Yves here. Many of you are Jamie Galbraith fans, and INET has just published a wide-ranging interview with him. Aside from discussing Galbraith’s latest book, which focuses on how flawed mainstream models ignore real resources and lead to distortions and bad policy. The talk also covers hot topics, including DOGE, Bitcoin, climate change and tariffs.
By Lynn Parramore, Senior Research Analyst at the Institute for New Economic Thinking. Originally published at the Institute for New Economic Thinking website
Economics has a dirty secret.
What if a core principle of mainstream economic theory—still taught in top universities, printed in textbooks, and shaping policy for all of us—is completely wrong? What if the iconic supply and demand chart every econ student knows by heart doesn’t actually capture the reality of how economies work?
That’s the claim of noted economist James K. Galbraith, who argues this is a “calamity.” His big problem with mainstream economics is its obsession with a 19th-century illusion: equilibrium.
Here’s the gist: general equilibrium theory—the foundation of modern economics—rests on the idea that markets naturally balance themselves over time. It assumes all economies are just a collection of independent markets, each one perfectly matching supply and demand. The theory leads to some wild assumptions: poor countries are poor because they’re “out of equilibrium,” and prosperity in developed countries is proof that free markets work—ignoring the fact that these markets sometimes crash, only to be brushed off as “unpredictable shocks.”
Galbraith thinks the real issue is that economists have been avoiding one essential truth: entropy. In physics, entropy is the force that drives the universe, and it’s fundamental to all living systems, including economies. But neoclassical economists have stubbornly stuck with equilibrium, even though entropy and equilibrium are incompatible. One is a universal law of nature; the other is just a convenient abstraction.
In his new book, Entropy Economics: The Living Basis of Value and Production (co-authored with Jing Chen), Galbraith makes the case for an economic model that embraces entropy, aligning economic theory with life processes and physical laws—something that has real implications for how we understand markets, power, and regulation. He spoke with the Institute for New Economic Thinking about the implications of entropy economics and how it helps us better understand real-world events—from the rise of crypto to climate change—by revealing them for what they truly are.
Lynn Parramore: How does entropy economics help us understand the distribution of power and control in the economy?
James K. Galbraith: The first point has to do with resources. Resources are front and center in our thinking — something which is really downplayed in mainstream economics. They tend to treat resources as being readily substitutable, one for another, and they’re not.
In order to do anything, you have to be able to extract resources and to do so with a certain degree of surplus, and that explains why it’s so difficult to move to the so-called energy transition when you’re looking at trying to replace a very efficient resource, whatever its difficulties may be, with something that’s far less efficient when the surplus is much smaller. The capacity to pull that off is really deeply questionable.
The second point has to do with the nature of the relationship between markets and government, and we make the point that all living systems and all mechanical systems are effectively regulated. That’s true of your body temperature, your blood pressure, it’s true of the pressure and the temperature in your engine, it’s true of the cooling systems on your reactors, and it’s true of the habits, regulations, laws, and constitutions that govern economic societies. These are all of the same general form, and without them, you get a meltdown or an explosion or a crash.
These are metaphors for what happens when you don’t have proper regulation in any kind of even remotely complex market system. Markets obviously exist, but they exist under the cover of governments, and as a result of that, if you try to operate a market without a government, it tends to run out of control, and it doesn’t last for very long. People will not operate in any kind of complicated market environment unless they feel that there is effective regulation.
LP: Trump has called himself the “First Crypto President.” How does what you’re saying tie into crypto, where the idea is to have money free from government control and regulation?
JG: We’re living in a very interesting experiment in which there’s a supposition that you’re going to replace state-managed monetary systems with one which is managed by the blockchain, essentially. The probable outcome is that you will see that an asset exists that people are prepared to hold as an underlying asset and mostly in the interstices of international transactions and things that people don’t want to expose to the air or don’t want to expose to regulatory authorities.
I’m not expecting Bitcoin to go away, but I think it’s implausible that it will turn into a stable exchange medium, like the very vast, very liquid, and highly reliable U.S. dollar – or, for that matter, the U.S. Treasury, a very stable core financial asset. We shall see, but I think it’s very implausible.
The price of Bitcoin goes up and down. It’s unstable and the scale is not there compared to the Treasury assets. I can see why this has a niche in the market, but it’s a different order of magnitude from the transactions on which the global economy actually relies. The interesting question there is, to what extent do the dollar and the Treasury bond and bill, the financial assets issued by a single national authority, continue to be the dominant medium for financial operations around the world? The United States has been in the business for the last few years, and not just under Trump but before, of calling into question its stability and reliability, and that is going to have, at least at the margins, the effect of reducing the hold that the U.S. has on the global financial system.
Obviously, if you’re going to say we’re going to exclude the Russians or the Iranians, and then maybe the Chinese, not to mention Venezuela and other places, then they’re obviously going to work on alternatives and workarounds. For the moment, that is not an existential threat to the prevalence of dollar-based transactions, and you can see this in the fact that the dollar’s value on exchange markets has been very stable, in fact, it’s going up relative to the euro, which is not a stable alternative. But nothing is guaranteed to last forever. There was a time not long ago when 70% of the world’s transactions were in the sterling zone. Of course, that’s not the case anymore. Now, the U.S. is not going to decline to the extent that Great Britain did, but, you know, you chip away at the credibility of your presence, the reliability of your regulatory structure, and ultimately people look for alternatives.
LP: Generally speaking, what do we miss when we adhere to general equilibrium theory, and what insight do we gain when applying the entropy economics model?
