In my previous post, I made a distinction between cybernetic theories, which address the internal decision making process of a system, and phenomenological theories, which identify stable correlations between observable properties. In that post, I suggested that we can use cybernetic theories to figure out which phenomenological theories can give us the most leverage with regards to changing outcomes; for example, indirectly controlling your body’s energy balance through changing what you eat is a more leveraged strategy than trying to directly control your calorie intake. The truth, however, gets even more complex: there are some phenomenological constructs that are so basic yet so shrouded by complexity that you cannot observe them in a very meaningful way: instead, they can only be used as a construction that makes more complex predictive theories logically sound. Here, I’d like to show that these two concepts, opacity and accounting identities, can illuminate how systems primarily manage and adapt to feedback in order to stay alive, and how this changes the way we should look at economics among many other fields.
In my nutrition example, I advocated for an approach to eating that emphasis what you eat rather than how much you eat. My explanation at the time had to do with the concept of leverage; and while it is true that there is more leverage in this approach, there was another fact that I simply left out: we don’t have an accurate idea of our calorie intake and expenditure. Despite the fact that people calorie count by logging what they eat, what they do at the gym, how much they walk, and so forth, it is still a very crude approximation. Not only do we not know exactly what is in our food and exactly how much any given exercise session will burn, we also need to account for all kinds of things such as resting metabolism (which is affected by all sorts of factors), thermogenesis, whether calories are going to fat or muscle, the calories burned by our brain (I’m hungrier at lunch on days where I have to concentrate a lot), where your energy comes from while exercising, how efficiently your body performs a specific exercise, and so on. You might think that even with all this, it’s reasonable to approximate; the problem with this is that it only takes 3500 excess calories to gain a pound of weight. That means that eating just 50 more calories per day (about 3% of your standard 2000 calorie diet, which is considered the margin of error for simple statistics and most likely an unrealisitcally low margin of error for something as imprecise as calorie counting) will mean gaining a pound in a little over two months. That on its own doesn’t sound like much, but consider that that means gaining 5 pounds per year, which would add up to quite a bit over a few years.
You might think that this is a simple matter of errors cancelling one another out, that you’ll have as many days where you’re 50 pounds below your target as you will where you’re 50 pounds above your target. In order to explain why this thinking is flawed, I’ll take a detour into different kinds of randomness. The most commonly known kind is Gaussian randomness. This kind of randomness is predictable and works as follows: imagine that you have a coin and decide to toss it 8 times. The odds of it coming up heads is always 50%, and it’s always the same on every toss. That means that you can easily get the odds of how many times you get heads out of the 8 tosses. The chances that there are no heads and no tails are pretty low (50% to the eighth power), because there is only one way to get to such a configuration. On the other hand, there’s a very high chance of getting four heads and four tails, or three heads and five tails, or five heads and three tails, because there are many different timelines that will get you to that configuration (maybe the first four tosses are heads and the second four are tails, or maybe it alternates, or any number of things.) In fact, the odds of getting all heads on as little as 8 coins are so low you should never worry about it (about 1 in 200). You can in fact see the probabilities of various outcomes (all tails on the left, all heads on the right) in a simple (and well known) curve:
Why is that important? Because you know that to a certain degree, your coin tosses will almost certainly cancel one another out. The problem is that the reason this works for the coins is the same reason it won’t work for other things: the outcomes of the coins are independent of one another. A coin coming up heads on one toss does not affect the probability of a coin coming up heads on the next toss. On the other hand, there’s no way you can measure this if factors interact. And this is exactly what the problem is with calorie expenditure: your diet and exercise is constantly interacting with the various processes in your body that are beyond your control, and even if you eat and exercise exactly as you’ve planned, your body will still be making decisions about all kinds of processes you don’t control. When you have these interactions, you have a curve that looks something more like this:
|Source: http://ross.typepad.com/blog/2004/03/power_laws_and_.html (please contact me if you are the owner and don’t want this image used.)|
If we used the dark blue curve for coin tosses, that would imply a higher probability for something like all 8 coins coming up heads–and it would actually be true if the outcomes of the coin tosses actually affected one another. What’s more important to note for our purposes is that there is not a guarantee that individual outcomes cancel one another out–which was the reason why in our original example we didn’t have to worry about getting 8 heads in a row. Note that I didn’t even add to all this the fact that our behavior is not totally in our control, and that even if we superficially maintain some rules, there will always be subtle ways to work around them (maybe you start running twice a week but then end up spending more of your spare time parked in front of the TV.)
