Entropy and Economics: Agglomeration

In the course of this series, we laid out the foundations of — and rationale for — adapting and applying the concepts of entropy to economic analysis. In this final entry, we will examine the concept from a “high level”, and lay out some areas where traditional economics has had explanatory trouble, and how the concept of entropy could help bring the picture into focus.

Longitude…and Latitude

When developing a theory of how the world works, whether it be in the physical or social sciences, it is helpful to keep in mind a rule set for judging the efficacy and validity of your theory — especially if it is meant to be an over-arching theory, that links many others together. Just like a theory which espouses to unite quantum gravity with general relativity, so too must a theory in economics which unites microeconomic behavior with macroeconomic phenomena be able to provide a “high-level” explanation which remains valid in both “worlds”, without making assumptions which do not mirror reality. Economics has had a very tough time meeting this criteria. To tie our theory of entropy into both microeconomic and macroeconomic phenomena, it must pass a few tests:

  • The theory must be able to explain phenomena that happens in the natural world. In our case, it must be able to start with a group of people and natural resources, and paint a picture of decreasing entropy and growing complexity (in products and organization), and wealth over time.
  • The theory must take us from our hunter-gatherer roots to our consumer-trader present, and do so in a way that is consistent with the punctuated equilibrium found in the historical record — preferably without any fuzzy concepts (like “happiness”, or “animal spirits”). Just as well, the theory should make a minimum of ad hoc assumptions, and remain stable without reaching outside itself.
  • Finally, the theory must be consistent with, and not contradict, other accepted laws in the sciences.

Does the our foundation of the theory of entropy in economics pass this test? Using the framework of physical and social technologies we can certainly sketch out the history of economic activity using our principles; which state that value creating activities are irreversible, involve waste in applying energy to matter and information, and human activity serves to create fit order — based upon the foundations of evolutionary preferences. The only exogenous dynamics are the growth in intellect of the human brain, the development of speech, and the evolution of basic preferences. The only exogenous factors are physical inputs of energy and matter, and the outputs of heat and waste. Given a galactic supercomputer, once the model was constructed, we could set off the search through the evolutionary design space of physical and social technologies, as well as business plans and individual preferences. and it would run in a very sustainable fashion, each design space evolving through niche construction. Each expanding exponentially with the discovery of evolutionary “good tricks”. We could then observe the emergence of a pattern of gradual initial development with punctuated periods of rapid change through the discovery of particularly useful physical/social technologies or business plans (“good tricks”), or S curves. We would see a world of lowering entropy, and increasing complexity and diversity. Most importantly, we would be able to observe increasing wealth — as all activity that creates wealth involves entropy!

All of that is not to say if we “played the tape” over again, we would end up with our exact circumstance. Natural evolution is nary more than million upon millions of tiny accidents. The same is true of economic evolution. Changing just a tiny portion of our history in the world would have a gigantic impact. Need proof? The most popular story is ancient China, which was a technological leader of no equal. The Middle East, as well. Both regions succumbed to regional conflict, and poor economic and governance systems (Communism, Sharia law).

Georgescu-Roegen’s insights, and the resulting G-R Conditions laid out in this series provide an important grounding point for any theory wishing to explain the development and evolution of economies. I would go as far as to say that irreversibility and decreasing entropy are laws of economics…even beyond the “laws” of supply and demand, and one price. The first two conditions are laws because they hold true without exception (indeed, if you can think of an exception to irreversibility or decreasing entropy, please let me know!).

Markets in…Disequilibrium?

There are two concepts in Traditional economics which are very much untrue, but also very persistent. Those two concepts are perfect rationality (now, bounded rationality), and equilibrium. Economists have spent the greater part of half-a-century relaxing the assumption of perfect rationality…but economics looks unlikely — or unwilling — to relax the equilibrium framework in which the study works. The concept allows equilibrium to be relaxed by giving time an arrow. Instead of snapping from one condition to the next, we can take an evolutionary biologists track and look at how systems (physical, social) feed off of eachother in an evolutionary cycle. After all, time waits for no one. The framework of entropic evolution would allow economists to become more inclusive to what were previously exogenous factors, like consumer tastes, technological innovation, productivity, and weather. It would allow us to no longer speak about “animal spirits”, or “consumer confidence”, or other mysterious forces of magic (like liquidity traps!). It would also allow us to view the economy as an accelerating, amplifying, self-reinforcing cycle, which is punctuated by periods of decelerations, dampening, and self-regulation (no, not what you think). It also allows us to rethink the “laws” of economics.