JG: In the first place, the general equilibrium theory is rooted in a view of the world that predates late 19th-century physics and biology. It predates Darwin. It predates the second law of thermodynamics. It’s a view of the world that there is some stable state to which we are tending and that we can figure out what it is.
So there may be short-term uncertainty, there may be shocks, but there’s no long-term uncertainty. Everything is going towards some stable state. And that’s a profoundly anti-scientific view. It is inconsistent with everything we know about physics, about biology, and about social societies, about human history, about everything.
The approach that we offer is not inconsistent with everything we know about everything else. And therefore, we’re sitting there in the intellectual, the scientific, the human tradition that the rest of the world has adopted well over a century ago and considers to be really the basis of human knowledge. Yes, you can anticipate things a short time into the future, but uncertainties increase as you go further out in time. They don’t diminish. So you’re always working, in making decisions, with a number of sort of basic questions that have to be resolved. For example, what’s the possibility of making a profit on a given endeavor? What are the resource costs? How much fixed investment do you have to make in advance in order to get the resources together and to apply the technologies? How much uncertainty do you face? How confident are you that things will work out? This is the kind of decision every business makes, every government makes, every household makes, in thinking about how they go from the present to the future. So we’re in line with all of that. We’re taking that as the basis for thinking about economic decisions, not something that’s just a gloss or something that can be treated as a set of, let’s say, random shocks or even chaotic motions. We’re putting in the foreground the fundamental decisions that any economic actor actually is faced with.
It’s not a complicated idea, but it’s fundamentally opposed to the notion that the world tends to a balance between the great forces of supply and demand, or however you want to characterize the textbook vision of things.
LP: How do those mistaken assumptions impact our view of tackling climate change?
JG: It tells you that if you’re going to actually do something about the challenge, you have to make decisions on a large scale, and frankly, they may not work out. I think our position on climate change tends to the pessimistic side because there are operating principles here. One of them is the Jevons Paradox, which has been around for now 160 years, that says that when you start moving to a new form of energy, it doesn’t reduce the amount that you continually use from the older sources. So when oil took over from coal for transportation, airplanes took over from the railroads, and electrical systems took over from the railroads, coal use did not diminish, it simply got concentrated in the creation of electric power. And we see that phenomenon again and again. We put together the windmills and the solar panels in order to produce renewable energy, and what crops up is that data centers and AI processors, absorb that energy, but they don’t reduce the overall use of fossil fuels for everything else.
The Germans have gone forward trying to replace not only their fossil fuels and Russian gas, but also their nuclear power with wind. They are discovering that the cost of doing so—and particularly the problem that arises when the wind dies—creates huge problems for the viability of their industry. This is something that will end up reducing living standards in Germany for a long period into the future; it’s a fundamental economic miscalculation.
So, in some sense, I think we are not being helpful by being in the cheerleading vanguard for those who think we can solve the climate change problem by making these kinds of investments. The history of experience shows that unless you have a truly dramatic shift to something much more efficient at extracting energy from resources—and nuclear power, of course, was the great example of that—you’re going to run into basic problems with the viability of the investments you’re proposing to make. We are now seeing this.
A major effect of the interest rate moves by the Federal Reserve over the last three years was to make a lot of offshore wind projects, which were extremely marginal economically when they were first proposed at a very, very low, say, 2% interest rate, no longer viable. With interest rates three times that they’re being canceled. Our view is, that you have to face the world as it actually is and not create mental bubbles.
LP: What are some of the benefits of applying entropy economics to policy over general equilibrium theory?
JG: The fact is, if you go down the road of basing policy on a fallacious body of theory, you’re going to get it wrong practically every time. Take the campaign of sanctions against the Russian Federation, which I’ve written on in another context. This is entirely based on the idea that the Russian economy consists of a set of essentially neoclassical building blocks—a capital market, a labor market—and that if you pull back and cut off those things, the Russian economy will collapse. Now, we applied a very comprehensive program of sanctions starting in 2014, which greatly accelerated in 2022.
Especially after 2022, we see the result, which was that the Russian economy didn’t collapse. In fact, it grew rather robustly. And why was that? Because we didn’t realize that by imposing these sanctions, we were creating conditions for Russian industry, Russian companies, to fill the gap in ways that were much more profitable than the conditions they faced before. We removed foreign competition. We gave them an environment where their internal resource costs were quite low. They weren’t getting the export earnings, but that’s secondary. Their internal costs were low.
Also, we forced them to develop their own internal payment system. We basically ended capital flight from Russia, so if you were a wealthy Russian oligarch, you either left altogether or you reinvested in your home country. And all of these things produced a robust and relatively balanced growth in the last few years in the Russian economy, according to really very basic business principles, but completely contrary to the neoclassical theory, which was explicitly underpinning what the Biden administration was doing and the previous administrations were thinking they were doing in trying to use this economic policy tool for a foreign policy purpose.
LP: How might this apply to what Trump has done vis-à-vis tariffs on China? Is it a continuation of neoclassical thinking?
JG: Yes, I do think so. I don’t know what they expect will happen, but what has happened already is that the Chinese have retaliated with their own tariffs, and the Chinese position is very strong. They will be able to sell their output in other markets. In fact, they’re already doing so. The U.S. share of Chinese exports has declined in recent years, and they will be able to source their raw materials, coal presumably, in other markets. So while the U.S. position with respect to Canada and Mexico is quite strong, the U.S. position with respect to China is not, and as a result, maybe they will come to some mutual agreement and back off from all of this. The idea that the U.S. has some enormous advantages over China has been tested again and again.