A fair question to ask right now is Alex, what’s the difference between this and what you were talking about yesterday? Isn’t this just more stuff about leverage? Not quite. In my previous entry, I was talking about how much control we have over a given variable. Here, I’m talking about how much knowledge we have of a given variable. It’s not just that we have little direct control over calorie intake, we can’t even get a reasonable approximation of how many calories we eat and expend in a short period of time. In other words, our energy balance is opaque.
So what makes this a phenomenological variable at all if we can’t observe it? The answer is that the phenomenon is (to an extent) observable; we know for a fact that the mathematics do work out such that organisms get bigger with calorie surpluses and smaller with calorie deficits, but when we look at the big picture we simply can’t know or predict the exact rate at which calories are entering and leaving the body at any given moment. The problem is that we believe we can; but let’s take a look at the actual definition of “energy balance”:
Energy Intake = Internal Heat Produced + External Work + Energy Stored
Note that all this does is take four variables and relate them to one another–that if you’re gaining weight (an increase in “energy stored”), then by definition we are either taking in more energy, producing less internal heat, or doing less external work. At no point is there any kind of inference happening–these variables simply describe what is happening. These definitions are important for making sure that any theory of weight gain or weight loss is consistent with thermodynamics, but that does not endow them with any kind of inferential power. What we are left with is an accounting identity, a mathematical definition that unfalsifiably relates variables to one another. Even though the laws of thermodynamics are actually falsifiable, for the purposes of nutrition, if we were to find ourselves gaining weight, we would not question the laws of thermodynamics; we know from this definition that it would either have to be a rise in energy intake, a drop in thermogenesis (body heat production), or a drop in exercise. And of course, that isn’t even addressing whether the extra weight is muscle or fat; most importantly, however, it does not predict anything.
And yet, despite all this opacity, we are all remarkably stable in our weights. This is not only true of people of an average weight–it’s also the case for people who are obese; they do not keep gaining weight indefinitely. As pointed out in my last entry, the body can regulate itself with a remarkable degree of sophistication; and it must–although we constantly speak of calories, it is absurd to forget that our diet requires many different nutrients at varying levels, which themselves control all of the processes that ultimately decide the flow of energy; and that’s just one of many nuances in our overall nutrition. If you believe in calorie counting, then I have one piece of advice: instead of thinking of it as “I’m going to try to control how many calories I eat”, instead think “I’m going to try to implement a pattern of eating and exercise that results in a calorie deficit.” In other words, use calories as a proxy for whether what you’re doing makes sense or not. If cutting down calories means feeling dizzy and irritable, you’re doing it wrong; your brain is not supposed to go on a diet.
But the significance of accounting identities and the opacity of the phenomena that they represent may apply much more deeply to a field whose language games are far more sinister than that of nutrition: economics.
The Elusive Concept of “Wealth”
When I was younger and knew even less about economics than the paltry amount that I know now, I found myself confused by the abstract numbers and concepts that seemed to dominate any discussion on the economy: GDP, inflation, interest rates, employment, and so forth. Although many of these numbers serve their purpose, I found, and continue to find, that many of them act as if there is absolutely no real world behind the economy from which we get finite resources and use them with our finite amounts of energy and time: a problem that is really the inverse of the “calorie fallacy” (impromptu name.) This led me to an analogy that I still continue to use to this day: talking about economics without natural resources is like talking about metabolism without food.
