing our com­plex soci­ety.

    A final com­pli­ca­tion which is related to entropy is that resources are not uni­formly con­cen­trated. The issue with cop­per pro­duc­tion, and all other min­er­als, is that the cost of extrac­tion is inversely related to the con­cen­tra­tion of the ore. The most con­cen­trated deposits are mined first at low cost. As the most con­cen­trated deposits are used up we move onto lower con­cen­tra­tion deposits which take more energy and so cost more. Web­Hub­bleTe­le­scope has some very detailed maths behind this which he calls the .

    At some point the con­cen­tra­tion of the ores will become so low that the energy required for extrac­tion will more than the energy we have avail­able and the ore becomes off lim­its (in vio­la­tion of ortho­dox eco­nomic “laws”!).

    It also means that the pos­si­ble rate of extrac­tion is finite, which means that at some point the rate of extrac­tion will peak and then decline. So as well as mod­el­ling the energy flows, it looks like it will be impor­tant to model the flow of pri­mary resources to get a real­is­tic feel for how the econ­omy may behave in the future.

    My read­ing of the evi­dence is that we are reach­ing the lim­its of energy flows and pri­mary resource extrac­tion flows, so it looks like we are really at the begin­ing of the Age of Con­se­quences …

  • Glenn Stehle

    @ Steve Keen

    Pleae don’t let peo­ple like F. Beard, Drew Priest or Ed Beau­gard get you all tripped up with their Pos­i­tivist faith, for, as John Gray points out, it “has all the trap­pings of reli­gion” and “is more emo­tional than ratio­nal.”

    ? F. BEARD:

    I don’t see a prob­lem with entropy for thou­sands of years at least since cur­rent energy tech­nolgy or near cur­rent tech­nolgy (espe­cially tho­rium reac­tors) should suf­fice till then.

    This is pie in the sky stuff. Here’s an alter­nate view­point:

    The dreams!” scoffs Norm Rubin, Energy Probe’s direc­tor of nuclear research and senior pol­icy ana­lyst.

    Tho­rium pitches are really just “appeals for pub­lic fund­ing,” he says: “Tho­rium reac­tors are only one of a sig­nif­i­cant num­ber of long-term dreams to plant soy­beans in Antarc­tica with the help of nuclear sun lamps. There is almost no limit to the dreams you can have with an end­less, too-cheap-to-meter source of clean, benign, what-could-pos­si­bly-go-wrong energy.”–thorium-touted-as-the-answer-to-our-energy-needs


    Accord­ing to the equa­tion for entropy of mix­tures given above, sep­a­rat­ing cop­per should only cost ~$0.04/lb [assum­ing $0.10/kwhr and 1000K smelt­ing temp.]

    Cur­rent spot price is $3.45/lb. This is much higher than what the 2nd thermo law requires.


    The major cost dri­vers are mine cap­i­tal equip­ment, labor, trans­porta­tion, and the energy required to break the chem­i­cal bonds of the ore, not the entropy change of extract­ing the ore.

    This ignores the fact that the man­u­fac­ture of cap­i­tal equip­ment, labor, trans­porta­tion and the energy required to break the chem­i­cal bonds of the ore all entail their own energy inputs. Each per­haps employs dif­fer­ent energy tech­nol­ogy, but each nev­er­the­less requires its own energy input and thus its own entropy change. The total entropy change nec­es­sary to refine cop­per would there­fore be the sum of all these.

    The crit­i­cism that has been lev­elled against the tho­rium Utopi­ans is also appro­pri­ate for Priest’s analy­sis:

    Tho­rium doesn’t elim­i­nate the prob­lems,” he con­tends. “If the nuclear industry’s prob­lem was afford­ing ura­nium, then switch­ing to tho­rium might solve their prob­lem. But that’s not their prob­lem. The fuel cost in today’s reac­tors is a tiny frac­tion of the total cost. That’s not what is giv­ing the Ontario gov­ern­ment sticker shock about the next two reac­tors at Dar­ling­ton. They’re solv­ing a non-prob­lem by sub­sti­tut­ing a cheaper fuel for ura­nium. Unless they solve the big prob­lems, they’ve got a curios­ity there instead of a prac­ti­cal solu­tion to anybody’s prob­lems.”–thorium-touted-as-the-answer-to-our-energy-needs


    What I mean is, what exactly is the dis­as­ter that will befall us in the very near future?