Another example along the same lines was the policy of denying China access to high-end NVIDIA graphics processing chips that underpin artificial intelligence. It had two consequences. One was to accelerate China’s development of its own advanced chips, which are now being incorporated into Huawei’s phones. The other was to accelerate and foster the reengineering of artificial intelligence protocols, which is why DeepSeek emerged. The idea is that there’s a combination of a neoclassical worldview—fundamentally the idea that there are, underlying production, some specific sources of technology and capital—and this is paired with a hubristic belief that we control these sources, while others don’t have access to the same resources, and that if you give others the opportunity and make it a profitable endeavor for them to exploit that opportunity, then somehow they’re not capable of doing it. And that’s effectively what’s not being considered here: What is essentially the market environment? What is the economic environment that’s being created for the other party by these policies?
That’s not a complicated thought, but somehow it doesn’t get assimilated by policymakers. And I suspect the reason is that they are absorbed with a particular worldview, which is consistent with the training and education they got when they studied economics or social studies or politics, philosophy, and economics at Oxford. They’ve got a rather basic textbook, a very superficial view of things, and they don’t really think outside of that box.
On tariffs, the first thing I would ask is, how do these measures change the decision environment facing a Chinese entrepreneur? And the first thing you have to recognize is that such things exist. This is not a country where everything’s decided by a state bureaucrat. It’s not done mechanically. It’s done by companies looking for opportunities to expand their markets, increase profitability, and net worth. So, this is an ordinary business decision. And now we’re creating an environment—particularly if we think the sanctions are going to be serious and enduring—in which, let’s say, they’re cut off from an easy source of, say, a particular technology. Again, the GPU chips are a nice example. We could probably come up with any number of them.
And of course, they say, well, we can simply close everything in this space and not do these things. Or maybe we can put some resources into fixed investments of our own that are autonomous and may be able to equal or exceed the performance of our competitor in a certain period of time. And that’s, in fact, what they did and are in the process of doing. And it’s not new. It’s what the United States did to Britain in the 19th century. It’s what Germany did in the 19th century. It’s what Japan did in the late 19th century. This is a process ongoing in the world. And again, nothing we say is out of line with very well-known history.
You have to ask: why is the policy implemented if the consequences are so clear-cut, as I think they are, when you think about it in the simple terms we offer? And the answer is they’re not thinking in those terms. They’re thinking in terms of some very mechanical notion of the production process and a strange idea that the flow of new technologies comes only from the sources that are presently—or at a given moment—providing the leading edge. Let’s say Silicon Valley and the Dutch lithography producers. That’s true at a given moment, but there’s nothing in history that suggests those things are enduring.
LP: Do you see any opportunities for entropy economics to be applied in the incoming administration?
JG: I’ve not been able to figure out what’s going on in the minds of the various people who are coming into power. I will say that a recognition that a certain part of your past investments are not profitable and should be curtailed and shut down is a good thing. I’ve long argued that the United States set out 30 years ago to dominate the world financially and militarily—particularly in the Clinton and George W. Bush period. This was a period of enormous expansion in the global ambitions of American elites, and they thought they would never face a peer competitor again, and in particular, that it wouldn’t be Russia. They didn’t anticipate and didn’t understand what was happening in the 1990s and 2000s in China. Now, we’re looking at a situation where we’re very heavily overextended. The fixed investments we made many years ago—bases, aircraft carriers, and the whole technology of American military power—have been shown to be not profitable anymore. It was once quite dominant, but technology moves on. New power centers emerge. You can get too close to someone else’s territory, and your advantage near that territory is not great, if it exists at all. That’s what we discovered in Ukraine. Trying to fight a proxy war against the Russians on the Russian border has never been a successful enterprise.
Understanding that requires making some very difficult decisions and going against large structures that have been built up but are dragging you down, and that’s a crucial, necessary step. In some sense, the Russians went through this 30 years ago. One of the things they did, and I had this explained to me once on a visit to Moscow, was they cut the military. They cut it by about 75%. The Soviet military, the Red Army, was based in a massive defensive posture in Eastern Europe and the western part of the Soviet Union. They essentially got rid of it, and then when they rebuilt their military, they rebuilt it in a different way, meeting the needs of the Russian Federation as it recovered economically.
We didn’t go through that. We simply built on what we had at the end of the Cold War, and now we’re spread out all over the world, spending a huge fraction of our national wealth on doing something that does not, in fact, assist the United States in maintaining its economic or political position. That seems to be a really basic argument. It’s not in the book, but it’s easy to see how the principles of the book apply to that.
LP: That seems to suggest that DOGE, for example, might focus on military spending if they really want to cut waste.
JG: Yeah, for sure. Military spending and the interest bill the Federal Reserve has confected for itself—paying interest on reserves to banks and paying out interest to bondholders. Why are we doing that? You want to save money? Let’s cut the interest rate. That would be another way of doing it. But yes, in terms of fixed investments—things that absorb real resources but bring a negative return—it seems to me the military budget is exactly where one should look, and so far as I can tell, they aren’t looking. There’s a cult of American power and an enormous political base for all of this spending – it’s the easiest thing to approve. And we’ll pay in terms of policy failures as we go forward into the future if they don’t come to grips with that. I understand the difficulty, but somebody has to start talking about it.
LP: Say a Harvard econ grad taught general equilibrium theory joins an AI company. How might their economic assumptions shape their approach in tech?
JG: My first reaction would be that anybody who goes on from a mainstream economics program into a real-world business environment is going to be rapidly disabused of what they were taught because the world simply doesn’t correspond to that line of instruction. They will be operating in a world that does in fact, reflect the principles of the arguments in fairly precise ways that we make in this book.