Rather than hearing much about things like the world’s supply of oil or the amount of energy needed to procure food, economists think in terms of prices, credit, liquidity, employment, and other factors that are not about the wealth itself but about the system that controls all of the wealth. To someone who has never read any economics, or perhaps has never lived in a society that has used money, this must seem absurd: isn’t what matters how much actual wealth we have? Well, yes; that, and our ability to allocate that wealth, are what actually matter. But this begs two questions: (1) what counts as “wealth”? How do we compare food and fuel, or luxury and necessity? Is a pound of corn of the same value as a pound of barley? What about less tangible things such as safety or the satisfaction of our emotional needs? (2) how do we, as a society, choose how to allocate our resources in such a way that we can meet our needs and grow our collective wealth?
As a tentative answer to question (1), I will define wealth as surplus thermodynamic energy. This may seem a bit strange, but it will make more sense upon explanation. For answer (2), I will have to go into a little bit of economic theory, explaining the concept of comparative advantage, which is arguably the cornerstone of classical economic theory. These two concepts, surplus energy and comparative advantage, are tightly linked and when put together illuminate a third concept that I would have trouble explaining otherwise.
So what do I mean by surplus energy? The definition of energy is quite simple: the ability to do work. In classical mechanics, work is defined as the ability to move an object that is in a state of rest or to stop an object that is in a state of emotion–in other words, the ability to overcome inertia. The thermodynamic definition of work is more nuanced and would be more comprehensive, but all we need to know for our purposes is that we need energy to grow food, to stay warm, to reproduce, to protect ourselves from predators, to maintain the rule of law, to conduct symphonies, etc. In fact, it’s required for any kind of activity, mental or physical. The more energy we have, the more of these things we can do.
In early agricultural societies, most of this work went to the bare necessities, staying fed, staying warm, and staying safe. Almost all of the energy provided by the food grown was spent on growing more food and doing anything else that was necessary for survival. With so little energy left, there isn’t much capacity for doing other things; so in a primitive society you may have a priest or a shaman of some kind for spiritual guidance, along with a few other simple specialists. On the other hand, should this society domesticate animals that are capable of doing heavy lifting, they’ll be able to grow more food with less energy, leaving spare energy for people to pursue more specialized pursuits and creating a more complex society. The same may happen with a labor saving device such as the plow or some fertilizer that makes crops more nutritious.
You may notice, however, that this is not simply “free energy” coming out of the ether. In the case of domesticated animals, the animals still have to be fed, or else they’ll starve and won’t be able to do any work at all. As for labor saving devices, someone still has to put in the work, just not quite as much. In other words, the surplus energy comes from the tribe becoming more efficient with the energy that they have. A horse may require food to run a plow, but running a horse with a plow gets much more food grown per calorie spent than having a human do the same thing with a simple shovel. This notion of efficiency is also the basis for comparative advantage, and by extension, for the entire science of economics.
So what is comparative advantage? This could best be described with a thought experiment. Let’s take two tribes, the Oomphs and the Bumps. The Oomphs are expert lumberjacks, chopping down trees with incredible efficiency and organization; but their farming system is quite inefficient, and so they spend that saved up energy on making up for their lackluster farming abilities. Meanwhile, the Bumps are most excellent farmers, but they are quite atrocious at cutting down trees. How can these two tribes improve their lot? The answer is easy: by trading. The Oomphs can buy food from the Bumps using their spare lumber. Since they are so much better at woodcutting than they are at farming, they’ll spend much less energy cutting down the extra lumber to trade than they would by growing the food themselves. Meanwhile, the Bumps can do the exact same thing with their food supply. This means that both tribes have much more energy to spare, which can be spent on all manner of things.
But what’s truly important is that this doesn’t just apply to trade between societies–it also is how a modern economy works on an individual level. Instead of having to grow my own food, prepare my own self-defense, and build my own house, I can simply pay someone else to do it, and earn the necessary money by doing what I’m good at. Note that this is basically what money is for: it allows people to offer to trade their services without having to precisely know exactly what other people want or need. Now, money is actually far more complicated than this simplified concept, but we can get to those questions later. What’s important to note as of now is that comparative advantage optimizes our use of energy, and in doing so, gives us energy to spare and allowing us to create a more complex society.