    And if ther­mo­dy­nam­ics was com­pletely true in the way we presently under­stand it, how did life begin at all?

    These state­ments are an indi­ca­tion of just how far this thread has slid into a ven­ti­la­tion of Pos­i­tivist the­ol­ogy.

    For a dis­cus­sion of some of the things that can go wrong to pre­clude the real­iza­tion of the worldly par­adise promised by the Pos­i­tivists, I rec­om­mend the two fol­low­ing lec­tures given at The Sci­ence Net­work:

  • Glenn Stehle

    I would also add that the type of energy tech­nol­ogy a civ­i­liza­tion employs goes a long way in deter­min­ing the type of eco­nomic and polit­i­cal orga­ni­za­tion it uses, as well as the reli­gious and sec­u­lar ide­olo­gies it con­structs in order to lend moral and intel­lec­tual legit­i­macy to that eco­nomic and polit­i­cal orga­ni­za­tion.

    As the his­to­rian Car­rol Quigley points out in The Evo­lu­tion of Civ­i­liza­tions, Clas­si­cal antiq­uity was based on slave power. Thus its basic ideology—-Pythagorean ratio­nal­ism and clas­si­cal philosophy—-was con­structed so as to give moral and intel­lec­tual legit­i­macy to slav­ery.

    But, as he goes on to point out, although West­ern civ­i­liza­tion emerged from the wreck­age of Clas­si­cal antiq­uity, its tech­nol­ogy dif­fered from it in very it in impor­tant ways. Even in its early stages, Quigley states, West­ern civ­i­liza­tion was “based on ani­mal power rather than on slav­ery.” Thus a dif­fer­ent reli­gious sys­tem and basic ide­ol­ogy were nec­es­sary, and clas­si­cal phi­los­o­phy gave way to the scholas­tic syn­the­sis of the medieval Roman Church. This pro­vided sort of a truce, or a half-way point between clas­si­cal phi­los­o­phy (with its real­ist ontol­ogy and syl­lo­gis­tic logic) and philo­soph­i­cal nom­i­nal­ism (with its rad­i­cal indi­vid­u­al­ism and nat­ural real­ity).

    As energy tech­nol­ogy evolved even fur­ther to include wind (e.g. sail­ing ves­sels and wind­mills), water (e.g. water mills and hydro­elec­tric), hydro­car­bons (steam engines, inter­nal com­bus­tion engines, tur­bines) and oth­ers (e.g. nuclear, solar), the scholas­tic syn­the­sis itself became obso­lete and slowly gave way to Mod­ernism. Modernity’s promise was and is to “make man mas­ter and pos­ses­sor of nature” and to free man from the curse of man­ual labor. And indeed, in the mod­ern era human labor power has come to play a pro­gres­sively smaller role in the civilization’s over­all energy pic­ture, thus par­tially ful­fill­ing Modernity’s promise. If some­thing should hap­pen to our plen­ti­ful and inex­pen­sive sup­ply of energy, how­ever, then the entire Mod­ernist enter­prise comes tum­bling down, includ­ing the clas­si­cal and neo­clas­si­cal eco­nomic ide­olo­gies that under­pin it and give it moral and intel­lec­tual legit­i­macy.

  • RobM

    Hi Steve, I am long time admirer of your work.

    You are one of the few econ­o­mists in the world try­ing to model the econ­omy based on phys­i­cal real­ity and sound math­e­mat­ics.

    I would like to bring your atten­tion to Tim­o­thy Gar­rett who is one of the few physi­cists in the world that is apply­ing physics to model the econ­omy.