To give you a concrete example, someone goes on to work for OpenAI, let’s say six months ago, and buys in at stock value at the time, and then discovers that the Chinese come in with something that is engineered much more efficiently and which therefore deprives the incumbents of market capitalization. There you’re talking about something that can be modeled quite precisely with the tools that we place in this book, but which is not anticipated by mainstream economics.
It’s not to say that there haven’t been important economists who were fully aware of this. There’s nothing really new in the world, and this is a Schumpeterian phenomenon, if you like, but these are phenomena that are deeply downplayed by mainstream economics, according to which technology kind of floats down like the gentle rain from heaven across all of the businesses and is absorbed by them in equal measure. The world simply doesn’t work that way.
LP: Is the DeepSeek surprise a bit like what entropy economics might look like to a mainstream economics professor?
JG: Well, you raise a very nice parallel. I would be happy to argue that the approach we’re taking is the sort of DeepSeek approach to the mainstream economic theory, in that it’s simple, it’s efficient, it accounts for a great many phenomena that occur in the real world in a very clear-cut way, and so it leaves a whole mass of textbook argumentation as essentially the realization of a set of ideas ought to fall rather dramatically. The real question is whether we’ll get a fair reading. We’ll see whether DeepSeek is being banned in order to preserve the market space of much less efficient operations. Intellectually, something similar is likely to happen here. It’s up to ordinary people to see whether they wish to cut through that particular problem.
49 comments
Comments are closed.
What if a core principle of mainstream economic theory—still taught in top universities…is completely wrong?…There’s nothing really new in the world, and this is a Schumpeterian phenomenon, if you like, but these are phenomena that are deeply downplayed by mainstream economics…I would be happy to argue that the approach we’re taking is the sort of DeepSeek approach to the mainstream economic theory,
What if the core narrative of history is wrong? I’ve been playing a game with DeepSeek. I ask it a series of questions that eventually result in a “server busy, try again later.” This happens when I ask DS to reconcile the contradictions that it has provided. It’s fun and reminds of an episode of Star Trek were the fearless Captain Kirk gets a computer to have a “mental breakdown.”
Can you share the best of these interactions?
This morning’s play (interaction) with DeepSeek had to do with geoengineering and Operation Popeye. It was very amusing.
I never heard of Operation Popeye, but Chinese weather records are extensive and available and the Chinese have used cloud seeding at times. Why not look for information about cloud seeding in China to begin with?
I don’t live in China, I’m a U.S. citizen and so my primary interest is geoengineering that may explain what I’m seeing in my sky when I look up.
Operation Popeye
DeepSeek is a Chinese AI system. The weather system that DeepSeek is especially familiar with is the superb Pangu.
I was merely trying to be helpful, which was evidently foolish. Forgive me.
No, very helpful, thanks…never heard of Pangu.
An artificial intelligence-powered weather prediction system called Pangu is garnering a lot of attention globally as it aims to revolutionize weather forecasting. Pangu, created by Huawei in China, creates high-resolution global forecasts for locations roughly 27 kilometers apart generated in under 10 seconds.
https://forecastwatch.com/2024/05/07/chinas-ai-model-pangu-makes-waves/
https://www.chinadaily.com.cn/a/202404/16/WS661dd61ca31082fc043c22ca.html
April 16, 2024
AI-powered forecasting proves a hit
Accuracy of Huawei’s Pangu weather prediction system wins global fans
By Yan Dongjie
Huawei’s artificial intelligence-powered Pangu weather prediction system is making waves with its potential to revolutionize weather forecasting, with high-resolution global forecasts for locations roughly 27 kilometers apart generated in under 10 seconds. *
Tian Qi, the leader of Pangu’s research and development team, said it uses neural network models for weather forecasting systems and achieves higher prediction accuracy than the world’s first similar AI weather forecasting model, Four-CastNet, which was released by Nvidia in 2022…
* https://www.nature.com/articles/s41586-023-06185-3
What core narrative of history? That USA invented freedom and democracy and equality, and won WWII too?
P.S. The other day I saw a video of some US politician/activist reading a speach in front of some formal assembly, and saying World War Eleven (that US won).
Core Narrative? Entropy? Memento mori.
I’ll try not to make a “speach” here but I get sick of hearing how the US won WWII — the Russians won World War Two, not the US. Read up on your history.
Hitler actually lost the war when he decided to invade the Soviet Union – he said it would take 3 weeks. The German army was decimated in that invasion and never recovered.
The US did not enter the European theatre until after Pearl Harbour in 1941
There was an episode of The Prisoner where Number Six ( played by Patrick McGoohan) gets a computer to melt itself down completely by asking The Unanswerable Question. I can’t remember which episode it is.
Here is a bunch of episodes.
https://www.youtube.com/playlist?list=PLHkRm9NxLxR1jCQbtpIcYWi3FHmY_lXtZ
> then discovers that the Chinese come in with something that is engineered much more efficiently and which therefore deprives the incumbents of market capitalization.
The late Don Lancaster worked amortization into his discussions of energy dynamics and solar energy. Gail Tverberg applies a similar perspective to energy source development. And Qiao Liang noted:
>> Americans are developing a huge, rapid global combat system: using ballistic missiles and supersonic aircraft five times or ten times faster than a supersonic cruise missile, it can quickly hit any area of high concentration of capital. Now the US claims it can hit any part of the world within 28 minutes, so no matter where capital is concentrated, it can hit anywhere in the world. As long as the United States does not want a particular place to have capital, a missile can get there in 28 minutes. And when the missile goes down, capital can be still quietly and nicely withdrawn.