But one can only optimize energy so much, whether through dividing labor or discovering other ways to use energy more efficiently, leaving the question of how further economic growth happens. There are two relatively simple answers: either grow the population, or discover new sources of energy. New sources of energy have been discovered throughout the entirety of human history: fire was discovered millions of years ago, and with it, we were able to cook our food, which metabolizes a lot of the food before we have to do any of the work ourselves; this meant we needed less time to digest our food and could devote more energy to other enhancements such as increased intelligence or better hunting abilities. The energy provided by the wind became the primary means of propulsion for ships and a way to mechanically grind grains. The examples go on and on, but the most potent one is the discovery of fossil fuels, or more accurately, the discovery of how to put fossil fuels to use through combustion. It’s no coincidence that since this discovery, economic growth has accelerated at an unbelievable pace; the amount of energy provided by a single gallon of gas is estimated to be around 500 man-hours of manual labor.
You may have noticed by now something else that’s important: if we want to discover new sources of energy, we’ll need surplus energy. The discovery and utilization of a new source of energy is an effort carried on by tons of trial and error on the part of scientists, entrepreneurs, tinkerers, and specialists of all kinds. Solar power has become increasingly advanced and affordable thanks to materials and designs of such complexity that it takes hundreds of people with extremely specific jobs all working together to develop them. Even the tinkerers that have found more simple ways could not have done so without the amount of spare time given to us by the conveniences of modern society. Even the extraction of crude oil now requires amazing complexity as more and more of what’s left is drilled out of reservoirs that exist thousands of feet beneath the sea.
Now that I’ve taken you through the process of specialization and the importance of surplus energy, one could easily identify that specialization is a cybernetic theory and surplus energy is a phenomenological theory. The problem, however, is that one can’t easily measure “surplus energy”: since a lot of it comes from increased efficiency, we can’t simply measure the amount of electricity, combustible energy, and dietary calories expended by a society in a given year. In addition, I’ve only been using the notion of “efficiency” in the context of the amount of energy that doesn’t simply get lost in transmission (every transfer of energy loses at least some of the energy to irreversible entropy), and have not considered that a person may just be spending the energy foolishly. Nor have we taken into account something else that is much more important: which natural resources will lead to more energy? Just like our body need many different nutrients and can’t use all calories in the same way, our society needs different raw materials and skills to do different things: rare earth metals for solar power and electric cars, rubber for creating tires, plastics for insulating circuits, etc. All of these resources work together in complex ways to determine what energy we can extract, what energy we can save, and how much energy it will cost to ultimately meet our real needs and preferences. Saying that the economy needs to take in more energy than it spends in order to grow is every bit as banal as saying that a person needs to take in less calories than they expend in order to lose weight.
No Accounting for Taste
Unlike the human body, however, we can’t even reliably use energy balance as an accounting identity because we simply have no real idea of what “efficiency” is, since we don’t have any true sense of what ultimately benefits us. In nutrition, we know that body fat is (up to a point) wasted energy, so we know that if we have less than 15% body fat, we have no serious problems with body composition (and even then, body composition does not tell the whole picture about health, there are all sorts of other illnesses and morbidities that can still occur.) Instead, we need a different accounting identity. In classical economics, this need was answered by the idea of utility: every person has a set of preferences for what they want; the only rule being that you can’t prefer apples to oranges, oranges to bananas, and bananas to apples all at the same time, since this would not be consistent.