    The two of you could do amaz­ing things together.

  • Steve Hum­mel

    I’m not a “true believer” in tech­nol­ogy either, but even with peak oil we’ve still got a win­dow, if a nar­row­ing and pre­car­i­ously nav­i­ga­ble win­dow, within which we can still main­tain a mod­ern civ­i­liza­tion. And the real key to not only giv­ing us more time to inno­vate in order to find the new source of rel­a­tively cheap energy is to apply the anti-entropic human tool of Wis­dom as the basis for both our self devel­op­ment AND our sys­tems. That way even if we fail in our quest for a new energy source we can more wisely learn to live with the con­se­quences.

    As I state in my Axioms of MESST (Mon­e­tary, Eco­nomic Sapi­ent Syn­the­sis Theory…whose philo­soph­i­cal basis for the con­crete process of pol­icy for­ma­tion is the fol­low­ing four con­den­sa­tions of human wis­dom: Faith as in Con­fi­dence, Hope, Love and Grace)

    Within a par­a­digm of Faith, Hope, Love and Grace con­sump­tion means the most effi­cient meet­ing of goods and ser­vices with con­sumers, not exces­sive or waste­ful pro­duc­tion or con­sump­tion.”

  • Dear Steve,
    If you get a moment, I would love to hear your response to my com­ment posted on the 21st Nov. Very much appre­ci­ate you valu­able time.
    With respect, Kalman.

  • drew priest

    Steve, thanks for the pos­i­tive response.

    I want to mod­ify my respone to say that resource spot prices also depend on change in debt, asset spec­u­la­tion, and sup­ply and demand, as you have clearly laid out in your many eco­nom­ics lec­tures. At $3.45/lb, this is an ele­vated price and has been encour­ag­ing min­ing invest­ment since the price is well above pro­duc­tion and deliv­ery cost.

    Please allow me to expand fur­ther on my pre­vi­ous com­ment…

    Another point on entropy and energy flows is that some­times energy losses to entropy are impor­tant, and some­times they are not.

    My min­ing exam­ple was to show that the entropy change of extract­ing puri­fied met­als from unre­al­is­ti­cally low grade ore is a neg­li­gi­ble loss that would usu­ally be lumped in the “mis­cel­la­neous” cat­e­gory. There is huge energy losses in that case, but it is not due to entropy changes of the mix­ture.

    Yet, for other cases entropy calcs are NOT neg­li­gi­ble. In the case of a heat engine (car, truck, min­ing equip­ment, loco­mo­tive, power plant, jet engine, etc) these entropy losses are very impor­tant.
    The energy of the fuel is used to run the engine to con­vert that stored chem­i­cal energy (fuel) to mechan­i­cal (turn wheels, hydraulic pump, water crops, run elec­tric gen­er­a­tor, etc). The con­ver­sion process is dom­i­nated by sig­nif­i­cant entropy changes. Put sim­ply, the heat engine’s energy flow is dom­i­nated by entropy changes.

    How does entropy dom­i­nate the energy flow of a heat engine? It goes out the exhaust –heat, cool­ing sys­tem –heat, and is loud –sound. The pri­mary losses (cool­ing and exhaust) carry away ~30–70% of the chem­i­cal energy of fuel. Cool­ing sys­tems make a lot of air slightly warmer. The exhaust makes a mod­er­ate amount of air mod­er­ately hot. This is the nature of increased entropy, these heat sources of radi­a­tor and exhaust air are very poor energy sources. It is very dif­fi­cult to recover energy from these sources. The pres­sures and tem­per­a­tures are just too low. Since the exhaust is mod­er­ately hot, we can get some energy out of it via tur­bocharg­ers, turbo com­pound­ing, and com­bined cycle designs. But this extra energy recov­ered from the exhaust is typ­i­cally a frac­tion of what is deliv­ered from the pri­mary design of the heat engine.