As Michael Hudson said of American salesmanship:
>> We can offer you not-bombing-you like we did in these countries. That’s all we can offer.
So a missile strike can be a capital strike. It’s not just that things are engineered more efficiently. See ‘Fearless commentary on finance, economics, politics and power’. Those are non-independent systems.
Please set down specific references when possible. Where, for instance, is the Qiao Liang reference to be found? What does the Michael Hudon quote mean? A further explanation of the comment would be helpful.
CA, I can only speak to the Michael Hudson quote, which is essentially saying that the U.S. will bomb the entity being negotiated with if that entity doesn’t do what the U.S. wants. Even shorter explanation is do what we say or we’ll bomb you.
I hope this helps.
Don Lancaster: archive.org/details/donlancaster_energfun
Gail Tverberg: ourfiniteworld.com/
Qiao Liang: limesonline.com/en/regions/one-belt-one-road-14720766/
Michael Hudson: michael-hudson.com/2023/04/all-or-nothing-foreign-policy/
(Too many links will sometimes cause moderation problems for me.)
Judy and Steve,
Thank you both, lots.
As someone with training in Physics and Engineering, I am uncomfortable with the invocation of “entropy” in economics. I suppose I will have to read the book to see how entropy is defined in an economic sense (entropy has a very precise definition in physics and in information theory). I could be wrong but I see this use of a physical concept as another version of “physics envy” in economics.
I do agree with Galbraith that equilibrium models are nonsense. I not long ago finished reading Benoit Mandelbrot’s classic work on markets (The Misbehavior of Markets) and came away with the view that markets/economies can be modeled, within limits, using random walk models (or in Mandelbrot’s case, using multi-fractal simulations). In his fractal formalism, it is possible to define a form of entropy, related to fractal dimension. The problem is, unlike in physics, how do you define the model? There are no conservation laws or physical symmetries and invariants that guide model development. In short, Mandelbrot implies that markets are at best “metastable” where you may have long periods of small market movements punctuated by periods of high volatility. (His concept of “fractal trading time” is quite interesting.)
I need to learn more about Galbraith’s work and would like to compare Galbraith’s Entropy Economics to Mandelbrot’s “Ten heresies of finance.”
Thank you. I was about to post the same thing. People love using the term “entropy” without knowing what it means. I have found that when economists try to apply the second law of thermodynamics they are often wrong, but whenever they apply the first law they are incredibly correct. We need to be including energy and mass balances in economic models as they provide physical limits to growth. Imo. It’s just like how when you design a chemical plant you include the economics in your calculations but reversed.
>> But neoclassical economists have stubbornly stuck with equilibrium, even though entropy and equilibrium are incompatible.
Within context, it’s a way to move the profession forward from equilibrium models. If you take entropy as simply the inverse of efficiency, and efficiency defined within economics, it does inform statements like this:
>> … it’s so difficult to move to the so-called energy transition when you’re looking at trying to replace a very efficient resource, whatever its difficulties may be, with something that’s far less efficient when the surplus is much smaller.
>> But neoclassical economists have stubbornly stuck with equilibrium, even though entropy and equilibrium are incompatible.
In the physical world, entropy and equilibrium are perfectly compatible. In fact, classical thermodynamics relies on “local equilibrium” in order to allow the various system variables to be defined (e.g. pressure, temperature, volume, concentration, etc.). An equilibrium situation is usually the one which maximizes entropy on a local level. There can also be more than one “equilibrium”. (In reality, equilibrium is a bit of a misnomer. I would say economic systems could mirror the physical in the sense that a metastable state is in reality one that orbits one of many “basins of attraction” like those defined for nonlinear dynamics – chaos theory). The trouble is defining the pertinent variables and the relationships between those variables (where the physics envy comes in. :-)) which define the system dynamics.
>> If you take entropy as simply the inverse of efficiency.
What does this even mean? How is this formulated? I am not an economist, so please excuse my ignorance. In information theory, a “high entropy” signal implies efficient transfer of information (removing redundancy, as in the case of data compression). So saying that high entropy implies low efficiency is a bit of a non-sequitur.
Thank you, BillS, I’ll untangle some of it for you. The first thing to know is that ‘entropy’ in information science is an obfuscatory phrase:
Shannon > My greatest concern was what to call it. I thought of calling it ‘information,’ but the word was overly used, so I decided to call it ‘uncertainty.’ When I discussed it with John von Neumann, he had a better idea. Von Neumann told me, ‘You should call it entropy, for two reasons. In the first place your uncertainty function has been used in statistical mechanics under that name, so it already has a name. In the second place, and more important, no one knows what entropy really is, so in a debate you will always have the advantage.’
The second thing is I don’t speak for Galbraith, and haven’t read the book yet. I’m framing it for you more overtly than many economists would be able to, and so expect some sloppiness in their definitions, as they are trying to re-orient their discipline to reality.
Mandelbrot’s student Taleb, who defined the economics use of the phrase Black Swan, would likely agree with your assessment of non/equilibrium and laugh about physics envy. I’ll stretch a bit, and suggest a Black Swan is a system suddenly snapping into a probabilistically unlikely state, which is why you can’t predict them.
When I say entropy is the inverse of efficiency, the thermodynamic phrasing is more precisely the Coefficient of Performance. But, again, in economics efficiency refers to marginal benefit to cost ratios and you’re using different units, so be cautious with assumptions.