Utility, however, can only be a theoretical construct. Ignoring for the moment that people don’t even have consistent needs and preferences, the concept of utility would also imply perfect information about the present and the future; something that only an omnipotent being could have. Instead, we use money; a highly unstable and crude signifier of wealth. It is a signifier (as opposed to an indicator) of wealth because, as we saw earlier, the concept of “wealth”, let alone “value” is intractable. But even if money can’t act as a gauge of wealth, it can still act as a unit of account by allowing us to create stable accounting identities for economics. Just like the rules of energy balance hold, so do the rules of monetary transaction; if you owe more than you have, you are in debt, and if you import more than you export, you have a trade deficit. If you have a trade deficit, it can only be shrunk by exporting more “wealth” as denominated in money; this may be done by devaluing the currency (you sell the goods for more money, but that money is worth less) or by consuming less and exporting the excess wealth, but no matter what the method, the money itself must unambiguously balance out. Another place identity that uses money as a unit of account is the “size” of an economy:
GDP = Consumption + Savings + Government Spending + (Exports – Imports) [net exports]
Note that this is just saying “the total amount of economic activity has to be the sum of how much money is spent, how much money is saved, and how much money is made from exports that wasn’t spent on imports. That last item on the list may be tricky to understand, but think about it this way: all imports are already accounted for as consumption, so if you counted the money made from exports that was spent on imports, you’d be double-counting the consumption of imports. What’s important to note is that this identity is not making any inferences, but only making the clear unambiguous rule that every dollar that goes through the economy must be categorized in one of these four variables.
Since currency does not actually signify wealth in any tractable way, this can be at best a rough approximation. Although economists talk about “real growth” and “real incomes” by “adjusting” for “inflation”, the truth is that the very concept of inflation is based on comparing money to wealth, which for reasons we’ve already been over is extremely problematic. So if GDP can’t measure growth or prosperity in any way at all, what’s the point of talking about this or any other money-based accounting identity? The answer is that we’re asking the wrong question. It’s not just money that’s the problem, it’s that the very concept of “growth” or “prosperity” is fundamentally the wrong way to think about economics. Along the same lines, “conservation” or “sustainability” is no better when we consider that we cannot anticipate our future needs any better. That’s not to say that we shouldn’t worry about the world’s supply of water, oil, topsoil, or food; but addressing those issues in a simplistic top-down manner won’t work because they are so phenomenologically opaque that the only accounting identities we have available to us have to use money as a unit of account.
So what is the right question if it’s not about how to grow or how to conserve? Before going into the answer, consider the function of money: it provides information to the economy and influences behavior. You, as an individual, know what you can and can’t own based on not just how much money you have on hand, but also by how expensive it is to borrow money and how available new revenue is. In other words, money is also a cybernetic entity; it provides feedback, which allows the economy to adapt to novel needs and challenges as they arise. The purpose of economics is adaptation, with money being one of many mechanisms that provide the information essential to this function. While more money does not translate to more adaptability, one should remember that calories are not unambiguously linked to health: instead, the dynamics of calories and the dynamics of money both provide us the necessary constraints to make further inferences. In the case of money, we’ll be able to use its mathematical constraints to illuminate how economies work as systems of adaptation:
The most simple form of feedback in economics is supply and demand, which itself is mediated by money. The price of something goes up if demand outpaces supply, and will continue to do so until either fewer people want it (for that price) or more of it is supplied (it goes without saying that this also applies vice-versa.) The same thing also happens with money itself: if there is more money, the “price” of money goes down–both in the form of borrowed money costing less interest and other goods costing more money. The closing of these gaps is a form of negative feedback, and could be considered the most basic kind of feedback in an economy. There are, however, more intense versions of this feedback, such as when some type of good or service is extremely overpriced (often because people see its price going up and want to try to buy it and re-sell it for a more expensive price) and then finally the price drops down to something more reasonable. Another more intense version may come from a change in the outside world, such as the price of some important item skyrocketing due to scarcity, in which case people must cut back their consumption of other goods or find alternatives to the item in question, leading to prices dropping elsewhere and an overall decrease in wealth that makes some sectors of the economy unsupportable. In all of these cases, the behavior of individuals will have to adjust, and in order to make that adjustment, the amount of money circulating in the economy will decrease since lost jobs, lost sales, failed investments, etc. will require that people conserve what they have. Keep in mind throughout all of this that we can simply think about this in terms of money, and do not have to think beyond a very rudimentary level about the material wealth underlying the money.