For example, if we estimate that the modern economy has 1010 SKUs (stock keeping units) in it, and if we assume that every decision in the economy is made at the speed of the worlds fastest supercomputer, then it would take something like 4 quintillion years for the economy to reach general equilibrium after an (each!) exogenous shock. Just a note, as a best-guess, the universe is only about twelve billion years old.

Supply and Demand (Sometimes Even in that Order!)

You may or may not understand the reference to the title of this section. However, the most famous “law” in economics is the law of supply and demand. These counterbalancing forces are the fundamental tenet that you hear espoused in layman’s terms as “there is no free lunch”. Despite claims to the contrary, the equilibrium of supply and demand is a very confusing concept. Indeed, Scott Sumner has mused on this subject. However, while Scott focuses upon the methods by which we teach supply and demand, I want to take a different track.

Supply and demand is so confusing because it’s embarrassingly imprecise. So imprecise, in fact, that studies exists which specifically refute the concept (Card and Krueger on minimum wage, for instance). Applying any fine grain at all will reveal that nearly all markets are naturally built around the assumption of disequilibrium in supply or demand. Thus, we have warehouses, shortages, backlogs, capacity underutilization, and middlemen. As we can learn from Sumner, reasoning from a price change is bad…however, I say (and believe Scott would probably agree) supply and demand are best used as approximations of phenomena. Indeed, Austrian economics is built upon a foundation which rejects the Traditional economic assumption that the existence of inventories and slack are “noise” around equilibrium. While I do find it highly useful to keep in mind that ceterus paribus, supply equals demand…all else is literally never equal.

There’s Only One Sheriff in This Town!

The second famous “law” of economics is the law of one price, which states that “In the absence of transportation costs and trade barriers, identical goods must sell at the same price in all markets.” This is the second corollary to the dictum “there is no free lunch”. In the aggregate, the “law” has much explanatory power. If it didn’t, instead of writing this blog post, I would be buying oil and gold in Europe and selling in the US (or visa versa?)…the fact that I can’t even pick the direction of the trade, of course, verifies this fact. However, subject to any level of detail (like non-liquid commodities, for instance), the approximation often breaks down…and this breakdown happens even at the macro level!

Quite possibly the most important test of the law of one price was the integration of European countries into the European Union. Given a single currency, low barriers to mobility, and greater price transparency; the EU should have experienced greater convergence of prices. In fact, the exact opposite happened. Between 1998 and 2003, the standard deviation in prices within the euro zone rose from 12.3% to 13.8%.[1] At a more fine level of detail, I have discussed the London ketchup market before, which showed deviations of up to 43% from the theoretical mean.

Of course, just like supply and demand, the idea behind the law of one price is astute and accurate; people do, in fact, have incentives to take advantage of arbitrage opportunities. However, barriers almost always exist (even on the internet![2])…even search costs. Scott Sumner does not like indices like CPI or PCE…and neither do I.

The Role of Entropy

I believe that entropy could play a major role in refining all of the theories listed above into a more general framework of economic evolution. I’m afraid that I have glossed over a lot of interesting work on market phenomenon in giving an overview (proving my thesis, Ph.D please?). However, entropy could give us more robust tools in observing and making sense of the world. Economists generally do not argue about the overarching narrative of the business cycle anymore (unless there happens to be a recession), but economists do have a lot of disagreement about various identification problems related to all sorts of phenomena. Would we have as many disagreements given a much stronger base of tools to draw upon?

Perhaps, it is social science, after all. But given a much more robust descriptive theory of the origin of economic activity[3], I think we could have much more productive debates.

I hope you found this series both enlightening and entertaining! It was a certainly a pleasure to write!

[1]Eurostat; T.L. Cook (2004)
[2]Exhibit A, Exhibit B. Note, there is free shipping from NewEgg, and it is $14 cheaper than the resale price. I wonder what is the barrier?
[3]Just as a little descriptive analysis: The theory of comparative advantage is one of the most robust theories in all of economics. Indeed, Paul Krugman states:

…”if there was an Economist’s Creed, it would surely contain the affirmations, “I believe in the Principle of Comparative Advantage,” and “I believe in free trade.”

Not surprisingly, trade is one area where economists as an entire group are largely in agreement! In a just world, comparative advantage would be a law!


One thought on “Entropy and Economics: Agglomeration

  1. Nice to see an accumulation of relevant info.I seriously hate Blogs with ad,your blog is something i always can describe a particular blog manufactured to provide Information. Greetings

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s