Finally, a high entropy information signal is one refers to the uniqueness of the signal and message, ranked so that a common message uses a lower energy signal. A time-sensitive message can also be shorter – my favorite example is the Klingon word for ‘fire missile’ is pronounced ‘bahk’, phonetically very short and sharp. But the basic equation only relates to message probability.
[esalq.usp.br/lepse/imgs/conteudo_thumb/Energy-and-Information.pdf]
ps: There are better specifics concerning the economic history in the comments down-thread.
BillS, I think there is an ambiguity in usage with information systems/theory. I’m referring to the Shannon Entropy Equation, but most of its questions were answered a century ago. Your usage is within data compression. A distinction in viewpoint is that, in data compression you want to reduce redundancy; while the perspective I saw was how much redundancy to introduce to overcome a noisy channel. In both cases, message integrity is the primary concern.
Thanks for your lengthy exposition, but I was really hoping for a reference to some mathematical description of entropy as it applies to economics. I am familiar with Shannon’s entropy but fail to see how it is relevant to economics. Entropy in economics seems more obfuscating than revealing..a bit like the fiction of the “rational actor”.
I’m a better ecologist than economist. Here’s the best I’ve got:
Tainter > Two concepts important to understanding the nature of complexity are inequality and heterogeneity… Inequality may be thought of as vertical differentiation, ranking, or unequal access to material and social resources. Heterogeneity is a subtler concept. It refers to the number of distinctive parts or components to a society, and at the same time to the ways in which a population is distributed among these parts. A population that is divided equally among the occupations and roles of a society is homogeneously distributed; the converse brings increasing heterogeneity and complexity (see also… A society with a great deal of heterogeneity, then, is one that is complex.
Heterogeneity = diversity = Shannon equation:
S = -Σ p log2(p)
where p refers to each species in the community.
In Human Ecology, p can refer to different roles or occupations.
Another ecological equation is the Verhulst equation:
dp/dt = rp(1-p)
where r is the reproduction rate, and (1-p) is the unexploited resource space. Pandit et al expand this via primatology research to:
(1-p)^i
where i is a rank-index based on power.
Pandit et al > we extend our model of primate coalitions and find that a division into a bottom class and an upper class is inevitable whenever fitness-enhancing resources, such as food or real estate, are exploitable or tradable and the members of the bottom class cannot easily leave the group. The model predicts that the bottom class has a near flat, low payoff and always comprises at least half the society.
That ‘half of society’ in economic (money) terms is adjacent to median wages.
Turchin > Over the past four decades, the median worker wage scaled by GDP per capita has declined to historically low levels. This means that the fruits of economic growth our society has experienced have not made the well-being of the majority of the population better (they primarily went to the upper 1 percent).
Turchin finds that as relative wage decreases, social instability increases, and can lead to collapse.
* * *
Tainter: The Collapse of Complex Societies
Pandit, Pradhan, & van Schaik: Why Class Formation Occurs in Humans but Not among Other Primates
Turchin: [prosocial.world/posts/ten-thousand-years-of-the-third-way-a-conversation-with-peter-turchin]
Economics is the fiction of the “rational actor”, told by unreliable narrator.
The concept of entropy is confusing because it arose in two different contexts. One context is that of classical thermodynamics, in which entropy arises in the definition of Gibbs free energy, as explained in this Wikipedia article. The other context is that of statistical mechanics, as explained in this Wikipedia article. The latter article also explains the connection between the two ways of looking at entropy.
These articles are both somewhat technical. The informal way to understand entropy is that it is a measure of the order in a closed system – low entropy corresponding to a high level of order (a low level of disorder) and high entropy corresponding to a low level of order (a high level of disorder). The second law of thermodynamics can be looked at as stating that in a closed system the entropy tends towards a maximum over time. Thus the final state of the system is one of maximum entropy – the “heat death” of the universe, which is indeed an equilibrium, of a sort.
The first propositions on resource depletion being entropic came from Nicholas Georgescu-Roegen, who did cross over between real science and the dismal science. So there has already been discussion of this with regard to economics, especially within the social ecology and degrowth movements.
We are left with another simple but related myth in conventional economic growth models, that we live on a finite planet so perpetual growth is simply impossible as an underlying principle, just as ‘homo economicus’ and the ‘invisible hand’ that inevitably produces market equilibirum are basically stupid ‘laissez fairy’ theoretical concepts.
As the decoupling of resource utilisation and consumption is also entirely ignored in economic theory, it’ll be interesting to see how the book deals with the degrowth movement, and the work of Nicholas Georgescu-Roegen and the ecological economics work of Herman Daly.
Of course we have already seen total energy consumption reduced, if not fully decoupled, in economic growth in some contexts, so the Jevons paradox is observable but not necesssarily immutable.
Two generations before Nicholas Georgescu-Roegen was Frederick Soddy, the scientist who discovered the potential power of radium (not an economist), who pointed out that economics had to account for entropy if it were to be at all useful. He also pointed out that interest is an arbitrary human creation, and national currencies should be free-floating instead of pegged. The Rôle of Money is a great book, way ahead of its time.
Apparently, Nicholas Georgescu-Roegen wasn’t aware of Soddy’s book, but his own book The Entropy Law and the Economic Process should answer any questions BillS has about the subject.
To me, economics is (or should be) the human expression of the law of entropy.
Wow! NC truly has the greatest commentariat. Thanks for this TiPi and Birch. I’ll have to dig up Roegen’s and Soddy’s work.
Ugh at Soddy … no background in history or the foundations of money/accountancy from a basic Natural History perspective. Shades of some Austrian econ sort divining the truth via deductive processes and call it natural or empiric.
The book 5000 Years of Debt would have compel him to take another look before proselytizing so forcefully.