From this point of view, recessions, while painful, are necessary feedback. If credit has been lent too freely, then interest rates (the price of acquiring credit) should commensurately go up; and if houses have become overpriced due to bubble behavior, then we should not continue to pay more for houses. The same goes for gasoline: high gas prices signal that we need to be wiser about how we use gas, or that we should look harder for new sources of energy. While this is all true, there’s one major problem: feedback does not exist in a vacuum. If the economy is harmed too much, it may compromise the very mechanisms that process this feedback. Consider, for example, lifting heavy weights at the gym. Up to a certain point, it will feel stressful and may even hurt a bit; you’re giving it your all and dripping sweat on the floor. After all this pain, you go walk it off and rest for a few days and come back to the gym able to lift an even heavier weight because of the adaptation. Now consider that this next time, you decide that you can do even more, and raise the weight by a much higher amount than usual. In the middle of your set, you feel a sharp pain and before you know it you’ve torn a muscle in your arm. Now you’ll certainly get weaker, at least in that arm, due to the fact that you won’t be able to do any heavy exercise with it for at least a few weeks. That’s the difference between just enough pain and too much pain.
With recessions, the same logic applies. For example, if too many people are out of work, they won’t be able to buy anything and more places either lay off workers or go out of business entirely. When that happens, it can turn into a vicious cycle; or, if you read my previous entry, a positive feedback loop. While some pain will correct the relative prices of goods and weed out irrelevant skills and unsustainable businesses, too much at once can lead to a runaway chain reaction. So we want harm, but not too much concentrated harm. More specifically, we want negative feedback, because that’s the kind of feedback that results in a correction, as opposed to positive feedback, where pain begets more pain. Even then, however, there’s a problem: we don’t necessarily know what’s going to spiral out of control and what’s going to ultimately act as beneficial feedback. In fact, we want the feedback to be sufficiently concentrated up to a point. To show why, let’s go back to the gym: this time, you’re benching 150 pounds. After about 10 repetitions, you can’t do another one, and you call it a day. Your friend next to you, although just as strong, benches 15 pounds and stops after 100 repetitions (for those who don’t believe me, I’ll give you a more extreme example: your friend benches 1.5 pounds 1000 times.) You both got feedback from the stressors, but you’ll benefit much much more than him because of the concentrated dose. What does this suggest? That intensity of feedback has accelerating benefits before it starts to cause harm, an idea that has been explored in more depth by Nassim Nicholas Taleb in his book Antifragile.
|Source: Antifragile by Nassim Nicholas Taleb|
So why should feedback work better if it’s concentrated if all that matters is eventually correcting discrepancies? Before getting into that, I need to address something that has been mostly ignored thus far: economic booms. Economic recessions almost always follow a time of rapid economic growth (denominated in whatever currency you’d like.) It is during this time that the discrepancies are built, since people have more money to spend, and this money ends up getting spent in inefficient and wasteful ways. Economists of the Austrian school call these built up discrepancies “malinvestments”: investments in which resources are wasted (or if you want more mathematical precision, investments in which resources are not invested optimally). As we noted before, these kinds of discrepancies are happening at all times, but oftentimes in very small amounts with booms and busts happening when many of these things happen at once; which happens more often and with more intensity than even many economists realize because of how interconnected economic events are (recall my spiel on probability distributions at the beginning of this post.)
Due to the intractability of both our present and our future needs, these malinvestments are inevitable. Fortunately, they are also desirable (to an extent) for the exact same reason. Consider the internet as it is now; it is extremely fast and ubiquitous, to the point where it is free to instantly communicate with somebody on the other side of the world. The infrastructure for this is in part made up by sprawling networks of fiber-optic cables that traverse entire oceans and continents. Many of these were built up during the dot-com bubble in the late 90s and early 2000s, and it was possible due to the amount of money people were foolishly willing to invest in all kinds of digital technologies. Eventually, most people lost their shirts in these investments and a recession followed, but not without making all of these fiber-optic cables dirt cheap due to investors’ needs to sell off what assets remain, providing the world with a whole new infrastructure.