    Speak­ing gener­i­cally, from mem­ory…
    For a gas tur­bine engine, roughly 15–35% of the fuel energy is deliv­ered to use­able work. That is the PEAK ther­mal effi­ciency at full load. When idling, the ther­mal effi­ciency is ZERO. This is true for all heat engines.
    For gasoline/petro engines it is 20–35% peak. Diesels are 25–50% peak. Steam tur­bines are 35–50% peak. Com­bined cycle [gas tur­bine + steam] is 50–70% peak. Human body is ~13–17% peak (tech­ni­cally the human body is not a heat engine and is more like a fuel cell, but I have fun com­par­ing it).

    In sum­mary, some­times entropy is an impor­tant calc, some­times it is not. The decider is math and physics.

  • Ed Beau­gard

    Hi Steve,

    Thanks for your com­ments.

    I’ve said this before, and I don’t mean to be a nag, but for me, it’s a mis­take to locate eco­nom­ics within ANY phys­i­cal sci­ence, ther­mo­dy­nam­ics, sys­tems engi­neer­ing, what have you.
    I see this all the time on INET, where you have peo­ple hap­pily apply­ing sta­tis­ti­cal or math­e­mat­i­cal tech­niques from the phys­i­cal sci­ences as if that’s new eco­nom­ics, which it isn’t. It’s really a con­tin­u­a­tion of the spirit of the neo-clas­si­cal pro­gram that Phil Mirowski talks about in his books.
    If econ­o­mists do that(apply physics to eco­nom­ics, for exam­ple), then do they believe that Keynes was wrong in think­ing that eco­nom­ics is a moral and not a nat­ural science(to steal an line from Robert Skidel­sky)? More likely, they prob­a­bly haven’t thought about that ques­tion at all. But it seems fun­da­men­tal to me.
    For what it’s worth, for me eco­nom­ics has noth­ing really to do with nat­ural sci­ence, and com­plex­ity and chaos the­ory have no explana­tory power within eco­nom­ics. Descrip­tive power yes, as you’ve shown in your bril­lant Min­sky model, but it doesn’t explain any­thing that’s not already in the eco­nomic ideas that lead to the model in the first place. It(chaos the­ory, com­plex­ity, sim­u­la­neous equa­tions, etc.) are just another form of ques­tion-beg­ging.
    That is, some­thing that seems to explain some­thing but really doesn’t.
    How­ever, I don’t have Diff Eq, so maybe I’m mis­un­der­stand­ing things in Min­sky.
    And of course, maybe I’m com­pletely wrong about all this, but no one’s ever really responded, even the peo­ple at INET

    Well, I’ve pretty much made my point, so I’ll be quiet now.
    But I’m a huge fan of yours and if I could afford it, I’d be one of your monthly sub­scribers.

    Ed B.

    P.S. have you ever thought of start­ing a Social-demo­c­ra­tic party?

  • Steve Hum­mel


    Your points are well taken. All eco­nom­ics and finance are cur­rently deriv­a­tive. The only thing that will change them in any deep way is to apply the anti-entropic force of Wis­dom to them. For instance, today we only have pri­vate and pub­lic credit cre­ation, and they are also only applied to indi­vid­u­als via exchange (work), or bor­row­ing (loans) and wel­fare (tax­a­tion and RE-dis­tri­b­u­tion). We require a new form of credit cre­ation, a social credit if you will, that is a gift (grace) and that is Dis­trib­u­tive (cre­ated ex nihilo) not RE-dis­trib­u­tive (taxed, some would say stolen, regard­less it is an unnec­es­sary bur­den). These are not deriv­a­tive, they are different/new ideas/policies.

  • Derek R

    It would be nice to have a main­stream eco­nomic the­ory explain­ing which of a range of actions was the most eth­i­cal one to take. Unfor­tu­nately we don’t even have a main­stream eco­nomic the­ory which can explain what the con­se­quences for a range of actions will be and get them cor­rect. Only when we get as far as being able to cor­rectly pre­dict that Action A will lead to Con­se­quence B whereas Action C will lead to Con­se­quence D, will we be in a posi­tion where we can com­pare the moral con­se­quences of a range of pol­icy options. 