Then some ponder the advent of the “Technocrat” aberration now being forced on everyone in America. Social constructs of whole cloth being handed down to the unwashed by squillionairs padding their pockets as the earned it and all should rejoice as the creator said so … it echos from antiquity yet some think its new …
Charles Walters Jr. of Acres USA respected Frederick Soddy very much. I tried reading Frederick Soddy’s “other book” ( Wealth, Virtual Wealth and Debt, the Solution of the Economic Paradox). I would have to read it again thirty-some years after I read it the first time . . . slowly and carefully so I could feel confident I understood it well enough to discuss it deeply.
What I remember is this . . . that Frederick Soddy was a Twice winner of the Nobel Prize in Chemistry. During the Great Depression he decided that the dominant schools of economics of the day had all failed to be able to provide organized mankind with any way to understand its own economic activity. He wondered what kind of theory of economics would be based on an understanding of the flow of energy through human civilization and society and how “money” could be constructed on this understanding. Did he get it right? I don’t know enough to even know. But Charles Walters of Acres USA thought Soddy’s thinking was very useful.
Here is a little Wikipedia stub “about” the book, presented as part of Wikipedia’s “Ecological Economics” series.
https://en.wikipedia.org/wiki/Wealth,_Virtual_Wealth_and_Debt
Here is a link to the entire book on pdf . . . page by page by page.
https://archive.org/details/soddy-f.-wealth-virtual-wealth-and-debt-1925/page/n1/mode/2up
( I bought my analog dead-tree copy many years ago at the Acres USA Bookstore when it still existed there. It is somewhere deep within my piles of books.)
Taleb : Read Nicolas Georgesu-Roetgen
The Germans have gone forward trying to replace not only their fossil fuels and Russian gas, but also their nuclear power with wind. They are discovering that the cost of doing so—and particularly the problem that arises when the wind dies—creates huge problems for the viability of their industry. This is something that will end up reducing living standards in Germany for a long period into the future; it’s a fundamental economic miscalculation…
— James Galbraith
This is a fine interview.
“The Germans have gone forward trying to replace not only their fossil fuels and Russian gas, but also their nuclear power with wind…”
The Chinese however have a greening policy that dictates having a fuel or energy source surplus, and developing and drawing from multiple energy sources, while greening. The Germans could have looked to and learned from the Chinese.
I have always taken the position that Marx used the classical labor theory of value to destroy the Classical School construct. It was left to the apologists Walras and Marshall et. al. to clean up the mess and construct a new apologia in the form of the Neo-Classical School underpinned by higher and higher mathematics. This evolution served a number of purposes: it dumped the Marxian idea of surplus value into a convenient ash can and would no longer taint the idea of “profit” belonging to the capitalist; by appropriating the tools of the physicists ‘Economists’ could now both conceal and justify social contradictions caused by market fluctuations; politics (Political Economy) no longer entered into this new, advanced, efficient, scientific paradigm and equilibrium now ruled the social/economic sphere.
Economists now became both the apologists of the oligarchs and adjudicators of the distribution of benefits from the system. These high priests insisted that their pseudo-science was “value free”, and they were simply diagnosing the economic system. It was for other to prescribe. On our way across the quad from a macro or micro class we used snicker how ‘value-free’ and ‘equilibrium’ were synonymous with obscurantist.
Entropy? As I used to say to my wife, “Keep me informed.”
John Kenneth Galbraith
Nicholas Georgescu-Roegen (The Entropy Law and the Economic Process, Harvard University Press, 1971) is to my knowledge the first high-grade economist to introduce the entropy concept in the realm of economics.
In a nutshell, he posits that the economic process “is not circular, but unidirectional (…) the economic process consists of a continuous transformation of low entropy into high entropy, that is, into irrevocable waste or, with a topical term, into pollution.” Similar to what happens with energy, all possible economic transformations are always in the sense of a “qualitative degradation”.
Challenging the relevance of the Cobb-Douglas production function, given its underlying assumption of unbounded resources productivity, he argues that thermodynamic laws set unsurpassable limits to substitutability and technical progress. Concepts such as “circular economies” are oxymora. A more efficient utilization of capital or labor can only reduce the losses of matter and energy, in other words the resulting waste of the production process. A better design may reduce the amount of matter and energy required, but there is always a limit – capital and labor cannot create the matter and energy on which they work.
The extrapolation of physics, in this case the second law of thermodynamics, to economics was a cause of concern for a number of Roegen’s colleagues (e.g. Solow, Stiglitz). It has been suggested that Roegen, notwithstanding his other merits, was never granted the Nobel prize, because of his failure to fully grasp the specifics of the second law of thermodynamics. Indeed the latter applies to energy, and energy only, whereas Roegen extended its scope to encompass both energy and matter. This scientific heresy imperiled the theoretical foundations of Roegen’s construct.
For those interested in learning more about Roegen’s ideas, you may start with https://www.uvm.edu/~jfarley/EEseminar/readings/energy%20myths.pdf
Thank you eduardo.
I’m confused by the last part though. Is an electron energy or matter? How could entropy apply to the energy aspects of an electron but not its matter aspects? I’m no expert, but it sounds like semantics to me. Matter breaks down, and spreads out, and can’t be reversed just like energy. This isn’t supposed to be entropic? In its logical conclusion, matter and energy all become the same amorphic ooze together. How are they different beyond our personal perception?
The main problem with applying entropy to economics is that economics, as currently practiced, isn’t scientific. So make it scientific.
Matter and energy are interchangeable (Einstein). Economics in its ultimate is about double entry bookkeeping (money) and the the conservation of mass/energy applied to our human construct for continuing existence.