But why the bust, you may ask? Can’t we just get this growth and try to cushion any fall that happens afterwards? The problem is that just because we don’t ultimately know what is wasteful, it doesn’t mean that there’s no such thing as waste. If a collapse in housing prices is propped up by the government giving subsidies to consumers, then the government will have to pay for it somewhere; if not by cutting costs elsewhere, then by raising taxes elsewhere or by printing money. While printing money may sound like the solution, one needs to remember that behind all of the money is a finite, though often growing, amount of material wealth, and buying more of one thing means buying less of another. The labor, raw materials, and loans that may have gone somewhere else are now tied up in a place where it wasn’t worth it. Just consider if every restaurant were propped up: there would be tons of real-estate, personnel, food, electricity, and gas tied up in restaurants that almost nobody wants to eat at.
The common retort is that this cushion doesn’t matter because growth will eventually outstrip it, but this neglects the possibility that misdirecting too many of our limited resources may in fact hamper future growth by not allowing adaptations to occur. I blame the common emphasis on the word “growth” for this misunderstanding: when the focus of economics becomes adaptation rather than growth, the boom and the bust are suddenly two sides of the same coin; both of them an essential part of making the changes that better suit us to both present and future needs. Consider, in addition, that when we measure “growth”, we are talking about it in terms of money, which is not a direct measurement of wealth but a feedback mechanism that follows certain basic constraints. Booms and busts are increases and decreases in the activity of money, so we should realize that what we’re looking at is not a pattern of abundance and scarcity per se, but signals of abundance and scarcity. This might seem contrary to ideas such as stagflation, but consider there that this is a phenomenon in which the purchasing power of money goes down while GDP, the amount of money circulating in the economy, stays stagnant.
Noting that these ideas of growth and recession are fundamentally about information, I can now make a big claim: it is not growth or atrophy that matters, it is the pattern of growth and atrophy. This statement, along with the fact that we patently need to both do stupid things and pay for our stupidity (rather than be smart), means that while an economy strives for adaptation, it does not do so through homeostasis, since it does not thrive by staying close to some equilibrium. The correct word is allostasis, long-term quasi-stability achieved through volatility. Without this volatility, the economy would be extremely brittle, as all of its decisions would be based on the market’s current (implicit) hypothesis about our current and future needs, allowing no room for the randomness that is necessary to compensate for what is unknown. More importantly, however much it may seem otherwise, the money itself is just information; our actual security, material wealth, and future challenges are a sea of chaos that is traversed through feedback and adaptation.
What then, makes a healthy economy? The answer is volatility above all other things.* Money does not provide knowledge, but it provides feedback. Volatility is an indicator of feedback in two ways: the negative feedback loops make corrections as errors come, while the positive feedback loops provide a level of randomness that appropriately handles the uncertainty of what’s unknown. So if you want to see whether or not an economy is doing well, don’t look at its growth, but rather at its variance; the more wide gaps between boom and bust, the better. The same also goes for living things: despite the craze for a low basal heart rate, the evidence seems to suggest that it is the variation in heart rate that may ultimately matter. But forget longevity for a second: anybody who isn’t completely neurotic understands that health is the ability to live a good life, not a long one; and living a good life means having the capacity for wild swings of both good times and bad times. That’s why in those pharmaceutical commercials you see those scenes with the dad going mountain biking with his kids because he finally got rid of his COPD–because he can now have more intense experiences without choking to death. A better measure, for that reason, might be metabolic range: by how much can you multiply your metabolic output? I don’t know much about the measurement, but there is a metric, and it seems to be linked to your peak physical capacity.
After all this, chances are that more questions than answers have come up; not least of which when we know the difference between out-of-control positive feedback and very large swings of negative feedback. A related concern is that there are probably many layers of feedback mechanisms rather than just one, such that less effective feedback mechanisms should be destroyed to make room for new ones–that alone is a headache to think about. While there is no simple answer to these things, we can still keep our sights in a reasonable range by remembering what Keynes once said: ”In the long run, we’re all dead.” At the same time understanding the centrality of allostasis may mean that we can finally get away from the clusterfuck that is occurring between the neo-Keynesians, the Austrians, the conservationists, and probably many more schools of thought that I’ve forgotten.
*For further discussion on volatility, I strongly recommend Antifragile, which I have cited multiple times here. It is a somewhat less theoretical, but much more empirical, treatment of many themes in this post’s subject matter