    Unfor­tu­nately we’re a loooong way from that today.

  • Steve Hum­mel


    In order to have an eco­nom­ics which explains/encourages eth­i­cal ends you have to have it based on some­thing other than profit (cap­i­tal­ism) or employ­ment (social­ism). With a sys­tem whose pri­mary pur­pose is profit how can we expect any­thing other than increase from exchange? In mod­ern life it col­ors most of what we do and think, unfor­tu­nately. Now if you based our eco­nomic, finan­cial and mon­e­tary sys­tems on the more uni­ver­sal ideas, val­ues, pur­poses and expe­ri­ences of Con­fi­dence, Hope, Love and Grace which are the con­den­sa­tions of all human wis­dom you could have profit and employ­ment take their proper place in the econ­omy and yet poli­cies which, if they were accu­rately bound back to these things, would actu­ally encour­age eth­i­cal results rather than inhibit them. No?

  • Derek R

    I’m not argu­ing with that, Steve H. I’m just say­ing that even those peo­ple who have designed sys­tems based on profit or employ­ment have done a pretty bad job of it if we go by the evi­dence of the Soviet Union or of the West. And to my mind that’s because the design­ers have been work­ing from flawed eco­nom­ics which can’t accu­rately pre­dict effects from causes in all nec­es­sary cases. 

    If you can design some­thing based on Con­fi­dence, Hope, Love and Grace then that’s great. As long as it can be used to work out con­se­quences from actions accu­rately. If it can do that then I’m all for it. That’s why I like Prof Keen’s work. It has some pre­dic­tive abil­ity on the mon­e­tary side. I still think that he needs to do more on the Land/Natural Resources side but that’s why I was very encour­aged by this post. It shows that he recog­nises that need and is work­ing towards ful­fill­ing it.

  • Hi Drew,

    Yes, that’s some­thing I teach my stu­dents (or used to when I had an hon­ours pro­gram!) as an intro­duc­tion to entropy–I take them through the Carnot heat engine.

    In response to a num­ber of other com­ments on this topic, I’m not sug­gest­ing import­ing physics con­cepts holus-bolus into eco­nom­ics. What I was argu­ing in that post is that the growth of pro­duc­tion over time has to be explained in the con­text of the laws of physics. So they prove to be a very use­ful way to cut through the crap over the­o­ries of value that has wasted so much time in eco­nom­ics. But whether they do or don’t apply as lim­its that affect cur­rent pro­duc­tion is very much an empir­i­cal ques­tion, as Drew out­lines here.

  • Steve Hum­mel

    So they prove to be a very use­ful way to cut through the crap over the­o­ries of value that has wasted so much time in eco­nom­ics.”

    Quite agree. We should be aware of the dan­gers of being BOTH “blinded by sci­ence” AND subjective/religious val­ues. Of course the inte­grated dis­ci­pline of think­ing and act­ing which is human wis­dom, and which is also that most accu­rate approx­i­ma­tion of our species des­ig­na­tion, homo sapi­ens, is pre­cisely what is needed in its most con­densed and yet uni­ver­sal basics. It syn­the­sizes and resolves both per­spec­tives men­tioned above. Sorry, but it bears repeat­ing. Maybe if I embed­ded a dif­fer­ent song or dance or other evoca­tive sphere of human life with each men­tion of it, it might ring more true and seem less sim­plis­tic. Just use your imag­i­na­tions.

  • Ed Beau­gard

    Hi Steve,

    What I was argu­ing in that post is that the growth of pro­duc­tion over time has to be explained in the con­text of the laws of physics.”

    Why are you locat­ing this out­side of human activ­ity? Or are you?

    So they prove to be a very use­ful way to cut through the crap over the­o­ries of value that has wasted so much time in eco­nom­ics.”


    Well, you’re def­i­nitely not a Wittgen­stein­ian.

    Okay, I really will shut up now. But very much look­ing for­ward to what you say about all this.

    Ed B.