You could be right. The point however is that the 2nd Law relates specifically to the transformation process of thermal energy into mechanical energy, and vice-versa. Carnot’s experiment can be replicated, always producing similar results — hence the derivation of an “entropy law” explaining the dissipation of energy. The economic processes by which material/labor/capital produce agricultural, manufacturing or service products offer a vast range of possibilities, all incurring the dissipation of matter, which Carnot’s principle of maximum energetic efficiency of a thermal engine cannot explain. Incidentally, Roegen himself acknowledged this fact by suggesting that a 4th Law would be needed. His views about the limits to the use of resources, as well as the irreversibility and the irrevocability of the economic processes seem to remain valid, but to speak of “entropy economics” sounds more like a literary analogy than as a scientific statement, which is what Roegen aimed at.
Michael Hudson, if you’re here, please comment on this review, or the book if you have read it.
Steve Keen has also done a lot of work on the centrality of energy to economics and how that is usually ignored by mainstream economists.
In fact, quite a few economists are thinking along these lines now.
https://surplusenergyeconomics.wordpress.com/
https://consciousnessofsheep.co.uk/
https://debunkingeconomics.com/
I think all these links have been shared here before. That’s how I first found them. Thanks again NC!
W.r.t equilibrium, in Capitalism: Competition, Conflict, Crisis Anwar Shaikh gives an impressive description of Capitalist market systems as chaotic systems always striving for equilibrium but constantly driven away from equilibrium.
A fun exchange of ideas between the article and the commentary. Nevertheless, it’s never a great idea to overstate one’s grasp of the laws of physics. There’s a big hole in the notion of entropy’s primacy. The concept of entropy helps explain how steam engines work but it doesn’t explain how living things, especially cells came to be. Considering that the complexity of a even the simplest cell is orders of magnitude greater than any human social arrangement or invention underscores the incompleteness of any statement that asserts entropy’s primacy in the order of things. And if you throw in all those strange “coincidences” in the alignment of forces that make the universe that we know possible, one should be a bit more cautious about making claims that “entropy is the force that drives the universe.”
I agree it doesn’t explain what entropy factors were relevant when living cells came to be. If it had been human minds commissioned to design them, they would have used up the lifetime energy output of 20 suns, if not more. Cell metabolism on the whole is more reliable than machines humans make. It can utilize long involved ornate processes to obtain energy, in photosynthesis eg storing energy by means of 3 phosphate group bonds in one ATP molecule and one hydrogen atom in an NADPH molecule. Because biological systems are so low energy consuming and painstakingly exact/without-much-error in their functioning, entropy doesn’t threaten a lot of’em…until over time their DNA-clock reduces these features [or maybe a morphic field clock? [I’m thinking here of leaf cells when there’s plenty of water and good soil]]. In regard to human technology, though, I think entropy is a huge, huge thing. If you were running your still with wood fire heat [required for 2 functions], and one day the nearby trees were all used up, you couldn’t stay there doing that thang if you had no means to transport wood. Human technology is like volcanoes compared to cell machinery.
Although the construct of entropy permits us the long overdue (re)introduction of the material world, and particularly physics, into the cloistered worldview of neoclassical economics…the construct that would open that world to the *life sciences* is not entropy, but negentropy. From this perspective, the capacity and central feature of living systems is to reverse, if always temporarily but ongoingly so, entropy.
On this view, entropy is neither an artificial abstraction nor a “natural” or “inevitable” state, nor even perhaps an optimal one. In living systems, equilibrium is not static, nor a moment in time, but an unfolding dynamic state, more or less stable within certain parameters, but, beyond these, subject to runaway. Holding a sub-system (e.g., the economy) in a stable equilibrium state within an evolving whole system (e.g., the ecosphere) can itself produce runaway and collapse.
Seems financialization exponentiates but evolution doesn’t. There is some control mechanism in nature that controls competition, maybe by facilitating an exponentiating vast variability instead of one massive useless frothing blob of financial profits with no viable ideas. In the biosphere entropy facilitates the dissipation of energy into an evolving variety of creatures and niches, it is actually generative instead of extractive. The economy here is decidedly restricted. Possibly without an objective except raw competitive survival governed by a zombie medium of exchange that replaces direct energy synthetically. I’d just like somebody to sit down and write up some reasonable economic goals and directives. Like, suggesting that financialized competition is a dead end without a focus on cooperation toward reasonable goals, one of which should be the preservation of the natural environment. Without a healthy natural environment even entropy will be precluded.
Earth is not a closed energy system, so entropy is more of a poetic notion than a controlling science. The enormous energy flowing from the sun drives the order of the biological system. Physically, Earth is closer to a closed system, which is what Galbraith states he is more concerned with. As our industrial system consumes finite raw materials of various types we approach an entropy limit as it becomes increasingly unprofitable to continue to extract resources. Consider the example of oil. Once we reach the point where it takes more energy to extract and refine oil than is found in the resulting fuel, we will have no reason to search for oil as a fuel, although we might still use some for rare and valuable chemicals.
I think that the laws of physics should always be treated as poetic suggestions for economics, and never as economic laws. Physics is an evolving system itself, and the understanding of our universe has changed considerably during recent centuries. Not even biology is a safe model, as is shown in the increasing gap between biological evolution and the applications of social Darwinism that are increasingly at odds with its alleged inspiration. Evolution is less about raw competition than once thought, and much more about coordination and cooperation. Complex systems arise everywhere in biology. Radical extraction and individualism are undermining the very foundations of life on Earth.